Simple python wrapper for Tesseract, an OCR engine to detect and read text from images. See https://pypi.org/project/pytesseract/wip/yesman
parent
6093167164
commit
76ad975840
@ -0,0 +1,31 @@ |
||||
{ buildPythonPackage, fetchPypi, lib, pillow, tesseract, substituteAll }: |
||||
|
||||
buildPythonPackage rec { |
||||
pname = "pytesseract"; |
||||
version = "0.2.5"; |
||||
|
||||
src = fetchPypi { |
||||
inherit pname version; |
||||
sha256 = "0wlz1vbg1k8cdrpzvrahjnbsfs4ki6xqhbkv17ycfchh7h6kfkfm"; |
||||
}; |
||||
|
||||
patches = [ |
||||
(substituteAll { |
||||
src = ./tesseract-binary.patch; |
||||
drv = "${tesseract}"; |
||||
}) |
||||
]; |
||||
|
||||
buildInputs = [ tesseract ]; |
||||
propagatedBuildInputs = [ pillow ]; |
||||
|
||||
# the package doesn't have any tests. |
||||
doCheck = false; |
||||
|
||||
meta = with lib; { |
||||
homepage = https://pypi.org/project/pytesseract/; |
||||
license = licenses.gpl3; |
||||
description = "A Python wrapper for Google Tesseract"; |
||||
maintainers = with maintainers; [ ma27 ]; |
||||
}; |
||||
} |
@ -0,0 +1,13 @@ |
||||
diff --git a/src/pytesseract.py b/src/pytesseract.py
|
||||
index 32713cf..5f9209d 100755
|
||||
--- a/src/pytesseract.py
|
||||
+++ b/src/pytesseract.py
|
||||
@@ -25,7 +25,7 @@ if numpy_installed:
|
||||
from numpy import ndarray
|
||||
|
||||
# CHANGE THIS IF TESSERACT IS NOT IN YOUR PATH, OR IS NAMED DIFFERENTLY
|
||||
-tesseract_cmd = 'tesseract'
|
||||
+tesseract_cmd = '@drv@/bin/tesseract'
|
||||
RGB_MODE = 'RGB'
|
||||
OSD_KEYS = {
|
||||
'Page number': ('page_num', int),
|
Loading…
Reference in new issue