def test_alternative_image_additions():
pcgts = PcGtsType(pcGtsId="foo")
assert pcgts.pcGtsId == 'foo'
# act
# Page/AlternativeImage
page = PageType()
pcgts.set_Page(page)
page.add_AlternativeImage(AlternativeImageType())
# TextRegion/AlternativeImage
region = TextRegionType()
page.add_TextRegion(region)
region.add_AlternativeImage(AlternativeImageType())
# TextLine/AlternativeImage
line = TextLineType()
region.add_TextLine(line)
line.add_AlternativeImage(AlternativeImageType())
# Word/AlternativeImage
word = WordType()
line.add_Word(word)
word.add_AlternativeImage(AlternativeImageType())
# Glyph/AlternativeImage
glyph = GlyphType()
word.add_Glyph(glyph)
glyph.add_AlternativeImage(AlternativeImageType())
def page_from_image(input_file):
"""
Create `OcrdPage </../../ocrd_models/ocrd_models.ocrd_page.html>`_
from an `OcrdFile </../../ocrd_models/ocrd_models.ocrd_file.html>`_
representing an image (i.e. should have ``mimetype`` starting with ``image/``).
Arguments:
* input_file (OcrdFile):
"""
if not input_file.local_filename:
raise ValueError("input_file must have 'local_filename' property")
if not Path(input_file.local_filename).exists():
raise FileNotFoundError("File not found: '%s' (%s)" %
(input_file.local_filename, input_file))
exif = exif_from_filename(input_file.local_filename)
now = datetime.now()
return PcGtsType(
Metadata=MetadataType(Creator="OCR-D/core %s" % VERSION,
Created=now,
LastChange=now),
Page=PageType(
imageWidth=exif.width,
imageHeight=exif.height,
# XXX brittle
imageFilename=input_file.url
if input_file.url is not None else input_file.local_filename))
def page_from_image(input_file):
"""
Create `OcrdPage </../../ocrd_models/ocrd_models.ocrd_page.html>`_
from an `OcrdFile </../../ocrd_models/ocrd_models.ocrd_file.html>`_
representing an image (i.e. should have ``mimetype`` starting with ``image/``).
Arguments:
* input_file (OcrdFile):
"""
if input_file.local_filename is None:
raise Exception("input_file must have 'local_filename' property")
exif = exif_from_filename(input_file.local_filename)
now = datetime.now()
return PcGtsType(
Metadata=MetadataType(
Creator="OCR-D/core %s" % VERSION,
Created=now,
LastChange=now
),
Page=PageType(
imageWidth=exif.width,
imageHeight=exif.height,
# XXX brittle
imageFilename=input_file.url if input_file.url is not None else 'file://' + input_file.local_filename
)
)
def render_all(self, pc_gts: PcGtsType) -> None:
page: PageType = pc_gts.get_Page()
self.render_type(page.get_PrintSpace())
self.render_type(page.get_Border())
def region_priority(region):
# often, regions overlap; since we don't alpha-composite,
# we should provide a useful default priority of what is
# rendered last (and thus unoccluded); document order is
# not helpful, because it depends on the workflow or geometry
# instead of visibility concerns; thus, ensure that:
# - SeparatorRegion is top-most
# - TextRegion is next
# - all others are next
if isinstance(region, SeparatorRegionType):
return 0
if isinstance(region, TextRegionType):
return -1
return -2
for region_ds in sorted(page.get_AllRegions(), key=region_priority):
self.render_type(region_ds)
if self.features & Feature.ORDER:
last_point: Optional[Point] = None
for region_ds in page.get_AllRegions(order='reading-order-only'):
region = self.region_factory.create(region_ds)
new_point = region.poly.representative_point()
if last_point:
self.operations.append(
ArrowOperation(last_point,
new_point,
color='#FF0000CF'))
last_point = new_point
def create_page_xml(imageFilename, height, width):
now = datetime.now()
pcgts = PcGtsType(Metadata=MetadataType(Creator='SBB_QURATOR',
Created=now,
LastChange=now),
Page=PageType(imageWidth=str(width),
imageHeight=str(height),
imageFilename=imageFilename,
readingDirection='left-to-right',
textLineOrder='top-to-bottom'))
return pcgts
def test_alternativeImage(self):
pcgts = PcGtsType(pcGtsId="foo")
self.assertEqual(pcgts.pcGtsId, 'foo')
# Page/AlternativeImage
page = PageType()
pcgts.set_Page(page)
page.add_AlternativeImage(AlternativeImageType())
# TextRegion/AlternativeImage
region = TextRegionType()
page.add_TextRegion(region)
region.add_AlternativeImage(AlternativeImageType())
# TextLine/AlternativeImage
line = TextLineType()
region.add_TextLine(line)
line.add_AlternativeImage(AlternativeImageType())
# Word/AlternativeImage
word = WordType()
line.add_Word(word)
word.add_AlternativeImage(AlternativeImageType())
# Glyph/AlternativeImage
glyph = GlyphType()
word.add_Glyph(glyph)
glyph.add_AlternativeImage(AlternativeImageType())
def page_from_image(input_file, with_tree=False):
"""
Create :py:class:`~ocrd_models.ocrd_page.OcrdPage`
from an :py:class:`~ocrd_models.ocrd_file.OcrdFile`
representing an image (i.e. should have ``@mimetype`` starting with ``image/``).
Arguments:
input_file (:py:class:`~ocrd_models.ocrd_file.OcrdFile`): file to open \
and produce a PAGE DOM for
Keyword arguments:
with_tree (boolean): whether to return XML node tree, element-node mapping \
and reverse mapping, too (cf. :py:func:`ocrd_models.ocrd_page.parseEtree`)
"""
if not input_file.local_filename:
raise ValueError("input_file must have 'local_filename' property")
if not Path(input_file.local_filename).exists():
raise FileNotFoundError("File not found: '%s' (%s)" %
(input_file.local_filename, input_file))
exif = exif_from_filename(input_file.local_filename)
now = datetime.now()
pcgts = PcGtsType(
Metadata=MetadataType(Creator="OCR-D/core %s" % VERSION,
Created=now,
LastChange=now),
Page=PageType(
imageWidth=exif.width,
imageHeight=exif.height,
# XXX brittle
imageFilename=input_file.url
if input_file.url is not None else input_file.local_filename),
pcGtsId=input_file.ID)
if not with_tree:
return pcgts
mapping = dict()
etree = pcgts.to_etree(mapping_=mapping)
revmap = dict(((node, element) for element, node in mapping.items()))
return pcgts, etree, mapping, revmap