qurator-spk
/
eynollah
mirror of https://github.com/qurator-spk/eynollah.git
1
0
Fork 0
Code Issues Projects Releases Wiki Activity

remove unncessary try-except, formatting

Browse Source
pull/19/head
Konstantin Baierer 4 years ago
parent 4480302b53
commit 48d8406fba
1 changed files with 49 additions and 83 deletions
Show Stats Download Patch File Download Diff File

132
sbb_newspapers_org_image/eynollah.py
Unescape Escape View File

@ -1310,7 +1310,7 @@ class eynollah:
id_indexer_l = 0 id_indexer_l = 0
if len(found_polygons_text_region) > 0: if len(found_polygons_text_region) > 0:
self.xml_reading_order(page, order_of_texts, id_of_texts, id_of_marginalia, 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'+str(id_indexer)) textregion.set('id', 'r'+str(id_indexer))
@ -1395,42 +1395,36 @@ class eynollah:
else: else:
points_co += str(int((all_found_texline_polygons_marginals[mm][j][l][0][0] + all_box_coord_marginals[mm][2] + page_coord[2]) / self.scale_x)) points_co += str(int((all_found_texline_polygons_marginals[mm][j][l][0][0] + all_box_coord_marginals[mm][2] + page_coord[2]) / self.scale_x))
points_co += ',' points_co += ','
points_co += str(int((all_found_texline_polygons_marginals[mm][j][l][0][1] + all_box_coord_marginals[mm][0] + page_coord[0])/self.scale_y)) points_co += str(int((all_found_texline_polygons_marginals[mm][j][l][0][1] + all_box_coord_marginals[mm][0] + page_coord[0])/self.scale_y))
else: else:
if len(all_found_texline_polygons_marginals[mm][j][l])==2: if len(all_found_texline_polygons_marginals[mm][j][l]) == 2:
points_co += str(int((all_found_texline_polygons_marginals[mm][j][l][0] + page_coord[2]) / self.scale_x)) points_co += str(int((all_found_texline_polygons_marginals[mm][j][l][0] + page_coord[2]) / self.scale_x))
points_co += ',' points_co += ','
points_co += str(int((all_found_texline_polygons_marginals[mm][j][l][1] + page_coord[0]) / self.scale_y)) points_co += str(int((all_found_texline_polygons_marginals[mm][j][l][1] + page_coord[0]) / self.scale_y))
else: else:
points_co += str(int((all_found_texline_polygons_marginals[mm][j][l][0][0] + page_coord[2]) / self.scale_x)) points_co += str(int((all_found_texline_polygons_marginals[mm][j][l][0][0] + page_coord[2]) / self.scale_x))
points_co += ',' points_co += ','
points_co += str(int((all_found_texline_polygons_marginals[mm][j][l][0][1] + page_coord[0]) / self.scale_y)) points_co += str(int((all_found_texline_polygons_marginals[mm][j][l][0][1] + page_coord[0]) / self.scale_y))
if l < len(all_found_texline_polygons_marginals[mm][j]) - 1: if l < len(all_found_texline_polygons_marginals[mm][j]) - 1:
points_co += ' ' points_co += ' '
coord.set('points',points_co) coord.set('points',points_co)
except: except:
pass pass
try:
id_indexer = len(found_polygons_text_region) + len(found_polygons_marginals)
for mm in range(len(found_polygons_text_region_img)):
textregion=ET.SubElement(page, 'ImageRegion')
textregion.set('id','r'+str(id_indexer)) id_indexer = len(found_polygons_text_region) + len(found_polygons_marginals)
id_indexer+=1 for mm in range(len(found_polygons_text_region_img)):
textregion=ET.SubElement(page, 'ImageRegion')
textregion.set('id', 'r%s' % id_indexer)
coord_text = ET.SubElement(textregion, 'Coords') id_indexer += 1
points_co='' coord_text = ET.SubElement(textregion, 'Coords')
for lmm in range(len(found_polygons_text_region_img[mm])): points_co = ''
points_co=points_co+str(int((found_polygons_text_region_img[mm][lmm,0,0] + page_coord[2]) / self.scale_x)) for lmm in range(len(found_polygons_text_region_img[mm])):
points_co=points_co+',' points_co += str(int((found_polygons_text_region_img[mm][lmm,0,0] + page_coord[2]) / self.scale_x))
points_co=points_co+str(int((found_polygons_text_region_img[mm][lmm,0,1] + page_coord[0]) / self.scale_y)) points_co += ','
if lmm < len(found_polygons_text_region_img[mm]) - 1: points_co += str(int((found_polygons_text_region_img[mm][lmm,0,1] + page_coord[0]) / self.scale_y))
points_co += ' ' if lmm < len(found_polygons_text_region_img[mm]) - 1:
coord_text.set('points', points_co) points_co += ' '
except: coord_text.set('points', points_co)
pass
self.logger.info("filename stem: '%s'", self.image_filename_stem) self.logger.info("filename stem: '%s'", self.image_filename_stem)
@ -1489,105 +1483,77 @@ class eynollah:
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:
for mm in range(len(found_polygons_marginals)): for mm in range(len(found_polygons_marginals)):
textregion = ET.SubElement(page, 'TextRegion') textregion = ET.SubElement(page, 'TextRegion')
textregion.set('id', id_of_marginalia[mm]) textregion.set('id', id_of_marginalia[mm])
textregion.set('type', 'marginalia')
textregion.set('type','marginalia')
#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_marginals, mm, page_coord)) coord_text.set('points', self.calculate_polygon_coords(found_polygons_marginals, mm, page_coord))
for j in range(len(all_found_texline_polygons_marginals[mm])): for j in range(len(all_found_texline_polygons_marginals[mm])):
textline=ET.SubElement(textregion, 'TextLine') textline = ET.SubElement(textregion, 'TextLine')
textline.set('id','l'+str(id_indexer_l)) textline.set('id', 'l%s' % 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')
uni.text = ' ' uni.text = ' '
points_co='' points_co=''
for l in range(len(all_found_texline_polygons_marginals[mm][j])): for l in range(len(all_found_texline_polygons_marginals[mm][j])):
if not self.curved_line: if not self.curved_line:
if len(all_found_texline_polygons_marginals[mm][j][l])==2: if len(all_found_texline_polygons_marginals[mm][j][l]) == 2:
points_co=points_co+str( int( (all_found_texline_polygons_marginals[mm][j][l][0] points_co += str(int((all_found_texline_polygons_marginals[mm][j][l][0] + all_box_coord_marginals[mm][2] + page_coord[2]) / self.scale_x))
+all_box_coord_marginals[mm][2]+page_coord[2])/self.scale_x) ) points_co += ','
points_co=points_co+',' points_co += str(int((all_found_texline_polygons_marginals[mm][j][l][1] + all_box_coord_marginals[mm][0] + page_coord[0]) / self.scale_y))
points_co=points_co+str( int( (all_found_texline_polygons_marginals[mm][j][l][1]
+all_box_coord_marginals[mm][0]+page_coord[0])/self.scale_y) )
else: else:
points_co=points_co+str( int( ( all_found_texline_polygons_marginals[mm][j][l][0][0] points_co += str(int((all_found_texline_polygons_marginals[mm][j][l][0][0] + all_box_coord_marginals[mm][2] + page_coord[2]) / self.scale_x))
+all_box_coord_marginals[mm][2]+page_coord[2])/self.scale_x ) ) points_co += ','
points_co=points_co+',' points_co+= str(int((all_found_texline_polygons_marginals[mm][j][l][0][1] + all_box_coord_marginals[mm][0] + page_coord[0]) / self.scale_y))
points_co=points_co+str( int( ( all_found_texline_polygons_marginals[mm][j][l][0][1]
+all_box_coord_marginals[mm][0]+page_coord[0])/self.scale_y) )
else: else:
if len(all_found_texline_polygons_marginals[mm][j][l])==2: if len(all_found_texline_polygons_marginals[mm][j][l])==2:
points_co=points_co+str( int( (all_found_texline_polygons_marginals[mm][j][l][0] points_co += str(int((all_found_texline_polygons_marginals[mm][j][l][0] + page_coord[2]) / self.scale_x))
+page_coord[2])/self.scale_x) ) points_co += ','
points_co=points_co+',' points_co += str(int((all_found_texline_polygons_marginals[mm][j][l][1] + page_coord[0]) / self.scale_y))
points_co=points_co+str( int( (all_found_texline_polygons_marginals[mm][j][l][1]
+page_coord[0])/self.scale_y) )
else: else:
points_co=points_co+str( int( ( all_found_texline_polygons_marginals[mm][j][l][0][0] points_co += str(int((all_found_texline_polygons_marginals[mm][j][l][0][0] + page_coord[2]) / self.scale_x))
+page_coord[2])/self.scale_x ) ) points_co += ','
points_co=points_co+',' points_co += str(int((all_found_texline_polygons_marginals[mm][j][l][0][1] + page_coord[0]) / self.scale_y))
points_co=points_co+str( int( ( all_found_texline_polygons_marginals[mm][j][l][0][1]
+page_coord[0])/self.scale_y) )
if l<(len(all_found_texline_polygons_marginals[mm][j])-1): if l<(len(all_found_texline_polygons_marginals[mm][j])-1):
points_co=points_co+' ' points_co=points_co+' '
#print(points_co)
coord.set('points',points_co) coord.set('points',points_co)
texteqreg = ET.SubElement(textregion, 'TextEquiv')
unireg = ET.SubElement(texteqreg, 'Unicode')
texteqreg=ET.SubElement(textregion, 'TextEquiv')
unireg=ET.SubElement(texteqreg, 'Unicode')
unireg.text = ' ' unireg.text = ' '
except: except:
pass pass
try: try:
id_indexer=len(contours_h)+len(contours)+len(found_polygons_marginals)+len(found_polygons_drop_capitals) id_indexer = len(found_polygons_text_region) + len(found_polygons_text_region_h) + len(found_polygons_marginals) + len(found_polygons_drop_capitals)
for mm in range(len(found_polygons_text_region_img)): for mm in range(len(found_polygons_text_region_img)):
textregion=ET.SubElement(page, 'ImageRegion') textregion=ET.SubElement(page, 'ImageRegion')
textregion.set('id','r%s' % id_indexer)
textregion.set('id','r'+str(id_indexer)) id_indexer += 1
id_indexer+=1
coord_text = ET.SubElement(textregion, 'Coords') coord_text = ET.SubElement(textregion, 'Coords')
coord_text.set('points', self.calculate_polygon_coords(found_polygons_text_region_img, mm, page_coord)) coord_text.set('points', self.calculate_polygon_coords(found_polygons_text_region_img, mm, page_coord))
except: except:
pass pass
try: try:
for mm in range(len(found_polygons_tables)): for mm in range(len(found_polygons_tables)):
textregion=ET.SubElement(page, 'TableRegion') textregion = ET.SubElement(page, 'TableRegion')
textregion.set('id', 'r%s' %id_indexer)
textregion.set('id','r'+str(id_indexer)) id_indexer += 1
id_indexer+=1
coord_text = ET.SubElement(textregion, 'Coords') coord_text = ET.SubElement(textregion, 'Coords')
coord_text.set('points', self.calculate_polygon_coords(found_polygons_tables, mm, page_coord)) coord_text.set('points', self.calculate_polygon_coords(found_polygons_tables, mm, page_coord))
except: except:
pass pass
##print(dir_of_image)
##print(self.f_name)
##print(os.path.join(dir_of_image, self.f_name) + ".xml")
##tree = ET.ElementTree(pcgts)
##tree.write(os.path.join(dir_of_image, self.image_filename_stem) + ".xml")
self.logger.info("filename stem: '%s'", self.image_filename_stem) self.logger.info("filename stem: '%s'", self.image_filename_stem)
# print(os.path.join(dir_of_image, self.image_filename_stem) + ".xml")
tree = ET.ElementTree(pcgts) tree = ET.ElementTree(pcgts)
tree.write(os.path.join(dir_of_image, self.image_filename_stem) + ".xml") tree.write(os.path.join(dir_of_image, self.image_filename_stem) + ".xml")
def get_regions_from_xy_2models(self,img,is_image_enhanced): def get_regions_from_xy_2models(self,img,is_image_enhanced):
self.logger.debug("enter get_regions_from_xy_2models") self.logger.debug("enter get_regions_from_xy_2models")

Write Preview
Loading…
Cancel
Save