def __init__(self, document: Document):
"""
Initializes the underlying ListStore and fills it with a row for each page, then start the lazy loading
The actual image and data loading happens in _load_row
"""
columns = {
self.COLUMN_PAGE_ID: str,
self.COLUMN_TOOLTIP: str,
self.COLUMN_FILENAME: str,
self.COLUMN_THUMB: GdkPixbuf.Pixbuf,
self.COLUMN_ORDER: int
# self.COLUMN_HASH: str file hash = filename + modified_time (gets added by LazyLoadingListStore)
}
super().__init__(*(columns.values()), init_row=self._init_row, load_row=self._load_row, hash_row=self._hash_row)
self.document = document
self.pixbufs: Dict[str, GdkPixbuf.Pixbuf] = {
icon_name: GdkPixbuf.Pixbuf.new_from_resource(
'/org/readmachine/ocrd-browser/icons/{}.png'.format(icon_name)
) for icon_name in ['page-loading', 'page-missing']
}
# TODO: make file_group selectable, see https://github.com/hnesk/browse-ocrd/issues/7#issuecomment-707851109
self.file_group = document.get_default_image_group(SETTINGS.file_groups.preferred_images)
file_lookup = document.get_image_paths(self.file_group)
order = count(start=1)
for page_id in self.document.page_ids:
file = file_lookup[page_id]
self.append((page_id, '', str(file) if file else None, None, next(order)))
GLib.timeout_add(10, self.start_loading)
def test_get_mime_types(self):
doc = Document.load(self.path)
self.assertEqual(
{
'application/vnd.prima.page+xml', 'image/tiff',
'application/alto+xml'
}, doc.mime_types)
def setUp(self):
self.root = Gtk.Box(name='container')
self.win = MagicMock(spec=MainWindow)
self.win.document = Document.create(self.win)
self.win.current_page_id = None
self.vm = ViewManager(self.win, self.root)
def test_path_with_spaces(self):
doc = Document.load(
(TEST_BASE_PATH /
'example/workspaces/heavy quoting/mets.xml').as_uri())
page = doc.page_for_id('PHYS_0017', 'OCR-D-GT-PAGE')
image = doc.workspace.image_from_page(page.page, 'PHYS_0017')
# Assert no exceptions happened and a sensible return value
self.assertGreater(image[0].height, 100)
def test_missing_image(self):
path = TEST_BASE_PATH / 'example/workspaces/kant_aufklaerung_1784_missing_image/mets.xml'
uri = path.as_uri()
doc = Document.load(uri)
page = doc.page_for_id('PHYS_0017', 'OCR-D-GT-PAGE')
image, info, exif = page.get_image(feature_selector='',
feature_filter='binarized')
# Assert no exceptions happened and no image returned
self.assertIsNone(image)
def test_save(self):
doc = Document.clone(self.path)
with TemporaryDirectory(prefix='browse-ocrd-tests') as directory:
saved_mets = directory + '/mets.xml'
doc.save_as(saved_mets)
saved = Document.load(saved_mets)
self.assertEqual(doc.file_groups, saved.file_groups)
self.assertEqual(doc.page_ids, saved.page_ids)
self.assertEqual(doc.workspace.mets.unique_identifier,
saved.workspace.mets.unique_identifier)
for page_id in doc.page_ids:
for file_group, mime in doc.file_groups_and_mimetypes:
original_file = doc.files_for_page_id(
page_id, file_group, mime)[0]
saved_file = saved.files_for_page_id(
page_id, file_group, mime)[0]
self.assertEqual(original_file, saved_file)
def test_path_ocrd_file(self):
doc = Document.load(self.path)
image_file = list(
doc.workspace.mets.find_files(pageId='PHYS_0017',
fileGrp='OCR-D-IMG'))[0]
self.assertEqual(
ASSETS_PATH /
'kant_aufklaerung_1784/data/OCR-D-IMG/INPUT_0017.tif',
doc.path(image_file))
def test_clone(self):
doc = Document.clone(self.path)
self.assertIn('browse-ocrd-clone-', doc.workspace.directory)
self.assertEqual(str(self.path), doc.baseurl_mets)
original_files = self.path.parent.rglob('*.*')
cloned_files = Path(doc.workspace.directory).rglob('*.*')
for original, cloned in zip(sorted(original_files),
sorted(cloned_files)):
self.assertEqual(original.read_bytes(), cloned.read_bytes())
def _open(self, uri: str) -> None:
# noinspection PyTypeChecker
self.document = Document.load(uri, emitter=self.emit)
self.page_list.set_document(self.document)
self.view_manager.set_document(self.document)
self.update_ui()
if len(self.document.page_ids):
self.on_page_activated(None, self.document.page_ids[0])
def test_page_for_id_with_multiple_images_for_page_and_fileGrp(self):
"""
returns first image and warns
"""
doc = Document.load(ASSETS_PATH /
'kant_aufklaerung_1784-complex/data/mets.xml')
# with self.assertLogs('ocrd_browser.model.document', level='WARNING') as log_watch:
page = doc.page_for_id('PHYS_0017', 'OCR-D-IMG-CLIP')
self.assertIsInstance(page, Page)
self.assertIsInstance(page.pc_gts, PcGtsType)
def test_can_call_get_image_if_supported(self):
page: LazyPage = Document.load(ASSETS_PATH / 'kant_aufklaerung_1784-binarized/data/mets.xml').page_for_id('P_0017', 'OCR-D-GT-WORD')
if IMAGE_FROM_PAGE_FILENAME_SUPPORT:
image_by_feature, _, _ = page.get_image(feature_selector={'binarized'}, feature_filter={'cropped'})
image_by_filename, _, _ = page.get_image(filename='OCR-D-IMG-BIN/BIN_0017.png', feature_filter={'cropped'})
self.assertEqual(image_by_feature, image_by_filename)
else:
try:
page.get_image(filename='OCR-D-GT-IMG-BIN/PAGE_2019.tif')
self.fail('IMAGE_FROM_PAGE_FILENAME_SUPPORT detected wrong')
except RuntimeError as e:
self.assertTrue(str(e).startswith('Parameter filename not supported in '))
def test_get_page_index(self):
doc = Document.load(self.path)
file_index = doc.get_file_index()
page17 = [
file for file in file_index.values()
if file.static_page_id == 'PHYS_0017'
]
alto = [
file for file in file_index.values()
if file.mimetype == 'application/alto+xml'
]
self.assertEqual(3, len(page17))
self.assertEqual(2, len(alto))
def test_page_for_id_with_multiple_images_for_page_and_fileGrp(self):
"""
returns first image and warns
"""
doc = Document.load(ASSETS_PATH /
'kant_aufklaerung_1784-complex/data/mets.xml')
with self.assertLogs('ocrd_browser.model.document',
level='WARNING') as log_watch:
page = doc.page_for_id('PHYS_0017', 'OCR-D-IMG-CLIP')
self.assertIsInstance(page, Page)
self.assertEqual(1, len(log_watch.records))
self.assertEqual(
"No PAGE-XML but 2 images for page 'PHYS_0017' in fileGrp 'OCR-D-IMG-CLIP'",
log_watch.records[0].msg)
def __init__(self, **kwargs: Any):
Gtk.ApplicationWindow.__init__(self, **kwargs)
# noinspection PyCallByClass,PyArgumentList
self.set_icon(
GdkPixbuf.Pixbuf.new_from_resource(
"/org/readmachine/ocrd-browser/icons/icon.png"))
self.view_manager = ViewManager(self, self.view_container)
self.current_page_id: Optional[str] = None
# noinspection PyTypeChecker
self.document = Document.create(emitter=self.emit)
self.actions = ActionRegistry(for_widget=self)
self.actions.create('close')
self.actions.create('goto_first')
self.actions.create('go_back')
self.actions.create('go_forward')
self.actions.create('goto_last')
self.actions.create('page_remove')
self.actions.create('page_properties')
self.actions.create('close_view', param_type=GLib.VariantType("s"))
self.actions.create('split_view', param_type=GLib.VariantType("(ssb)"))
self.actions.create('create_view', param_type=GLib.VariantType("s"))
self.actions.create('replace_view',
param_type=GLib.VariantType("(ss)"))
self.actions.create('toggle_edit_mode', state=GLib.Variant('b', False))
self.actions.create('save')
self.actions.create('save_as')
self.connect('delete-event', self.on_delete_event)
self.page_list = PagePreviewList(self.document)
self.page_list_scroller.add(self.page_list)
self.page_list.connect('page_activated', self.on_page_activated)
self.page_list.connect('pages_selected', self.on_pages_selected)
for id_, view in self.view_registry.get_view_options().items():
menu_item = Gtk.ModelButton(visible=True,
centered=False,
halign=Gtk.Align.FILL,
label=view,
hexpand=True)
menu_item.set_detailed_action_name(
'win.create_view("{}")'.format(id_))
self.view_menu_box.pack_start(menu_item, True, True, 0)
self.view_manager.set_root_view(ViewPage)
# self.view_manager.split(None, ViewPage, False)
self.update_ui()
def test_page_for_id_with_no_images_for_page_and_fileGrp(self):
"""
Issue #4: list index out of range on non-XML fileGrp
https://github.com/hnesk/browse-ocrd/issues/4
"""
doc = Document.load(ASSETS_PATH /
'kant_aufklaerung_1784-complex/data/mets.xml')
with self.assertLogs('ocrd_browser.model.document',
level='WARNING') as log_watch:
page = doc.page_for_id('PHYS_0020', 'OCR-D-IMG-CLIP')
self.assertIsNone(page)
self.assertEqual(1, len(log_watch.records))
self.assertEqual(
"No PAGE-XML and no image for page 'PHYS_0020' in fileGrp 'OCR-D-IMG-CLIP'",
log_watch.records[0].msg)
def test_page_for_id_with_nothing_for_page_and_fileGrp(self):
"""
Issue #4 again: This time for missing PAGE-XMLs
https://github.com/hnesk/browse-ocrd/issues/4
"""
doc = Document.load(
ASSETS_PATH /
'../example/workspaces/kant_aufklaerung_1784_missing_xml/mets.xml')
with self.assertLogs('ocrd_browser.model.document',
level='WARNING') as log_watch:
page = doc.page_for_id('PHYS_0020', 'OCR-D-GT-PAGE')
self.assertIsNone(page)
self.assertEqual(1, len(log_watch.records))
self.assertEqual(
"No PAGE-XML and no image for page 'PHYS_0020' in fileGrp 'OCR-D-GT-PAGE'",
log_watch.records[0].msg)
def test_path_path(self):
doc = Document.load(self.path)
self.assertEqual(
ASSETS_PATH / 'kant_aufklaerung_1784/data/OCR-D-DIR/lala.xml',
doc.path(Path('OCR-D-DIR/lala.xml')))
def test_path_string(self):
doc = Document.load(self.path)
self.assertEqual(ASSETS_PATH / 'kant_aufklaerung_1784/data/lala.xml',
doc.path('lala.xml'))
def test_get_default_image_group_with_missing_ocr_d_img(self):
doc = Document.load(
ASSETS_PATH / '../example/workspaces/no_ocrd_d_img_group/mets.xml')
file_group = doc.get_default_image_group()
self.assertEqual('OCR-D-IMG-PNG', file_group)
def test_get_default_image_group_no_preference(self):
doc = Document.load(ASSETS_PATH /
'kant_aufklaerung_1784-complex/data/mets.xml')
file_group = doc.get_default_image_group()
self.assertEqual('OCR-D-IMG', file_group)
def test_get_default_image_group(self):
doc = Document.load(ASSETS_PATH /
'kant_aufklaerung_1784-complex/data/mets.xml')
file_group = doc.get_default_image_group(
['OCR-D-IMG-BIN', 'OCR-D-IMG.*'])
self.assertEqual('OCR-D-IMG-BIN', file_group)
def test_delete(self):
doc = Document.clone(self.path)
doc.delete_page('PHYS_0017')
self.assertEqual(['PHYS_0020'], doc.page_ids)
def test_get_file_groups_and_mimetypes(self):
doc = Document.load(self.path)
expected = [('OCR-D-IMG', 'image/tiff'),
('OCR-D-GT-PAGE', 'application/vnd.prima.page+xml'),
('OCR-D-GT-ALTO', 'application/alto+xml')]
self.assertEqual(expected, doc.file_groups_and_mimetypes)
def test_modify_when_editable(self):
doc = Document.clone(self.path)
doc.reorder(['PHYS_0020', 'PHYS_0017'])
def test_modify_when_not_editable(self):
doc = Document.load(self.path)
with self.assertRaises(PermissionError):
doc.reorder(['PHYS_0020', 'PHYS_0017'])
def test_get_page_ids(self):
doc = Document.load(self.path)
self.assertEqual(['PHYS_0017', 'PHYS_0020'], doc.page_ids)
def test_derive_backup_directory(self):
self.assertEqual(
Path('/home/jk/.bak.important_project.20200813-184321'),
Document._derive_backup_directory(
Path('/home/jk/important_project'),
datetime(2020, 8, 13, 18, 43, 21)))
def test_reorder(self):
doc = Document.clone(self.path)
doc.reorder(['PHYS_0020', 'PHYS_0017'])
self.assertEqual(['PHYS_0020', 'PHYS_0017'], doc.page_ids)
def test_reorder_with_wrong_ids_raises_value_error(self):
doc = Document.clone(self.path)
with self.assertRaises(ValueError) as context:
doc.reorder(['PHYS_0021', 'PHYS_0017'])
self.assertIn('page_ids do not match', str(context.exception))
def test_get_image_paths(self):
doc = Document.load(self.path)
image_paths = doc.get_image_paths('OCR-D-IMG')
self.assertEqual(2, len(image_paths))
self.assertEqual('INPUT_0017.tif', image_paths['PHYS_0017'].name)
self.assertEqual('INPUT_0020.tif', image_paths['PHYS_0020'].name)
