menekis/OCRmyPDF
Public Access
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Move Ghostscript text analysis into its own module

Browse Source
This commit is contained in:
James R. Barlow
2018-11-04 14:55:48 -08:00
parent c023cae299
commit 995fc58466
2 changed files with 97 additions and 72 deletions
Download Patch File Download Diff File Expand all files Collapse all files
src/ocrmypdf/pdfinfo/__init__.py
+3 -72
View File
@@ -23,13 +23,13 @@ from math import hypot, isclose
from pathlib import Path
from unittest.mock import Mock
import re
import xml.etree.ElementTree as ET
from pikepdf import PdfMatrix
import pikepdf
from .ghosttext import extract_text_xml
from .layout import get_page_analysis, get_text_boxes
from ..exec import ghostscript
from ..helpers import fspath
@@ -40,17 +40,6 @@ Encoding = Enum('Encoding',
'ccitt jpeg jpeg2000 jbig2 asciihex ascii85 lzw flate ' + \
'runlength')
# Forgive me for I have sinned
# I am using regular expressions to parse XML. However the XML in this case,
# generated by Ghostscript, is self-consistent enough to be parseable.
regex_remove_char_tags = re.compile(br"""
<char\b
(?: [^>] # anything single character but >
| \">\" # special case: trap ">"
)*
/> # terminate with '/>'
""", re.VERBOSE)
FRIENDLY_COLORSPACE = {
'/DeviceGray': Colorspace.gray,
'/CalGray': Colorspace.gray,
@@ -526,43 +515,6 @@ def _process_content_streams(*, pdf, container, shorthand=None):
yield from _find_form_xobject_images(pdf, container, contentsinfo)
def _page_get_textblocks(infile, pageno, xmltext):
"""Smarter text detection"""
root = xmltext
if not hasattr(xmltext, 'findall'):
return []
def blocks():
for span in root.findall('.//span'):
bbox_str = span.attrib['bbox']
font_size = span.attrib['size']
pts = [int(pt) for pt in bbox_str.split()]
pts[1] = pts[1] - int(float(font_size) + 0.5)
bbox = tuple(pts)
yield bbox
def joined_blocks():
prev = None
for bbox in blocks():
if prev is None:
prev = bbox
if bbox[1] == prev[1] and bbox[3] == prev[3]:
gap = prev[2] - bbox[0]
height = bbox[3] - bbox[1]
if gap < height:
# Join boxes
prev = (prev[0], prev[1], bbox[2], bbox[3])
continue
# yield previously joined bboxes and start anew
yield prev
prev = bbox
if prev is not None:
yield prev
return [block for block in joined_blocks()]
def _page_has_text(text_blocks, page_width, page_height):
"""Smarter text detection that ignores text in margins"""
@@ -668,28 +620,7 @@ def _pdf_get_all_pageinfo(infile, log=None):
log = Mock()
pdf = pikepdf.open(infile)
existing_text = ghostscript.extract_text(infile, pageno=None)
existing_text = regex_remove_char_tags.sub(b' ', existing_text)
try:
root = ET.fromstringlist([
b'<document>\n', existing_text, b'</document>\n'
])
page_xml = root.findall('page')
except ET.ParseError as e:
log.error(
"An error occurred while attempting to retrieve existing text in "
"the input file. Will attempt to continue assuming that there is "
"no existing text in the file. The error was:")
log.error(e)
page_xml = [None] * len(pdf.pages)
page_count_difference = len(pdf.pages) - len(page_xml)
if page_count_difference != 0:
log.error("The number of pages in the input file is inconsistent.")
if page_count_difference > 0:
page_xml.extend([None] * page_count_difference)
page_xml = extract_text_xml(infile, pdf, pageno=None, log=log)
pages = []
for n in range(len(pdf.pages)):
src/ocrmypdf/pdfinfo/ghosttext.py
+94
View File
@@ -0,0 +1,94 @@
# © 2018 James R. Barlow: github.com/jbarlow83
#
# This file is part of OCRmyPDF.
#
# OCRmyPDF is free software: you can redistribute it and/or modify
# it under the terms of the GNU General Public License as published by
# the Free Software Foundation, either version 3 of the License, or
# (at your option) any later version.
#
# OCRmyPDF is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License
# along with OCRmyPDF. If not, see <http://www.gnu.org/licenses/>.
import re
import xml.etree.ElementTree as ET
from ..exec import ghostscript
# Forgive me for I have sinned
# I am using regular expressions to parse XML. However the XML in this case,
# generated by Ghostscript, is self-consistent enough to be parseable.
regex_remove_char_tags = re.compile(br"""
<char\b
(?: [^>] # anything single character but >
| \">\" # special case: trap ">"
)*
/> # terminate with '/>'
""", re.VERBOSE)
def _page_get_textblocks(infile, pageno, xmltext):
"""Smarter text detection"""
root = xmltext
if not hasattr(xmltext, 'findall'):
return []
def blocks():
for span in root.findall('.//span'):
bbox_str = span.attrib['bbox']
font_size = span.attrib['size']
pts = [int(pt) for pt in bbox_str.split()]
pts[1] = pts[1] - int(float(font_size) + 0.5)
bbox = tuple(pts)
yield bbox
def joined_blocks():
prev = None
for bbox in blocks():
if prev is None:
prev = bbox
if bbox[1] == prev[1] and bbox[3] == prev[3]:
gap = prev[2] - bbox[0]
height = bbox[3] - bbox[1]
if gap < height:
# Join boxes
prev = (prev[0], prev[1], bbox[2], bbox[3])
continue
# yield previously joined bboxes and start anew
yield prev
prev = bbox
if prev is not None:
yield prev
return [block for block in joined_blocks()]
def extract_text_xml(infile, pdf, pageno=None, log=None):
existing_text = ghostscript.extract_text(infile, pageno=None)
existing_text = regex_remove_char_tags.sub(b' ', existing_text)
try:
root = ET.fromstringlist([
b'<document>\n', existing_text, b'</document>\n'
])
page_xml = root.findall('page')
except ET.ParseError as e:
log.error(
"An error occurred while attempting to retrieve existing text in "
"the input file. Will attempt to continue assuming that there is "
"no existing text in the file. The error was:")
log.error(e)
page_xml = [None] * len(pdf.pages)
page_count_difference = len(pdf.pages) - len(page_xml)
if page_count_difference != 0:
log.error("The number of pages in the input file is inconsistent.")
if page_count_difference > 0:
page_xml.extend([None] * page_count_difference)
return page_xml