Context Navigation

Back to WO1_Euclid_1607

WO1_Euclid_1607: Euclid_1607.pl

File Euclid_1607.pl, 30.9 KB (added by Wolfgang Schmidle, 15 years ago)

Line
1	#!/usr/bin/perl -w
2	use strict;
3	use warnings;
4	use utf8;
5	binmode STDOUT, ':utf8';
6	use integer;
7
8
9	# Euclid_1607.pl
10	#
11	# author: Wolfgang Schmidle
12	# (c) Max Planck Institute for the History of Science, Berlin, Germany
13	# version 1.5, 2009-04-07
14	#
15	# This script processes the raw text of Euclid (1607), i.e. it turns (a local copy of)
16	# http://pythia.mpiwg-berlin.mpg.de/department1/mpdl/raw-texts/WO1_Euclid_1607.txt/WO1_Euclid_1607_V1.txt
17	# into a valid XML file.
18	#
19	# For Euclid (1607) see also
20	# https://itgroup.mpiwg-berlin.mpg.de:8080/tracs/mpdl-project-content/wiki/WO1_Euclid_1607
21	#
22	# Usage: perl Euclid_1607.txt > Euclid_1607.xml
23
24
25	my $rawtext = shift @ARGV;
26
27
28	# text-specific preparations
29
30
31	# global variables
32
33	my @text;
34
35	# counters
36	my $line;
37	my $doNotCheckBeforeLine = 0;
38	my $pb = 0;
39	my $def = 0;
40	my $post = 0;
41	my $comm = 0;
42	my $propos = 0;
43
44	my %divTypeCount;
45
46	# booleans
47	my $inH = 0;
48	my $inP = 0;
49	my $inMath = 0;
50	my $inMg = 0;
51	my $italicsH = 0;
52	my $italicsP = 0;
53	my $italicsTb = 0;
54	my $italicsMg = 0;
55	my $lineStartsWithS = 0;
56	my $inTb = 0;
57	my $mgInItalics = 0;
58
59	my $inDiv3 = 0;
60	my $inDef = 0;
61	my $defOnHold = 0;
62	my $inPost = 0;
63	my $inComm = 0;
64	my $inPropos = 0;
65
66	my $checkForDiv4 = 0;
67
68	# pointers
69	my $inMgSince = 0;
70
71	# others
72	my $pbTag = "";
73	my $div4 = "";
74
75
76	# read in the raw text
77	# and immediate emendations (things that break the processing, and changes of the line structure)
78
79	$line = 0;
80
81	open (RAWTEXT, "<:utf8", $rawtext) or die "Can't open the raw text file \"$rawtext\"!\n";
82	while (<RAWTEXT>) {
83
84	$line++;
85
86	s!\r\n!!; # remove Windows line ends
87
88	next if $line == 73677; # the last line contains the forbidden character SUBSTITUTE (U+001A)
89
90	# <pb> and <h> on separate lines (line 7, 303)
91	# (note that this emendation changes the line numbering!)
92	if (m!<pb><h!) {
93	push @text, "<pb>";
94	s!<pb>!!;
95	}
96
97	# resolve ten occurrences of </mg[lr]></p>
98	# (this code would fail if <mgr>...</mgr><p> is in one line)
99	if (m!</mg[lr]></p>!) {
100	s!</p>!!;
101	my $beginningOfNote = $line-2 + 2; # +2 because of lines 7 and 303, and it doesn't happen in one-line notes
102	while (!($text[$beginningOfNote] =~ m!<mg[lr]!)) { $beginningOfNote--; }
103	$text[$beginningOfNote-1] .= '</p>';
104	}
105
106	# remove library stamp
107	if (($line >= 26) && ($line <= 31)) { $_ = ""; }
108
109	push @text, $_;
110	}
111	close (RAWTEXT);
112
113
114	# emendations of the raw text before XMLifying
115
116	$line = 0;
117
118	for (@text) {
119
120	$line++;
121
122	# $ --> ſ (long s)
123	s!\$!ſ!g;
124
125	# & is a reserved character in XML
126	s!&!&!g;
127
128	# provisional treatment of a table which fills a whole page
129	# The problem is that one sentence starts on p.501 and continues on p.503, and p.502 is a table with caption.
130	# The first "solution" (<cap> --> <h>) was problematic because it meant that <head> could be inside <s>.
131	# Now we provisionally solve the problem by inserting a period, i.e. by artificially splitting up one sentence into two.
132	# An alternative would be to accept <cap> and <tb> within <s>, but there is only this single table within any <s>.
133	# Another alternative would be to move the last two lines of the paragraph from p.503 to p.501 (all other tables are between two <p>).
134	s!Al<007>as proportiones eiuſdem!Al<007>as proportiones eiuſdem.!; # line 23791
135	s!<cap>TABVLA PYTHA GORICA.</cap>!<cap>TABVLA PYTHAGORICA.</cap>!; # line 23792: remove a space
136
137	# mid dot after fractions should be normal period
138	s!·!.!g;
139
140	# centered formulas that look like tables or headings, and similar cases
141	s!<tb>16. # 24. # 36. # 54. # 81.</tb>!<math><seq>16, 24, 36, 54, 81</seq></math>!; # line 25489
142	s!<tb>3. # 4. # 12. # 16. # 48. # 64. # 192. # 256.</tb>!<math><seq>3, 4, 12, 16, 48, 64, 192, 256</seq></math>!; # line 25510
143	if (s!<h>([\d\@{/}]+\.( [\d\@{/}]+\.)*( &c\.)?)</h>!<math><seq>$1</seq></math>!) {
144	s!\.!,!g;
145	s!,</seq>!</seq>!g;
146	}
147	s!<tb it>7 # R. q. 63. # 9.</tb>!<math>7 # _R. q._ 63. # 9.</math>!; # line 25749
148
149	# typos that are a problem for the script
150	s!<scG</sc>!<sc>G</sc>!; # line 57290; must be before "obviously missing spaces between words"
151	s!R Q\. Q T,!R Q, Q T,!; # line 61676: would produce an artifact sentence
152
153	# old-style numerals typed as letters
154	# (must be before "obviously missing spaces between words")
155	s!PROPOS !PROPOS. !; # normalize the period
156	s!PROPOS\. I!PROPOS. 1!;
157	s!PROPOS\. II!PROPOS. 11!;
158	s!<mgl aI>5\.!<mgl a>15.!; # line 38201
159	s!<mgl bI>I\.!<mgl b>11.!; # line 38206
160	s!<mgl dI>5\.!<mgl d>15.!; # line 38226
161	# s!(<mg[lr].*?)>II\.!$1>11.!; # this may be wrong!
162
163	# obviously missing spaces between words
164	# (part 1: remove the exceptions)
165	s!peZ!pez!g; # z in italics has often been typed as capital Z (trapeZ...)
166	s!pe Z!pez!; # (by the way: remove some superfluous spaces)
167	s!ClaZomenius!Clazomenius!; # line 4741
168	s!θaμaIoIIoιηπnὴ!θαυματοποιητικὴ!; # line 4673: a single Greek word going very wrong (corrected version without ligature braces)
169	s!oſtend<007>mW!oſtendimus!; # line 20009: badly printed
170	s!lxIx!lxix!; # line 54383: dotless i typed as capital I in roman number
171	s!lxIv!lxiv!; # line 54416: dotless i typed as capital I in roman number
172	s!X CIV!xciv!; # (line 55865: superfluous space in roman number)
173	s!lXXXIX!lxxxix!; # line 55866: small and capital letters in roman number
174	s!paraHelæ!parallelæ!; # line 61171: ll --> H
175	s!kL M!K L M!; # line 63775: typesetter's fault
176	s!eCùm!<n e> Cùm!; # line 64564: the anchor is difficult to identify
177	s!AbCD!ABCD!; # line 68929: typesetter's fault
178	s!CBkHAE!CBKHAE!; # line 71905: typesetter's fault
179	# (part 2: apply the rule)
180	s!([a-zæ])([A-Z])!$1 $2!; # insert space between small letter and capital letter
181
182	s! exdefin\.! ex defin.!g;
183
184	# long s --> f
185	s! deſin\.! defin.!g;
186	s! fint! ſint!g; # I didn't check whether the typesetter or the typists got it wrong; likely the latter
187
188	# running heads
189	# (all 1451 running heads have the form <rh>[A-Z. Æ´:_]+</rh>)
190	s!LIBER I\.´!LIBER I.!;
191	s!LIBER Y!LIBER I!;
192	s!LIBER XII:!LIBER XII.!;
193	s!<rh>_(.+?)_(\.?</rh>)!<rh>$1$2!; # remove italics in the running head (rather pointless information in the running head)
194
195	# italics
196	s!<h>_(.+?)_</h>!<h it>$1</h>!; # replace _ _ by it in one-line headings
197	s!_K_!K!g; # 264 occurrences of K in italics: typesetter's fault
198
199	# notes in italics: _ _ --> it (problem of the Specs 1.1.2)
200	if (m!<mg[lr]!) {
201	$inMgSince = $line;
202	$mgInItalics = 1;
203	}
204	if ($inMgSince) {
205	# examples of lines that count as italics: <mgl>_text_<mgl>; <mgl>_text_.<mgl>; <mgl>12.; <mgl>12. _pri_.</mgl>
206	# does not catch: 2 2. and a few other cases
207	unless (m!(<mg[lr].?>)?(\d+\. )?_.+?_[.-]?(</mg[lr])?! \|\| m!<mg[lr].?>[\d\@]+\.?!) { $mgInItalics = 0; }
208	if (m!</mg[lr]>!) {
209	if ($mgInItalics) {
210	$text[$inMgSince-1] =~ s!<(mg[lr])(.*?)>!<$1 it$2>!;
211	for my $i ($inMgSince-1 .. $line-1) { $text[$i] =~ s!_!!g; } # remove ALL italics markers in these notes
212	}
213	$inMgSince = 0;
214	}
215	}
216
217	# incorrect periods, which would produce artifact sentence tags
218	# (a regex such as "\. [a-z]" would not work well)
219	# line ends
220	s!3 DVABVS datis rectis lineis inæqualibus\.!3 DVABVS datis rectis lineis inæqualibus,!; # badly printed comma
221	s!reliquis trianguli lateribus deſcribuntur\.!reliquis trianguli lateribus deſcribuntur,!; # line 516: badly printed comma
222	s!ex denominatori\.!ex denominatori-!; # line 24587
223	s! Et\.! Et!; # line 2179
224	s! Non\.! Non!; # line 3671
225	s!pun\.!pun-!; # line 16033
226	# others
227	s!4c\.!40.!; # line 33222
228	s!duo\. us!duobus!; # line 33949
229	s!commenſura\.!commenſura-!; # line 54153
230
231	# resolve punctuation clusters (which would cause artifact sentences)
232	# (be careful with & !)
233	s!\.\.!.!;
234
235	# emendations of headings
236	s!<h>SCHOL!<h it>SCHOL!; # lines 6820, 12246, 59058, 66953: I haven't checked the originals; call it normalization
237	s!<h>_SCHOLIVM_!<h it>SCHOLIVM!; # 5 times; not checked
238	s!EXCAMPANO!EX CAMPANO!;
239	s!<h>EX CAMPANO!<h it>EX CAMPANO!; # not checked
240	s!EXPROCLO!EX PROCLO!;
241	s!<h>EX PROCLO!<h it>EX PROCLO!; # not checked
242	s!EXORONTIO!EX ORONTIO.!;
243	s!<h>EX CARDANO!<h it>EX CARDANO!; # not checked
244	s!EX PELET ARIO!EX PELETARIO!;
245	s!<h>EX PELETARIO!<h it>EX PELETARIO!; # not checked
246	s!<h>PRAXIS!<h it>PRAXIS!; # not checked
247	s!<h>EX FEDERICO COMMANDINO!<h it>EX FEDERICO COMMANDINO!; # not checked
248
249	s!PROBLEMA I\.!PROBLEMA 1.!;
250	s!PROPOSITIO I\.!PROPOSITIO 1.!;
251	s!PROPOS\. 1I\.!PROPOS. 11.!;
252	s!PROPOS\. I\.!PROPOS. 1.!;
253
254	s!CONVERSV M!CONVERSVM!;
255	s!COROLLARIVMI!COROLLARIVM I!;
256
257	s!<h>THEOR\. 13\. PROPOS\. 19</h>!<h>THEOR. 13. PROPOS. 19.</h>!; # line 34061
258
259	if ($line == 13631 + 2) { s!SECVNDVM\.!SECVNDVM.</h>!; }
260	if ($line == 13632 + 2) { s!_DEFINITIONES_\.!<h it>DEFINITIONES.</h>!; }
261	if ($line == 13633 + 2) { s!I\.</h>!<h>I.</h>!; }
262
263	if ($line == 16232 + 2) { s!TERTIVM\.!TERTIVM.</h>!; }
264	if ($line == 16233 + 2) { s!_DEFINITIONES\._</h>!<h it>DEFINITIONES.</h>!; }
265
266	if ($line == 21217 + 2) { s!QVARTVM\.!QVARTVM.</h>!; }
267	if ($line == 21218 + 2) { s!_DEFINITIONES_\.!<h it>DEFINITIONES.</h>!; }
268	if ($line == 21219 + 2) { s!I\.</h>!<h>I.</h>!; }
269
270	if ($line == 38501 + 2) { s!SEPTIMVM\.!SEPTIMVM.</h>!; }
271	if ($line == 38502 + 2) { s!_DEFINITIONES\._</h>!<h it>DEFINITIONES.</h>!; }
272	# these changes are tedious in a script; better emend it directly in the txt file!
273
274	s!<h it>_DEFINITIONES_\.</h>!<h it>DEFINITIONES.</h>!;
275	s!<h it>_DEFINITIONES\._</h>!<h it>DEFINITIONES.</h>!;
276
277	if ($line == 63713) { s!<h>THEOR. 16. PROPOS. 16.</h>!<h>THEOR. 16. PROPOS. 18.</h>!; } # silent correction of obvious typo!
278	if ($line == 66659) { s!PROOPS!PROPOS!; } # silent correction of obvious typo in the book
279
280	s!<h>INVENIRE quintam Apotomen\.</h>!<p>INVENIRE quintam Apotomen.</p>!; # line 55545: would create artifact div4
281
282	# superscript marked as subscript (there is no <^> in Clavius)
283	s!<_>e</_>x E\.!ex E!; # line 11142 (p.0250): typesetter's mistake: remove superscript and dot
284	s!2<_>a</_>!<reg orig="2a" type="abbrev">secunda</reg>!; # line 20594 (p.0435)
285
286	# misc. typos I stumbled upon
287	s!I hilo!Philo!; # badly printed P
288	s!Mænum!Mœnum!; # æ --> œ
289	s!mθnſtratione!monſtratione!; # θ --> o
290	s!IV M!IVM!; # superfluous space as in SCHOLIV M, COROLLARIV M
291	s!SCHOLI M!SCHOLIVM!; # missing V
292	s!ZVART!QVART!; # ZV --> QV
293	s!7. quind. Septima libri quindecimi.!7. quind. # Septima libri quindecimi.!; # line 6769: missing # in table
294
295	}
296
297	# raw text --> XML
298
299	# metadata: 30 additional lines
300	print <<'METADATA';
301	<?xml version="1.0" encoding="UTF-8"?>
302	<echo xmlns="http://www.mpiwg-berlin.mpg.de/ns/echo/1.0/"
303	xmlns:dc="http://purl.org/dc/elements/1.1/"
304	xmlns:dcq="http://purl.org/dc/qualifiers/1.0/"
305	xmlns:dct="http://purl.org/dc/terms/1.0/"
306	xmlns:echo="http://www.mpiwg-berlin.mpg.de/ns/echo/1.0/"
307	xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#"
308	xmlns:xhtml="http://www.w3.org/1999/xhtml"
309	xmlns:xlink="http://www.w3.org/1999/xlink"
310	xmlns:xml="http://www.w3.org/XML/1998/namespace">
311	<metadata>
312	<dc:creator>Clavius</dc:creator>
313	<dc:title xml:lang="la">Elementorum Libri XV</dc:title>
314	<dc:date>
315	<rdf:Description>
316	<dcq:dateScheme>ISO 8601</dcq:dateScheme>
317	<rdf:value>1607</rdf:value>
318	</rdf:Description>
319	</dc:date>
320	<dc:language>
321	<rdf:Description>
322	<dcq:languageScheme>ISO 639-2</dcq:languageScheme>
323	<rdf:value>lat</rdf:value>
324	</rdf:Description>
325	</dc:language>
326	<dc:identifier>test.xml</dc:identifier>
327	<dc:rights>open access</dc:rights>
328	<dct:license>http://echo.mpiwg-berlin.mpg.de/policy/oa_basics/declaration</dct:license>
329	</metadata>
330	<text xml:lang="la">
331	METADATA
332
333	$line = 0;
334
335	for (@text) {
336
337	$line++;
338
339
340	# ad hoc tagging of the higher-level div structure
341	# including some additional lowest-level closing div's
342	# note that single lines have been inserted before lines 7 and 303 !
343
344	# frontmatter
345	if ($line == 1) { print '<front><div level="1" type="title" n="1">'; }
346	if ($line == 33) { print '</div><div level="1" type="dedication" n="1">'; } # from here on: offset 1
347	if ($line == 223) { print '</div><div level="1" type="preface" n="1">'; }
348	if ($line == 304) { print '</div><div level="1" type="index" n="1">'; }
349	if ($line == 386) { print '</div><div level="1" type="index" n="2">'; } # from here on: offset 2
350	if ($line == 2087) { print '</div><div level="1" type="index" n="3">'; }
351	if ($line == 4377) { print '</div><div level="1" type="preface" n="2">'; }
352	if ($line == 4505) { print '</div><div level="1" type="dedication" n="2">'; }
353	if ($line == 4563) { print '</div><div level="1" type="introduction" n="1">'; }
354
355	# book 1
356	if ($line == 5225) { print '</div></front><body><div level="1" type="book" n="1"><div level="2" type="booktitle" n="1">'; }
357	if ($line == 5230) { print '</div><div level="2" type="bookhead" n="1">'; $inDiv3 = 0; $inDef = 1; $def = 0; }
358	if ($line == 6226) { print '</div></div>'; $inDiv3 = 0; $inDef = 0; $inPost = 1; $post = 0; }
359	if ($line == 6283) { print '</div></div>'; $inDiv3 = 0; $inPost = 0; $inComm = 1; $comm = 0; }
360	if ($line == 6777) { print '</div></div></div><div level="2" type="bookmain" n="1">'; $inDiv3 = 0; $inComm = 0; $inPropos = 1; $propos = 0; }
361
362	# book 2
363	if ($line == 13629) { print '</div></div></div></div><div level="1" type="book" n="2"><div level="2" type="booktitle" n="1">'; }
364	if ($line == 13634) { print '</div><div level="2" type="bookhead" n="1">'; $inDiv3 = 0; $inDef = 1; $def = 0; }
365	if ($line == 13806) { print '</div></div></div><div level="2" type="bookmain" n="1">'; $inDiv3 = 0; $inDef = 0; $inPropos = 1; $propos = 0; }
366
367	# book 3
368	if ($line == 16230) { print '</div></div></div></div><div level="1" type="book" n="3"><div level="2" type="booktitle" n="1">'; }
369	if ($line == 16235) { print '</div><div level="2" type="bookhead" n="1">'; $inDiv3 = 0; $inDef = 1; $def = 0; }
370	if ($line == 16481) { print '</div></div></div><div level="2" type="bookmain" n="1">'; $inDiv3 = 0; $inDef = 0; $inPropos = 1; $propos = 0; }
371
372	# book 4
373	if ($line == 21215) { print '</div></div></div></div><div level="1" type="book" n="4"><div level="2" type="booktitle" n="1">'; }
374	if ($line == 21220) { print '</div><div level="2" type="bookhead" n="1">'; $inDiv3 = 0; $inDef = 1; $def = 0; }
375	if ($line == 21301) { print '</div></div></div><div level="2" type="bookmain" n="1">'; $inDiv3 = 0; $inDef = 0; $inPropos = 1; $propos = 0; }
376
377	# book 5
378	if ($line == 23058) { print '</div></div></div></div><div level="1" type="book" n="5"><div level="2" type="booktitle" n="1">'; }
379	if ($line == 23063) { print '</div><div level="2" type="bookhead" n="1">'; $inDiv3 = 0; $inDef = 1; $def = 0; }
380	if ($line == 23194) { print '</div><div level="3" type="addendum" n="1">'; $defOnHold = 1; }
381	if ($line == 27240) { $defOnHold = 0; }
382	if ($line == 27611) { print '</div><div level="3" type="addendum" n="2">'; $defOnHold = 1; }
383	if ($line == 28186) { $defOnHold = 0; }
384	if ($line == 28707) { print '</div></div></div><div level="2" type="bookmain" n="1">'; $inDiv3 = 0; $inDef = 0; $inPropos = 1; $propos = 0; }
385
386	# book 6
387	if ($line == 31048) { print '</div></div></div></div><div level="1" type="book" n="6"><div level="2" type="booktitle" n="1">'; }
388	if ($line == 31053) { print '</div><div level="2" type="bookhead" n="1">'; $inDiv3 = 0; $inDef = 1; $def = 0; }
389	if ($line == 31640) { print '</div></div></div><div level="2" type="bookmain" n="1">'; $inDiv3 = 0; $inDef = 0; $inPropos = 1; $propos = 0; }
390
391	# second book title
392	if ($line == 38391) { print '</div></div></div></div><div level="1" type="index" n="1">'; $inPropos = 0; $inDiv3 = 0; } # quick'n'dirty !!!
393	if ($line == 38480) { print '</div><div level="1" type="title" n="1">'; }
394
395	# book 7
396	if ($line == 38499) { print '</div><div level="1" type="book" n="7"><div level="2" type="booktitle" n="1">'; }
397	if ($line == 38504) { print '</div><div level="2" type="bookhead" n="1">'; $inDiv3 = 0; $inDef = 1; $def = 0; }
398	if ($line == 39239) { print '</div></div>'; $inDiv3 = 0; $inDef = 0; $inPost = 1; $post = 0; }
399	if ($line == 39250) { print '</div></div>'; $inDiv3 = 0; $inPost = 0; $inComm = 1; $comm = 0; }
400	if ($line == 39355) { print '</div></div></div><div level="2" type="bookmain" n="1">'; $inDiv3 = 0; $inComm = 0; $inPropos = 1; $propos = 0; }
401
402	# book 8
403	if ($line == 42003) { print '</div></div></div></div><div level="1" type="book" n="8"><div level="2" type="booktitle" n="1">'; }
404	if ($line == 42008) { print '</div><div level="2" type="bookmain" n="1">'; $inDiv3 = 0; $inPropos = 1; $propos = 0; }
405
406	# book 9
407	if ($line == 44322) { print '</div></div></div></div><div level="1" type="book" n="9"><div level="2" type="booktitle" n="1">'; }
408	if ($line == 44327) { print '</div><div level="2" type="bookmain" n="1">'; $inDiv3 = 0; $inPropos = 1; $propos = 0; }
409
410	# book 10
411	if ($line == 48379) { print '</div></div></div></div><div level="1" type="book" n="10"><div level="2" type="booktitle" n="1">'; }
412	if ($line == 48384) { print '</div><div level="2" type="bookhead" n="1">'; $inDiv3 = 0; $inDef = 1; $def = 0; }
413	if ($line == 48638) { print '</div>'; $inDef = 0; }
414	if ($line == 48639) { print '<div type="postulatum" level="3" n="1">'; } # a single postulate, without number "I."
415	if ($line == 48650) { print '</div></div>'; $inDiv3 = 0; $inComm = 1; $comm = 0; }
416	if ($line == 48677) { print '</div></div></div><div level="2" type="bookmain" n="1">'; $inDiv3 = 0; $inComm = 0; $inPropos = 1; $propos = 0; }
417
418	if ($line == 52978) { print '</div></div></div><div level="2" type="bookhead" n="2">'; $inDiv3 = 0; $inPropos = 0; $inDef = 1; $def = 0; }
419	if ($line == 53045) { print '</div></div></div><div level="2" type="bookmain" n="2">'; $inDiv3 = 0; $inDef = 0; $inPropos = 1; $propos = 0; }
420	if ($line == 55353) { print '</div></div></div><div level="2" type="bookhead" n="3">'; $inDiv3 = 0; $inPropos = 0; $inDef = 1; $def = 0; }
421	if ($line == 55386) { print '</div></div></div><div level="2" type="bookmain" n="3">'; $inDiv3 = 0; $inDef = 0; $inPropos = 1; $propos = 0; }
422
423	# book 11
424	if ($line == 57419) { print '</div></div></div></div><div level="1" type="book" n="11"><div level="2" type="booktitle" n="1">'; }
425	if ($line == 57425) { print '</div><div level="2" type="bookhead" n="1">'; $inDiv3 = 0; $inDef = 1; $def = 0; }
426	if ($line == 58209) { print '</div></div></div><div level="2" type="bookmain" n="1">'; $inDiv3 = 0; $inDef = 0; $inPropos = 1; $propos = 0; }
427
428	# book 12
429	if ($line == 61352) { print '</div></div></div></div><div level="1" type="book" n="12"><div level="2" type="booktitle" n="1">'; }
430	if ($line == 61358) { print '</div><div level="2" type="bookmain" n="1">'; $inDiv3 = 0; $inPropos = 1; $propos = 0; }
431
432	# book 13
433	if ($line == 63790) { print '</div></div></div></div><div level="1" type="book" n="13"><div level="2" type="booktitle" n="1">'; }
434	if ($line == 63796) { print '</div><div level="2" type="bookmain" n="1">'; $inDiv3 = 0; $inPropos = 1; $propos = 0; }
435
436	# book 14
437	if ($line == 66389) { print '</div></div></div></div><div level="1" type="book" n="14"><div level="2" type="booktitle" n="1">'; }
438	if ($line == 66398) { print '<div level="2" type="prooemium" n="1">'; }
439	if ($line == 66461) { print '</div></div><div level="2" type="bookmain" n="1">'; $inDiv3 = 0; $inDef = 0; $inPropos = 1; $propos = 0; }
440
441	# book 15
442	if ($line == 68896) { print '</div></div></div></div><div level="1" type="book" n="15"><div level="2" type="booktitle" n="1">'; $inDiv3 = 0; } # $inDiv3 = 0: quick'n'dirty solution !!!
443	if ($line == 68905) { print '</div><div level="2" type="bookmain" n="1">'; $inDiv3 = 0; $inPropos = 1; $propos = 0; }
444
445	# book 16
446	if ($line == 70684) { print '</div></div></div></div><div level="1" type="book" n="16"><div level="2" type="booktitle" n="1">'; }
447	if ($line == 70705) { print '</div><div level="2" type="bookmain" n="1">'; $inDiv3 = 0; $inPropos = 1; $propos = 0; }
448
449	if ($line == 72170) { print '</div></div></div><div level="2" type="addendum" n="1">'; $inPropos = 0; $inDiv3 = 0; } # $inDiv3 = 0: ignore the three addenda
450	if ($line == 73209) { print '</div><div level="2" type="addendum" n="2">'; }
451	if ($line == 73378) { print '</div><div level="2" type="addendum" n="3">'; }
452
453
454	# resolve common abbreviations
455	# (problem of &c.: The period may or may not mark the end of a sentence. Heuristics will be: etc. plus capital letter --> <s>)
456
457	s!([a-z])propoſ\.!$1 propoſ.!g; # 24 occurrences of propo$. without preceding space, all checked (all other: space, _, beginning of line)
458	s!([Aa]d) (propoſ)\.!$1 <reg orig="MURP" type="abbrev">$2itionem</reg>!g; # ad propo$itionem
459	s!([Pp]ropoſ)\.!<reg orig="$1." type="abbrev">$1itio</reg>!g; # propo$itio (may be the wrong casus in some cases, e.g. ex propo$.)
460	s!MURP!propoſ.!g;
461	# does not catch e.g. pro-_ // poſ.
462
463	s!([a-z])coroll\.!$1 coroll.!g; # 15 occurrences of coroll. without preceding space, all checked
464	s!([Cc]oroll)\.!<reg orig="$1." type="abbrev">$1arium</reg>!g; # (may sometimes be the wrong casus, e.g. ex coroll.)
465
466	s!lib\.!<reg orig="lib." type="abbrev">liber</reg>!g; # liber (does that make sense? It may be e.g. "libro".)
467	# missing: defin. def. cap. and others (see below)
468
469	# resolve ligatures
470	s!\\’q;!{que}!g; # -que ligature with preceding \’
471	s!q;!{que}!g; # -que ligature without preceding \’
472	s!{(.+?)}!$1<\!-- {$1} -->!g;
473
474	# unknown characters
475	s!<007>!i<\!-- 007 -->!g; # <007> (dotless i) is resolved as i<!-- 007 -->
476	s!<(\d\d\d)>!<unknown code=\"$1\"/>!g;
477	s!<\?>!<unsure/>!g;
478	s!<gap>!<gap/>!g;
479
480	# figures
481	# (must be before the tagging of div's to avoid </div></fig>)
482	# (the internal ordering is important for successive figures)
483	s!<fig>!<figure/>!g;
484
485	# # more elaborate version for texts which have been sent with DESpecs 2 or higher
486	# if ($inFig) {
487	# if (m!<cap>! \|\| m!<desc>! \|\| m!<var>!) {
488	# print "$_\n";
489	# next;
490	# }
491	# print "</figure>";
492	# $inFig = 0;
493	# }
494	# if (s!<fig>!<figure>!) {
495	# $inFig = 1;
496	# print "$_\n";
497	# next;
498	# }
499
500
501	# handwritten material
502	s!<hd>!<hd/>!;
503
504
505	# tables
506	# (must be before paragraphs)
507	# many tables are preceded by <cap> </cap>, which can be one or two lines long; we ignore <cap> for the moment
508
509	if (m!</?cap>!) { # all captions are one or two lines long; this code would break with a longer caption!
510	print "$_\n";
511	next;
512	}
513
514	if (s!<tb it>!<tb>!) { $italicsTb = 1; }
515	if (m!<tb>!) { $inTb = 1; }
516
517	if ($inTb) {
518	unless (s!<(/?)tb>!<$1xhtml:table>!) {
519	$_ = "<xhtml:tr><xhtml:td>$_</xhtml:td></xhtml:tr>"; # this breaks if the table consists of only one line!
520	s!#!</xhtml:td><xhtml:td>!g;
521	}
522	if ($italicsTb) { s!<xhtml:table>!<xhtml:table style="italics">!; }
523	print "$_\n";
524	if (m!</xhtml:table>!) { $inTb = 0; $italicsTb = 0; }
525	next;
526	}
527
528
529	# mathematical formulas
530
531	if (m!<math>!) { $inMath = 1; }
532
533	if ($inMath) {
534	print "$_\n";
535	if (m!</math>!) { $inMath = 0; }
536	next;
537	}
538
539
540	# page breaks
541
542	if (m!<pb!) {
543	$pb++;
544
545	my $pagenumber = (m!<pb (.+?)>! ? " o=\"$1\"" : "" );
546	my $rhead = (m!<rh>(.+?)</rh>! ? " rhead=\"$1\"" : "" );
547	my $link = 'http://echo.mpiwg-berlin.mpg.de/ECHOdocuView/ECHOzogiLib?'.
548	'url=/mpiwg/online/permanent/library/2QTVUHDT/pageimg&mode=imagepath&pn='.$pb;
549
550	$_ = "<pb$pagenumber$rhead n=\"$pb\" xlink:href='$link'/>";
551
552	$pbTag = "";
553	if (($inDef) && (!$defOnHold)) { if ( $text[$line] =~ m!<h>[XVI]+\.</h>!) { $pbTag = $_; }}
554	if ($inPost) { if ( $text[$line] =~ m!<h>[XVI]+\.</h>!) { $pbTag = $_; }}
555	if ($inComm) { if ( $text[$line] =~ m!<h>[XVI]+\.</h>!) { $pbTag = $_; }}
556	if ($inPropos) { if ( $text[$line] =~ m!<h>.+? PROPOS\. \d+\.</h>!) { $pbTag = $_; }}
557
558	print "$_" unless $pbTag ne "";
559	print "\n";
560	next;
561	}
562
563
564	# anchors
565	s!<n (.+?)>!<anchor symbol="$1" anchorLocation="given"/>!g; # <n> never occours without anchor
566
567
568	# marginal notes
569	# (must be before paragraphs)
570	if (s!(<mg[lr]) it!$1!) { $italicsMg = 1; }
571	if (s!<mgl>!<note place="left">!) { $inMg = 1; }
572	if (s!<mgr>!<note place="right">!) { $inMg = 1; }
573	if (s!<mgl (.+?)>!<note place="left" anchor="$1">!) { $inMg = 1; }
574	if (s!<mgr (.+?)>!<note place="right" anchor="$1">!) { $inMg = 1; }
575	if ($italicsMg) {
576	s!(<note.*?)>!$1 style="italics">!;
577	$italicsMg = 0;
578	}
579	if ($inMg) {
580	if (s!</mg[lr]>!</note>!) { $inMg = 0; } else { $_ .= "<lb/>"; }
581	print "$_\n";
582	next;
583	}
584
585
586	sub resetDiv4 {
587	$div4 = "";
588	$divTypeCount{"comment"} = 0;
589	$divTypeCount{"euclid"} = 0;
590	$divTypeCount{"euclidComment"} = 0;
591	$divTypeCount{"pseudoEuclid"} = 0;
592	$divTypeCount{"pseudoEuclidComment"} = 0;
593	$divTypeCount{"corollarium"} = 0;
594	$divTypeCount{"corollariumComment"} = 0;
595	}
596
597	sub addDiv4 {
598	if ($div4) { print '</div>'; }
599	$div4 = shift;
600	$divTypeCount{$div4}++;
601	print '<div type="'.$div4.'" level="4" n="'.$divTypeCount{$div4}.'">';
602	}
603
604	sub checkForNewDiv4 {
605
606	return unless $inDiv3;
607	return unless $line >= $doNotCheckBeforeLine;
608	return unless (m!<head>! \|\| m!<p>!);
609
610	# corollarium
611	if (m!<head>COR!) { # <head>COR always triggers a new div4
612	addDiv4("corollarium");
613	return;
614	}
615
616	# headings: Always start a new div4, but ignore upright/italics in the heading and take upright/italics from the next paragraph instead
617	if (m!<head>!) {
618	my $nextP = $line;
619	while (!($text[$nextP] =~ m!<p[ >]!)) {
620	return if m!<pb!; # do not look further than the next page break: don't start a new div4 in this (hopefully rare) case
621	$nextP++;
622	}
623	$doNotCheckBeforeLine = $nextP;
624
625	# upright: always pseudo-Euclid
626	if ($text[$nextP] =~ m!<p>!) {
627	addDiv4("pseudoEuclid");
628	return;
629	}
630	# italics: either a comment div on the preceding upright div, or the same comment div type as before
631	if ($div4 =~ m![cC]omment!) {
632	addDiv4($div4);
633	return;
634	}
635	addDiv4($div4."Comment"); # we assume that there is no "comment", i.e. a comment without preceding upright p
636	return;
637	}
638
639	# paragraph in italics
640	if ($italicsP) {
641	if ($div4 eq "") {
642	addDiv4("comment"); # this does not exist in Clavius
643	return;
644	}
645	return if $div4 =~ m![cC]omment!; # return if we are already in a comment div
646
647	addDiv4($div4."Comment");
648	return;
649	}
650
651	# euclid and pseudo-euclid
652	return if $div4 =~ m![eE]uclid$!; # return if we are already in a euclid or pseudoEuclid div
653	return if $div4 =~ m!corollarium$!; # return if we are already in a corollarium div
654
655	# the first euclid part
656	if ($div4 eq "") {
657	addDiv4("euclid");
658	return;
659	}
660
661	# every euclid part apart from the first is called pseudo-euclid
662	addDiv4("pseudoEuclid");
663	return;
664	}
665
666
667	# headings
668
669	if (s!<h it>!<h>!) { $italicsH = 1; }
670	if (s!<h>!<head>!) { $inH = 1; }
671
672	if ($inH) {
673
674	s!</h>!</head>!;
675
676	# headings at level 3
677	if (($inDef) && (!$defOnHold)) {
678	if (m!<head>[XVI]+\.</head>!) {
679	$def++;
680	if ($inDiv3) { print '</div></div>'; }
681	print '<div type="definitio" level="3" n="'.$def.'">';
682	if ($pbTag) { print $pbTag; $pbTag = ""; }
683	$inDiv3 = 1;
684	resetDiv4;
685	$doNotCheckBeforeLine = $line + 1;
686	}
687	}
688	if ($inPost) {
689	if (m!<head>[XVI]+\.</head>!) {
690	$post++;
691	if ($inDiv3) { print '</div></div>'; }
692	print '<div type="postulatum" level="3" n="'.$post.'">';
693	if ($pbTag) { print $pbTag; $pbTag = ""; }
694	$inDiv3 = 1;
695	resetDiv4;
696	$doNotCheckBeforeLine = $line + 1;
697	}
698	}
699	if ($inComm) {
700	if (m!<head>[XVI]+\.</head>!) {
701	$comm++;
702	if ($inDiv3) { print '</div></div>'; }
703	print '<div type="axioma" level="3" n="'.$comm.'">';
704	if ($pbTag) { print $pbTag; $pbTag = ""; }
705	$inDiv3 = 1;
706	resetDiv4;
707	$doNotCheckBeforeLine = $line + 1;
708	}
709	}
710	if ($inPropos) {
711	if ((m!<head>PROBLEMA 1\.</head>!) \|\| (m!<head>.+? PROPOS\. \d+\.</head>!)) {
712	$propos++;
713	if ($inDiv3) { print '</div></div>'; }
714	print '<div type="propositio" level="3" n="'.$propos.'">';
715	if ($pbTag) { print $pbTag; $pbTag = ""; }
716	$inDiv3 = 1;
717	resetDiv4;
718	$doNotCheckBeforeLine = $line + 1;
719	if (m!<head>PROBLEMA 1\.</head>!) { $doNotCheckBeforeLine++; } # allow for the following PROPOSITIO 1. line
720	}
721	}
722
723	checkForNewDiv4;
724
725	# end each line with <lb/>
726	$_ .= "<lb/>";
727	s!</head><lb/>!</head>!;
728
729	# italics
730	if ($italicsH) { s!<head>!<head style="italics">! }
731
732	# print all headings
733	print "$_\n";
734
735	if (m!</head>!) { $inH = 0; $italicsH = 0; }
736	next;
737	}
738
739
740	# paragraphs
741
742	if (s!<p it>!<p>!) { $italicsP = 1; }
743	if (m!<p>!) { $inP = 1; }
744
745	if ($inP) {
746
747	checkForNewDiv4;
748
749	# normalize the periods: there should always be a period before </p>
750	s!([^.])</p>!$1\.</p>!g;
751
752
753	# tagging sentences
754	# (A period . may or may not indicate a new sentence; see "handle exceptions".)
755	# (A colon : always indicates a new sentence.)
756	# (A semicolon ; indicates a new sentence, probably unless there seems to be a variable before it.)
757
758	# - handle exceptions
759	# period
760	s!\."!MURP"!g; # avoid artifact sentences in xml attributes, especially resolved abbreviations
761	s!(\d+)\.!$1MURP!g; # assumption: no new sentence after an arabic number
762	s!([a-zA-Z])\. (\d+)!$1MURP $2!g; # assumption: no new sentence before an arabic number
763	s!}\.!}MURP!g; # no new sentence after fractions
764	s!<p>([IVXLC]+)\.!<p>$1MURP!; # roman number at the beginning of the paragraph
765
766	s! defin\.! definMURP!g; # some ad hoc MURPing; better properly resolve these abbreviations?
767	s! def\.! defMURP!g;
768	s! cap\.! capMURP!g;
769	s! Rep\.! RepMURP!g;
770	s! quib\.! quibMURP!g;
771	s! duob\.! duobMURP!g;
772	s! proportionib.! proportionibMURP!g;
773	s! verticib.! verticibMURP!g;
774	s! ſchol\.! ſcholMURP!g;
775	s! Regiom\.! RegiomMURP!g; # Regiomontanus
776
777	s!li<\!-- 007 -->b\.!li<\!-- 007 -->bMURP!g;
778	s! ([A-Z])\. ([^A-Z])! $1MURP $2!g; # catches, among others, " R. q."
779	if (m! [A-Z]\.$!) {
780	if ($text[$line] =~ m!^[^A-Z]!) { s!( [A-Z])\.$!$1MURP!; } # same rule at line end
781	}
782
783	# semicolon
784	s!& c\.!&c.!g; # normalize &c.
785	s!&c\. ([^A-Z])!&cMURP $1!g; # problem of &c., see above.
786	if (m!&c\.$!) { # &c. at line end
787	if ($text[$line] =~ m!^[^A-Z]!) { s!&c\.$!&cMURP!; }
788	}
789	# (This rule will inevitably produce some incorrect tags, e.g. "&c. _Ad" in lines 2802, 3414.)
790
791	s!q;!qHUBA!g; # -que ligature
792	# s!( [a-zA-Z]);!$1HUBA!g; # variables before the ;
793	s!( [a-zA-Z]); ([a-zA-Z] )!$1HUBA $2!g; # semicolon between two apparent variables
794	if (m! [a-zA-Z];$!) {
795	if ($text[$line] =~ m!^[a-zA-Z] !) { s!( [a-zA-Z]);!$1HUBA!; } # same rule at line end
796	}
797	s!&!&ampHUBA!g; # &
798
799
800	# - other preparations
801	s!([.:;])</sc>!</sc>$1!g;
802
803
804	# - main
805	s!<p>!<p><s>!;
806	if ($lineStartsWithS) { $_ = "<s>$_"; $lineStartsWithS = 0; }
807	s![.:;]!$&</s><s>!g;
808	if (s!<s>$!!) { $lineStartsWithS = 1; }
809
810	# - corrections
811	s!([.:;])</s><s>_!$1_</s><s>!g; # (the underscore _ should be made into italics!)
812
813	# - change back non-sentence-ending punctuation marks
814	s!MURP!.!g; # period
815	s!HUBA!;!g; # semicolon
816
817	# - final corrections
818	s!([.:;])</p>!$1</s></p>!;
819	s!<s></p>!</p>!;
820	# s!<s> !<s>!g;
821
822	# mark italics
823	# (must be after the sentence tagging)
824	if ($italicsP) { s!<p([ >])!<p style="italics"$1!; } # however, in Clavius there are no attributes other than it
825
826	# end each line with <lb/>
827	$_ .= "<lb/>" unless (($line < $#text) && (($text[$line] =~ m!<tb>!) \|\| ($text[$line] =~ m!<pb!)));
828	s!</p><lb/>!</p>!;
829	s!<s><lb/>!<lb/><s>!g;
830
831	# corrections after a first discussion
832	s!</s><s>\)!)</s><s>!g;
833	s!</s><s> ! </s><s>!g; # beachte Zeilenumbruch??
834
835	print "$_\n";
836
837	if (m!</p>!) { $inP = 0; $italicsP = 0; }
838	next;
839	}
840
841
842	# everything else
843
844	print "$_\n";
845	}
846
847
848	print "</div></div></body></text></echo>\n";

Download in other formats:

Original Format