qurator-spk/eynollah
1
0
Fork 0
mirror of https://github.com/qurator-spk/eynollah.git synced 2025-06-11 05:09:53 +02:00
Code Issues Projects Releases Wiki Activity

allow passing pcgts to eynollah and writer

Browse source
This commit is contained in:
Konstantin Baierer 2021-04-14 10:38:07 +02:00
parent 2bc34891a5
commit 8c4e9b6068
3 changed files with 12 additions and 10 deletions

4
qurator/eynollah/eynollah.py
View file

@ -95,6 +95,7 @@ class Eynollah:
headers_off=False, headers_off=False,
override_dpi=None, override_dpi=None,
logger=None, logger=None,
pcgts=None,
): ):
self.image_filename = image_filename self.image_filename = image_filename
self.dir_out = dir_out self.dir_out = dir_out
@ -117,7 +118,8 @@ class Eynollah:
self.writer = EynollahXmlWriter( self.writer = EynollahXmlWriter(
dir_out=self.dir_out, dir_out=self.dir_out,
image_filename=self.image_filename, image_filename=self.image_filename,
curved_line=self.curved_line) curved_line=self.curved_line,
pcgts=pcgts)
self.logger = logger if logger else getLogger('eynollah') self.logger = logger if logger else getLogger('eynollah')
self.dir_models = dir_models self.dir_models = dir_models

14
qurator/eynollah/processor.py
View file

@ -30,10 +30,10 @@ class EynollahProcessor(Processor):
assert_file_grp_cardinality(self.output_file_grp, 1) assert_file_grp_cardinality(self.output_file_grp, 1)
for n, input_file in enumerate(self.input_files): for n, input_file in enumerate(self.input_files):
page_id = input_file.pageId or input_file.ID page_id = input_file.pageId or input_file.ID
LOG.info("INPUT FILE %s / %s ", page_id, len(self.input_files)) LOG.info("INPUT FILE %s (%d/%d) ", page_id, n + 1, len(self.input_files))
pcgts_in = page_from_file(self.workspace.download_file(input_file)) pcgts = page_from_file(self.workspace.download_file(input_file))
self.add_metadata(pcgts_in) self.add_metadata(pcgts)
page = pcgts_in.get_Page() page = pcgts.get_Page()
page_image, _, _ = self.workspace.image_from_page(page, page_id, feature_filter='binarized') page_image, _, _ = self.workspace.image_from_page(page, page_id, feature_filter='binarized')
file_id = make_file_id(input_file, self.output_file_grp) file_id = make_file_id(input_file, self.output_file_grp)
with NamedTemporaryFile(buffering=0, suffix='.tif') as f: with NamedTemporaryFile(buffering=0, suffix='.tif') as f:
@ -47,13 +47,13 @@ class EynollahProcessor(Processor):
'headers_off': self.parameter['headers_off'], 'headers_off': self.parameter['headers_off'],
'override_dpi': self.parameter['dpi'] if self.parameter['dpi'] > 0 else None, 'override_dpi': self.parameter['dpi'] if self.parameter['dpi'] > 0 else None,
'logger': LOG, 'logger': LOG,
'pcgts': pcgts,
'image_filename': f.name} 'image_filename': f.name}
pcgts_out = Eynollah(**eynollah_kwargs).run() Eynollah(**eynollah_kwargs).run()
pcgts_out.get_Page().imageFilename = pcgts_in.get_Page().imageFilename
self.workspace.add_file( self.workspace.add_file(
ID=file_id, ID=file_id,
file_grp=self.output_file_grp, file_grp=self.output_file_grp,
pageId=page_id, pageId=page_id,
mimetype=MIMETYPE_PAGE, mimetype=MIMETYPE_PAGE,
local_filename=join(self.output_file_grp, file_id) + '.xml', local_filename=join(self.output_file_grp, file_id) + '.xml',
content=to_xml(pcgts_out)) content=to_xml(pcgts))

4
qurator/eynollah/writer.py
View file

@ -141,7 +141,7 @@ class EynollahXmlWriter():
self.logger.debug('enter build_pagexml_no_full_layout') self.logger.debug('enter build_pagexml_no_full_layout')
# create the file structure # create the file structure
pcgts = create_page_xml(self.image_filename, self.height_org, self.width_org) pcgts = self.pcgts if self.pcgts else create_page_xml(self.image_filename, self.height_org, self.width_org)
page = pcgts.get_Page() page = pcgts.get_Page()
page.set_Border(BorderType(Coords=CoordsType(points=self.calculate_page_coords(cont_page)))) page.set_Border(BorderType(Coords=CoordsType(points=self.calculate_page_coords(cont_page))))
@ -181,7 +181,7 @@ class EynollahXmlWriter():
self.logger.debug('enter build_pagexml_full_layout') self.logger.debug('enter build_pagexml_full_layout')
# create the file structure # create the file structure
pcgts = create_page_xml(self.image_filename, self.height_org, self.width_org) pcgts = self.pcgts if self.pcgts else create_page_xml(self.image_filename, self.height_org, self.width_org)
page = pcgts.get_Page() page = pcgts.get_Page()
page.set_Border(BorderType(Coords=CoordsType(points=self.calculate_page_coords(cont_page)))) page.set_Border(BorderType(Coords=CoordsType(points=self.calculate_page_coords(cont_page))))