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

Set minimum Tesseract to 4.1.1

Browse Source
This commit is contained in:
James R. Barlow
2022-08-02 15:20:29 -07:00
parent 80b7cf6330
commit acc70036cc
10 changed files with 26 additions and 59 deletions
Download Patch File Download Diff File Expand all files Collapse all files
src/ocrmypdf/_exec/tesseract.py
+1 -10
View File
@@ -33,7 +33,7 @@ HOCR_TEMPLATE = """<?xml version="1.0" encoding="UTF-8"?>
<head>
<title></title>
<meta http-equiv="Content-Type" content="text/html;charset=utf-8" />
<meta name='ocr-system' content='tesseract 4.0.0' />
<meta name='ocr-system' content='tesseract 4.1.1' />
<meta name='ocr-capabilities' content='ocr_page ocr_carea ocr_par ocr_line ocrx_word ocrp_wconf'/>
</head>
<body>
@@ -114,15 +114,6 @@ def version() -> str:
return get_version('tesseract', regex=r'tesseract\s(.+)')
def has_user_words() -> bool:
"""Does Tesseract have --user-words capability?
Not available in 4.0, but available in 4.1. Also available in 3.x, but
we no longer support 3.x.
"""
return version() >= '4.1'
def has_thresholding() -> bool:
"""Does Tesseract have -c thresholding method capability?"""
return version() >= '5.0'
src/ocrmypdf/builtin_plugins/tesseract_ocr.py
+2 -7
View File
@@ -43,7 +43,7 @@ def add_options(parser):
metavar='MODE',
choices=range(0, 4),
help=(
"Set Tesseract 4.0+ OCR engine mode: "
"Set Tesseract 4+ OCR engine mode: "
"0 - original Tesseract only; "
"1 - neural nets LSTM only; "
"2 - Tesseract + LSTM; "
@@ -93,7 +93,7 @@ def check_options(options):
program='tesseract',
package={'linux': 'tesseract-ocr'},
version_checker=tesseract.version,
need_version='4.0.0-beta.1', # using backport for Travis CI
need_version='4.1.1', # Ubuntu 20.04 version
version_parser=tesseract.TesseractVersion,
)
@@ -101,11 +101,6 @@ def check_options(options):
if options.pdf_renderer == 'auto':
options.pdf_renderer = 'sandwich'
if not tesseract.has_user_words() and (options.user_words or options.user_patterns):
log.warning(
"Tesseract 4.0 (which you have installed) ignores --user-words and "
"--user-patterns, so these arguments have no effect."
)
if not tesseract.has_thresholding() and options.tesseract_thresholding != 0:
log.warning(
"The installed version of Tesseract does not support changes to its "