Delinting

This commit is contained in:
James R. Barlow
2019-01-02 13:34:45 -08:00
parent 68fbd9fcc9
commit 7d330afd81
14 changed files with 40 additions and 53 deletions
-3
View File
@@ -581,7 +581,6 @@ def select_ocr_image(infiles, output_file, log, context):
with Image.open(image) as im:
from PIL import ImageColor
from PIL import ImageDraw
from decimal import Decimal
white = ImageColor.getcolor('#ffffff', im.mode)
# pink = ImageColor.getcolor('#ff0080', im.mode)
@@ -800,8 +799,6 @@ def get_docinfo(base_pdf, options):
def generate_postscript_stub(input_file, output_file, log, context):
options = context.get_options()
pdf = pikepdf.open(input_file)
generate_pdfa_ps(output_file)
+1 -1
View File
@@ -611,7 +611,7 @@ class Pix(LeptonicaObject):
0,
)
)
except (LeptonicaError, ValueError, IndexError) as e:
except (LeptonicaError, ValueError, IndexError):
return
finally:
with suppress(FileNotFoundError):
+1 -1
View File
@@ -148,7 +148,7 @@ def extract_image_generic(*, pike, root, log, image, xref, options):
def extract_images(pike, root, log, options, extract_fn):
"""Extract image using extract_fn
extract_fn decides where the image is interesting in this case
extract_fn decides whether the image is interesting in this case
"""
include_xrefs = set()
+22 -23
View File
@@ -25,25 +25,11 @@ import pdfminer.pdfdevice
import pdfminer.pdfinterp
from pdfminer.converter import PDFLayoutAnalyzer
from pdfminer.glyphlist import glyphname2unicode
from pdfminer.layout import (
LAParams,
LTChar,
LTContainer,
LTLayoutContainer,
LTPage,
LTTextBox,
LTTextLine,
)
from pdfminer.layout import LAParams, LTChar, LTPage, LTTextBox
from pdfminer.pdfdocument import PDFTextExtractionNotAllowed
from pdfminer.pdffont import (
PDFCIDFont,
PDFFont,
PDFSimpleFont,
PDFType3Font,
PDFUnicodeNotDefined,
)
from pdfminer.pdffont import PDFFont, PDFSimpleFont, PDFUnicodeNotDefined
from pdfminer.pdfpage import PDFPage
from pdfminer.utils import bbox2str, fsplit, matrix2str
from pdfminer.utils import bbox2str, matrix2str
from ..exceptions import EncryptedPdfError
@@ -161,11 +147,21 @@ class LTStateAwareChar(LTChar):
text,
textwidth,
textdisp,
ncs,
graphicstate,
textstate,
*args,
):
super().__init__(
matrix, font, fontsize, scaling, rise, text, textwidth, textdisp, *args
matrix,
font,
fontsize,
scaling,
rise,
text,
textwidth,
textdisp,
ncs,
graphicstate,
)
self.rendermode = textstate.render
@@ -223,11 +219,13 @@ class TextPositionTracker(PDFLayoutAnalyzer):
self.pageno += 1
self.receive_layout(self.cur_item)
def render_string(self, textstate, seq, *args):
def render_string(self, textstate, seq, ncs, graphicstate):
self.textstate = textstate.copy()
super().render_string(self.textstate, seq, *args)
super().render_string(self.textstate, seq, ncs, graphicstate)
def render_char(self, matrix, font, fontsize, scaling, rise, cid, *args):
def render_char(
self, matrix, font, fontsize, scaling, rise, cid, ncs, graphicstate
):
try:
text = font.to_unichr(cid)
assert isinstance(text, str), str(type(text))
@@ -244,8 +242,9 @@ class TextPositionTracker(PDFLayoutAnalyzer):
text,
textwidth,
textdisp,
ncs,
graphicstate,
self.textstate,
*args,
)
self.cur_item.add(item)
return item.adv