|
1
|
|
|
# -*- coding: utf-8 -*- |
|
2
|
|
|
|
|
3
|
|
|
from os import chdir, curdir, walk, stat, chmod, umask |
|
4
|
|
|
import shutil |
|
5
|
|
|
import logging |
|
6
|
|
|
from stat import filemode |
|
7
|
|
|
from os.path import join, exists, abspath, basename, dirname |
|
8
|
|
|
from shutil import copyfile, copytree as copytree_, rmtree |
|
9
|
|
|
from pathlib import Path |
|
10
|
|
|
from gzip import open as gzip_open |
|
11
|
|
|
|
|
12
|
|
|
from PIL import Image |
|
13
|
|
|
import numpy as np |
|
14
|
|
|
|
|
15
|
|
|
import pytest |
|
16
|
|
|
|
|
17
|
|
|
from tests.base import ( |
|
18
|
|
|
assets, |
|
19
|
|
|
main, |
|
20
|
|
|
FIFOIO |
|
21
|
|
|
) |
|
22
|
|
|
|
|
23
|
|
|
from ocrd_models import ( |
|
24
|
|
|
OcrdFile, |
|
25
|
|
|
OcrdMets |
|
26
|
|
|
) |
|
27
|
|
|
from ocrd_models.ocrd_page import parseString |
|
28
|
|
|
from ocrd_models.ocrd_page import TextRegionType, CoordsType, AlternativeImageType |
|
29
|
|
|
from ocrd_utils import polygon_mask, xywh_from_polygon, bbox_from_polygon, points_from_polygon |
|
30
|
|
|
from ocrd_modelfactory import page_from_file |
|
31
|
|
|
from ocrd.resolver import Resolver |
|
32
|
|
|
from ocrd.workspace import Workspace |
|
33
|
|
|
from ocrd.workspace_backup import WorkspaceBackupManager |
|
34
|
|
|
from ocrd_validators import WorkspaceValidator |
|
35
|
|
|
|
|
36
|
|
|
TMP_FOLDER = '/tmp/test-core-workspace' |
|
37
|
|
|
SRC_METS = assets.path_to('kant_aufklaerung_1784/data/mets.xml') |
|
38
|
|
|
|
|
39
|
|
|
SAMPLE_FILE_FILEGRP = 'OCR-D-IMG' |
|
40
|
|
|
SAMPLE_FILE_ID = 'INPUT_0017' |
|
41
|
|
|
SAMPLE_FILE_URL = join(SAMPLE_FILE_FILEGRP, '%s.tif' % SAMPLE_FILE_ID) |
|
42
|
|
|
|
|
43
|
|
|
|
|
44
|
|
|
def copytree(src, dst, *args, **kwargs): |
|
45
|
|
|
rmtree(dst) |
|
46
|
|
|
copytree_(src, dst, *args, **kwargs) |
|
47
|
|
|
|
|
48
|
|
|
|
|
49
|
|
|
def count_files(d): return sum(len(files) for _, _, files in walk(d)) |
|
50
|
|
|
|
|
51
|
|
|
|
|
52
|
|
|
@pytest.fixture(name='plain_workspace') |
|
53
|
|
|
def _fixture_plain_workspace(tmp_path): |
|
54
|
|
|
resolver = Resolver() |
|
55
|
|
|
ws = resolver.workspace_from_nothing(directory=tmp_path) |
|
56
|
|
|
prev_dir = abspath(curdir) |
|
57
|
|
|
chdir(tmp_path) |
|
58
|
|
|
yield ws |
|
59
|
|
|
chdir(prev_dir) |
|
60
|
|
|
|
|
61
|
|
|
def test_workspace_add_file(plain_workspace): |
|
62
|
|
|
fpath = plain_workspace.directory / 'ID1.tif' |
|
63
|
|
|
|
|
64
|
|
|
# act |
|
65
|
|
|
plain_workspace.add_file( |
|
66
|
|
|
'GRP', |
|
67
|
|
|
file_id='ID1', |
|
68
|
|
|
mimetype='image/tiff', |
|
69
|
|
|
content='CONTENT', |
|
70
|
|
|
page_id=None, |
|
71
|
|
|
local_filename=fpath |
|
72
|
|
|
) |
|
73
|
|
|
f = plain_workspace.mets.find_all_files()[0] |
|
74
|
|
|
|
|
75
|
|
|
# assert |
|
76
|
|
|
assert f.ID == 'ID1' |
|
77
|
|
|
assert f.mimetype == 'image/tiff' |
|
78
|
|
|
assert not f.url |
|
79
|
|
|
assert f.local_filename == str(fpath) |
|
80
|
|
|
assert Path(f.local_filename).exists() |
|
81
|
|
|
|
|
82
|
|
|
|
|
83
|
|
|
def test_workspace_add_file_overwrite(plain_workspace): |
|
84
|
|
|
fpath = plain_workspace.directory / 'ID1.tif' |
|
85
|
|
|
|
|
86
|
|
|
# act |
|
87
|
|
|
plain_workspace.add_file('GRP', file_id='ID1', mimetype='image/tiff', content='CONTENT', page_id='phys1', local_filename=fpath) |
|
88
|
|
|
with pytest.raises(FileExistsError) as fn_exc: |
|
89
|
|
|
plain_workspace.add_file('GRP', file_id='ID1', mimetype='image/tiff', content='CONTENT', page_id=None, local_filename=fpath) |
|
90
|
|
|
assert str(fn_exc.value) == "File with file_id='ID1' already exists" |
|
91
|
|
|
with pytest.raises(FileExistsError) as fn_exc: |
|
92
|
|
|
plain_workspace.add_file('GRP', file_id='ID1', mimetype='image/tiff', content='CONTENT', page_id='phys2', local_filename=fpath, force=True) |
|
93
|
|
|
assert 'cannot mitigate' in str(fn_exc.value) |
|
94
|
|
|
plain_workspace.add_file('GRP', file_id='ID1', mimetype='image/tiff', content='CONTENT2', page_id='phys1', local_filename=fpath, force=True) |
|
95
|
|
|
|
|
96
|
|
|
f = plain_workspace.mets.find_all_files()[0] |
|
97
|
|
|
assert f.ID == 'ID1' |
|
98
|
|
|
assert f.mimetype == 'image/tiff' |
|
99
|
|
|
assert not f.url |
|
100
|
|
|
assert f.local_filename == str(fpath) |
|
101
|
|
|
assert f.pageId == 'phys1' |
|
102
|
|
|
assert fpath.exists() |
|
103
|
|
|
|
|
104
|
|
|
|
|
105
|
|
|
def test_workspace_add_file_basename_no_content(plain_workspace): |
|
106
|
|
|
plain_workspace.add_file('GRP', file_id='ID1', mimetype='image/tiff', page_id=None) |
|
107
|
|
|
f = next(plain_workspace.mets.find_files()) |
|
108
|
|
|
assert f.url == '' |
|
109
|
|
|
|
|
110
|
|
|
def test_workspace_add_file_binary_content(plain_workspace): |
|
111
|
|
|
fpath = join(plain_workspace.directory, 'subdir', 'ID1.tif') |
|
112
|
|
|
plain_workspace.add_file('GRP', file_id='ID1', content=b'CONTENT', local_filename=fpath, url='http://foo/bar', page_id=None) |
|
113
|
|
|
|
|
114
|
|
|
# assert |
|
115
|
|
|
assert exists(fpath) |
|
116
|
|
|
|
|
117
|
|
|
|
|
118
|
|
|
def test_workspacec_add_file_content_wo_local_filename(plain_workspace): |
|
119
|
|
|
# act |
|
120
|
|
|
with pytest.raises(Exception) as fn_exc: |
|
121
|
|
|
plain_workspace.add_file('GRP', file_id='ID1', content=b'CONTENT', page_id='foo1234') |
|
122
|
|
|
|
|
123
|
|
|
assert "'content' was set but no 'local_filename'" in str(fn_exc.value) |
|
124
|
|
|
|
|
125
|
|
|
|
|
126
|
|
|
def test_workspacec_add_file_content_wo_pageid(plain_workspace): |
|
127
|
|
|
# act |
|
128
|
|
|
with pytest.raises(ValueError) as val_err: |
|
129
|
|
|
plain_workspace.add_file('GRP', file_id='ID1', content=b'CONTENT', local_filename='foo') |
|
130
|
|
|
|
|
131
|
|
|
assert "workspace.add_file must be passed a 'page_id' kwarg, even if it is None." in str(val_err.value) |
|
132
|
|
|
|
|
133
|
|
|
|
|
134
|
|
|
def test_workspace_str(plain_workspace): |
|
135
|
|
|
|
|
136
|
|
|
# act |
|
137
|
|
|
plain_workspace.save_mets() |
|
138
|
|
|
plain_workspace.reload_mets() |
|
139
|
|
|
|
|
140
|
|
|
# assert |
|
141
|
|
|
ws_dir = plain_workspace.directory |
|
142
|
|
|
assert str(plain_workspace) == 'Workspace[remote=False, directory=%s, baseurl=None, file_groups=[], files=[]]' % ws_dir |
|
143
|
|
|
|
|
144
|
|
|
|
|
145
|
|
|
def test_workspace_backup(plain_workspace): |
|
146
|
|
|
|
|
147
|
|
|
# act |
|
148
|
|
|
plain_workspace.automatic_backup = WorkspaceBackupManager(plain_workspace) |
|
149
|
|
|
plain_workspace.save_mets() |
|
150
|
|
|
plain_workspace.reload_mets() |
|
151
|
|
|
|
|
152
|
|
|
# TODO |
|
153
|
|
|
# changed test semantics |
|
154
|
|
|
assert exists(join(plain_workspace.directory, '.backup')) |
|
155
|
|
|
|
|
156
|
|
|
|
|
157
|
|
|
def _url_to_file(the_path): |
|
158
|
|
|
dummy_mets = OcrdMets.empty_mets() |
|
159
|
|
|
dummy_url = abspath(the_path) |
|
160
|
|
|
return dummy_mets.add_file('TESTGRP', ID=Path(dummy_url).name, url=dummy_url) |
|
161
|
|
|
|
|
162
|
|
|
|
|
163
|
|
|
def test_download_very_self_file(plain_workspace): |
|
164
|
|
|
the_file = _url_to_file(abspath(__file__)) |
|
165
|
|
|
fn = plain_workspace.download_file(the_file) |
|
166
|
|
|
assert fn, join('TESTGRP', basename(__file__)) |
|
167
|
|
|
assert fn == the_file |
|
168
|
|
|
|
|
169
|
|
|
|
|
170
|
|
|
def test_download_url_without_baseurl_raises_exception(tmp_path): |
|
171
|
|
|
# arrange |
|
172
|
|
|
dst_mets = join(tmp_path, 'mets.xml') |
|
173
|
|
|
copyfile(SRC_METS, dst_mets) |
|
174
|
|
|
ws1 = Resolver().workspace_from_url(dst_mets) |
|
175
|
|
|
the_file = _url_to_file(SAMPLE_FILE_URL) |
|
176
|
|
|
|
|
177
|
|
|
# act |
|
178
|
|
|
with pytest.raises(Exception) as exc: |
|
179
|
|
|
ws1.download_file(the_file) |
|
180
|
|
|
|
|
181
|
|
|
# assert exception message contents |
|
182
|
|
|
assert "File path passed as 'url' to download_to_directory does not exist:" in str(exc.value) |
|
183
|
|
|
|
|
184
|
|
|
|
|
185
|
|
|
def test_download_url_with_baseurl(tmp_path): |
|
186
|
|
|
# arrange |
|
187
|
|
|
dst_mets = join(tmp_path, 'mets.xml') |
|
188
|
|
|
copyfile(SRC_METS, dst_mets) |
|
189
|
|
|
tif_dir = tmp_path / 'OCR-D-IMG' |
|
190
|
|
|
tif_dir.mkdir() |
|
191
|
|
|
dst_tif = join(tmp_path, SAMPLE_FILE_URL) |
|
192
|
|
|
copyfile(join(dirname(SRC_METS), SAMPLE_FILE_URL), dst_tif) |
|
193
|
|
|
ws1 = Resolver().workspace_from_url(dst_mets, src_baseurl=dirname(SRC_METS)) |
|
194
|
|
|
the_file = _url_to_file(dst_tif) |
|
195
|
|
|
|
|
196
|
|
|
# act |
|
197
|
|
|
# TODO |
|
198
|
|
|
# semantics changed from .download_url to .download_file |
|
199
|
|
|
# and from context path 'DEPRECATED' to 'OCR-D-IMG' |
|
200
|
|
|
f = Path(ws1.download_file(the_file).local_filename) |
|
201
|
|
|
|
|
202
|
|
|
# assert |
|
203
|
|
|
assert f.name == f'{SAMPLE_FILE_ID}.tif' |
|
204
|
|
|
assert f.parent.name == 'TESTGRP' |
|
205
|
|
|
assert Path(ws1.directory, f).exists() |
|
206
|
|
|
|
|
207
|
|
|
|
|
208
|
|
|
def test_from_url_dst_dir_download(plain_workspace): |
|
209
|
|
|
""" |
|
210
|
|
|
https://github.com/OCR-D/core/issues/319 |
|
211
|
|
|
""" |
|
212
|
|
|
ws_dir = join(plain_workspace.directory, 'non-existing-for-good-measure') |
|
213
|
|
|
# Create a relative path to trigger #319 |
|
214
|
|
|
src_path = str(Path(assets.path_to('kant_aufklaerung_1784/data/mets.xml'))) |
|
215
|
|
|
plain_workspace.resolver.workspace_from_url(src_path, dst_dir=ws_dir, download=True) |
|
216
|
|
|
|
|
217
|
|
|
# assert |
|
218
|
|
|
assert Path(ws_dir, 'mets.xml').exists() # sanity check, mets.xml must exist |
|
219
|
|
|
assert Path(ws_dir, 'OCR-D-GT-PAGE/PAGE_0017_PAGE.xml').exists() |
|
220
|
|
|
|
|
221
|
|
|
|
|
222
|
|
|
def test_superfluous_copies_in_ws_dir(tmp_path): |
|
223
|
|
|
""" |
|
224
|
|
|
https://github.com/OCR-D/core/issues/227 |
|
225
|
|
|
""" |
|
226
|
|
|
# arrange |
|
227
|
|
|
src_path = assets.path_to('SBB0000F29300010000/data/mets_one_file.xml') |
|
228
|
|
|
dst_path = join(tmp_path, 'mets.xml') |
|
229
|
|
|
copyfile(src_path, dst_path) |
|
230
|
|
|
ws1 = Workspace(Resolver(), tmp_path) |
|
231
|
|
|
|
|
232
|
|
|
# assert directory files |
|
233
|
|
|
assert count_files(tmp_path) == 1 |
|
234
|
|
|
|
|
235
|
|
|
# act |
|
236
|
|
|
for file in ws1.mets.find_all_files(): |
|
237
|
|
|
ws1.download_file(file) |
|
238
|
|
|
|
|
239
|
|
|
# assert |
|
240
|
|
|
assert count_files(tmp_path) == 2 |
|
241
|
|
|
assert exists(join(tmp_path, 'OCR-D-IMG/FILE_0005_IMAGE.tif')) |
|
242
|
|
|
|
|
243
|
|
|
|
|
244
|
|
|
@pytest.fixture(name='sbb_data_tmp') |
|
245
|
|
|
def _fixture_sbb_data_tmp(tmp_path): |
|
246
|
|
|
copytree(assets.path_to('SBB0000F29300010000/data'), str(tmp_path)) |
|
247
|
|
|
yield str(tmp_path) |
|
248
|
|
|
|
|
249
|
|
|
|
|
250
|
|
|
@pytest.fixture(name='sbb_data_workspace') |
|
251
|
|
|
def _fixture_sbb_data(sbb_data_tmp): |
|
252
|
|
|
resolver = Resolver() |
|
253
|
|
|
workspace = Workspace(resolver, directory=sbb_data_tmp) |
|
254
|
|
|
yield workspace |
|
255
|
|
|
|
|
256
|
|
|
|
|
257
|
|
|
def test_remove_file_not_existing_raises_error(sbb_data_workspace): |
|
258
|
|
|
|
|
259
|
|
|
# act |
|
260
|
|
|
with pytest.raises(FileNotFoundError) as fnf_err: |
|
261
|
|
|
sbb_data_workspace.remove_file('non-existing-id') |
|
262
|
|
|
|
|
263
|
|
|
# assert |
|
264
|
|
|
assert "not found" in str(fnf_err.value) |
|
265
|
|
|
|
|
266
|
|
|
|
|
267
|
|
|
def test_remove_file_force(sbb_data_workspace): |
|
268
|
|
|
"""Enforce removal of non-existing-id doesn't yield any error |
|
269
|
|
|
but also returns no ocrd-file identifier""" |
|
270
|
|
|
|
|
271
|
|
|
# TODO check semantics - can a non-existent thing be removed? |
|
272
|
|
|
assert not sbb_data_workspace.remove_file('non-existing-id', force=True) |
|
273
|
|
|
# should also succeed |
|
274
|
|
|
sbb_data_workspace.overwrite_mode = True |
|
275
|
|
|
assert not sbb_data_workspace.remove_file('non-existing-id', force=False) |
|
276
|
|
|
|
|
277
|
|
|
|
|
278
|
|
|
def test_remove_file_remote_not_available_raises_exception(plain_workspace): |
|
279
|
|
|
plain_workspace.add_file('IMG', file_id='page1_img', mimetype='image/tiff', url='http://remote', page_id=None) |
|
280
|
|
|
with pytest.raises(Exception) as not_avail_exc: |
|
281
|
|
|
plain_workspace.remove_file('page1_img') |
|
282
|
|
|
|
|
283
|
|
|
assert "not locally available" in str(not_avail_exc.value) |
|
284
|
|
|
|
|
285
|
|
|
|
|
286
|
|
|
def test_remove_file_remote(plain_workspace): |
|
287
|
|
|
|
|
288
|
|
|
# act |
|
289
|
|
|
plain_workspace.add_file('IMG', file_id='page1_img', mimetype='image/tiff', url='http://remote', page_id=None) |
|
290
|
|
|
|
|
291
|
|
|
# must succeed because removal is enforced |
|
292
|
|
|
assert plain_workspace.remove_file('page1_img', force=True) |
|
293
|
|
|
|
|
294
|
|
|
# TODO check returned value |
|
295
|
|
|
# should also "succeed", because overwrite_mode is set which also sets 'force' to 'True' |
|
296
|
|
|
plain_workspace.overwrite_mode = True |
|
297
|
|
|
assert not plain_workspace.remove_file('page1_img') |
|
298
|
|
|
|
|
299
|
|
|
|
|
300
|
|
|
def test_rename_file_group(tmp_path): |
|
301
|
|
|
# arrange |
|
302
|
|
|
copytree(assets.path_to('kant_aufklaerung_1784-page-region-line-word_glyph/data'), tmp_path) |
|
303
|
|
|
workspace = Workspace(Resolver(), directory=tmp_path) |
|
304
|
|
|
|
|
305
|
|
|
# before act |
|
306
|
|
|
# TODO clear semantics |
|
307
|
|
|
# requires rather odd additional path-setting because root path from |
|
308
|
|
|
# workspace is not propagated - works only if called inside workspace |
|
309
|
|
|
# which can be achieved with pushd_popd functionalities |
|
310
|
|
|
ocrd_file = next(workspace.mets.find_files(ID='OCR-D-GT-SEG-WORD_0001')) |
|
311
|
|
|
relative_name = ocrd_file.local_filename |
|
312
|
|
|
ocrd_file.local_filename = tmp_path / relative_name |
|
313
|
|
|
pcgts_before = page_from_file(ocrd_file) |
|
314
|
|
|
# before assert |
|
315
|
|
|
assert pcgts_before.get_Page().imageFilename == 'OCR-D-IMG/INPUT_0017.tif' |
|
316
|
|
|
|
|
317
|
|
|
# act |
|
318
|
|
|
workspace.rename_file_group('OCR-D-IMG', 'FOOBAR') |
|
319
|
|
|
next_ocrd_file = next(workspace.mets.find_files(ID='OCR-D-GT-SEG-WORD_0001')) |
|
320
|
|
|
next_ocrd_file.local_filename = str(tmp_path / relative_name) |
|
321
|
|
|
pcgts_after = page_from_file(next_ocrd_file) |
|
322
|
|
|
|
|
323
|
|
|
# assert |
|
324
|
|
|
assert pcgts_after.get_Page().imageFilename == 'FOOBAR/INPUT_0017.tif' |
|
325
|
|
|
assert Path(tmp_path / 'FOOBAR/INPUT_0017.tif').exists() |
|
326
|
|
|
assert not Path('OCR-D-IMG/OCR-D-IMG_0001.tif').exists() |
|
327
|
|
|
assert workspace.mets.get_physical_pages(for_fileIds=['OCR-D-IMG_0001']) == [None] |
|
328
|
|
|
assert workspace.mets.get_physical_pages(for_fileIds=['FOOBAR_0001']) == ['phys_0001'] |
|
329
|
|
|
|
|
330
|
|
|
|
|
331
|
|
|
def test_remove_file_group_invalid_raises_exception(sbb_data_workspace): |
|
332
|
|
|
with pytest.raises(Exception) as no_fg_exc: |
|
333
|
|
|
# should fail |
|
334
|
|
|
sbb_data_workspace.remove_file_group('I DO NOT EXIST') |
|
335
|
|
|
assert "No such fileGrp" in str(no_fg_exc.value) |
|
336
|
|
|
|
|
337
|
|
|
|
|
338
|
|
|
def test_remove_file_group_force(sbb_data_workspace): |
|
339
|
|
|
|
|
340
|
|
|
# TODO |
|
341
|
|
|
# check function and tests semantics |
|
342
|
|
|
# should succeed |
|
343
|
|
|
assert not sbb_data_workspace.remove_file_group('I DO NOT EXIST', force=True) |
|
344
|
|
|
# should also succeed |
|
345
|
|
|
sbb_data_workspace.overwrite_mode = True |
|
346
|
|
|
assert not sbb_data_workspace.remove_file_group('I DO NOT EXIST', force=False) |
|
347
|
|
|
|
|
348
|
|
|
|
|
349
|
|
|
def test_remove_file_group_rmdir(sbb_data_tmp, sbb_data_workspace): |
|
350
|
|
|
assert exists(join(sbb_data_tmp, 'OCR-D-IMG')) |
|
351
|
|
|
sbb_data_workspace.remove_file_group('OCR-D-IMG', recursive=True) |
|
352
|
|
|
assert not exists(join(sbb_data_tmp, 'OCR-D-IMG')) |
|
353
|
|
|
|
|
354
|
|
|
|
|
355
|
|
|
def test_remove_file_group_flat(plain_workspace): |
|
356
|
|
|
""" |
|
357
|
|
|
https://github.com/OCR-D/core/issues/728 |
|
358
|
|
|
""" |
|
359
|
|
|
|
|
360
|
|
|
# act |
|
361
|
|
|
added_res = plain_workspace.add_file('FOO', file_id='foo', mimetype='foo/bar', local_filename='file.ext', content='foo', page_id=None).local_filename |
|
362
|
|
|
# requires additional prepending of current path because not pushd_popd-magic at work |
|
363
|
|
|
added_filename = join(plain_workspace.directory, added_res) |
|
364
|
|
|
|
|
365
|
|
|
# assert |
|
366
|
|
|
assert Path(added_filename).exists() |
|
367
|
|
|
plain_workspace.remove_file_group('FOO', recursive=True) |
|
368
|
|
|
|
|
369
|
|
|
|
|
370
|
|
|
@pytest.fixture(name='kant_complex_workspace') |
|
371
|
|
|
def _fixture_kant_complex(tmp_path): |
|
372
|
|
|
copytree(assets.path_to('kant_aufklaerung_1784-complex/data'), str(tmp_path)) |
|
373
|
|
|
yield Workspace(Resolver, directory=tmp_path) |
|
374
|
|
|
|
|
375
|
|
|
|
|
376
|
|
|
def test_remove_file_page_recursive0(kant_complex_workspace): |
|
377
|
|
|
assert len(kant_complex_workspace.mets.find_all_files()) == 119 |
|
378
|
|
|
kant_complex_workspace.remove_file('OCR-D-OCR-OCRO-fraktur-SEG-LINE-tesseract-ocropy-DEWARP_0001', page_recursive=True, page_same_group=False, keep_file=True) |
|
379
|
|
|
assert len(kant_complex_workspace.mets.find_all_files()) == 83 |
|
380
|
|
|
kant_complex_workspace.remove_file('PAGE_0017_ALTO', page_recursive=True) |
|
381
|
|
|
|
|
382
|
|
|
|
|
383
|
|
|
def test_remove_file_page_recursive_keep_file(kant_complex_workspace): |
|
384
|
|
|
before = count_files(kant_complex_workspace.directory) |
|
385
|
|
|
kant_complex_workspace.remove_file('OCR-D-IMG-BINPAGE-sauvola_0001', page_recursive=True, page_same_group=False, force=True) |
|
386
|
|
|
after = count_files(kant_complex_workspace.directory) |
|
387
|
|
|
assert after == (before - 2), '2 files deleted' |
|
388
|
|
|
|
|
389
|
|
|
|
|
390
|
|
|
def test_remove_file_page_recursive_same_group(kant_complex_workspace): |
|
391
|
|
|
before = count_files(kant_complex_workspace.directory) |
|
392
|
|
|
kant_complex_workspace.remove_file('OCR-D-IMG-BINPAGE-sauvola_0001', page_recursive=True, page_same_group=True, force=False) |
|
393
|
|
|
after = count_files(kant_complex_workspace.directory) |
|
394
|
|
|
assert after == before - 1, '1 file deleted' |
|
395
|
|
|
|
|
396
|
|
|
|
|
397
|
|
|
def test_download_to_directory_from_workspace_download_file(plain_workspace): |
|
398
|
|
|
""" |
|
399
|
|
|
https://github.com/OCR-D/core/issues/342 |
|
400
|
|
|
""" |
|
401
|
|
|
f1 = plain_workspace.add_file('IMG', file_id='page1_img', mimetype='image/tiff', local_filename='test.tif', content='', page_id=None) |
|
402
|
|
|
f2 = plain_workspace.add_file('GT', file_id='page1_gt', mimetype='text/xml', local_filename='test.xml', content='', page_id=None) |
|
403
|
|
|
|
|
404
|
|
|
assert not f1.url |
|
405
|
|
|
assert not f2.url |
|
406
|
|
|
|
|
407
|
|
|
# these should be no-ops |
|
408
|
|
|
plain_workspace.download_file(f1) |
|
409
|
|
|
plain_workspace.download_file(f2) |
|
410
|
|
|
|
|
411
|
|
|
assert f1.local_filename == 'test.tif' |
|
412
|
|
|
assert f2.local_filename == 'test.xml' |
|
413
|
|
|
|
|
414
|
|
|
|
|
415
|
|
|
def test_save_image_file_invalid_mimetype_raises_exception(plain_workspace): |
|
416
|
|
|
img = Image.new('RGB', (1000, 1000)) |
|
417
|
|
|
|
|
418
|
|
|
# act raise |
|
419
|
|
|
with pytest.raises(KeyError) as key_exc: |
|
420
|
|
|
plain_workspace.save_image_file(img, 'page1_img', 'IMG', 'page1', 'ceci/nest/pas/une/mimetype') |
|
421
|
|
|
|
|
422
|
|
|
assert "'ceci/nest/pas/une/mimetype'" == str(key_exc.value) |
|
423
|
|
|
|
|
424
|
|
|
|
|
425
|
|
|
def test_save_image_file(plain_workspace): |
|
426
|
|
|
|
|
427
|
|
|
# arrange |
|
428
|
|
|
img = Image.new('RGB', (1000, 1000)) |
|
429
|
|
|
|
|
430
|
|
|
# act |
|
431
|
|
|
assert plain_workspace.save_image_file(img, 'page1_img', 'IMG', 'page1', 'image/jpeg') |
|
432
|
|
|
assert exists(join(plain_workspace.directory, 'IMG', 'page1_img.jpg')) |
|
433
|
|
|
# should succeed |
|
434
|
|
|
assert plain_workspace.save_image_file(img, 'page1_img', 'IMG', 'page1', 'image/jpeg', force=True) |
|
435
|
|
|
# should also succeed |
|
436
|
|
|
plain_workspace.overwrite_mode = True |
|
437
|
|
|
assert plain_workspace.save_image_file(img, 'page1_img', 'IMG', 'page1', 'image/jpeg') |
|
438
|
|
|
|
|
439
|
|
|
|
|
440
|
|
|
@pytest.fixture(name='workspace_kant_aufklaerung') |
|
441
|
|
|
def _fixture_workspace_kant_aufklaerung(tmp_path): |
|
442
|
|
|
copytree(assets.path_to('kant_aufklaerung_1784/data/'), str(tmp_path)) |
|
443
|
|
|
resolver = Resolver() |
|
444
|
|
|
ws = resolver.workspace_from_url(join(tmp_path, 'mets.xml'), src_baseurl=tmp_path) |
|
445
|
|
|
prev_dir = abspath(curdir) |
|
446
|
|
|
chdir(tmp_path) |
|
447
|
|
|
yield ws |
|
448
|
|
|
chdir(prev_dir) |
|
449
|
|
|
|
|
450
|
|
|
|
|
451
|
|
|
def test_resolve_image_exif(workspace_kant_aufklaerung): |
|
452
|
|
|
|
|
453
|
|
|
tif_path = 'OCR-D-IMG/INPUT_0017.tif' |
|
454
|
|
|
|
|
455
|
|
|
# act |
|
456
|
|
|
exif = workspace_kant_aufklaerung.resolve_image_exif(tif_path) |
|
457
|
|
|
|
|
458
|
|
|
# assert |
|
459
|
|
|
assert exif.compression == 'jpeg' |
|
460
|
|
|
assert exif.width == 1457 |
|
461
|
|
|
|
|
462
|
|
|
|
|
463
|
|
|
def test_resolve_image_as_pil(workspace_kant_aufklaerung): |
|
464
|
|
|
img = workspace_kant_aufklaerung._resolve_image_as_pil('OCR-D-IMG/INPUT_0017.tif') |
|
465
|
|
|
assert img.width == 1457 |
|
466
|
|
|
img = workspace_kant_aufklaerung._resolve_image_as_pil('OCR-D-IMG/INPUT_0017.tif', coords=([100, 100], [50, 50])) |
|
467
|
|
|
assert img.width == 50 |
|
468
|
|
|
|
|
469
|
|
|
|
|
470
|
|
|
@pytest.fixture(name='workspace_gutachten_data') |
|
471
|
|
|
def _fixture_workspace_gutachten_data(tmp_path): |
|
472
|
|
|
copytree(assets.path_to('gutachten/data'), str(tmp_path)) |
|
473
|
|
|
resolver = Resolver() |
|
474
|
|
|
ws = resolver.workspace_from_url(join(str(tmp_path), 'mets.xml')) |
|
475
|
|
|
prev_path = abspath(curdir) |
|
476
|
|
|
chdir(tmp_path) |
|
477
|
|
|
yield ws |
|
478
|
|
|
chdir(prev_path) |
|
479
|
|
|
|
|
480
|
|
|
|
|
481
|
|
|
def test_image_from_page_basic(workspace_gutachten_data): |
|
482
|
|
|
# arrange |
|
483
|
|
|
with open(assets.path_to('gutachten/data/TEMP1/PAGE_TEMP1.xml'), 'r') as f: |
|
484
|
|
|
pcgts = parseString(f.read().encode('utf8'), silence=True) |
|
485
|
|
|
|
|
486
|
|
|
# act + assert |
|
487
|
|
|
_, info, _ = workspace_gutachten_data.image_from_page(pcgts.get_Page(), page_id='PHYS_0017', feature_selector='clipped', feature_filter='cropped') |
|
488
|
|
|
assert info['features'] == 'binarized,clipped' |
|
489
|
|
|
_, info, _ = workspace_gutachten_data.image_from_page(pcgts.get_Page(), page_id='PHYS_0017') |
|
490
|
|
|
assert info['features'] == 'binarized,clipped' |
|
491
|
|
|
|
|
492
|
|
|
|
|
493
|
|
|
@pytest.fixture(name='workspace_sample_features') |
|
494
|
|
|
def _fixture_workspace_sample_features(tmp_path): |
|
495
|
|
|
copytree('tests/data/sample-features', str(tmp_path)) |
|
496
|
|
|
resolver = Resolver() |
|
497
|
|
|
ws = resolver.workspace_from_url(join(str(tmp_path), 'mets.xml')) |
|
498
|
|
|
prev_path = abspath(curdir) |
|
499
|
|
|
chdir(tmp_path) |
|
500
|
|
|
yield ws |
|
501
|
|
|
chdir(prev_path) |
|
502
|
|
|
|
|
503
|
|
|
|
|
504
|
|
|
def test_image_feature_selectoro(workspace_sample_features): |
|
505
|
|
|
# arrange |
|
506
|
|
|
with open(Path(workspace_sample_features.directory) / 'image_features.page.xml', 'r', encoding='utf-8') as f: |
|
507
|
|
|
pcgts = parseString(f.read().encode('utf-8')) |
|
508
|
|
|
|
|
509
|
|
|
# richest feature set is not last: |
|
510
|
|
|
_, info, _ = workspace_sample_features.image_from_page(pcgts.get_Page(), page_id='page1', feature_selector='dewarped') |
|
511
|
|
|
# recropped because foo4 contains cropped+deskewed but not recropped yet: |
|
512
|
|
|
assert info['features'] == 'cropped,dewarped,binarized,despeckled,deskewed' |
|
513
|
|
|
# richest feature set is also last: |
|
514
|
|
|
_, info, _ = workspace_sample_features.image_from_page(pcgts.get_Page(), page_id='page1', feature_selector='dewarped', feature_filter='binarized') |
|
515
|
|
|
# no deskewing here, thus no recropping: |
|
516
|
|
|
assert info['features'] == 'cropped,dewarped,despeckled' |
|
517
|
|
|
|
|
518
|
|
|
def test_deskewing(plain_workspace): |
|
519
|
|
|
#from ocrd_utils import initLogging, setOverrideLogLevel |
|
520
|
|
|
#setOverrideLogLevel('DEBUG') |
|
521
|
|
|
size = (3000, 4000) |
|
522
|
|
|
poly = [[1403, 2573], [1560, 2573], [1560, 2598], [2311, 2598], [2311, 2757], |
|
523
|
|
|
[2220, 2757], [2220, 2798], [2311, 2798], [2311, 2908], [1403, 2908]] |
|
524
|
|
|
xywh = xywh_from_polygon(poly) |
|
525
|
|
|
bbox = bbox_from_polygon(poly) |
|
526
|
|
|
skew = 4.625 |
|
527
|
|
|
image = Image.new('L', size) |
|
528
|
|
|
image = polygon_mask(image, poly) |
|
529
|
|
|
#image.show(title='image') |
|
530
|
|
|
pixels = np.count_nonzero(np.array(image) > 0) |
|
531
|
|
|
name = 'foo0' |
|
532
|
|
|
assert plain_workspace.save_image_file(image, name, 'IMG') |
|
533
|
|
|
pcgts = page_from_file(next(plain_workspace.mets.find_files(ID=name))) |
|
534
|
|
|
page = pcgts.get_Page() |
|
535
|
|
|
region = TextRegionType(id='nonrect', |
|
536
|
|
|
Coords=CoordsType(points=points_from_polygon(poly)), |
|
537
|
|
|
orientation=-skew) |
|
538
|
|
|
page.add_TextRegion(region) |
|
539
|
|
|
page_image, page_coords, _ = plain_workspace.image_from_page(page, '') |
|
540
|
|
|
#page_image.show(title='page_image') |
|
541
|
|
|
assert list(image.getdata()) == list(page_image.getdata()) |
|
542
|
|
|
assert np.all(page_coords['transform'] == np.eye(3)) |
|
543
|
|
|
reg_image, reg_coords = plain_workspace.image_from_segment(region, page_image, page_coords, |
|
544
|
|
|
feature_filter='deskewed', fill=0) |
|
545
|
|
|
assert list(image.crop(bbox).getdata()) == list(reg_image.getdata()) |
|
546
|
|
|
assert reg_image.width == xywh['w'] == 908 |
|
547
|
|
|
assert reg_image.height == xywh['h'] == 335 |
|
548
|
|
|
assert reg_coords['transform'][0, 2] == -xywh['x'] |
|
549
|
|
|
assert reg_coords['transform'][1, 2] == -xywh['y'] |
|
550
|
|
|
# same fg after cropping to minimal bbox |
|
551
|
|
|
reg_pixels = np.count_nonzero(np.array(reg_image) > 0) |
|
552
|
|
|
assert pixels == reg_pixels |
|
553
|
|
|
# now with deskewing (test for size after recropping) |
|
554
|
|
|
reg_image, reg_coords = plain_workspace.image_from_segment(region, page_image, page_coords, fill=0) |
|
555
|
|
|
#reg_image.show(title='reg_image') |
|
556
|
|
|
assert reg_image.width == 932 > xywh['w'] |
|
557
|
|
|
assert reg_image.height == 382 > xywh['h'] |
|
558
|
|
|
assert reg_coords['transform'][0, 1] != 0 |
|
559
|
|
|
assert reg_coords['transform'][1, 0] != 0 |
|
560
|
|
|
assert 'deskewed' in reg_coords['features'] |
|
561
|
|
|
# same fg after cropping to minimal bbox (roughly - due to aliasing) |
|
562
|
|
|
reg_pixels = np.count_nonzero(np.array(reg_image) > 0) |
|
563
|
|
|
assert np.abs(pixels - reg_pixels) / pixels < 0.005 |
|
564
|
|
|
reg_array = np.array(reg_image) > 0 |
|
565
|
|
|
# now via AlternativeImage |
|
566
|
|
|
path = plain_workspace.save_image_file(reg_image, region.id + '_img', 'IMG') |
|
567
|
|
|
region.add_AlternativeImage(AlternativeImageType(filename=path, comments=reg_coords['features'])) |
|
568
|
|
|
logger_capture = FIFOIO(256) |
|
569
|
|
|
logger_handler = logging.StreamHandler(logger_capture) |
|
570
|
|
|
#logger_handler.setFormatter(logging.Formatter(fmt=LOG_FORMAT, datefmt=LOG_TIMEFMT)) |
|
571
|
|
|
logger = logging.getLogger('ocrd.utils.crop_image') |
|
572
|
|
|
logger.addHandler(logger_handler) |
|
573
|
|
|
reg_image2, reg_coords2 = plain_workspace.image_from_segment(region, page_image, page_coords, fill=0) |
|
574
|
|
|
#reg_image2.show(title='reg_image2') |
|
575
|
|
|
logger_output = logger_capture.getvalue() |
|
576
|
|
|
logger_capture.close() |
|
577
|
|
|
assert logger_output == '' |
|
578
|
|
|
assert reg_image2.width == reg_image.width |
|
579
|
|
|
assert reg_image2.height == reg_image.height |
|
580
|
|
|
assert np.allclose(reg_coords2['transform'], reg_coords['transform']) |
|
581
|
|
|
assert reg_coords2['features'] == reg_coords['features'] |
|
582
|
|
|
# same fg after cropping to minimal bbox (roughly - due to aliasing) |
|
583
|
|
|
reg_pixels2 = np.count_nonzero(np.array(reg_image) > 0) |
|
584
|
|
|
assert reg_pixels2 == reg_pixels |
|
585
|
|
|
reg_array2 = np.array(reg_image2) > 0 |
|
586
|
|
|
assert 0.98 < np.sum(reg_array == reg_array2) / reg_array.size <= 1.0 |
|
587
|
|
|
|
|
588
|
|
|
def test_downsample_16bit_image(plain_workspace): |
|
589
|
|
|
# arrange image |
|
590
|
|
|
img_path = Path(plain_workspace.directory, '16bit.tif') |
|
591
|
|
|
with gzip_open(Path(__file__).parent / 'data/OCR-D-IMG_APBB_Mitteilungen_62.0002.tif.gz', 'rb') as gzip_in: |
|
592
|
|
|
with open(img_path, 'wb') as tif_out: |
|
593
|
|
|
tif_out.write(gzip_in.read()) |
|
594
|
|
|
|
|
595
|
|
|
# act |
|
596
|
|
|
plain_workspace.add_file('IMG', file_id='foo', local_filename=img_path, mimetype='image/tiff', page_id=None) |
|
597
|
|
|
|
|
598
|
|
|
# assert |
|
599
|
|
|
pil_before = Image.open(img_path) |
|
600
|
|
|
assert pil_before.mode == 'I;16' |
|
601
|
|
|
pil_after = plain_workspace._resolve_image_as_pil(img_path) |
|
602
|
|
|
assert pil_after.mode == 'L' |
|
603
|
|
|
|
|
604
|
|
|
|
|
605
|
|
|
def test_mets_permissions(plain_workspace): |
|
606
|
|
|
plain_workspace.save_mets() |
|
607
|
|
|
mets_path = join(plain_workspace.directory, 'mets.xml') |
|
608
|
|
|
mask = umask(0) |
|
609
|
|
|
umask(mask) |
|
610
|
|
|
assert (stat(mets_path).st_mode) == 0o100664 & ~mask |
|
611
|
|
|
chmod(mets_path, 0o777) |
|
612
|
|
|
plain_workspace.save_mets() |
|
613
|
|
|
assert filemode(stat(mets_path).st_mode) == '-rwxrwxrwx' |
|
614
|
|
|
|
|
615
|
|
|
|
|
616
|
|
|
def test_merge0(tmp_path): |
|
617
|
|
|
|
|
618
|
|
|
# arrange |
|
619
|
|
|
dst_path1 = tmp_path / 'kant_aufklaerung' |
|
620
|
|
|
dst_path1.mkdir() |
|
621
|
|
|
dst_path2 = tmp_path / 'sbb' |
|
622
|
|
|
dst_path2.mkdir() |
|
623
|
|
|
copytree(assets.path_to('kant_aufklaerung_1784/data'), dst_path1) |
|
624
|
|
|
copytree(assets.path_to('SBB0000F29300010000/data'), dst_path2) |
|
625
|
|
|
|
|
626
|
|
|
ws1 = Workspace(Resolver(), dst_path1) |
|
627
|
|
|
ws2 = Workspace(Resolver(), dst_path2) |
|
628
|
|
|
|
|
629
|
|
|
# assert number of files before |
|
630
|
|
|
assert len(ws1.mets.find_all_files()) == 6 |
|
631
|
|
|
assert len(ws2.mets.find_all_files()) == 35 |
|
632
|
|
|
|
|
633
|
|
|
# act |
|
634
|
|
|
ws1.merge(ws2, overwrite=True) |
|
635
|
|
|
|
|
636
|
|
|
# assert |
|
637
|
|
|
assert len(ws1.mets.find_all_files()) == 41 |
|
638
|
|
|
assert exists(join(dst_path1, 'OCR-D-IMG/INPUT_0017.tif')) |
|
639
|
|
|
|
|
640
|
|
|
def test_merge_no_copy_files(tmp_path): |
|
641
|
|
|
|
|
642
|
|
|
# arrange |
|
643
|
|
|
dst_path1 = tmp_path / 'ws1' |
|
644
|
|
|
dst_path1.mkdir() |
|
645
|
|
|
dst_path2 = dst_path1 / 'ws2' |
|
646
|
|
|
dst_path2.mkdir() |
|
647
|
|
|
|
|
648
|
|
|
ws1 = Resolver().workspace_from_nothing(directory=dst_path1) |
|
649
|
|
|
ws2 = Resolver().workspace_from_nothing(directory=dst_path2) |
|
650
|
|
|
|
|
651
|
|
|
ws2.add_file('GRP2', page_id='p01', mimetype='text/plain', file_id='f1', local_filename='GRP2/f1', content='ws2') |
|
652
|
|
|
|
|
653
|
|
|
ws1.merge(ws2, copy_files=False, fileId_mapping={'f1': 'f1_copy_files'}) |
|
654
|
|
|
|
|
655
|
|
|
assert next(ws1.mets.find_files(ID='f1_copy_files')).local_filename == 'ws2/GRP2/f1' |
|
656
|
|
|
|
|
657
|
|
|
with pytest.raises(FileExistsError): |
|
658
|
|
|
ws1.merge(ws2, copy_files=True, fileId_mapping={'f1': 'f1_copy_files'}) |
|
659
|
|
|
ws1.merge(ws2, copy_files=True, fileId_mapping={'f1': 'f1_copy_files'}, force=True) |
|
660
|
|
|
assert next(ws1.mets.find_files(ID='f1_copy_files')).local_filename == 'GRP2/f1' |
|
661
|
|
|
|
|
662
|
|
|
def test_merge_overwrite(tmp_path): |
|
663
|
|
|
# arrange |
|
664
|
|
|
dst_path1 = tmp_path / 'ws1' |
|
665
|
|
|
dst_path1.mkdir() |
|
666
|
|
|
dst_path2 = dst_path1 / 'ws2' |
|
667
|
|
|
dst_path2.mkdir() |
|
668
|
|
|
|
|
669
|
|
|
ws1 = Resolver().workspace_from_nothing(directory=dst_path1) |
|
670
|
|
|
ws2 = Resolver().workspace_from_nothing(directory=dst_path2) |
|
671
|
|
|
|
|
672
|
|
|
with pytest.raises(Exception) as exc: |
|
673
|
|
|
ws1.add_file('X', page_id='X', mimetype='X', file_id='id123', local_filename='X/X', content='ws1') |
|
674
|
|
|
ws2.add_file('X', page_id='X', mimetype='X', file_id='id456', local_filename='X/X', content='ws2') |
|
675
|
|
|
ws1.merge(ws2) |
|
676
|
|
|
assert "would overwrite" == str(exc.value) |
|
677
|
|
|
|
|
678
|
|
|
def test_merge_with_filter(plain_workspace, tmp_path): |
|
679
|
|
|
# arrange |
|
680
|
|
|
page_id1, file_id1, file_grp1 = 'page1', 'ID1', 'GRP1' |
|
681
|
|
|
plain_workspace.add_file(file_grp1, file_id='ID1', mimetype='image/tiff', page_id='page1') |
|
682
|
|
|
|
|
683
|
|
|
dst_path2 = tmp_path / 'foo' |
|
684
|
|
|
resolver = Resolver() |
|
685
|
|
|
ws2 = resolver.workspace_from_nothing(directory=dst_path2) |
|
686
|
|
|
page_id2, file_id2, file_grp2 = 'page2', 'ID2', 'GRP2' |
|
687
|
|
|
ws2.add_file('GRP2', file_id=file_id2, mimetype='image/tiff', page_id=page_id2, url='bar') |
|
688
|
|
|
ws2.add_file('GRP2', file_id='ID2-2', mimetype='image/tiff', page_id='page3', url='bar') |
|
689
|
|
|
|
|
690
|
|
|
# act |
|
691
|
|
|
plain_workspace.merge(ws2, copy_files=False, page_id=page_id2, file_id=file_id2, |
|
692
|
|
|
file_grp=file_grp2, filegrp_mapping={file_grp2: file_grp1}) |
|
693
|
|
|
|
|
694
|
|
|
# assert: |
|
695
|
|
|
files = list(plain_workspace.find_files()) |
|
696
|
|
|
assert len(files) == 2 |
|
697
|
|
|
|
|
698
|
|
|
for f in files: |
|
699
|
|
|
assert f.fileGrp == file_grp1 |
|
700
|
|
|
assert f.pageId in [page_id1, page_id2] |
|
701
|
|
|
assert f.ID in [file_id1, file_id2] |
|
702
|
|
|
|
|
703
|
|
|
def test_merge_force(plain_workspace, tmp_path): |
|
704
|
|
|
resolver = Resolver() |
|
705
|
|
|
|
|
706
|
|
|
# target ws |
|
707
|
|
|
page_id1, file_id1, file_grp1 = 'page1', 'ID1', 'GRP1' |
|
708
|
|
|
plain_workspace.add_file(file_grp1, file_id=file_id1, mimetype='image/tiff', page_id=page_id1) |
|
709
|
|
|
|
|
710
|
|
|
# source ws |
|
711
|
|
|
dst_path2 = tmp_path / 'foo' |
|
712
|
|
|
ws2 = resolver.workspace_from_nothing(directory=dst_path2) |
|
713
|
|
|
page_id2, file_id2, file_grp2 = 'page1', 'ID1', 'GRP1' |
|
714
|
|
|
ws2.add_file(file_grp2, file_id=file_id2, mimetype='image/tiff', page_id=page_id2, url='bar') |
|
715
|
|
|
|
|
716
|
|
|
# fails because force is false |
|
717
|
|
|
with pytest.raises(Exception) as fn_exc: |
|
718
|
|
|
plain_workspace.merge(ws2, force=False) |
|
719
|
|
|
|
|
720
|
|
|
# works because force overrides ID clash |
|
721
|
|
|
plain_workspace.merge(ws2, force=True) |
|
722
|
|
|
|
|
723
|
|
|
files = list(plain_workspace.find_files()) |
|
724
|
|
|
assert len(files) == 1 |
|
725
|
|
|
|
|
726
|
|
|
@pytest.fixture(name='workspace_metsDocumentID') |
|
727
|
|
|
def _fixture_metsDocumentID(tmp_path): |
|
728
|
|
|
resolver = Resolver() |
|
729
|
|
|
mets_content = (Path(__file__).parent / "data/mets-with-metsDocumentID.xml").read_text() |
|
730
|
|
|
with open(tmp_path / 'mets.xml', 'w', encoding='utf-8') as f: |
|
731
|
|
|
f.write(mets_content) |
|
732
|
|
|
yield Workspace(Resolver, directory=tmp_path) |
|
733
|
|
|
|
|
734
|
|
|
def test_agent_before_metsDocumentID(workspace_metsDocumentID): |
|
735
|
|
|
report = WorkspaceValidator.validate(Resolver(), mets_url=workspace_metsDocumentID.mets_target) |
|
736
|
|
|
assert report.is_valid |
|
737
|
|
|
workspace_metsDocumentID.mets.add_agent('foo bar v0.0.1', 'OTHER', 'OTHER', 'OTHER') |
|
738
|
|
|
workspace_metsDocumentID.save_mets() |
|
739
|
|
|
report = WorkspaceValidator.validate(Resolver(), mets_url=workspace_metsDocumentID.mets_target) |
|
740
|
|
|
print(report.errors) |
|
741
|
|
|
assert report.is_valid |
|
742
|
|
|
|
|
743
|
|
|
if __name__ == '__main__': |
|
744
|
|
|
main(__file__) |
|
745
|
|
|
|
OCR-D /
core
