Skip to content
This repository has been archived by the owner on Jun 14, 2018. It is now read-only.

Update pyocr to use psm_parameter based on tesseract version #102

Open
wants to merge 6 commits into
base: master
Choose a base branch
from
Open

Update pyocr to use psm_parameter based on tesseract version #102

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
6 commits
Select commit Hold shift + click to select a range
f5ee17a
Simplify psm_parameter() with a conditional expression.
ddddavidmartin Jun 11, 2018
ddfe4dd
Use absolute import to work around circular builder/tesseract import …
ddddavidmartin Jun 11, 2018
adfff7b
Use absolute import in tesseract.py to avoid circular import issue.
ddddavidmartin Jun 11, 2018
e34cb4d
Import psm_parameter later to avoid circular import issues.
ddddavidmartin Jun 13, 2018
cc3c0b9
Revert circular import workarounds from tesseract.py.
ddddavidmartin Jun 13, 2018
a6e06b4
Merge branch 'master' of github.com:openpaperwork/pyocr into update_d…
ddddavidmartin Jun 13, 2018
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
9 changes: 6 additions & 3 deletions src/pyocr/builders.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -304,8 +304,9 @@ class TextBuilder(BaseBuilder):

def __init__(self, tesseract_layout=3, cuneiform_dotmatrix=False,
cuneiform_fax=False, cuneiform_singlecolumn=False):
from .tesseract import psm_parameter
tess_flags = [psm_parameter(), str(tesseract_layout)]
file_ext = ["txt"]
tess_flags = ["-psm", str(tesseract_layout)]
cun_args = ["-f", "text"]
# Add custom cuneiform parameters if needed
for par, arg in [(cuneiform_dotmatrix, "--dotmatrix"),
Expand Down Expand Up @@ -561,8 +562,9 @@ class WordBoxBuilder(BaseBuilder):
"""

def __init__(self, tesseract_layout=1):
from .tesseract import psm_parameter
tess_flags = [psm_parameter(), str(tesseract_layout)]
file_ext = ["html", "hocr"]
tess_flags = ["-psm", str(tesseract_layout)]
tess_conf = ["hocr"]
cun_args = ["-f", "hocr"]
super(WordBoxBuilder, self).__init__(file_ext, tess_flags, tess_conf,
Expand Down Expand Up @@ -637,8 +639,9 @@ class LineBoxBuilder(BaseBuilder):
"""

def __init__(self, tesseract_layout=1):
from .tesseract import psm_parameter
tess_flags = [psm_parameter(), str(tesseract_layout)]
file_ext = ["html", "hocr"]
tess_flags = ["-psm", str(tesseract_layout)]
tess_conf = ["hocr"]
cun_args = ["-f", "hocr"]
super(LineBoxBuilder, self).__init__(file_ext, tess_flags, tess_conf,
Expand Down
8 changes: 7 additions & 1 deletion src/pyocr/tesseract.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -161,6 +161,12 @@ def can_detect_orientation():
)


def psm_parameter():
"""Return the psm option string depending on the Tesseract version."""
version = get_version()
return "--psm" if version[0] > 3 else "-psm"


def detect_orientation(image, lang=None):
"""
Arguments:
Expand All @@ -178,7 +184,7 @@ def detect_orientation(image, lang=None):
"""
_set_environment()
with temp_dir() as tmpdir:
command = [TESSERACT_CMD, "input.bmp", 'stdout', "-psm", "0"]
command = [TESSERACT_CMD, "input.bmp", 'stdout', psm_parameter(), "0"]
version = get_version()
if version[0] >= 4:
# XXX: temporary fix to remove once Tesseract 4 is stable
Expand Down