qurator-spk/eynollah
1
0
Fork 0
mirror of https://github.com/qurator-spk/eynollah.git synced 2025-06-09 04:09:54 +02:00
Code Issues Projects Releases Wiki Activity

id_of_marginalia still necessary

Browse source
This commit is contained in:
Konstantin Baierer 2021-03-02 17:47:06 +01:00
parent 56b688befe
commit d95fcf14c0
1 changed files with 7 additions and 5 deletions
Download patch file Download diff file Expand all files Collapse all files

12
qurator/eynollah/writer.py
View file

@ -140,8 +140,9 @@ class EynollahXmlWriter():
counter_textregions = EynollahIdCounter() counter_textregions = EynollahIdCounter()
counter_marginals = EynollahIdCounter(region_idx=len(order_of_texts)) counter_marginals = EynollahIdCounter(region_idx=len(order_of_texts))
id_of_marginalia = [counter_marginals.next_region_id for _ in found_polygons_marginals]
if len(found_polygons_text_region) > 0: if len(found_polygons_text_region) > 0:
xml_reading_order(page, order_of_texts, id_of_texts, [counter_marginals.next_region_id for _ in found_polygons_marginals]) xml_reading_order(page, order_of_texts, id_of_texts, id_of_marginalia)
for mm in range(len(found_polygons_text_region)): for mm in range(len(found_polygons_text_region)):
textregion = ET.SubElement(page, 'TextRegion') textregion = ET.SubElement(page, 'TextRegion')
@ -152,12 +153,12 @@ class EynollahXmlWriter():
self.serialize_lines_in_region(textregion, all_found_texline_polygons, mm, page_coord, all_box_coord, slopes, counter_textregions) self.serialize_lines_in_region(textregion, all_found_texline_polygons, mm, page_coord, all_box_coord, slopes, counter_textregions)
add_textequiv(textregion) add_textequiv(textregion)
for idx_marginal, _ in enumerate(found_polygons_marginals): for idx_marginal, marginal_polygon in enumerate(found_polygons_marginals):
marginal = ET.SubElement(page, 'TextRegion') marginal = ET.SubElement(page, 'TextRegion')
marginal.set('id', id_of_marginalia[idx_marginal]) marginal.set('id', id_of_marginalia[idx_marginal])
marginal.set('type', 'marginalia') marginal.set('type', 'marginalia')
coord_text = ET.SubElement(marginal, 'Coords') coord_text = ET.SubElement(marginal, 'Coords')
coord_text.set('points', self.calculate_polygon_coords(found_polygons_marginals[mm], page_coord)) coord_text.set('points', self.calculate_polygon_coords(marginal_polygon, page_coord))
self.serialize_lines_in_marginal(marginal, all_found_texline_polygons_marginals, mm, page_coord, all_box_coord_marginals, slopes_marginals, counter_textregions) self.serialize_lines_in_marginal(marginal, all_found_texline_polygons_marginals, mm, page_coord, all_box_coord_marginals, slopes_marginals, counter_textregions)
for mm in range(len(found_polygons_text_region_img)): for mm in range(len(found_polygons_text_region_img)):
@ -183,10 +184,11 @@ class EynollahXmlWriter():
coord_page = ET.SubElement(page_print_sub, "Coords") coord_page = ET.SubElement(page_print_sub, "Coords")
coord_page.set('points', self.calculate_page_coords(cont_page)) coord_page.set('points', self.calculate_page_coords(cont_page))
counter_marginals = EynollahIdCounter(region_idx=len(order_of_texts))
counter_textregions = EynollahIdCounter() counter_textregions = EynollahIdCounter()
counter_marginals = EynollahIdCounter(region_idx=len(order_of_texts))
xml_reading_order(page, order_of_texts, id_of_texts, [counter_marginals.next_region_id for _ in found_polygons_marginals]) id_of_marginalia = [counter_marginals.next_region_id for _ in found_polygons_marginals]
xml_reading_order(page, order_of_texts, id_of_texts, id_of_marginalia)
for mm in range(len(found_polygons_text_region)): for mm in range(len(found_polygons_text_region)):
textregion=ET.SubElement(page, 'TextRegion') textregion=ET.SubElement(page, 'TextRegion')