forked from johnjung/ocr_converters
-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathbuild_ia_bookreader_ocr.py
523 lines (438 loc) · 18.6 KB
/
build_ia_bookreader_ocr.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
#!/usr/bin/env python3
"""Usage:
build_ia_bookreader_ocr <identifier> <min-year> <max-year> [<shrink_to_height>]
"""
import os, requests, sqlite3, sys
import xml.etree.ElementTree as ElementTree
from PIL import Image
from docopt import docopt
def get_number_of_images(ark):
noid = ark.replace('ark:61001/', '').replace('ark:/61001/', '')
p = '/data/digital_collections_ocfl/ark_data/' + '/'.join([noid[i:i+2] for i in range(0, len(noid), 2)])
count = 0
for f in os.listdir(p + '/v1/content'):
if f.startswith('0000'):
count += 1
return count
def list_mvol_files(local_root, identifier, subdir):
'''Get a list of complete paths to mvol files in ALTO, JPEG, POS or TIFF
directories
'''
subdir_path = '{}{}{}{}{}'.format(
local_root,
os.sep,
identifier.replace('-', os.sep),
os.sep,
subdir
)
files = []
for f in os.listdir(subdir_path):
files.append('{}{}{}'.format(subdir_path, os.sep, f))
return sorted(files)
class OCRBuilder():
def __init__(self, file_dict, min_year, max_year, shrink_to_height):
if not all([k in file_dict for k in ('dc', 'ocr_files', 'tifs', 'txt')]):
raise KeyError
self.file_dict = file_dict
self.min_year = min_year
self.max_year = max_year
try:
self.shrink_to_height = int(shrink_to_height)
except TypeError:
self.shrink_to_height = None
self.dc = ElementTree.fromstring(requests.get(self.file_dict['dc']).content)
ElementTree.register_namespace('xtf', 'http://cdlib.org/xtf')
def get_tif_size(self, i):
''' Get the size of a tiff from the self.file_dict['tifs'] list. Note i starts at 0.
'''
ocr = ElementTree.fromstring(requests.get(self.file_dict['ocr_files'][i]).content)
p = ocr.find('./{http://www.loc.gov/standards/alto/ns-v2#}Layout/{http://www.loc.gov/standards/alto/ns-v2#}Page')
return (p.attrib['WIDTH'], p.attrib['HEIGHT'])
def get_jpg_tif_ratio(self, i):
''' Get the amount that jpegs "shrunk" compared to tiffs, so we
can rescale OCR data.
'''
if self.shrink_to_height == None:
return 1
else:
return float(self.shrink_to_height) / float(self.get_tif_size(i)[1])
def get_decade(self):
return "%s0s" % self.get_year()[0:3]
def get_publication_type(self):
return 'student' if self.dc.find('{http://purl.org/dc/elements/1.1/}title').text in ('Cap and Gown', 'Daily Maroon') else 'university'
def get_volume_number(self):
return int(self.dc.find('{http://purl.org/dc/elements/1.1/}identifier').text.split('-')[2].lstrip('0'))
def get_year(self):
''' Get the year from a date string embedded in DC data- e.g.
2020-04-22 to 2020.
'''
return self.dc.find('{http://purl.org/dc/elements/1.1/}date').text.split('/')[0].split('-')[0]
def get_human_readable_date(self):
months = {
'01': 'January',
'02': 'February',
'03': 'March',
'04': 'April',
'05': 'May',
'06': 'June',
'07': 'July',
'08': 'August',
'09': 'September',
'10': 'October',
'11': 'November',
'12': 'December'
}
d = self.dc.find('{http://purl.org/dc/elements/1.1/}date').text.split('/')[0].split('-')
if len(d) == 1: # year
return d[0]
elif len(d) == 2: # month year
return '%s %s' % (months[d[1]], d[0])
else:
return '%s %d, %s' % (months[d[1]], int(d[2]), d[0])
def get_meta(self):
meta = ElementTree.Element('{http://cdlib.org/xtf}meta')
ElementTree.SubElement(meta, 'facet-sidebartitle', attrib={
'{http://cdlib.org/xtf}meta': 'true',
'{http://cdlib.org/xtf}facet': 'yes',
'{http://cdlib.org/xtf}tokenize': 'no'
}).text = '%s::Volume %02d' % (self.dc.find('{http://purl.org/dc/elements/1.1/}title').text, self.get_volume_number())
ElementTree.SubElement(meta, 'facet-title', attrib={
'{http://cdlib.org/xtf}meta': 'true',
'{http://cdlib.org/xtf}facet': 'yes',
'{http://cdlib.org/xtf}tokenize': 'no'
}).text = '%s::Volume %02d' % (self.dc.find('{http://purl.org/dc/elements/1.1/}title').text, self.get_volume_number())
ElementTree.SubElement(meta, 'browse-title', attrib={
'{http://cdlib.org/xtf}meta': 'true',
'{http://cdlib.org/xtf}facet': 'no',
'{http://cdlib.org/xtf}tokenize': 'yes'
}).text = '%s Volume %02d' % (self.dc.find('{http://purl.org/dc/elements/1.1/}title').text, self.get_volume_number())
ElementTree.SubElement(meta, 'facet-date', attrib={
'{http://cdlib.org/xtf}meta': 'true',
'{http://cdlib.org/xtf}facet': 'yes',
'{http://cdlib.org/xtf}tokenize': 'no'
}).text = '%s::%s' % (self.get_decade(), self.get_year())
ElementTree.SubElement(meta, 'browse-date', attrib={
'{http://cdlib.org/xtf}meta': 'true',
'{http://cdlib.org/xtf}facet': 'no',
'{http://cdlib.org/xtf}tokenize': 'yes'
}).text = '%s %s' % (self.get_decade(), self.get_year())
ElementTree.SubElement(meta, 'range-date', attrib={
'{http://cdlib.org/xtf}meta': 'true',
'{http://cdlib.org/xtf}facet': 'no',
'{http://cdlib.org/xtf}tokenize': 'no'
}).text = self.get_year()
ElementTree.SubElement(meta, 'facet-category', attrib={
'{http://cdlib.org/xtf}meta': 'true',
'{http://cdlib.org/xtf}facet': 'yes',
'{http://cdlib.org/xtf}tokenize': 'no'
}).text = '%s::%s' % (self.get_publication_type(), self.dc.find('{http://purl.org/dc/elements/1.1/}title').text)
ElementTree.SubElement(meta, 'browse-category', attrib={
'{http://cdlib.org/xtf}meta': 'true',
'{http://cdlib.org/xtf}facet': 'no',
'{http://cdlib.org/xtf}tokenize': 'yes'
}).text = self.get_publication_type()
ElementTree.SubElement(meta, 'sort-identifier', attrib={
'{http://cdlib.org/xtf}meta': 'true',
'{http://cdlib.org/xtf}facet': 'no',
'{http://cdlib.org/xtf}tokenize': 'no'
}).text = self.dc.find('{http://purl.org/dc/elements/1.1/}identifier').text
ElementTree.SubElement(meta, 'display-title', attrib={
'{http://cdlib.org/xtf}meta': 'true',
'{http://cdlib.org/xtf}facet': 'no',
'{http://cdlib.org/xtf}tokenize': 'no'
}).text = '%s (%s)' % (self.dc.find('{http://purl.org/dc/elements/1.1/}title').text, '%s-%s' % (self.min_year, self.max_year))
ElementTree.SubElement(meta, 'display-item', attrib={
'{http://cdlib.org/xtf}meta': 'true',
'{http://cdlib.org/xtf}facet': 'no',
'{http://cdlib.org/xtf}tokenize': 'no'
}).text = 'Volume %s (%s)' % (self.get_volume_number(), self.get_year())
try:
ElementTree.SubElement(meta, 'browse-description', attrib={
'{http://cdlib.org/xtf}meta': 'true',
'{http://cdlib.org/xtf}facet': 'no',
'{http://cdlib.org/xtf}tokenize': 'yes'
}).text = self.dc.find('{http://purl.org/dc/elements/1.1/}description').text
except AttributeError:
pass
ElementTree.SubElement(meta, 'year', attrib={
'{http://cdlib.org/xtf}meta': 'true',
'{http://cdlib.org/xtf}facet': 'no',
'{http://cdlib.org/xtf}tokenize': 'no'
}).text = self.get_year()
ElementTree.SubElement(meta, 'facet-volume', attrib={
'{http://cdlib.org/xtf}meta': 'true',
'{http://cdlib.org/xtf}facet': 'no',
'{http://cdlib.org/xtf}tokenize': 'no'
}).text = 'Volume %d' % self.get_volume_number()
ElementTree.SubElement(meta, 'human-readable-date', attrib={
'{http://cdlib.org/xtf}meta': 'true',
'{http://cdlib.org/xtf}facet': 'no',
'{http://cdlib.org/xtf}tokenize': 'no'
}).text = self.get_human_readable_date()
return meta
def get_position_data_from_pos(self, data_str, scale):
'''
Note: in PHP I used mb_convert_encoding($fields[4], "UTF-8", "ISO-8859-1") on 'text'
'''
output = []
for line in data_str.split('\n'):
fields = line.split('\t')
try:
output.append({
'x': int(float(fields[0]) * scale),
'y': int(float(fields[1]) * scale),
'w': int(float(fields[2]) * scale),
'h': int(float(fields[3]) * scale),
'text': fields[4]
})
except ValueError:
continue
return output
def get_position_data_from_alto(self, xml, scale):
'''
Note: in PHP I used mb_convert_encoding($fields[4], "UTF-8", "ISO-8859-1") on 'text'
'''
fields = []
strings = xml.findall('.//{http://www.loc.gov/standards/alto/ns-v2#}String') or \
xml.findall('.//{http://www.loc.gov/standards/alto/ns-v3#}String')
for string in strings:
fields.append({
'x': int(float(string.get('HPOS')) * scale),
'y': int(float(string.get('VPOS')) * scale),
'w': int(float(string.get('WIDTH')) * scale),
'h': int(float(string.get('HEIGHT')) * scale),
'text': string.get('CONTENT')
})
return fields
def get_position_data(self, i):
scale = self.get_jpg_tif_ratio(i)
data_str = requests.get(self.file_dict['ocr_files'][i]).content
# handle UnicodeDecodeError?
try:
xml = ElementTree.fromstring(data_str)
except ElementTree.ParseError:
return self.get_position_data_from_pos(data_str, scale)
return self.get_position_data_from_alto(xml, scale)
def get_structural_dict(self):
data_str = requests.get(self.file_dict['txt']).content
output = {}
for line in data_str.split('\n'):
fields = line.split('\t')
if len(fields) < 2:
continue
if fields[0][0] == '0':
output[fields[0]] = fields[1]
return output
def newline_between_words(self, word1, word2):
''' The words are associative arrays with x, y, w, and h parameters.
Check to see if there should be a newline between them.
If the left edge of word2 is to the left of the right edge of
word1, there should be a newline between these words.
If any part of word2 is above word1, there should be a
newline between the words.
+--------+ +--------+
| WORD 1 | | WORD 2 | = false
+--------+ +--------+
+--------+ +--------+
| | | WORD 1 |
+--------+ +--------+
= true
+--------+ +--------+
| WORD 2 | | |
+--------+ +--------+
'''
if word2['x'] < (word1['x'] + word1['w']):
return True
if word2['y'] > (word1['y'] + word1['h']):
return True;
return False;
def pos_data_line_groups(self, i):
pos_data = self.get_position_data(i)
lines = []
line = []
while True:
# When posData is empty we're done.
if len(pos_data) == 0:
# If there is a line to add, add it.
if len(line) > 0:
lines.append(line)
break
# If the current line is empty, get an element from pos_data.
if len(line) == 0:
line.append(pos_data.pop(0))
continue
# If the next word is on the same line as the previous word... */
if not self.newline_between_words(line[-1], pos_data[0]):
line.append(pos_data.pop(0))
else:
lines.append(line)
line = []
return lines
def leaf_data_from_pos_data(self, i):
pos_data_line_groups = self.pos_data_line_groups(i)
records = []
for pos_data_lines in pos_data_line_groups:
rs = []
bs = []
for pos_data_line in pos_data_lines:
rs.append(pos_data_line['x'] + pos_data_line['w'])
bs.append(pos_data_line['y'] + pos_data_line['h'])
records.append({
't': min([line['y'] for line in pos_data_lines]),
'b': max(bs),
'l': min([line['x'] for line in pos_data_lines]),
'r': max(rs),
'words': self.line_data_from_pos_data(pos_data_lines)
})
return records
def line_data_from_pos_data(self, pos_data_lines):
words = []
for pos_data_line in pos_data_lines:
words.append({
'l': pos_data_line['x'],
'r': pos_data_line['x'] + pos_data_line['w'],
't': pos_data_line['y'],
'b': pos_data_line['y'] + pos_data_line['h'],
'text': pos_data_line['text']
})
return words
def scale_leaf_data(self, leaf_data, scale):
l = 0
while l < len(leaf_data):
for k in ('l', 't', 'r', 'b'):
leaf_data[l][k] = (int)(leaf_data[l][k] * scale)
w = 0
while w < len(leaf_data[l]['words']):
for k in ('l', 't', 'r', 'b'):
leaf_data[l]['words'][w][k] = (int)(leaf_data[l]['words'][w][k] * scale)
w = w + 1
l = l + 1
return leaf_data
def get_line_spacing(self, leaf_data, l):
spacing = []
w = 0
while w < len(leaf_data[l]['words']):
word_width = leaf_data[l]['words'][w]['r'] - leaf_data[l]['words'][w]['l']
spacing.append(word_width)
if w < len(leaf_data[l]['words']) - 1:
space_width = leaf_data[l]['words'][w + 1]['l'] - leaf_data[l]['words'][w]['r']
spacing.append(space_width)
w = w + 1
return spacing
def get_spacing_coords(self, spacing, w):
offset_l = 0
s = 0
while s < w * 2:
offset_l = offset_l + spacing[s]
s = s + 1
offset_r = 0
s = 0
while s <= w * 2:
offset_r = offset_r + spacing[s]
s = s + 1
return (offset_l, offset_r)
def get_line_text(self, line):
if 'words' in line:
return ' '.join([word['text'] for word in line['words']])
else:
return ''
def spacing_to_string(self, spacing):
return ' '.join([str(space) for space in spacing])
def get_leaf(self, n):
leaf_num = '%08d' % (n + 1)
human_readable_leaf_num = str(n + 1)
if self.shrink_to_height == None:
jpg_size = self.get_tif_size(n)
else:
s = self.get_tif_size(n)
r = self.get_jpg_tif_ratio(n)
jpg_size = [float(d) * r for d in s]
leaf = ElementTree.Element('leaf', attrib={
'leafNum': human_readable_leaf_num,
'type': '',
'access': 'true',
'imgFile': '%s.jpg' % leaf_num,
'x': str(jpg_size[0]),
'y': str(jpg_size[1]),
'humanReadableLeafNum': human_readable_leaf_num,
'{http://cdlib.org/xtf}sectionType': leaf_num
})
ElementTree.SubElement(leaf, 'cropBox', attrib={
'x': str(jpg_size[0]),
'y': str(jpg_size[1]),
'w': str(jpg_size[0]),
'h': str(jpg_size[1])
})
leaf_data = self.leaf_data_from_pos_data(n)
#scale = self.get_jpg_tif_ratio(n)
#leaf_data = self.scale_leaf_data(leaf_data, scale)
# Add line spacing info to each line. This has to happen after scaling.
i = 0
while i < len(leaf_data):
leaf_data[i]['spacing'] = self.get_line_spacing(leaf_data, i)
i = i + 1
l = 0
for line in leaf_data:
ElementTree.SubElement(leaf, 'line', attrib={
'l': str(line['l']),
't': str(line['t']),
'r': str(line['r']),
'b': str(line['b']),
'spacing': self.spacing_to_string(line['spacing'])
}).text = self.get_line_text(line)
l = l + 1
return leaf
def get_xtf_converted_book(self):
x = ElementTree.Element('xtf-converted-book')
x.append(self.get_meta())
for i in range(len(self.file_dict['ocr_files'])):
x.append(self.get_leaf(i))
return x
if __name__=='__main__':
arguments = docopt(__doc__)
identifier = arguments['<identifier>']
con = sqlite3.connect('/data/digital_collections_ocfl/ark_data.db')
cur = con.cursor()
cur.execute('SELECT ark FROM arks WHERE original_identifier = ?', (identifier,))
ark = cur.fetchall()[0][0]
# get OCR type.
ocr_type = 'ALTO'
dir_type = 'uppercase'
'''
if os.path.isdir('{}/ALTO'.format(mvol_path)):
ocr_type = 'ALTO'
dir_type = 'uppercase'
elif os.path.isdir('{}/POS'.format(mvol_path)):
ocr_type = 'POS'
dir_type = 'uppercase'
elif os.path.isdir('{}/xml'.format(mvol_path)):
ocr_type = 'xml'
dir_type = 'lowercase'
elif os.path.isdir('{}/pos'.format(mvol_path)):
ocr_type = 'pos'
dir_type = 'lowercase'
else:
sys.stderr.write('no OCR data for {}\n'.format(arguments['<identifier>']))
sys.exit(1)
'''
dc_xml = 'http://ocfl.lib.uchicago.edu/{}/file.dc.xml'.format(ark)
ocr_files = []
for i in range(1, get_number_of_images(ark) + 1):
ocr_files.append('http://ocfl.lib.uchicago.edu/{}/{:08d}/file.xml'.format(ark, i))
struct_txt = 'http://ocfl.lib.uchicago.edu/{}/file.struct.txt'.format(ark)
tifs = []
for i in range(1, get_number_of_images(ark) + 1):
tifs.append('http://ocfl.lib.uchicago.edu/{}/{:08d}/file.tif'.format(ark, i))
o = OCRBuilder(
{
'dc': dc_xml,
'ocr_files': ocr_files,
'tifs': tifs,
'txt': struct_txt
},
int(arguments['<min-year>']),
int(arguments['<max-year>']),
arguments['<shrink_to_height>']
)
xtf_converted_book = o.get_xtf_converted_book()
print(ElementTree.tostring(xtf_converted_book, encoding='utf-8', method='xml').decode('utf-8'))