annotate src/main/java/edu/harvard/iq/dataverse/dataaccess/TabularSubsetGenerator.java @ 10:a50cf11e5178

Rewrite LGDataverse completely upgrading to dataverse4.0
author Zoe Hong <zhong@mpiwg-berlin.mpg.de>
date Tue, 08 Sep 2015 17:00:21 +0200
parents
children
Ignore whitespace changes - Everywhere: Within whitespace: At end of lines:
rev   line source
10
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1 /*
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
2 Copyright (C) 2005-2012, by the President and Fellows of Harvard College.
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
3
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
4 Licensed under the Apache License, Version 2.0 (the "License");
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
5 you may not use this file except in compliance with the License.
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
6 You may obtain a copy of the License at
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
7
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
8 http://www.apache.org/licenses/LICENSE-2.0
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
9
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
10 Unless required by applicable law or agreed to in writing, software
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
11 distributed under the License is distributed on an "AS IS" BASIS,
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
12 WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
13 See the License for the specific language governing permissions and
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
14 limitations under the License.
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
15
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
16 Dataverse Network - A web application to share, preserve and analyze research data.
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
17 Developed at the Institute for Quantitative Social Science, Harvard University.
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
18 Version 3.0.
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
19 */
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
20
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
21 package edu.harvard.iq.dataverse.dataaccess;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
22
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
23 import edu.harvard.iq.dataverse.DataFile;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
24 import edu.harvard.iq.dataverse.datavariable.DataVariable;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
25 import java.util.*;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
26 import java.util.Scanner;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
27 import java.util.logging.*;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
28 import java.io.*;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
29 import java.io.FileNotFoundException;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
30 import java.math.BigDecimal;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
31 import java.math.MathContext;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
32 import java.math.RoundingMode;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
33 import java.nio.ByteBuffer;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
34 import java.nio.channels.FileChannel;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
35 import java.nio.file.Paths;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
36 import java.nio.file.StandardOpenOption;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
37 import java.util.regex.Matcher;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
38
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
39
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
40 import org.apache.commons.lang.*;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
41
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
42
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
43 /**
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
44 *
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
45 * @author Leonid Andreev
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
46 * original author:
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
47 * @author a.sone
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
48 */
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
49
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
50 public class TabularSubsetGenerator implements SubsetGenerator {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
51
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
52 private static Logger dbgLog = Logger.getLogger(TabularSubsetGenerator.class.getPackage().getName());
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
53
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
54 private static int COLUMN_TYPE_STRING = 1;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
55 private static int COLUMN_TYPE_LONG = 2;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
56 private static int COLUMN_TYPE_DOUBLE = 3;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
57 private static int COLUMN_TYPE_FLOAT = 4;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
58
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
59 private static int MAX_COLUMN_BUFFER = 8192;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
60
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
61 private FileChannel fileChannel = null;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
62
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
63 private int varcount;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
64 private int casecount;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
65 private int subsetcount;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
66
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
67 private byte[][] columnEntries = null;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
68
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
69
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
70 private ByteBuffer[] columnByteBuffers;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
71 private int[] columnBufferSizes;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
72 private int[] columnBufferOffsets;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
73
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
74 private long[] columnStartOffsets;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
75 private long[] columnTotalOffsets;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
76 private long[] columnTotalLengths;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
77
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
78 public TabularSubsetGenerator() {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
79
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
80 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
81
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
82 public TabularSubsetGenerator (DataFile datafile, List<DataVariable> variables) throws IOException {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
83 if (!datafile.isTabularData()) {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
84 throw new IOException("DataFile is not tabular data.");
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
85 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
86
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
87 setVarCount(datafile.getDataTable().getVarQuantity().intValue());
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
88 setCaseCount(datafile.getDataTable().getCaseQuantity().intValue());
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
89
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
90 File tabfile = datafile.getFileSystemLocation().toFile();
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
91 File rotatedImageFile = getRotatedImage(tabfile, getVarCount(), getCaseCount());
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
92 long[] columnEndOffsets = extractColumnOffsets(rotatedImageFile, getVarCount(), getCaseCount());
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
93
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
94 fileChannel = (FileChannel.open(Paths.get(rotatedImageFile.getAbsolutePath()), StandardOpenOption.READ));
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
95
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
96 if (variables == null || variables.size() < 1 || variables.size() > getVarCount()) {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
97 throw new IOException("Illegal number of variables in the subset request");
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
98 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
99
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
100 subsetcount = variables.size();
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
101 columnTotalOffsets = new long[subsetcount];
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
102 columnTotalLengths = new long[subsetcount];
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
103 columnByteBuffers = new ByteBuffer[subsetcount];
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
104
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
105
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
106
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
107 if (subsetcount == 1) {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
108 if (!datafile.getDataTable().getId().equals(variables.get(0).getDataTable().getId())) {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
109 throw new IOException("Variable in the subset request does not belong to the datafile.");
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
110 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
111 dbgLog.fine("single variable subset; setting fileChannel position to "+extractColumnOffset(columnEndOffsets, variables.get(0).getFileOrder()));
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
112 fileChannel.position(extractColumnOffset(columnEndOffsets, variables.get(0).getFileOrder()));
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
113 columnTotalLengths[0] = extractColumnLength(columnEndOffsets, variables.get(0).getFileOrder());
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
114 columnTotalOffsets[0] = 0;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
115 } else {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
116 columnEntries = new byte[subsetcount][];
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
117
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
118 columnBufferSizes = new int[subsetcount];
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
119 columnBufferOffsets = new int[subsetcount];
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
120 columnStartOffsets = new long[subsetcount];
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
121
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
122 int i = 0;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
123 for (DataVariable var : variables) {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
124 if (!datafile.getDataTable().getId().equals(var.getDataTable().getId())) {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
125 throw new IOException("Variable in the subset request does not belong to the datafile.");
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
126 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
127 columnByteBuffers[i] = ByteBuffer.allocate(MAX_COLUMN_BUFFER);
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
128 columnTotalLengths[i] = extractColumnLength(columnEndOffsets, var.getFileOrder());
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
129 columnStartOffsets[i] = extractColumnOffset(columnEndOffsets, var.getFileOrder());
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
130 if (columnTotalLengths[i] < MAX_COLUMN_BUFFER) {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
131 columnByteBuffers[i].limit((int)columnTotalLengths[i]);
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
132 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
133 fileChannel.position(columnStartOffsets[i]);
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
134 columnBufferSizes[i] = fileChannel.read(columnByteBuffers[i]);
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
135 columnBufferOffsets[i] = 0;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
136 columnTotalOffsets[i] = columnBufferSizes[i];
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
137 i++;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
138 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
139 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
140 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
141
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
142 private int getVarCount() {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
143 return varcount;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
144 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
145
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
146 private void setVarCount(int varcount) {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
147 this.varcount = varcount;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
148 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
149
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
150 private int getCaseCount() {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
151 return casecount;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
152 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
153
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
154 private void setCaseCount(int casecount) {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
155 this.casecount = casecount;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
156 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
157
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
158
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
159 /*
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
160 * Note that this method operates on the *absolute* column number, i.e.
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
161 * the number of the physical column in the tabular file. This is stored
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
162 * in DataVariable.FileOrder.
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
163 * This "column number" should not be confused with the number of column
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
164 * in the subset request; a user can request any number of variable
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
165 * columns, in an order that doesn't have to follow the physical order
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
166 * of the columns in the file.
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
167 */
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
168 private long extractColumnOffset(long[] columnEndOffsets, int column) throws IOException {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
169 if (columnEndOffsets == null || columnEndOffsets.length <= column) {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
170 throw new IOException("Offsets table not initialized; or column out of bounds.");
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
171 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
172 long columnOffset;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
173
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
174 if (column > 0) {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
175 columnOffset = columnEndOffsets[column - 1];
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
176 } else {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
177 columnOffset = getVarCount() * 8;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
178 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
179 return columnOffset;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
180 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
181
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
182 /*
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
183 * See the comment for the method above.
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
184 */
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
185 private long extractColumnLength(long[] columnEndOffsets, int column) throws IOException {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
186 if (columnEndOffsets == null || columnEndOffsets.length <= column) {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
187 throw new IOException("Offsets table not initialized; or column out of bounds.");
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
188 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
189 long columnLength;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
190
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
191 if (column > 0) {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
192 columnLength = columnEndOffsets[column] - columnEndOffsets[column - 1];
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
193 } else {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
194 columnLength = columnEndOffsets[0] - varcount * 8;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
195 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
196
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
197 return columnLength;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
198 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
199
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
200
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
201 private void bufferMoreColumnBytes(int column) throws IOException {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
202 if (columnTotalOffsets[column] >= columnTotalLengths[column]) {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
203 throw new IOException("attempt to buffer bytes past the column boundary");
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
204 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
205 fileChannel.position(columnStartOffsets[column] + columnTotalOffsets[column]);
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
206
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
207 columnByteBuffers[column].clear();
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
208 if (columnTotalLengths[column] < columnTotalOffsets[column] + MAX_COLUMN_BUFFER) {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
209 dbgLog.fine("Limiting the buffer to "+(columnTotalLengths[column] - columnTotalOffsets[column])+" bytes");
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
210 columnByteBuffers[column].limit((int) (columnTotalLengths[column] - columnTotalOffsets[column]));
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
211 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
212 columnBufferSizes[column] = fileChannel.read(columnByteBuffers[column]);
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
213 dbgLog.fine("Read "+columnBufferSizes[column]+" bytes for subset column "+column);
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
214 columnBufferOffsets[column] = 0;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
215 columnTotalOffsets[column] += columnBufferSizes[column];
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
216 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
217
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
218 /*
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
219 do not use this method!
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
220 there's a high potential for the "UTF8 character split between buffers" error!
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
221 public String readColumnEntry(int column) {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
222 String ret = null;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
223 int currentbyte;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
224
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
225 if (columnBufferOffsets[column] >= columnBufferSizes[column]) {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
226 try {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
227 bufferMoreColumnBytes(column);
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
228 } catch (IOException ioe) {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
229 return null;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
230 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
231 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
232
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
233 currentbyte = columnBufferOffsets[column];
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
234 try {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
235 while (columnByteBuffers[column].array()[currentbyte] != '\n') {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
236 currentbyte++;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
237 if (currentbyte == columnBufferSizes[column]) {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
238 // save the leftover:
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
239 if (ret == null) {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
240 ret = new String(columnByteBuffers[column].array(), columnBufferOffsets[column], columnBufferSizes[column] - columnBufferOffsets[column], "UTF8");
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
241 } else {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
242 ret = ret.concat(new String(columnByteBuffers[column].array(), columnBufferOffsets[column], columnBufferSizes[column] - columnBufferOffsets[column], "UTF8"));
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
243 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
244 // read more bytes:
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
245 bufferMoreColumnBytes(column);
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
246 currentbyte = 0;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
247 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
248 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
249
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
250 // presumably, we have found our '\n':
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
251 if (ret == null) {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
252 ret = new String(columnByteBuffers[column].array(), columnBufferOffsets[column], currentbyte - columnBufferOffsets[column], "UTF8");
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
253 } else {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
254 ret = ret.concat(new String(columnByteBuffers[column].array(), columnBufferOffsets[column], currentbyte - columnBufferOffsets[column], "UTF8"));
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
255 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
256
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
257 } catch (IOException ioe) {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
258 return null;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
259 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
260
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
261 columnBufferOffsets[column] += (currentbyte + 1);
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
262
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
263 return ret;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
264 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
265 */
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
266
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
267 public byte[] readColumnEntryBytes(int column) {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
268 return readColumnEntryBytes(column, true);
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
269 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
270
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
271
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
272 public byte[] readColumnEntryBytes(int column, boolean addTabs) {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
273 byte[] leftover = null;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
274 byte[] ret = null;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
275
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
276 if (columnBufferOffsets[column] >= columnBufferSizes[column]) {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
277 try {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
278 bufferMoreColumnBytes(column);
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
279 if (columnBufferSizes[column] < 1) {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
280 return null;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
281 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
282 } catch (IOException ioe) {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
283 return null;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
284 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
285 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
286
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
287 int byteindex = columnBufferOffsets[column];
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
288 try {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
289 while (columnByteBuffers[column].array()[byteindex] != '\n') {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
290 byteindex++;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
291 if (byteindex == columnBufferSizes[column]) {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
292 // save the leftover:
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
293 if (leftover == null) {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
294 leftover = new byte[columnBufferSizes[column] - columnBufferOffsets[column]];
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
295 System.arraycopy(columnByteBuffers[column].array(), columnBufferOffsets[column], leftover, 0, columnBufferSizes[column] - columnBufferOffsets[column]);
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
296 } else {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
297 byte[] merged = new byte[leftover.length + columnBufferSizes[column]];
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
298
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
299 System.arraycopy(leftover, 0, merged, 0, leftover.length);
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
300 System.arraycopy(columnByteBuffers[column].array(), 0, merged, leftover.length, columnBufferSizes[column]);
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
301 leftover = merged;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
302 merged = null;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
303 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
304 // read more bytes:
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
305 bufferMoreColumnBytes(column);
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
306 if (columnBufferSizes[column] < 1) {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
307 return null;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
308 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
309 byteindex = 0;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
310 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
311 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
312
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
313 // presumably, we have found our '\n':
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
314 if (leftover == null) {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
315 ret = new byte[byteindex - columnBufferOffsets[column] + 1];
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
316 System.arraycopy(columnByteBuffers[column].array(), columnBufferOffsets[column], ret, 0, byteindex - columnBufferOffsets[column] + 1);
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
317 } else {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
318 ret = new byte[leftover.length + byteindex + 1];
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
319 System.arraycopy(leftover, 0, ret, 0, leftover.length);
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
320 System.arraycopy(columnByteBuffers[column].array(), 0, ret, leftover.length, byteindex + 1);
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
321 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
322
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
323 } catch (IOException ioe) {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
324 return null;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
325 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
326
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
327 columnBufferOffsets[column] = (byteindex + 1);
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
328
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
329 if (column < columnBufferOffsets.length - 1) {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
330 ret[ret.length - 1] = '\t';
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
331 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
332 return ret;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
333 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
334
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
335 public int readSingleColumnSubset(byte[] buffer) throws IOException {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
336 if (columnTotalOffsets[0] == columnTotalLengths[0]) {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
337 return -1;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
338 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
339
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
340 if (columnByteBuffers[0] == null) {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
341 dbgLog.fine("allocating single column subset buffer.");
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
342 columnByteBuffers[0] = ByteBuffer.allocate(buffer.length);
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
343 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
344
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
345 int bytesread = fileChannel.read(columnByteBuffers[0]);
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
346 dbgLog.fine("single column subset: read "+bytesread+" bytes.");
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
347 if (columnTotalOffsets[0] + bytesread > columnTotalLengths[0]) {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
348 bytesread = (int)(columnTotalLengths[0] - columnTotalOffsets[0]);
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
349 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
350 System.arraycopy(columnByteBuffers[0].array(), 0, buffer, 0, bytesread);
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
351
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
352 columnTotalOffsets[0] += bytesread;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
353 columnByteBuffers[0].clear();
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
354 return bytesread > 0 ? bytesread : -1;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
355 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
356
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
357
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
358 public byte[] readSubsetLineBytes() throws IOException {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
359 byte[] ret = null;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
360 int total = 0;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
361
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
362 for (int i = 0; i < subsetcount; i++) {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
363 columnEntries[i] = readColumnEntryBytes(i);
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
364 if (columnEntries[i] == null) {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
365 throw new IOException("Failed to read subset line entry");
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
366 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
367 total += columnEntries[i].length;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
368 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
369
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
370 ret = new byte[total];
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
371 int offset = 0;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
372 for (int i = 0; i < subsetcount; i++) {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
373 System.arraycopy(columnEntries[i], 0, ret, offset, columnEntries[i].length);
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
374 offset += columnEntries[i].length;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
375 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
376 dbgLog.fine("line: "+new String(ret));
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
377 return ret;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
378 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
379
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
380
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
381 public void close() {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
382 if (fileChannel != null) {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
383 try {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
384 fileChannel.close();
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
385 } catch (IOException ioe) {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
386 // don't care.
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
387 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
388 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
389 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
390
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
391 public void subsetFile(String infile, String outfile, Set<Integer> columns, Long numCases) {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
392 subsetFile(infile, outfile, columns, numCases, "\t");
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
393 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
394
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
395 public void subsetFile(String infile, String outfile, Set<Integer> columns, Long numCases,
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
396 String delimiter) {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
397 try {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
398 subsetFile(new FileInputStream(new File(infile)), outfile, columns, numCases, delimiter);
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
399 } catch (IOException ex) {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
400 throw new RuntimeException("Could not open file "+infile);
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
401 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
402 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
403
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
404
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
405 public void subsetFile(InputStream in, String outfile, Set<Integer> columns, Long numCases,
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
406 String delimiter) {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
407 try {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
408 Scanner scanner = new Scanner(in);
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
409 scanner.useDelimiter("\\n");
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
410
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
411 BufferedWriter out = new BufferedWriter(new FileWriter(outfile));
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
412 for (long caseIndex = 0; caseIndex < numCases; caseIndex++) {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
413 if (scanner.hasNext()) {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
414 String[] line = (scanner.next()).split(delimiter,-1);
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
415 List<String> ln = new ArrayList<String>();
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
416 for (Integer i : columns) {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
417 ln.add(line[i]);
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
418 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
419 out.write(StringUtils.join(ln,"\t")+"\n");
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
420 } else {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
421 throw new RuntimeException("Tab file has fewer rows than the determined number of cases.");
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
422 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
423 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
424
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
425 while (scanner.hasNext()) {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
426 if (!"".equals(scanner.next()) ) {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
427 throw new RuntimeException("Tab file has extra nonempty rows than the determined number of cases.");
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
428
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
429 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
430 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
431
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
432 scanner.close();
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
433 out.close();
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
434
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
435 } catch (FileNotFoundException e) {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
436 e.printStackTrace();
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
437 } catch (IOException e) {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
438 e.printStackTrace();
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
439 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
440
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
441 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
442
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
443 /*
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
444 * Straightforward method for subsetting a column; inefficient on large
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
445 * files, OK to use on small files:
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
446 */
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
447
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
448 public static Double[] subsetDoubleVector(InputStream in, int column, int numCases) {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
449 Double[] retVector = new Double[numCases];
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
450 Scanner scanner = new Scanner(in);
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
451 scanner.useDelimiter("\\n");
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
452
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
453 for (int caseIndex = 0; caseIndex < numCases; caseIndex++) {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
454 if (scanner.hasNext()) {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
455 String[] line = (scanner.next()).split("\t", -1);
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
456 try {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
457 retVector[caseIndex] = new Double(line[column]);
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
458 } catch (NumberFormatException ex) {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
459 retVector[caseIndex] = null; // missing value
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
460 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
461 } else {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
462 scanner.close();
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
463 throw new RuntimeException("Tab file has fewer rows than the stored number of cases!");
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
464 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
465 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
466
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
467 int tailIndex = numCases;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
468 while (scanner.hasNext()) {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
469 String nextLine = scanner.next();
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
470 if (!"".equals(nextLine)) {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
471 scanner.close();
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
472 throw new RuntimeException("Column "+column+": tab file has more nonempty rows than the stored number of cases ("+numCases+")! current index: "+tailIndex+", line: "+nextLine);
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
473 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
474 tailIndex++;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
475 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
476
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
477 scanner.close();
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
478 return retVector;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
479
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
480 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
481
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
482 /*
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
483 * Straightforward method for subsetting a tab-delimited data file, extracting
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
484 * all the columns representing continuous variables and returning them as
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
485 * a 2-dimensional array of Doubles;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
486 * Inefficient on large files, OK to use on small ones.
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
487 */
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
488 public static Double[][] subsetDoubleVectors(InputStream in, Set<Integer> columns, int numCases) throws IOException {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
489 Double[][] retVector = new Double[columns.size()][numCases];
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
490 Scanner scanner = new Scanner(in);
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
491 scanner.useDelimiter("\\n");
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
492
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
493 for (int caseIndex = 0; caseIndex < numCases; caseIndex++) {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
494 if (scanner.hasNext()) {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
495 String[] line = (scanner.next()).split("\t", -1);
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
496 int j = 0;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
497 for (Integer i : columns) {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
498 try {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
499 // TODO: verify that NaN and +-Inf are going to be
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
500 // handled correctly here! -- L.A.
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
501 // NO, "+-Inf" is not handled correctly; see the
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
502 // comment further down below.
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
503 retVector[j][caseIndex] = new Double(line[i]);
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
504 } catch (NumberFormatException ex) {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
505 retVector[j][caseIndex] = null; // missing value
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
506 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
507 j++;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
508 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
509 } else {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
510 scanner.close();
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
511 throw new IOException("Tab file has fewer rows than the stored number of cases!");
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
512 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
513 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
514
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
515 int tailIndex = numCases;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
516 while (scanner.hasNext()) {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
517 String nextLine = scanner.next();
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
518 if (!"".equals(nextLine)) {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
519 scanner.close();
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
520 throw new IOException("Tab file has more nonempty rows than the stored number of cases ("+numCases+")! current index: "+tailIndex+", line: "+nextLine);
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
521 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
522 tailIndex++;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
523 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
524
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
525 scanner.close();
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
526 return retVector;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
527
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
528 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
529
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
530 public String[] subsetStringVector(DataFile datafile, int column) throws IOException {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
531 return (String[])subsetObjectVector(datafile, column, COLUMN_TYPE_STRING);
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
532 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
533
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
534 public Double[] subsetDoubleVector(DataFile datafile, int column) throws IOException {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
535 return (Double[])subsetObjectVector(datafile, column, COLUMN_TYPE_DOUBLE);
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
536 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
537
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
538 public Long[] subsetLongVector(DataFile datafile, int column) throws IOException {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
539 return (Long[])subsetObjectVector(datafile, column, COLUMN_TYPE_LONG);
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
540 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
541
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
542 // Float methods are temporary;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
543 // In normal operations we'll be treating all the floating point types as
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
544 // doubles. I need to be able to handle floats for some 4.0 vs 3.* ingest
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
545 // tests. -- L.A.
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
546
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
547 public Float[] subsetFloatVector(DataFile datafile, int column) throws IOException {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
548 return (Float[])subsetObjectVector(datafile, column, COLUMN_TYPE_FLOAT);
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
549 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
550
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
551 public String[] subsetStringVector(File tabfile, int column, int varcount, int casecount) throws IOException {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
552 return (String[])subsetObjectVector(tabfile, column, varcount, casecount, COLUMN_TYPE_STRING);
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
553 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
554
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
555 public Double[] subsetDoubleVector(File tabfile, int column, int varcount, int casecount) throws IOException {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
556 return (Double[])subsetObjectVector(tabfile, column, varcount, casecount, COLUMN_TYPE_DOUBLE);
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
557 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
558
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
559 public Long[] subsetLongVector(File tabfile, int column, int varcount, int casecount) throws IOException {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
560 return (Long[])subsetObjectVector(tabfile, column, varcount, casecount, COLUMN_TYPE_LONG);
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
561 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
562
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
563 public Float[] subsetFloatVector(File tabfile, int column, int varcount, int casecount) throws IOException {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
564 return (Float[])subsetObjectVector(tabfile, column, varcount, casecount, COLUMN_TYPE_FLOAT);
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
565 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
566
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
567 public Object[] subsetObjectVector(DataFile dataFile, int column, int columntype) throws IOException {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
568 if (!dataFile.isTabularData()) {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
569 throw new IOException("DataFile is not tabular data.");
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
570 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
571
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
572 int varcount = dataFile.getDataTable().getVarQuantity().intValue();
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
573 int casecount = dataFile.getDataTable().getCaseQuantity().intValue();
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
574
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
575 if (column >= varcount) {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
576 throw new IOException("Column "+column+" is out of bounds.");
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
577 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
578
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
579 File tabfile = dataFile.getFileSystemLocation().toFile();
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
580
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
581 if (columntype == COLUMN_TYPE_STRING) {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
582 String filename = dataFile.getFileMetadata().getLabel();
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
583 if (filename != null) {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
584 filename = filename.replaceFirst("^_", "");
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
585 Integer fnumvalue = null;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
586 try {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
587 fnumvalue = new Integer(filename);
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
588 } catch (Exception ex){
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
589 fnumvalue = null;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
590 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
591 if (fnumvalue != null) {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
592 //if ((fnumvalue.intValue() < 112497)) { // && (fnumvalue.intValue() > 60015)) {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
593 if ((fnumvalue.intValue() < 111931)) { // && (fnumvalue.intValue() > 60015)) {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
594 if (!(fnumvalue.intValue() == 60007
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
595 || fnumvalue.intValue() == 59997
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
596 || fnumvalue.intValue() == 60015
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
597 || fnumvalue.intValue() == 59948
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
598 || fnumvalue.intValue() == 60012
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
599 || fnumvalue.intValue() == 52585
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
600 || fnumvalue.intValue() == 60005
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
601 || fnumvalue.intValue() == 60002
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
602 || fnumvalue.intValue() == 59954
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
603 || fnumvalue.intValue() == 60008
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
604 || fnumvalue.intValue() == 54972
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
605 || fnumvalue.intValue() == 55010
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
606 || fnumvalue.intValue() == 54996
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
607 || fnumvalue.intValue() == 53527
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
608 || fnumvalue.intValue() == 53546
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
609 || fnumvalue.intValue() == 55002
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
610 || fnumvalue.intValue() == 55006
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
611 || fnumvalue.intValue() == 54998
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
612 || fnumvalue.intValue() == 52552
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
613 // SPSS/SAV cases with similar issue - compat mode must be disabled
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
614 //|| fnumvalue.intValue() == 101826 // temporary - tricky file with accents and v. 16...
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
615 || fnumvalue.intValue() == 54618 // another SAV file, with long strings...
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
616 || fnumvalue.intValue() == 54619 // [same]
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
617 || fnumvalue.intValue() == 57983
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
618 || fnumvalue.intValue() == 58262
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
619 || fnumvalue.intValue() == 58288
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
620 || fnumvalue.intValue() == 58656
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
621 || fnumvalue.intValue() == 59144
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
622 // || fnumvalue.intValue() == 69626 [nope!]
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
623 )) {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
624 dbgLog.info("\"Old\" file name detected; using \"compatibility mode\" for a character vector subset;");
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
625 return subsetObjectVector(tabfile, column, varcount, casecount, columntype, true);
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
626 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
627 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
628 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
629 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
630 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
631
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
632 return subsetObjectVector(tabfile, column, varcount, casecount, columntype);
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
633 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
634
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
635 public Object[] subsetObjectVector(File tabfile, int column, int varcount, int casecount, int columntype) throws IOException {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
636 return subsetObjectVector(tabfile, column, varcount, casecount, columntype, false);
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
637 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
638
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
639
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
640
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
641 public Object[] subsetObjectVector(File tabfile, int column, int varcount, int casecount, int columntype, boolean compatmode) throws IOException {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
642
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
643 Object[] retVector = null;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
644
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
645 boolean isString = false;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
646 boolean isDouble = false;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
647 boolean isLong = false;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
648 boolean isFloat = false;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
649
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
650 //Locale loc = new Locale("en", "US");
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
651
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
652 if (columntype == COLUMN_TYPE_STRING) {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
653 isString = true;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
654 retVector = new String[casecount];
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
655 } else if (columntype == COLUMN_TYPE_DOUBLE) {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
656 isDouble = true;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
657 retVector = new Double[casecount];
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
658 } else if (columntype == COLUMN_TYPE_LONG) {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
659 isLong = true;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
660 retVector = new Long[casecount];
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
661 } else if (columntype == COLUMN_TYPE_FLOAT){
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
662 isFloat = true;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
663 retVector = new Float[casecount];
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
664 } else {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
665 throw new IOException("Unsupported column type: "+columntype);
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
666 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
667
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
668 File rotatedImageFile = getRotatedImage(tabfile, varcount, casecount);
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
669 long[] columnEndOffsets = extractColumnOffsets(rotatedImageFile, varcount, casecount);
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
670 long columnOffset = 0;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
671 long columnLength = 0;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
672
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
673 if (column > 0) {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
674 columnOffset = columnEndOffsets[column - 1];
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
675 columnLength = columnEndOffsets[column] - columnEndOffsets[column - 1];
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
676 } else {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
677 columnOffset = varcount * 8;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
678 columnLength = columnEndOffsets[0] - varcount * 8;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
679 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
680
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
681 FileChannel fc = (FileChannel.open(Paths.get(rotatedImageFile.getAbsolutePath()), StandardOpenOption.READ));
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
682 fc.position(columnOffset);
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
683 int MAX_COLUMN_BUFFER = 8192;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
684
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
685 ByteBuffer in = ByteBuffer.allocate(MAX_COLUMN_BUFFER);
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
686
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
687 if (columnLength < MAX_COLUMN_BUFFER) {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
688 in.limit((int)(columnLength));
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
689 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
690
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
691 long bytesRead = 0;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
692 long bytesReadTotal = 0;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
693 int caseindex = 0;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
694 int byteoffset = 0;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
695 byte[] leftover = null;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
696
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
697 while (bytesReadTotal < columnLength) {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
698 bytesRead = fc.read(in);
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
699 byte[] columnBytes = in.array();
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
700 int bytecount = 0;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
701
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
702
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
703 while (bytecount < bytesRead) {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
704 if (columnBytes[bytecount] == '\n') {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
705 /*
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
706 String token = new String(columnBytes, byteoffset, bytecount-byteoffset, "UTF8");
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
707
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
708 if (leftover != null) {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
709 String leftoverString = new String (leftover, "UTF8");
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
710 token = leftoverString + token;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
711 leftover = null;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
712 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
713 */
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
714 /*
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
715 * Note that the way I was doing it at first - above -
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
716 * was not quite the correct way - because I was creating UTF8
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
717 * strings from the leftover bytes, and the bytes in the
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
718 * current buffer *separately*; which means, if a multi-byte
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
719 * UTF8 character got split in the middle between one buffer
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
720 * and the next, both chunks of it would become junk
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
721 * characters, on each side!
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
722 * The correct way of doing it, of course, is to create a
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
723 * merged byte buffer, and then turn it into a UTF8 string.
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
724 * -- L.A. 4.0
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
725 */
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
726 String token = null;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
727
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
728 if (leftover == null) {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
729 token = new String(columnBytes, byteoffset, bytecount-byteoffset, "UTF8");
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
730 } else {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
731 byte[] merged = new byte[leftover.length + bytecount-byteoffset];
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
732
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
733 System.arraycopy(leftover, 0, merged, 0, leftover.length);
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
734 System.arraycopy(columnBytes, byteoffset, merged, leftover.length, bytecount-byteoffset);
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
735 token = new String (merged, "UTF8");
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
736 leftover = null;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
737 merged = null;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
738 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
739
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
740 if (isString) {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
741 if ("".equals(token)) {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
742 // An empty string is a string missing value!
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
743 // An empty string in quotes is an empty string!
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
744 retVector[caseindex] = null;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
745 } else {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
746 // Strip the outer quotes:
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
747 token = token.replaceFirst("^\\\"", "");
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
748 token = token.replaceFirst("\\\"$", "");
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
749
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
750 // We need to restore the special characters that
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
751 // are stored in tab files escaped - quotes, new lines
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
752 // and tabs. Before we do that however, we need to
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
753 // take care of any escaped backslashes stored in
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
754 // the tab file. I.e., "foo\t" should be transformed
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
755 // to "foo<TAB>"; but "foo\\t" should be transformed
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
756 // to "foo\t". This way new lines and tabs that were
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
757 // already escaped in the original data are not
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
758 // going to be transformed to unescaped tab and
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
759 // new line characters!
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
760
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
761 String[] splitTokens = token.split(Matcher.quoteReplacement("\\\\"), -2);
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
762
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
763 // (note that it's important to use the 2-argument version
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
764 // of String.split(), and set the limit argument to a
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
765 // negative value; otherwise any trailing backslashes
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
766 // are lost.)
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
767
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
768 for (int i = 0; i < splitTokens.length; i++) {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
769 splitTokens[i] = splitTokens[i].replaceAll(Matcher.quoteReplacement("\\\""), "\"");
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
770 splitTokens[i] = splitTokens[i].replaceAll(Matcher.quoteReplacement("\\t"), "\t");
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
771 splitTokens[i] = splitTokens[i].replaceAll(Matcher.quoteReplacement("\\n"), "\n");
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
772 splitTokens[i] = splitTokens[i].replaceAll(Matcher.quoteReplacement("\\r"), "\r");
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
773 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
774 // TODO:
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
775 // Make (some of?) the above optional; for ex., we
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
776 // do need to restore the newlines when calculating UNFs;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
777 // But if we are subsetting these vectors in order to
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
778 // create a new tab-delimited file, they will
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
779 // actually break things! -- L.A. Jul. 28 2014
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
780
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
781 token = StringUtils.join(splitTokens, '\\');
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
782
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
783 // "compatibility mode" - a hack, to be able to produce
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
784 // unfs identical to those produced by the "early"
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
785 // unf5 jar; will be removed in production 4.0.
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
786 // -- L.A. (TODO: ...)
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
787 if (compatmode && !"".equals(token)) {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
788 if (token.length() > 128) {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
789 if ("".equals(token.trim())) {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
790 // don't ask...
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
791 token = token.substring(0, 129);
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
792 } else {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
793 token = token.substring(0, 128);
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
794 //token = String.format(loc, "%.128s", token);
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
795 token = token.trim();
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
796 //dbgLog.info("formatted and trimmed: "+token);
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
797 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
798 } else {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
799 if ("".equals(token.trim())) {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
800 // again, don't ask;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
801 // - this replicates some bugginness
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
802 // that happens inside unf5;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
803 token = "null";
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
804 } else {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
805 token = token.trim();
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
806 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
807 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
808 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
809
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
810 retVector[caseindex] = token;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
811 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
812 } else if (isDouble) {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
813 try {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
814 // TODO: verify that NaN and +-Inf are
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
815 // handled correctly here! -- L.A.
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
816 // Verified: new Double("nan") works correctly,
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
817 // resulting in Double.NaN;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
818 // Double("[+-]Inf") doesn't work however;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
819 // (the constructor appears to be expecting it
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
820 // to be spelled as "Infinity", "-Infinity", etc.
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
821 if ("inf".equalsIgnoreCase(token) || "+inf".equalsIgnoreCase(token)) {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
822 retVector[caseindex] = java.lang.Double.POSITIVE_INFINITY;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
823 } else if ("-inf".equalsIgnoreCase(token)) {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
824 retVector[caseindex] = java.lang.Double.NEGATIVE_INFINITY;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
825 } else if (token == null || token.equals("")) {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
826 // missing value:
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
827 retVector[caseindex] = null;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
828 } else {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
829 retVector[caseindex] = new Double(token);
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
830 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
831 } catch (NumberFormatException ex) {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
832 dbgLog.warning("NumberFormatException thrown for "+token+" as Double");
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
833
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
834 retVector[caseindex] = null; // missing value
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
835 // TODO: ?
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
836 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
837 } else if (isLong) {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
838 try {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
839 retVector[caseindex] = new Long(token);
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
840 } catch (NumberFormatException ex) {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
841 retVector[caseindex] = null; // assume missing value
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
842 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
843 } else if (isFloat) {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
844 try {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
845 if ("inf".equalsIgnoreCase(token) || "+inf".equalsIgnoreCase(token)) {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
846 retVector[caseindex] = java.lang.Float.POSITIVE_INFINITY;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
847 } else if ("-inf".equalsIgnoreCase(token)) {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
848 retVector[caseindex] = java.lang.Float.NEGATIVE_INFINITY;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
849 } else if (token == null || token.equals("")) {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
850 // missing value:
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
851 retVector[caseindex] = null;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
852 } else {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
853 retVector[caseindex] = new Float(token);
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
854 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
855 } catch (NumberFormatException ex) {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
856 dbgLog.warning("NumberFormatException thrown for "+token+" as Float");
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
857 retVector[caseindex] = null; // assume missing value (TODO: ?)
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
858 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
859 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
860 caseindex++;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
861
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
862 if (bytecount == bytesRead - 1) {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
863 byteoffset = 0;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
864 } else {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
865 byteoffset = bytecount + 1;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
866 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
867 } else {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
868 if (bytecount == bytesRead - 1) {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
869 // We've reached the end of the buffer;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
870 // This means we'll save whatever unused bytes left in
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
871 // it - i.e., the bytes between the last new line
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
872 // encountered and the end - in the leftover buffer.
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
873
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
874 // *EXCEPT*, there may be a case of a very long String
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
875 // that is actually longer than MAX_COLUMN_BUFFER, in
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
876 // which case it is possible that we've read through
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
877 // an entire buffer of bytes without finding any
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
878 // new lines... in this case we may need to add this
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
879 // entire byte buffer to an already existing leftover
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
880 // buffer!
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
881 if (leftover == null) {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
882 leftover = new byte[(int)bytesRead - byteoffset];
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
883 System.arraycopy(columnBytes, byteoffset, leftover, 0, (int)bytesRead - byteoffset);
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
884 } else {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
885 if (byteoffset != 0) {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
886 throw new IOException("Reached the end of the byte buffer, with some leftover left from the last read; yet the offset is not zero!");
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
887 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
888 byte[] merged = new byte[leftover.length + (int)bytesRead];
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
889
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
890 System.arraycopy(leftover, 0, merged, 0, leftover.length);
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
891 System.arraycopy(columnBytes, byteoffset, merged, leftover.length, (int)bytesRead);
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
892 //leftover = null;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
893 leftover = merged;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
894 merged = null;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
895 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
896 byteoffset = 0;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
897
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
898 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
899 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
900 bytecount++;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
901 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
902
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
903 bytesReadTotal += bytesRead;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
904 in.clear();
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
905 if (columnLength - bytesReadTotal < MAX_COLUMN_BUFFER) {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
906 in.limit((int)(columnLength - bytesReadTotal));
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
907 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
908 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
909
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
910 fc.close();
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
911
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
912 if (caseindex != casecount) {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
913 throw new IOException("Faile to read "+casecount+" tokens for column "+column);
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
914 //System.out.println("read "+caseindex+" tokens instead of expected "+casecount+".");
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
915 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
916
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
917 return retVector;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
918 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
919
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
920 private long[] extractColumnOffsets (File rotatedImageFile, int varcount, int casecount) throws IOException {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
921 BufferedInputStream rotfileStream = new BufferedInputStream(new FileInputStream(rotatedImageFile));
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
922
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
923 byte[] offsetHeader = new byte[varcount * 8];
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
924 long[] byteOffsets = new long[varcount];
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
925
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
926
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
927 int readlen = rotfileStream.read(offsetHeader);
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
928
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
929 if (readlen != varcount * 8) {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
930 throw new IOException ("Could not read "+varcount*8+" header bytes from the rotated file.");
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
931 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
932
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
933 for (int varindex = 0; varindex < varcount; varindex++) {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
934 byte[] offsetBytes = new byte[8];
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
935 System.arraycopy(offsetHeader, varindex*8, offsetBytes, 0, 8);
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
936
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
937 ByteBuffer offsetByteBuffer = ByteBuffer.wrap(offsetBytes);
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
938 byteOffsets[varindex] = offsetByteBuffer.getLong();
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
939
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
940 //System.out.println(byteOffsets[varindex]);
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
941 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
942
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
943 rotfileStream.close();
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
944
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
945 return byteOffsets;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
946 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
947
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
948 private File getRotatedImage(File tabfile, int varcount, int casecount) throws IOException {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
949 String fileName = tabfile.getAbsolutePath();
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
950 String rotatedImageFileName = fileName + ".90d";
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
951 File rotatedImageFile = new File(rotatedImageFileName);
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
952 if (rotatedImageFile.exists()) {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
953 //System.out.println("Image already exists!");
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
954 return rotatedImageFile;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
955 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
956
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
957 return generateRotatedImage(tabfile, varcount, casecount);
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
958
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
959 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
960
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
961 private File generateRotatedImage (File tabfile, int varcount, int casecount) throws IOException {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
962 // TODO: throw exceptions if bad file, zero varcount, etc. ...
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
963
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
964 String fileName = tabfile.getAbsolutePath();
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
965 String rotatedImageFileName = fileName + ".90d";
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
966
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
967 int MAX_OUTPUT_STREAMS = 32;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
968 int MAX_BUFFERED_BYTES = 10 * 1024 * 1024; // 10 MB - for now?
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
969 int MAX_COLUMN_BUFFER = 8 * 1024;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
970
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
971 // offsetHeader will contain the byte offsets of the individual column
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
972 // vectors in the final rotated image file
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
973 byte[] offsetHeader = new byte[varcount * 8];
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
974 int[] bufferedSizes = new int[varcount];
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
975 long[] cachedfileSizes = new long[varcount];
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
976 File[] columnTempFiles = new File[varcount];
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
977
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
978 for (int i = 0; i < varcount; i++) {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
979 bufferedSizes[i] = 0;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
980 cachedfileSizes[i] = 0;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
981 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
982
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
983 // TODO: adjust MAX_COLUMN_BUFFER here, so that the total size is
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
984 // no more than MAX_BUFFERED_BYTES (but no less than 1024 maybe?)
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
985
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
986 byte[][] bufferedColumns = new byte [varcount][MAX_COLUMN_BUFFER];
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
987
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
988 // read the tab-delimited file:
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
989
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
990 FileInputStream tabfileStream = new FileInputStream(tabfile);
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
991
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
992 Scanner scanner = new Scanner(tabfileStream);
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
993 scanner.useDelimiter("\\n");
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
994
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
995 for (int caseindex = 0; caseindex < casecount; caseindex++) {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
996 if (scanner.hasNext()) {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
997 String[] line = (scanner.next()).split("\t", -1);
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
998 // TODO: throw an exception if there are fewer tab-delimited
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
999 // tokens than the number of variables specified.
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1000 String token = "";
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1001 int tokensize = 0;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1002 for (int varindex = 0; varindex < varcount; varindex++) {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1003 // TODO: figure out the safest way to convert strings to
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1004 // bytes here. Is it going to be safer to use getBytes("UTF8")?
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1005 // we are already making the assumption that the values
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1006 // in the tab file are in UTF8. -- L.A.
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1007 token = line[varindex] + "\n";
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1008 tokensize = token.getBytes().length;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1009 if (bufferedSizes[varindex]+tokensize > MAX_COLUMN_BUFFER) {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1010 // fill the buffer and dump its contents into the temp file:
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1011 // (do note that there may be *several* MAX_COLUMN_BUFFERs
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1012 // worth of bytes in the token!)
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1013
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1014 int tokenoffset = 0;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1015
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1016 if (bufferedSizes[varindex] != MAX_COLUMN_BUFFER) {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1017 tokenoffset = MAX_COLUMN_BUFFER-bufferedSizes[varindex];
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1018 System.arraycopy(token.getBytes(), 0, bufferedColumns[varindex], bufferedSizes[varindex], tokenoffset);
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1019 } // (otherwise the buffer is already full, and we should
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1020 // simply dump it into the temp file, without adding any
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1021 // extra bytes to it)
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1022
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1023 File bufferTempFile = columnTempFiles[varindex];
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1024 if (bufferTempFile == null) {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1025 bufferTempFile = File.createTempFile("columnBufferFile", "bytes");
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1026 columnTempFiles[varindex] = bufferTempFile;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1027 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1028
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1029 // *append* the contents of the buffer to the end of the
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1030 // temp file, if already exists:
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1031 BufferedOutputStream outputStream = new BufferedOutputStream(new FileOutputStream (bufferTempFile, true));
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1032 outputStream.write(bufferedColumns[varindex], 0, MAX_COLUMN_BUFFER);
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1033 cachedfileSizes[varindex] += MAX_COLUMN_BUFFER;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1034
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1035 // keep writing MAX_COLUMN_BUFFER-size chunks of bytes into
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1036 // the temp file, for as long as there's more than MAX_COLUMN_BUFFER
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1037 // bytes left in the token:
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1038
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1039 while (tokensize - tokenoffset > MAX_COLUMN_BUFFER) {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1040 outputStream.write(token.getBytes(), tokenoffset, MAX_COLUMN_BUFFER);
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1041 cachedfileSizes[varindex] += MAX_COLUMN_BUFFER;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1042 tokenoffset += MAX_COLUMN_BUFFER;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1043 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1044
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1045 outputStream.close();
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1046
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1047 // buffer the remaining bytes and reset the buffered
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1048 // byte counter:
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1049
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1050 System.arraycopy(token.getBytes(),
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1051 tokenoffset,
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1052 bufferedColumns[varindex],
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1053 0,
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1054 tokensize - tokenoffset);
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1055
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1056 bufferedSizes[varindex] = tokensize - tokenoffset;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1057
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1058 } else {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1059 // continue buffering
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1060 System.arraycopy(token.getBytes(), 0, bufferedColumns[varindex], bufferedSizes[varindex], tokensize);
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1061 bufferedSizes[varindex] += tokensize;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1062 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1063 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1064 } else {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1065 scanner.close();
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1066 throw new IOException("Tab file has fewer rows than the stored number of cases!");
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1067 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1068
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1069 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1070
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1071 // OK, we've created the individual byte vectors of the tab file columns;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1072 // they may be partially saved in temp files and/or in memory.
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1073 // We now need to go through all these buffers and create the final
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1074 // rotated image file.
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1075
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1076 BufferedOutputStream finalOut = new BufferedOutputStream(new FileOutputStream (new File(rotatedImageFileName)));
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1077
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1078 // but first we should create the offset header and write it out into
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1079 // the final file; because it should be at the head, doh!
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1080
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1081 long columnOffset = varcount * 8;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1082 // (this is the offset of the first column vector; it is equal to the
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1083 // size of the offset header, i.e. varcount * 8 bytes)
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1084
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1085 for (int varindex = 0; varindex < varcount; varindex++) {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1086 long totalColumnBytes = cachedfileSizes[varindex] + bufferedSizes[varindex];
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1087 columnOffset+=totalColumnBytes;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1088 //totalColumnBytes;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1089 byte[] columnOffsetByteArray = ByteBuffer.allocate(8).putLong(columnOffset).array();
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1090 System.arraycopy(columnOffsetByteArray, 0, offsetHeader, varindex * 8, 8);
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1091 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1092
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1093 finalOut.write(offsetHeader, 0, varcount * 8);
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1094
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1095 for (int varindex = 0; varindex < varcount; varindex++) {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1096 long cachedBytesRead = 0;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1097
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1098 // check if there is a cached temp file:
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1099
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1100 File cachedTempFile = columnTempFiles[varindex];
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1101 if (cachedTempFile != null) {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1102 byte[] cachedBytes = new byte[MAX_COLUMN_BUFFER];
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1103 BufferedInputStream cachedIn = new BufferedInputStream(new FileInputStream(cachedTempFile));
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1104 int readlen = 0;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1105 while ((readlen = cachedIn.read(cachedBytes)) > -1) {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1106 finalOut.write(cachedBytes, 0, readlen);
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1107 cachedBytesRead += readlen;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1108 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1109 cachedIn.close();
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1110 // delete the temp file:
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1111 cachedTempFile.delete();
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1112
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1113 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1114
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1115 if (cachedBytesRead != cachedfileSizes[varindex]) {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1116 finalOut.close();
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1117 throw new IOException("Could not read the correct number of bytes cached for column "+varindex+"; "+
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1118 cachedfileSizes[varindex] + " bytes expected, "+cachedBytesRead+" read.");
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1119 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1120
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1121 // then check if there are any bytes buffered for this column:
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1122
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1123 if (bufferedSizes[varindex] > 0) {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1124 finalOut.write(bufferedColumns[varindex], 0, bufferedSizes[varindex]);
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1125 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1126
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1127 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1128
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1129 finalOut.close();
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1130 return new File(rotatedImageFileName);
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1131
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1132 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1133
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1134 /*
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1135 * Test method for taking a "rotated" image, and reversing it, reassembling
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1136 * all the columns in the original order. Which should result in a file
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1137 * byte-for-byte identical file to the original tab-delimited version.
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1138 *
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1139 * (do note that this method is not efficiently implemented; it's only
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1140 * being used for experiments so far, to confirm the accuracy of the
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1141 * accuracy of generateRotatedImage(). It should not be used for any
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1142 * practical means in the application!)
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1143 */
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1144 private void reverseRotatedImage (File rotfile, int varcount, int casecount) throws IOException {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1145 // open the file, read in the offset header:
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1146 BufferedInputStream rotfileStream = new BufferedInputStream(new FileInputStream(rotfile));
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1147
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1148 byte[] offsetHeader = new byte[varcount * 8];
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1149 long[] byteOffsets = new long[varcount];
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1150
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1151 int readlen = rotfileStream.read(offsetHeader);
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1152
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1153 if (readlen != varcount * 8) {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1154 throw new IOException ("Could not read "+varcount*8+" header bytes from the rotated file.");
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1155 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1156
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1157 for (int varindex = 0; varindex < varcount; varindex++) {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1158 byte[] offsetBytes = new byte[8];
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1159 System.arraycopy(offsetHeader, varindex*8, offsetBytes, 0, 8);
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1160
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1161 ByteBuffer offsetByteBuffer = ByteBuffer.wrap(offsetBytes);
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1162 byteOffsets[varindex] = offsetByteBuffer.getLong();
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1163
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1164 //System.out.println(byteOffsets[varindex]);
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1165 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1166
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1167 String [][] reversedMatrix = new String[casecount][varcount];
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1168
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1169 long offset = varcount * 8;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1170 byte[] columnBytes;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1171
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1172 for (int varindex = 0; varindex < varcount; varindex++) {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1173 long columnLength = byteOffsets[varindex] - offset;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1174
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1175
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1176
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1177 columnBytes = new byte[(int)columnLength];
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1178 readlen = rotfileStream.read(columnBytes);
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1179
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1180 if (readlen != columnLength) {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1181 throw new IOException ("Could not read "+columnBytes+" bytes for column "+varindex);
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1182 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1183 /*
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1184 String columnString = new String(columnBytes);
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1185 //System.out.print(columnString);
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1186 String[] values = columnString.split("\n", -1);
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1187
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1188 if (values.length < casecount) {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1189 throw new IOException("count mismatch: "+values.length+" tokens found for column "+varindex);
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1190 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1191
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1192 for (int caseindex = 0; caseindex < casecount; caseindex++) {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1193 reversedMatrix[caseindex][varindex] = values[caseindex];
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1194 }*/
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1195
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1196 int bytecount = 0;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1197 int byteoffset = 0;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1198 int caseindex = 0;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1199 //System.out.println("generating value vector for column "+varindex);
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1200 while (bytecount < columnLength) {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1201 if (columnBytes[bytecount] == '\n') {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1202 String token = new String(columnBytes, byteoffset, bytecount-byteoffset);
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1203 reversedMatrix[caseindex++][varindex] = token;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1204 byteoffset = bytecount + 1;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1205 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1206 bytecount++;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1207 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1208
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1209 if (caseindex != casecount) {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1210 throw new IOException("count mismatch: "+caseindex+" tokens found for column "+varindex);
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1211 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1212 offset = byteOffsets[varindex];
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1213 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1214
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1215 for (int caseindex = 0; caseindex < casecount; caseindex++) {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1216 for (int varindex = 0; varindex < varcount; varindex++) {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1217 System.out.print(reversedMatrix[caseindex][varindex]);
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1218 if (varindex < varcount-1) {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1219 System.out.print("\t");
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1220 } else {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1221 System.out.print("\n");
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1222 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1223 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1224 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1225
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1226 rotfileStream.close();
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1227
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1228
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1229 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1230
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1231 /**
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1232 * main() method, for testing
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1233 * usage: java edu.harvard.iq.dataverse.dataaccess.TabularSubsetGenerator testfile.tab varcount casecount column type
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1234 * make sure the CLASSPATH contains ...
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1235 *
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1236 */
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1237
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1238 public static void main(String[] args) {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1239
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1240 String tabFileName = args[0];
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1241 int varcount = new Integer(args[1]).intValue();
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1242 int casecount = new Integer(args[2]).intValue();
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1243 int column = new Integer(args[3]).intValue();
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1244 String type = args[4];
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1245
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1246 File tabFile = new File(tabFileName);
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1247 File rotatedImageFile = null;
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1248
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1249 TabularSubsetGenerator subsetGenerator = new TabularSubsetGenerator();
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1250
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1251 /*
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1252 try {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1253 rotatedImageFile = subsetGenerator.getRotatedImage(tabFile, varcount, casecount);
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1254 } catch (IOException ex) {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1255 System.out.println(ex.getMessage());
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1256 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1257 */
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1258
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1259 //System.out.println("\nFinished generating \"rotated\" column image file.");
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1260
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1261 //System.out.println("\nOffsets:");
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1262
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1263 MathContext doubleMathContext = new MathContext(15, RoundingMode.HALF_EVEN);
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1264 String FORMAT_IEEE754 = "%+#.15e";
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1265
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1266 try {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1267 //subsetGenerator.reverseRotatedImage(rotatedImageFile, varcount, casecount);
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1268 //String[] columns = subsetGenerator.subsetStringVector(tabFile, column, varcount, casecount);
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1269 if ("string".equals(type)) {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1270 String[] columns = subsetGenerator.subsetStringVector(tabFile, column, varcount, casecount);
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1271 for (int i = 0; i < casecount; i++) {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1272 System.out.println(columns[i]);
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1273 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1274 } else {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1275
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1276 Double[] columns = subsetGenerator.subsetDoubleVector(tabFile, column, varcount, casecount);
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1277 for (int i = 0; i < casecount; i++) {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1278 if (columns[i] != null) {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1279 BigDecimal outBigDecimal = new BigDecimal(columns[i], doubleMathContext);
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1280 System.out.println(String.format(FORMAT_IEEE754, outBigDecimal));
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1281 } else {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1282 System.out.println("NA");
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1283 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1284 //System.out.println(columns[i]);
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1285 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1286 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1287 } catch (IOException ex) {
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1288 System.out.println(ex.getMessage());
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1289 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1290 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1291 }
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1292
a50cf11e5178 Rewrite LGDataverse completely upgrading to dataverse4.0
Zoe Hong <zhong@mpiwg-berlin.mpg.de>
parents:
diff changeset
1293