def resource_example():
from pdfminer.pdffont import CFFFont, TrueTypeFont
from pdfminer.pdffont import PDFFont, PDFSimpleFont, PDFType1Font, PDFTrueTypeFont, PDFType3Font, PDFCIDFont
from pdfminer.psparser import literal_name
from pdfminer.pdftypes import PDFObjRef
from pdfminer.pdftypes import list_value, dict_value, stream_value
from pdfminer.pdfcolor import PDFColorSpace
from pdfminer.pdfcolor import PREDEFINED_COLORSPACE
font_filepath = '/path/to/font.ttf'
with open(font_filepath, 'rb') as fp:
#font = CFFFont(font_filepath, fp)
font = TrueTypeFont(font_filepath, fp)
print('Font type = {}.'.format(font.fonttype))
print('Font fp = {}.'.format(font.fp))
print('Font name = {}.'.format(font.name))
print('Font tables = {}.'.format(font.tables))
#--------------------
pdf_filepath = '/path/to/sample.pdf'
fp = None
try:
# Open a PDF file.
fp = open(pdf_filepath, 'rb')
# Create a PDF resource manager object that stores shared resources.
rsrcmgr = PDFResourceManager()
pages = PDFPage.get_pages(
fp, pagenos=None, maxpages=0, password=b''
) # pagenos uses zero-based indices. pagenos is sorted inside the function.
page = next(pages)
if page:
resources, contents = page.resources, page.contents
if not resources:
print('No resource.')
return
if contents:
print('Contents: {}.'.format(contents))
#for ct in contents:
# print(ct.resolve())
# REF [function] >> pdfminer.pdfinterp.PDFPageInterpreter.init_resources()
def get_colorspace(spec):
if isinstance(spec, list):
name = literal_name(spec[0])
else:
name = literal_name(spec)
if name == 'ICCBased' and isinstance(spec,
list) and 2 <= len(spec):
return PDFColorSpace(name, stream_value(spec[1])['N'])
elif name == 'DeviceN' and isinstance(spec,
list) and 2 <= len(spec):
return PDFColorSpace(name, len(list_value(spec[1])))
else:
return PREDEFINED_COLORSPACE[name]
fontmap, xobjmap = dict(), dict()
csmap = PREDEFINED_COLORSPACE.copy()
for (k, v) in dict_value(resources).items():
#if 2 <= self.debug:
# print >>stderr, 'Resource: %r: %r' % (k,v)
if k == 'Font':
for (font_id, spec) in dict_value(v).items():
obj_id = None
if isinstance(spec, PDFObjRef):
obj_id = spec.objid
spec = dict_value(spec)
fontmap[font_id] = rsrcmgr.get_font(obj_id, spec)
elif k == 'ColorSpace':
for (cs_id, spec) in dict_value(v).items():
csmap[cs_id] = get_colorspace(resolve1(spec))
elif k == 'ProcSet':
rsrcmgr.get_procset(list_value(v))
elif k == 'XObject':
for (xobj_id, xobjstrm) in dict_value(v).items():
xobjmap[xobj_id] = xobjstrm
#spec = ...
#if 'FontDescriptor' in spec:
# print('FontDescriptor: {}.'.format(spec['FontDescriptor'].resolve()))
font = PDFType1Font(rsrcmgr, spec)
font = PDFTrueTypeFont(rsrcmgr, spec)
#font = PDFType3Font(rsrcmgr, spec)
font = PDFCIDFont(rsrcmgr, spec)
for font_id, font in fontmap.items():
print(
'------------------------------------------------------------'
)
print('Descriptor: {}.'.format(font.descriptor))
print('\tFont name: {}, Font type: {}.'.format(
font.fontname,
type(font).__name__))
if hasattr(font, 'basefont'):
print('\tBase font: {}.'.format(font.basefont))
if hasattr(font, 'flags'):
print('\tFlags = {}.'.format(font.flags))
if hasattr(font, 'default_width') and hasattr(font, 'widths'):
print('\tDefault width = {}, Widths = {}.'.format(
font.default_width, font.widths))
print('\tAscent: {}, {}.'.format(font.ascent,
font.get_ascent()))
print('\tDescent: {}, {}.'.format(font.descent,
font.get_descent()))
if hasattr(font, 'hscale') and hasattr(font, 'vscale'):
print('\tScale: {}, {}.'.format(font.hscale, font.vscale))
if hasattr(font, 'leading') and hasattr(font, 'italic_angle'):
print('\tLeading = {}, Italic angle = {}.'.format(
font.leading, font.italic_angle))
print('\tBbox = {}.'.format(font.bbox))
if hasattr(font, 'get_width') and hasattr(font, 'get_height'):
print('\t(width, height) = ({}, {}).'.format(
font.get_width(), font.get_height()))
if hasattr(font, 'is_multibyte') and hasattr(
font, 'is_vertical'):
print('\tis_multibyte = {}, is_vertical = {}.'.format(
font.is_multibyte(), font.is_vertical()))
if hasattr(font, 'cid2unicode') and hasattr(
font, 'unicode_map'):
print('\tcid2unicode = {}, unicode_map = {}.'.format(
font.cid2unicode, font.unicode_map))
#if hasattr(font, 'char_disp'):
# print('\tchar_disp({}) = {}.'.format(cid, font.char_disp(cid)))
#if hasattr(font, 'to_unichr'):
# print('\tto_unichr({}) = {}.'.format(cid, font.to_unichr(cid)))
#if hasattr(font, 'char_width') and hasattr(font, 'string_width'):
# print('\tchar_width({}) = {}, string_width({}) = {}.'.format(cid, font.char_width(cid), s, font.string_width(s)))
for cs_id, cs in csmap.items():
print('CS ID: {}.'.format(cs_id))
print('\t{}.'.format(cs))
for xobj_id, xobj in xobjmap.items():
print('XObj ID: {}.'.format(xobj_id))
print('\t{}.'.format(xobj))
except FileNotFoundError as ex:
print('File not found, {}: {}.'.format(pdf_filepath, ex))
except Exception as ex:
print('Unknown exception raised in {}: {}.'.format(pdf_filepath, ex))
finally:
if fp: fp.close()
def get_font(self, objid, spec):
font = PDFResourceManager.get_font(self, objid, spec)
# Correct broken fond - either it has an Encoding or a Unicode_map for text extraction
if literal_name(spec['Encoding']) == 'WinAnsiEncoding':
font.unicode_map = None
return font
def get_font(self, objid, spec):
font = PDFResourceManager.get_font(self, objid, spec)
font.unicode_map = None
return font
