sporada/ocr
1
0
Fork 0
mirror of https://gitlab.ub.uni-bielefeld.de/sfb1288inf/ocr.git synced 2024-12-27 05:34:18 +00:00
Code Issues Projects Releases Wiki Activity

Use argparse in hocrtotei

Browse Source
This commit is contained in:
Patrick Jentsch 2019-05-16 14:21:01 +02:00
parent c39edec1ab
commit b81ad4cc67
2 changed files with 27 additions and 21 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

37
hocrtotei
View File

@ -1,21 +1,23 @@
#!/usr/bin/env python3.5 #!/usr/bin/env python3.5
# coding=utf-8 # coding=utf-8
import xml.etree.ElementTree as ET
from xml.sax.saxutils import escape from xml.sax.saxutils import escape
import os import argparse
import re import xml.etree.ElementTree as ET
import sys
input_files = sorted( parser = argparse.ArgumentParser()
filter( parser.add_argument(
lambda x: x.endswith(".hocr"), 'i',
os.listdir(sys.argv[1]) help='The input files.',
), nargs='*',
key=lambda x: int(re.search(r'\d+', x).group(0))
) )
# "page-1.hocr" -> "1" parser.add_argument(
output_file = open(sys.argv[2], "w") 'o',
help='The output file.',
)
args = parser.parse_args()
output_file = open(args.o, "w")
output_file.write( output_file.write(
'<?xml version="1.0" encoding="UTF-8"?>\n' '<?xml version="1.0" encoding="UTF-8"?>\n'
@ -32,23 +34,20 @@ output_file.write(
+ ' <text>\n' + ' <text>\n'
+ ' <body>\n' + ' <body>\n'
) )
for index, input_file in enumerate(args.i):
for input_file in input_files: tree = ET.parse(input_file)
tree = ET.parse(os.path.join(sys.argv[1], input_file)) output_file.write(' <pb n="%i"/>\n' % (index + 1))
page_number = int(re.search(r'\d+', input_file.split(".")[0]).group(0))
output_file.write(' <pb n="%i"/>\n' % (page_number))
for para in tree.findall(".//*[@class='ocr_par']"): for para in tree.findall(".//*[@class='ocr_par']"):
output_file.write(' <p>\n') output_file.write(' <p>\n')
for line in para.findall(".//*[@class='ocr_line']"): for line in para.findall(".//*[@class='ocr_line']"):
first_word_in_line = True first_word_in_line = True
for word in line.findall(".//*[@class='ocrx_word']"): for word in line.findall(".//*[@class='ocrx_word']"):
if word.text is not None: if word.text is not None:
output_file.write((" " if first_word_in_line else " ") + escape(word.text.strip())) output_file.write((' ' if first_word_in_line else ' ') + escape(word.text.strip()))
first_word_in_line = False first_word_in_line = False
if not first_word_in_line: if not first_word_in_line:
output_file.write('<lb/>\n') output_file.write('<lb/>\n')
output_file.write(' </p>\n') output_file.write(' </p>\n')
output_file.write( output_file.write(
' </body>\n' ' </body>\n'
+ ' </text>\n' + ' </text>\n'

11
ocr
View File

@ -307,8 +307,15 @@ class OCRWorkflow(WorkflowRunner):
''' '''
hocr_to_tei_jobs = [] hocr_to_tei_jobs = []
for index, job in enumerate(self.jobs): for index, job in enumerate(self.jobs):
cmd = 'hocrtotei "%s" "%s"' % ( files = os.listdir(os.path.join(job['output_dir'], 'tmp'))
os.path.join(job['output_dir'], 'tmp'), files = filter(lambda x: x.endswith('.hocr'), files)
files.sort(key=lambda x: int(re.search(r'\d+', x).group(0)))
files = map(
lambda x: '"' + os.path.join(job['output_dir'], 'tmp', x) + '"',
files
)
cmd = 'hocrtotei %s "%s"' % (
' '.join(files),
os.path.join( os.path.join(
job['output_dir'], job['output_dir'],
os.path.join(job['output_dir'], job['name'] + '.xml') os.path.join(job['output_dir'], job['name'] + '.xml')