Mercurial > hg > extraction-interface
view geotemco/lib/jszip/jszip-load.js @ 24:b55f5d3564c4
add some regular expressions and wordlist for topic ??
author | Zoe Hong <zhong@mpiwg-berlin.mpg.de> |
---|---|
date | Fri, 27 Feb 2015 16:35:59 +0100 |
parents | b12c99b7c3f0 |
children |
line wrap: on
line source
/** JSZip - A Javascript class for generating and reading zip files <http://stuartk.com/jszip> (c) 2011 David Duponchel <d.duponchel@gmail.com> Dual licenced under the MIT license or GPLv3. See LICENSE.markdown. **/ /*global JSZip,JSZipBase64 */ (function () { var MAX_VALUE_16BITS = 65535; var MAX_VALUE_32BITS = -1; // well, "\xFF\xFF\xFF\xFF\xFF\xFF\xFF\xFF" is parsed as -1 /** * Prettify a string read as binary. * @param {string} str the string to prettify. * @return {string} a pretty string. */ var pretty = function (str) { var res = '', code, i; for (i = 0; i < (str||"").length; i++) { code = str.charCodeAt(i); res += '\\x' + (code < 16 ? "0" : "") + code.toString(16).toUpperCase(); } return res; }; /** * Find a compression registered in JSZip. * @param {string} compressionMethod the method magic to find. * @return {Object|null} the JSZip compression object, null if none found. */ var findCompression = function (compressionMethod) { for (var method in JSZip.compressions) { if( !JSZip.compressions.hasOwnProperty(method) ) { continue; } if (JSZip.compressions[method].magic === compressionMethod) { return JSZip.compressions[method]; } } return null; }; // class StreamReader {{{ /** * Read bytes from a stream. * Developer tip : when debugging, a watch on pretty(this.reader.stream.slice(this.reader.index)) * is very useful :) * @constructor * @param {String|ArrayBuffer|Uint8Array} stream the stream to read. */ function StreamReader(stream) { this.stream = ""; if (JSZip.support.uint8array && stream instanceof Uint8Array) { this.stream = JSZip.utils.uint8Array2String(stream); } else if (JSZip.support.arraybuffer && stream instanceof ArrayBuffer) { var bufferView = new Uint8Array(stream); this.stream = JSZip.utils.uint8Array2String(bufferView); } else { this.stream = JSZip.utils.string2binary(stream); } this.index = 0; } StreamReader.prototype = { /** * Check that the offset will not go too far. * @param {string} offset the additional offset to check. * @throws {Error} an Error if the offset is out of bounds. */ checkOffset : function (offset) { this.checkIndex(this.index + offset); }, /** * Check that the specifed index will not be too far. * @param {string} newIndex the index to check. * @throws {Error} an Error if the index is out of bounds. */ checkIndex : function (newIndex) { if (this.stream.length < newIndex || newIndex < 0) { throw new Error("End of stream reached (stream length = " + this.stream.length + ", asked index = " + (newIndex) + "). Corrupted zip ?"); } }, /** * Change the index. * @param {number} newIndex The new index. * @throws {Error} if the new index is out of the stream. */ setIndex : function (newIndex) { this.checkIndex(newIndex); this.index = newIndex; }, /** * Skip the next n bytes. * @param {number} n the number of bytes to skip. * @throws {Error} if the new index is out of the stream. */ skip : function (n) { this.setIndex(this.index + n); }, /** * Get the byte at the specified index. * @param {number} i the index to use. * @return {number} a byte. */ byteAt : function(i) { return this.stream.charCodeAt(i); }, /** * Get the next number with a given byte size. * @param {number} size the number of bytes to read. * @return {number} the corresponding number. */ readInt : function (size) { var result = 0, i; this.checkOffset(size); for(i = this.index + size - 1; i >= this.index; i--) { result = (result << 8) + this.byteAt(i); } this.index += size; return result; }, /** * Get the next string with a given byte size. * @param {number} size the number of bytes to read. * @return {string} the corresponding string. */ readString : function (size) { this.checkOffset(size); // this will work because the constructor applied the "& 0xff" mask. var result = this.stream.slice(this.index, this.index + size); this.index += size; return result; }, /** * Get the next date. * @return {Date} the date. */ readDate : function () { var dostime = this.readInt(4); return new Date( ((dostime >> 25) & 0x7f) + 1980, // year ((dostime >> 21) & 0x0f) - 1, // month (dostime >> 16) & 0x1f, // day (dostime >> 11) & 0x1f, // hour (dostime >> 5) & 0x3f, // minute (dostime & 0x1f) << 1); // second } }; // }}} end of StreamReader // class ZipEntry {{{ /** * An entry in the zip file. * @constructor * @param {Object} options Options of the current file. * @param {Object} loadOptions Options for loading the stream. */ function ZipEntry(options, loadOptions) { this.options = options; this.loadOptions = loadOptions; } ZipEntry.prototype = { /** * say if the file is encrypted. * @return {boolean} true if the file is encrypted, false otherwise. */ isEncrypted : function () { // bit 1 is set return (this.bitFlag & 0x0001) === 0x0001; }, /** * say if the file has utf-8 filename/comment. * @return {boolean} true if the filename/comment is in utf-8, false otherwise. */ useUTF8 : function () { // bit 11 is set return (this.bitFlag & 0x0800) === 0x0800; }, /** * Read the local part of a zip file and add the info in this object. * @param {StreamReader} reader the reader to use. */ readLocalPart : function(reader) { var compression, localExtraFieldsLength; // we already know everything from the central dir ! // If the central dir data are false, we are doomed. // On the bright side, the local part is scary : zip64, data descriptors, both, etc. // The less data we get here, the more reliable this should be. // Let's skip the whole header and dash to the data ! reader.skip(22); // in some zip created on windows, the filename stored in the central dir contains \ instead of /. // Strangely, the filename here is OK. // I would love to treat these zip files as corrupted (see http://www.info-zip.org/FAQ.html#backslashes // or APPNOTE#4.4.17.1, "All slashes MUST be forward slashes '/'") but there are a lot of bad zip generators... // Search "unzip mismatching "local" filename continuing with "central" filename version" on // the internet. // // I think I see the logic here : the central directory is used to display // content and the local directory is used to extract the files. Mixing / and \ // may be used to display \ to windows users and use / when extracting the files. // Unfortunately, this lead also to some issues : http://seclists.org/fulldisclosure/2009/Sep/394 this.fileNameLength = reader.readInt(2); localExtraFieldsLength = reader.readInt(2); // can't be sure this will be the same as the central dir this.fileName = reader.readString(this.fileNameLength); reader.skip(localExtraFieldsLength); if (this.compressedSize == -1 || this.uncompressedSize == -1) { throw new Error("Bug or corrupted zip : didn't get enough informations from the central directory " + "(compressedSize == -1 || uncompressedSize == -1)"); } this.compressedFileData = reader.readString(this.compressedSize); compression = findCompression(this.compressionMethod); if (compression === null) { // no compression found throw new Error("Corrupted zip : compression " + pretty(this.compressionMethod) + " unknown (inner file : " + this.fileName + ")"); } this.uncompressedFileData = compression.uncompress(this.compressedFileData); if (this.uncompressedFileData.length !== this.uncompressedSize) { throw new Error("Bug : uncompressed data size mismatch"); } if (this.loadOptions.checkCRC32 && JSZip.prototype.crc32(this.uncompressedFileData) !== this.crc32) { throw new Error("Corrupted zip : CRC32 mismatch"); } }, /** * Read the central part of a zip file and add the info in this object. * @param {StreamReader} reader the reader to use. */ readCentralPart : function(reader) { this.versionMadeBy = reader.readString(2); this.versionNeeded = reader.readInt(2); this.bitFlag = reader.readInt(2); this.compressionMethod = reader.readString(2); this.date = reader.readDate(); this.crc32 = reader.readInt(4); this.compressedSize = reader.readInt(4); this.uncompressedSize = reader.readInt(4); this.fileNameLength = reader.readInt(2); this.extraFieldsLength = reader.readInt(2); this.fileCommentLength = reader.readInt(2); this.diskNumberStart = reader.readInt(2); this.internalFileAttributes = reader.readInt(2); this.externalFileAttributes = reader.readInt(4); this.localHeaderOffset = reader.readInt(4); if (this.isEncrypted()) { throw new Error("Encrypted zip are not supported"); } this.fileName = reader.readString(this.fileNameLength); this.readExtraFields(reader); this.parseZIP64ExtraField(reader); this.fileComment = reader.readString(this.fileCommentLength); // warning, this is true only for zip with madeBy == DOS (plateform dependent feature) this.dir = this.externalFileAttributes & 0x00000010 ? true : false; }, /** * Parse the ZIP64 extra field and merge the info in the current ZipEntry. * @param {StreamReader} reader the reader to use. */ parseZIP64ExtraField : function(reader) { if(!this.extraFields[0x0001]) { return; } // should be something, preparing the extra reader var extraReader = new StreamReader(this.extraFields[0x0001].value); // I really hope that these 64bits integer can fit in 32 bits integer, because js // won't let us have more. if(this.uncompressedSize === MAX_VALUE_32BITS) { this.uncompressedSize = extraReader.readInt(8); } if(this.compressedSize === MAX_VALUE_32BITS) { this.compressedSize = extraReader.readInt(8); } if(this.localHeaderOffset === MAX_VALUE_32BITS) { this.localHeaderOffset = extraReader.readInt(8); } if(this.diskNumberStart === MAX_VALUE_32BITS) { this.diskNumberStart = extraReader.readInt(4); } }, /** * Read the central part of a zip file and add the info in this object. * @param {StreamReader} reader the reader to use. */ readExtraFields : function(reader) { var start = reader.index, extraFieldId, extraFieldLength, extraFieldValue; this.extraFields = this.extraFields || {}; while (reader.index < start + this.extraFieldsLength) { extraFieldId = reader.readInt(2); extraFieldLength = reader.readInt(2); extraFieldValue = reader.readString(extraFieldLength); this.extraFields[extraFieldId] = { id: extraFieldId, length: extraFieldLength, value: extraFieldValue }; } }, /** * Apply an UTF8 transformation if needed. */ handleUTF8 : function() { if (this.useUTF8()) { this.fileName = JSZip.prototype.utf8decode(this.fileName); this.fileComment = JSZip.prototype.utf8decode(this.fileComment); } } }; // }}} end of ZipEntry // class ZipEntries {{{ /** * All the entries in the zip file. * @constructor * @param {String|ArrayBuffer|Uint8Array} data the binary stream to load. * @param {Object} loadOptions Options for loading the stream. */ function ZipEntries(data, loadOptions) { this.files = []; this.loadOptions = loadOptions; if (data) { this.load(data); } } ZipEntries.prototype = { /** * Check that the reader is on the speficied signature. * @param {string} expectedSignature the expected signature. * @throws {Error} if it is an other signature. */ checkSignature : function(expectedSignature) { var signature = this.reader.readString(4); if (signature !== expectedSignature) { throw new Error("Corrupted zip or bug : unexpected signature " + "(" + pretty(signature) + ", expected " + pretty(expectedSignature) + ")"); } }, /** * Read the end of the central directory. */ readBlockEndOfCentral : function () { this.diskNumber = this.reader.readInt(2); this.diskWithCentralDirStart = this.reader.readInt(2); this.centralDirRecordsOnThisDisk = this.reader.readInt(2); this.centralDirRecords = this.reader.readInt(2); this.centralDirSize = this.reader.readInt(4); this.centralDirOffset = this.reader.readInt(4); this.zipCommentLength = this.reader.readInt(2); this.zipComment = this.reader.readString(this.zipCommentLength); }, /** * Read the end of the Zip 64 central directory. * Not merged with the method readEndOfCentral : * The end of central can coexist with its Zip64 brother, * I don't want to read the wrong number of bytes ! */ readBlockZip64EndOfCentral : function () { this.zip64EndOfCentralSize = this.reader.readInt(8); this.versionMadeBy = this.reader.readString(2); this.versionNeeded = this.reader.readInt(2); this.diskNumber = this.reader.readInt(4); this.diskWithCentralDirStart = this.reader.readInt(4); this.centralDirRecordsOnThisDisk = this.reader.readInt(8); this.centralDirRecords = this.reader.readInt(8); this.centralDirSize = this.reader.readInt(8); this.centralDirOffset = this.reader.readInt(8); this.zip64ExtensibleData = {}; var extraDataSize = this.zip64EndOfCentralSize - 44, index = 0, extraFieldId, extraFieldLength, extraFieldValue; while(index < extraDataSize) { extraFieldId = this.reader.readInt(2); extraFieldLength = this.reader.readInt(4); extraFieldValue = this.reader.readString(extraFieldLength); this.zip64ExtensibleData[extraFieldId] = { id: extraFieldId, length: extraFieldLength, value: extraFieldValue }; } }, /** * Read the end of the Zip 64 central directory locator. */ readBlockZip64EndOfCentralLocator : function () { this.diskWithZip64CentralDirStart = this.reader.readInt(4); this.relativeOffsetEndOfZip64CentralDir = this.reader.readInt(8); this.disksCount = this.reader.readInt(4); if (this.disksCount > 1) { throw new Error("Multi-volumes zip are not supported"); } }, /** * Read the local files, based on the offset read in the central part. */ readLocalFiles : function() { var i, file; for(i = 0; i < this.files.length; i++) { file = this.files[i]; this.reader.setIndex(file.localHeaderOffset); this.checkSignature(JSZip.signature.LOCAL_FILE_HEADER); file.readLocalPart(this.reader); file.handleUTF8(); } }, /** * Read the central directory. */ readCentralDir : function() { var file; this.reader.setIndex(this.centralDirOffset); while(this.reader.readString(4) === JSZip.signature.CENTRAL_FILE_HEADER) { file = new ZipEntry({ zip64: this.zip64 }, this.loadOptions); file.readCentralPart(this.reader); this.files.push(file); } }, /** * Read the end of central directory. */ readEndOfCentral : function() { var offset = this.reader.stream.lastIndexOf(JSZip.signature.CENTRAL_DIRECTORY_END); if (offset === -1) { throw new Error("Corrupted zip : can't find end of central directory"); } this.reader.setIndex(offset); this.checkSignature(JSZip.signature.CENTRAL_DIRECTORY_END); this.readBlockEndOfCentral(); /* extract from the zip spec : 4) If one of the fields in the end of central directory record is too small to hold required data, the field should be set to -1 (0xFFFF or 0xFFFFFFFF) and the ZIP64 format record should be created. 5) The end of central directory record and the Zip64 end of central directory locator record must reside on the same disk when splitting or spanning an archive. */ if ( this.diskNumber === MAX_VALUE_16BITS || this.diskWithCentralDirStart === MAX_VALUE_16BITS || this.centralDirRecordsOnThisDisk === MAX_VALUE_16BITS || this.centralDirRecords === MAX_VALUE_16BITS || this.centralDirSize === MAX_VALUE_32BITS || this.centralDirOffset === MAX_VALUE_32BITS ) { this.zip64 = true; /* Warning : the zip64 extension is supported, but ONLY if the 64bits integer read from the zip file can fit into a 32bits integer. This cannot be solved : Javascript represents all numbers as 64-bit double precision IEEE 754 floating point numbers. So, we have 53bits for integers and bitwise operations treat everything as 32bits. see https://developer.mozilla.org/en-US/docs/JavaScript/Reference/Operators/Bitwise_Operators and http://www.ecma-international.org/publications/files/ECMA-ST/ECMA-262.pdf section 8.5 */ // should look for a zip64 EOCD locator offset = this.reader.stream.lastIndexOf(JSZip.signature.ZIP64_CENTRAL_DIRECTORY_LOCATOR); if (offset === -1) { throw new Error("Corrupted zip : can't find the ZIP64 end of central directory locator"); } this.reader.setIndex(offset); this.checkSignature(JSZip.signature.ZIP64_CENTRAL_DIRECTORY_LOCATOR); this.readBlockZip64EndOfCentralLocator(); // now the zip64 EOCD record this.reader.setIndex(this.relativeOffsetEndOfZip64CentralDir); this.checkSignature(JSZip.signature.ZIP64_CENTRAL_DIRECTORY_END); this.readBlockZip64EndOfCentral(); } }, /** * Read a zip file and create ZipEntries. * @param {String|ArrayBuffer|Uint8Array} data the binary string representing a zip file. */ load : function(data) { this.reader = new StreamReader(data); this.readEndOfCentral(); this.readCentralDir(); this.readLocalFiles(); } }; // }}} end of ZipEntries /** * Implementation of the load method of JSZip. * It uses the above classes to decode a zip file, and load every files. * @param {String|ArrayBuffer|Uint8Array} data the data to load. * @param {Object} options Options for loading the stream. * options.base64 : is the stream in base64 ? default : false */ JSZip.prototype.load = function(data, options) { var files, zipEntries, i, input; options = options || {}; if(options.base64) { data = JSZipBase64.decode(data); } zipEntries = new ZipEntries(data, options); files = zipEntries.files; for (i = 0; i < files.length; i++) { input = files[i]; this.file(input.fileName, input.uncompressedFileData, { binary:true, optimizedBinaryString:true, date:input.date, dir:input.dir }); } return this; }; }()); // enforcing Stuk's coding style // vim: set shiftwidth=3 softtabstop=3 foldmethod=marker: