qurator-spk/eynollah
1
0
Fork 0
mirror of https://github.com/qurator-spk/eynollah.git synced 2025-06-10 04:39:54 +02:00
Code Issues Projects Releases Wiki Activity

keep id_indexer_l thru serialize_lines_in_region

Browse source
This commit is contained in:
Konstantin Baierer 2021-01-28 18:39:16 +01:00
parent e11d9adfea
commit 4d31a21c3e
1 changed files with 6 additions and 5 deletions
Download patch file Download diff file Expand all files Collapse all files

9
sbb_newspapers_org_image/eynollah.py
View file

@ -1404,11 +1404,11 @@ class eynollah:
poly.put(poly_sub) poly.put(poly_sub)
box_sub.put(boxes_sub_new) box_sub.put(boxes_sub_new)
def serialize_lines_in_region(self, textregion, all_found_texline_polygons, region_idx, page_coord, all_box_coord, slopes): def serialize_lines_in_region(self, textregion, all_found_texline_polygons, region_idx, page_coord, all_box_coord, slopes, id_indexer_l):
for j in range(len(all_found_texline_polygons[region_idx])): for j in range(len(all_found_texline_polygons[region_idx])):
textline=ET.SubElement(textregion, 'TextLine') textline=ET.SubElement(textregion, 'TextLine')
textline.set('id','l'+str(id_indexer_l)) textline.set('id','l'+str(id_indexer_l))
id_indexer_l+=1 id_indexer_l += 1
coord = ET.SubElement(textline, 'Coords') coord = ET.SubElement(textline, 'Coords')
texteq = ET.SubElement(textline, 'TextEquiv') texteq = ET.SubElement(textline, 'TextEquiv')
uni = ET.SubElement(texteq, 'Unicode') uni = ET.SubElement(texteq, 'Unicode')
@ -1480,6 +1480,7 @@ class eynollah:
if l<(len(all_found_texline_polygons[region_idx][j])-1): if l<(len(all_found_texline_polygons[region_idx][j])-1):
points_co=points_co+' ' points_co=points_co+' '
coord.set('points',points_co) coord.set('points',points_co)
return id_indexer_l
def write_into_page_xml_full(self, contours, contours_h, page_coord, dir_of_image, order_of_texts, id_of_texts, all_found_texline_polygons, all_found_texline_polygons_h, all_box_coord, all_box_coord_h, found_polygons_text_region_img, found_polygons_tables, found_polygons_drop_capitals, found_polygons_marginals, all_found_texline_polygons_marginals, all_box_coord_marginals, slopes, slopes_marginals): def write_into_page_xml_full(self, contours, contours_h, page_coord, dir_of_image, order_of_texts, id_of_texts, all_found_texline_polygons, all_found_texline_polygons_h, all_box_coord, all_box_coord_h, found_polygons_text_region_img, found_polygons_tables, found_polygons_drop_capitals, found_polygons_marginals, all_found_texline_polygons_marginals, all_box_coord_marginals, slopes, slopes_marginals):
@ -1550,7 +1551,7 @@ class eynollah:
#print(points_co) #print(points_co)
coord_text.set('points',points_co) coord_text.set('points',points_co)
self.serialize_lines_in_region(textregion, all_found_texline_polygons, mm, page_coord, all_box_coord) id_indexer_l = self.serialize_lines_in_region(textregion, all_found_texline_polygons, mm, page_coord, all_box_coord, slopes, id_indexer_l)
texteqreg=ET.SubElement(textregion, 'TextEquiv') texteqreg=ET.SubElement(textregion, 'TextEquiv')
unireg=ET.SubElement(texteqreg, 'Unicode') unireg=ET.SubElement(texteqreg, 'Unicode')
@ -1578,7 +1579,7 @@ class eynollah:
coord_text = ET.SubElement(textregion, 'Coords') coord_text = ET.SubElement(textregion, 'Coords')
self.serialize_lines_in_region(textregion, all_found_texline_polygons_h, mm, page_coord, all_box_coord, slopes) id_indexer_l = self.serialize_lines_in_region(textregion, all_found_texline_polygons_h, mm, page_coord, all_box_coord, slopes, id_indexer_l)
texteqreg=ET.SubElement(textregion, 'TextEquiv') texteqreg=ET.SubElement(textregion, 'TextEquiv')
unireg=ET.SubElement(texteqreg, 'Unicode') unireg=ET.SubElement(texteqreg, 'Unicode')