From 8c4e9b6068b522e52cb722c65f7a1c548b0b995b Mon Sep 17 00:00:00 2001 From: Konstantin Baierer Date: Wed, 14 Apr 2021 10:38:07 +0200 Subject: [PATCH] allow passing pcgts to eynollah and writer --- qurator/eynollah/eynollah.py | 4 +++- qurator/eynollah/processor.py | 14 +++++++------- qurator/eynollah/writer.py | 4 ++-- 3 files changed, 12 insertions(+), 10 deletions(-) diff --git a/qurator/eynollah/eynollah.py b/qurator/eynollah/eynollah.py index d59fe89..c6b4096 100644 --- a/qurator/eynollah/eynollah.py +++ b/qurator/eynollah/eynollah.py @@ -95,6 +95,7 @@ class Eynollah: headers_off=False, override_dpi=None, logger=None, + pcgts=None, ): self.image_filename = image_filename self.dir_out = dir_out @@ -117,7 +118,8 @@ class Eynollah: self.writer = EynollahXmlWriter( dir_out=self.dir_out, 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.dir_models = dir_models diff --git a/qurator/eynollah/processor.py b/qurator/eynollah/processor.py index 07d7ab2..68da037 100644 --- a/qurator/eynollah/processor.py +++ b/qurator/eynollah/processor.py @@ -30,10 +30,10 @@ class EynollahProcessor(Processor): assert_file_grp_cardinality(self.output_file_grp, 1) for n, input_file in enumerate(self.input_files): page_id = input_file.pageId or input_file.ID - LOG.info("INPUT FILE %s / %s ", page_id, len(self.input_files)) - pcgts_in = page_from_file(self.workspace.download_file(input_file)) - self.add_metadata(pcgts_in) - page = pcgts_in.get_Page() + LOG.info("INPUT FILE %s (%d/%d) ", page_id, n + 1, len(self.input_files)) + pcgts = page_from_file(self.workspace.download_file(input_file)) + self.add_metadata(pcgts) + page = pcgts.get_Page() page_image, _, _ = self.workspace.image_from_page(page, page_id, feature_filter='binarized') file_id = make_file_id(input_file, self.output_file_grp) with NamedTemporaryFile(buffering=0, suffix='.tif') as f: @@ -47,13 +47,13 @@ class EynollahProcessor(Processor): 'headers_off': self.parameter['headers_off'], 'override_dpi': self.parameter['dpi'] if self.parameter['dpi'] > 0 else None, 'logger': LOG, + 'pcgts': pcgts, 'image_filename': f.name} - pcgts_out = Eynollah(**eynollah_kwargs).run() - pcgts_out.get_Page().imageFilename = pcgts_in.get_Page().imageFilename + Eynollah(**eynollah_kwargs).run() self.workspace.add_file( ID=file_id, file_grp=self.output_file_grp, pageId=page_id, mimetype=MIMETYPE_PAGE, local_filename=join(self.output_file_grp, file_id) + '.xml', - content=to_xml(pcgts_out)) + content=to_xml(pcgts)) diff --git a/qurator/eynollah/writer.py b/qurator/eynollah/writer.py index 7ddfea5..5b42043 100644 --- a/qurator/eynollah/writer.py +++ b/qurator/eynollah/writer.py @@ -141,7 +141,7 @@ class EynollahXmlWriter(): self.logger.debug('enter build_pagexml_no_full_layout') # 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.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') # 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.set_Border(BorderType(Coords=CoordsType(points=self.calculate_page_coords(cont_page))))