qurator-spk/eynollah
1
0
Fork 0
mirror of https://github.com/qurator-spk/eynollah.git synced 2026-05-01 03:32:00 +02:00
Code Issues Projects Releases Wiki Activity

writer: also annotate col-classifier result…

Browse source 
both notations:
- in `/PcGts/Page/@custom` (CSS-style)
- in `/PcGts/Metadata/Comment` (qurator-style)
This commit is contained in:
Robert Sachunsky 2026-04-23 23:46:23 +02:00
parent e3720d6623
commit 9723dfeb73
2 changed files with 9 additions and 0 deletions
Download patch file Download diff file Expand all files Collapse all files

4
src/eynollah/eynollah.py
View file

@ -2253,6 +2253,7 @@ class Eynollah:
conf_contours_textregions =[0] conf_contours_textregions =[0]
pcgts = writer.build_pagexml_no_full_layout( pcgts = writer.build_pagexml_no_full_layout(
num_col=num_col_classifier,
found_polygons_text_region=cont_page, found_polygons_text_region=cont_page,
page_coord=page_coord, page_coord=page_coord,
order_of_texts=order_text_new, order_of_texts=order_text_new,
@ -2323,6 +2324,7 @@ class Eynollah:
self.logger.info("No columns detected - generating empty PAGE-XML") self.logger.info("No columns detected - generating empty PAGE-XML")
pcgts = writer.build_pagexml_no_full_layout( pcgts = writer.build_pagexml_no_full_layout(
num_col=0,
found_polygons_text_region=[], found_polygons_text_region=[],
page_coord=page_coord, page_coord=page_coord,
page_slope=slope_deskew, page_slope=slope_deskew,
@ -2612,6 +2614,7 @@ class Eynollah:
if self.full_layout: if self.full_layout:
pcgts = writer.build_pagexml_full_layout( pcgts = writer.build_pagexml_full_layout(
num_col=num_col_classifier,
found_polygons_text_region=polygons_of_textregions, found_polygons_text_region=polygons_of_textregions,
found_polygons_text_region_h=polygons_of_textregions_h, found_polygons_text_region_h=polygons_of_textregions_h,
page_coord=page_coord, page_coord=page_coord,
@ -2642,6 +2645,7 @@ class Eynollah:
) )
else: else:
pcgts = writer.build_pagexml_no_full_layout( pcgts = writer.build_pagexml_no_full_layout(
num_col=num_col_classifier,
found_polygons_text_region=polygons_of_textregions, found_polygons_text_region=polygons_of_textregions,
page_coord=page_coord, page_coord=page_coord,
page_slope=slope_deskew, page_slope=slope_deskew,

5
src/eynollah/writer.py
View file

@ -71,6 +71,7 @@ class EynollahXmlWriter:
def build_pagexml_no_full_layout( def build_pagexml_no_full_layout(
self, self,
*, *,
num_col,
found_polygons_text_region, found_polygons_text_region,
page_coord, page_coord,
page_slope, page_slope,
@ -98,6 +99,7 @@ class EynollahXmlWriter:
conf_tables=None, conf_tables=None,
): ):
return self.build_pagexml_full_layout( return self.build_pagexml_full_layout(
num_col=num_col,
found_polygons_text_region=found_polygons_text_region, found_polygons_text_region=found_polygons_text_region,
found_polygons_text_region_h=[], found_polygons_text_region_h=[],
page_coord=page_coord, page_coord=page_coord,
@ -131,6 +133,7 @@ class EynollahXmlWriter:
def build_pagexml_full_layout( def build_pagexml_full_layout(
self, self,
*, *,
num_col,
found_polygons_text_region, found_polygons_text_region,
found_polygons_text_region_h, found_polygons_text_region_h,
page_coord, page_coord,
@ -171,6 +174,8 @@ class EynollahXmlWriter:
pcgts = self.pcgts if self.pcgts else create_page_xml( pcgts = self.pcgts if self.pcgts else create_page_xml(
self.image_filename, self.image_height, self.image_width) self.image_filename, self.image_height, self.image_width)
page = pcgts.get_Page() page = pcgts.get_Page()
pcgts.Metadata.Comments = "num_col %d" % num_col
page.set_custom('layout {num_col:%d;} ' % num_col)
page.set_orientation(-page_slope) page.set_orientation(-page_slope)
if len(cont_page): if len(cont_page):
page.set_Border(BorderType(Coords=CoordsType(points=self.calculate_points(cont_page[0])))) page.set_Border(BorderType(Coords=CoordsType(points=self.calculate_points(cont_page[0]))))