qurator-spk/eynollah
1
0
Fork 0
mirror of https://github.com/qurator-spk/eynollah.git synced 2025-10-15 19:09:58 +02:00
Code Issues Projects Releases Wiki Activity

XML encoding should be utf-8 not utf8

Browse source 
... and  should use OCR-D's generateDS PAGE API consistently
This commit is contained in:
kba 2025-10-10 16:39:16 +02:00
parent 2056a8bdb9
commit 745cf3be48
2 changed files with 3 additions and 3 deletions
Download patch file Download diff file Expand all files Collapse all files

4
src/eynollah/eynollah.py
View file

@ -5284,7 +5284,7 @@ class Eynollah_ocr:
##unicode_textpage.text = tot_page_text ##unicode_textpage.text = tot_page_text
ET.register_namespace("",name_space) ET.register_namespace("",name_space)
tree1.write(out_file_ocr,xml_declaration=True,method='xml',encoding="utf8",default_namespace=None) tree1.write(out_file_ocr,xml_declaration=True,method='xml',encoding="utf-8",default_namespace=None)
else: else:
###max_len = 280#512#280#512 ###max_len = 280#512#280#512
###padding_token = 1500#299#1500#299 ###padding_token = 1500#299#1500#299
@ -5833,5 +5833,5 @@ class Eynollah_ocr:
##unicode_textpage.text = tot_page_text ##unicode_textpage.text = tot_page_text
ET.register_namespace("",name_space) ET.register_namespace("",name_space)
tree1.write(out_file_ocr,xml_declaration=True,method='xml',encoding="utf8",default_namespace=None) tree1.write(out_file_ocr,xml_declaration=True,method='xml',encoding="utf-8",default_namespace=None)
#print("Job done in %.1fs", time.time() - t0) #print("Job done in %.1fs", time.time() - t0)

2
src/eynollah/mb_ro_on_layout.py
View file

@ -805,7 +805,7 @@ class machine_based_reading_order_on_layout:
tree_xml.write(os.path.join(dir_out, file_name+'.xml'), tree_xml.write(os.path.join(dir_out, file_name+'.xml'),
xml_declaration=True, xml_declaration=True,
method='xml', method='xml',
encoding="utf8", encoding="utf-8",
default_namespace=None) default_namespace=None)
#sys.exit() #sys.exit()