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

🎨 simplify

Browse source
This commit is contained in:
Konstantin Baierer 2021-02-23 13:49:48 +01:00
parent bfb6e77db1
commit fdbcfe44a0
1 changed files with 16 additions and 31 deletions
Download patch file Download diff file Expand all files Collapse all files

35
sbb_newspapers_org_image/eynollah.py
View file

@ -1269,8 +1269,10 @@ class eynollah:
id_indexer = 0 id_indexer = 0
id_indexer_l = 0 id_indexer_l = 0
id_of_marginalia = []
if len(contours) > 0: if len(contours) > 0:
self.xml_reading_order(page, order_of_texts, id_of_texts, found_polygons_marginals) self.xml_reading_order(page, order_of_texts, id_of_texts, id_of_marginalia, found_polygons_marginals)
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')
textregion.set('id', 'r%s' % id_indexer) textregion.set('id', 'r%s' % id_indexer)
@ -1297,38 +1299,19 @@ class eynollah:
unireg = ET.SubElement(texteqreg, 'Unicode') unireg = ET.SubElement(texteqreg, 'Unicode')
unireg.text = ' ' unireg.text = ' '
if len(found_polygons_drop_capitals) > 0: if len(found_polygons_drop_capitals) > 0:
id_indexer = len(contours_h) + len(contours) + len(found_polygons_marginals) id_indexer = len(contours_h) + len(contours) + len(found_polygons_marginals)
for mm in range(len(found_polygons_drop_capitals)): for mm in range(len(found_polygons_drop_capitals)):
textregion=ET.SubElement(page, 'TextRegion') textregion=ET.SubElement(page, 'TextRegion')
textregion.set('id',' r%s' % id_indexer)
#id_indexer_l=id_indexer_l
textregion.set('id','r'+str(id_indexer))
id_indexer += 1 id_indexer += 1
textregion.set('type','drop-capital') textregion.set('type','drop-capital')
#if mm==0:
# textregion.set('type','header')
#else:
# textregion.set('type','paragraph')
coord_text = ET.SubElement(textregion, 'Coords') coord_text = ET.SubElement(textregion, 'Coords')
coord_text.set('points', self.calculate_polygon_coords(found_polygons_drop_capitals, mm, page_coord)) coord_text.set('points', self.calculate_polygon_coords(found_polygons_drop_capitals, mm, page_coord))
texteqreg = ET.SubElement(textregion, 'TextEquiv') texteqreg = ET.SubElement(textregion, 'TextEquiv')
unireg=ET.SubElement(texteqreg, 'Unicode') unireg=ET.SubElement(texteqreg, 'Unicode')
unireg.text = ' ' unireg.text = ' '
try: try:
try: try:
id_indexer_l=id_indexer_l id_indexer_l=id_indexer_l
except: except:
@ -1449,7 +1432,10 @@ class eynollah:
points_page_print = points_page_print + ' ' points_page_print = points_page_print + ' '
return points_page_print return points_page_print
def xml_reading_order(self, page, order_of_texts, id_of_texts, found_polygons_marginals): def xml_reading_order(self, page, order_of_texts, id_of_texts, id_of_marginalia, found_polygons_marginals):
"""
XXX side-effect: extends id_of_marginalia
"""
region_order = ET.SubElement(page, 'ReadingOrder') region_order = ET.SubElement(page, 'ReadingOrder')
region_order_sub = ET.SubElement(region_order, 'OrderedGroup') region_order_sub = ET.SubElement(region_order, 'OrderedGroup')
region_order_sub.set('id', "ro357564684568544579089") region_order_sub.set('id', "ro357564684568544579089")
@ -1460,7 +1446,6 @@ class eynollah:
name.set('index', str(indexer_region)) name.set('index', str(indexer_region))
name.set('regionRef', id_of_texts[vj]) name.set('regionRef', id_of_texts[vj])
indexer_region+=1 indexer_region+=1
id_of_marginalia=[]
for vm in range(len(found_polygons_marginals)): for vm in range(len(found_polygons_marginals)):
id_of_marginalia.append('r%s' % indexer_region) id_of_marginalia.append('r%s' % indexer_region)
name = "coord_text_%s" % indexer_region name = "coord_text_%s" % indexer_region
@ -1468,11 +1453,11 @@ class eynollah:
name.set('index', str(indexer_region)) name.set('index', str(indexer_region))
name.set('regionRef', 'r%s' % indexer_region) name.set('regionRef', 'r%s' % indexer_region)
indexer_region += 1 indexer_region += 1
return id_of_marginalia
def write_into_page_xml(self, contours, page_coord, dir_of_image, order_of_texts, id_of_texts, all_found_texline_polygons, all_box_coord, found_polygons_text_region_img, found_polygons_marginals, all_found_texline_polygons_marginals, all_box_coord_marginals, curved_line, slopes, slopes_marginals): def write_into_page_xml(self, contours, page_coord, dir_of_image, order_of_texts, id_of_texts, all_found_texline_polygons, all_box_coord, found_polygons_text_region_img, found_polygons_marginals, all_found_texline_polygons_marginals, all_box_coord_marginals, curved_line, slopes, slopes_marginals):
self.logger.debug('enter write_into_page_xml') self.logger.debug('enter write_into_page_xml')
id_of_marginalia
found_polygons_text_region = contours found_polygons_text_region = contours
@ -1484,7 +1469,7 @@ class eynollah:
if len(contours) > 0: if len(contours) > 0:
id_of_marginalia = self.xml_reading_order(page, order_of_texts, id_of_texts, found_polygons_marginals) self.xml_reading_order(page, order_of_texts, id_of_texts, id_of_marginalia, found_polygons_marginals)
id_indexer = 0 id_indexer = 0
id_indexer_l = 0 id_indexer_l = 0