2020-04-06 07:21:52 +00:00
|
|
|
#!/usr/bin/env python3.7
|
2018-10-09 12:43:23 +00:00
|
|
|
# coding=utf-8
|
|
|
|
|
|
2021-03-17 15:58:13 +00:00
|
|
|
""""Convert hOCR to TEI XML."""
|
2021-02-19 12:04:03 +00:00
|
|
|
|
2018-10-09 12:43:23 +00:00
|
|
|
from xml.sax.saxutils import escape
|
2020-04-03 15:35:30 +00:00
|
|
|
from argparse import ArgumentParser
|
2021-03-15 11:45:05 +00:00
|
|
|
import re
|
2019-05-16 12:21:01 +00:00
|
|
|
import xml.etree.ElementTree as ET
|
2018-10-09 12:43:23 +00:00
|
|
|
|
2021-03-17 15:58:13 +00:00
|
|
|
parser = ArgumentParser(description='Convert hOCR to TEI XML.')
|
2021-03-18 09:32:55 +00:00
|
|
|
parser.add_argument('input', metavar='Path to hOCR input file')
|
|
|
|
|
parser.add_argument('output', metavar='Path to TEI output file')
|
2019-05-16 12:21:01 +00:00
|
|
|
args = parser.parse_args()
|
|
|
|
|
|
2021-03-17 15:58:13 +00:00
|
|
|
tei = ''
|
|
|
|
|
tei += '<TEI xmlns="http://www.tei-c.org/ns/1.0">\n'
|
|
|
|
|
tei += ' <teiHeader>\n'
|
|
|
|
|
tei += ' <fileDesc>\n'
|
|
|
|
|
tei += ' <titleStmt>\n'
|
|
|
|
|
tei += ' <title></title>\n'
|
|
|
|
|
tei += ' </titleStmt>\n'
|
|
|
|
|
tei += ' <publicationStmt>\n'
|
|
|
|
|
tei += ' <p></p>\n'
|
|
|
|
|
tei += ' </publicationStmt>\n'
|
|
|
|
|
tei += ' <sourceDesc>\n'
|
|
|
|
|
tei += ' <p></p>\n'
|
|
|
|
|
tei += ' </sourceDesc>\n'
|
|
|
|
|
tei += ' </fileDesc>\n'
|
|
|
|
|
tei += ' </teiHeader>\n'
|
|
|
|
|
tei += ' <text>\n'
|
|
|
|
|
tei += ' <body>\n'
|
|
|
|
|
# Conversion start
|
2021-03-18 09:32:55 +00:00
|
|
|
hocr = ET.parse(args.input)
|
2021-03-17 15:58:13 +00:00
|
|
|
for page in hocr.findall('.//*[@class="ocr_page"]'):
|
2021-03-15 11:45:05 +00:00
|
|
|
page_properties = page.attrib.get('title')
|
|
|
|
|
facsimile = re.search(r'image \"(.*?)\"', page_properties).group(1)
|
|
|
|
|
page_number = re.search(r'ppageno (\d+)', page_properties).group(1)
|
2021-03-17 15:58:13 +00:00
|
|
|
tei += ' <pb facs="{}" n="{}"/>\n'.format(facsimile, page_number)
|
2021-03-15 11:45:05 +00:00
|
|
|
for para in page.findall('.//*[@class="ocr_par"]'):
|
2021-03-17 15:58:13 +00:00
|
|
|
tei += ' <p>\n'
|
2019-05-20 09:10:40 +00:00
|
|
|
for line in para.findall('.//*[@class="ocr_line"]'):
|
2021-03-17 15:58:13 +00:00
|
|
|
tei += ' <lb/>'
|
2021-03-15 11:45:05 +00:00
|
|
|
indent = ''
|
2019-05-20 09:10:40 +00:00
|
|
|
for word in line.findall('.//*[@class="ocrx_word"]'):
|
2018-10-09 12:43:23 +00:00
|
|
|
if word.text is not None:
|
2021-03-17 15:58:13 +00:00
|
|
|
tei += indent + escape(word.text.strip())
|
2021-03-15 11:45:05 +00:00
|
|
|
indent = ' '
|
2021-03-17 15:58:13 +00:00
|
|
|
tei += '\n'
|
|
|
|
|
tei += ' </p>\n'
|
|
|
|
|
# Conversion end
|
|
|
|
|
tei += ' </body>\n'
|
|
|
|
|
tei += ' </text>\n'
|
|
|
|
|
tei += '</TEI>\n'
|
|
|
|
|
|
2021-03-18 09:32:55 +00:00
|
|
|
with open(args.output, 'w') as tei_file:
|
2021-03-17 15:58:13 +00:00
|
|
|
tei_file.write(tei)
|
