def do_Do(self, xobjid):
xobjid = literal_name(xobjid)
try:
xobj = stream_value(self.xobjmap[xobjid])
except KeyError:
if STRICT:
raise PDFInterpreterError('Undefined xobject id: %r' % xobjid)
return
if 1 <= self.debug:
print >>stderr, 'Processing xobj: %r' % xobj
subtype = xobj.dic.get('Subtype')
if subtype is LITERAL_FORM and 'BBox' in xobj.dic:
interpreter = self.dup()
bbox = list_value(xobj.dic['BBox'])
matrix = list_value(xobj.dic.get('Matrix', MATRIX_IDENTITY))
self.device.begin_figure(xobjid, bbox, matrix)
interpreter.render_contents(dict_value(xobj.dic.get('Resources')), [xobj], ctm=mult_matrix(matrix, self.ctm))
self.device.end_figure(xobjid)
elif subtype is LITERAL_IMAGE and 'Width' in xobj.dic and 'Height' in xobj.dic:
self.device.begin_figure(xobjid, (0,0,1,1), MATRIX_IDENTITY)
(w,h) = (xobj.dic['Width'], xobj.dic['Height'])
self.device.render_image(xobj, (w,h))
self.device.end_figure(xobjid)
else:
# unsupported xobject type.
pass
return
def do_Do(self, xobjid):
# the base of this function is basically copy-pasted from ancestor; unfortunately, I found no better solution
xobjid = literal_name(xobjid)
try:
xobj = stream_value(self.xobjmap[xobjid])
except KeyError:
if STRICT:
raise PDFInterpreterError("Undefined xobject id: %r" % xobjid)
return
if self.debug:
logging.info("Processing xobj: %r" % xobj)
subtype = xobj.get("Subtype")
if subtype is LITERAL_FORM and "BBox" in xobj:
interpreter = self.dup()
interpreter.is_first_level_call = None
bbox = list_value(xobj["BBox"])
matrix = list_value(xobj.get("Matrix", MATRIX_IDENTITY))
# According to PDF reference 1.7 section 4.9.1, XObjects in
# earlier PDFs (prior to v1.2) use the page's Resources entry
# instead of having their own Resources entry.
resources = dict_value(xobj.get("Resources")) or self.resources.copy()
self.device.begin_figure(xobjid, bbox, matrix)
interpreter.render_contents(resources, [xobj], ctm=mult_matrix(matrix, self.ctm))
self.device.end_figure(xobjid)
# for (k,v) in interpreter.text_lines.iteritems():
# self.text_sequences[k + self.keyword_count] = v
self.keyword_count += interpreter.keyword_count
print "Included %i keywords" % interpreter.keyword_count
else:
# ignored xobject type.
pass
return
def draw_cid(self, ts, cid, force_space=False):
verbose("drawing cid: ", cid)
Trm = utils.mult_matrix((ts.Tfs * ts.Th, 0, 0, ts.Tfs, 0, ts.Trise),
ts.Tm)
if Trm[1] != 0:
return
if Trm[2] != 0:
return
verbose('Trm', Trm)
if cid == 32 or force_space:
Tw = ts.Tw
else:
Tw = 0
try:
if force_space:
unichar = ' '
else:
try:
unichar = ts.Tf.to_unichr(cid)
except Exception as e:
verbose(f"Failed to process {cid = }: {e}")
unichar = ' '
except PDFUnicodeNotDefined:
if MISSING_CHAR:
unichar = MISSING_CHAR
else:
raise
(gx, gy) = utils.apply_matrix_pt(Trm, (0, 0))
verbose("drawing unichar: '", unichar, "' @", gx, ",", gy)
tfs = Trm[0]
if self.current_block is None:
self.current_block = (ts.Tf, tfs, gx, gy, [unichar])
elif ((self.current_block[0] == ts.Tf)
and (self.current_block[1] == tfs)):
self.current_block[4].append(unichar)
else:
self.blocks.append(self.current_block)
self.current_block = (ts.Tf, tfs, gx, gy, [unichar])
verbose('current block: ', self.current_block)
verbose('blocks: ', self.blocks)
if force_space:
pass
else:
w = ts.Tf.char_width(cid)
if ts.Tf.is_vertical():
tx = 0
ty = self.new_ty(w, 0, ts.Tfs, ts.Tc, Tw)
else:
tx = self.new_tx(w, 0, ts.Tfs, ts.Tc, Tw, ts.Th)
ty = 0
ts.Tm = utils.translate_matrix(ts.Tm, (tx, ty))
def render_string(self, textstate, seq):
matrix = mult_matrix(textstate.matrix, self.ctm)
font = textstate.font
fontsize = textstate.fontsize
scaling = textstate.scaling * .01
charspace = textstate.charspace * scaling
wordspace = textstate.wordspace * scaling
rise = textstate.rise
if font.is_multibyte():
wordspace = 0
dxscale = .001 * fontsize * scaling
if font.is_vertical():
textstate.linematrix = self.render_string_vertical(
seq, matrix, textstate.linematrix, font, fontsize,
scaling, charspace, wordspace, rise, dxscale)
else:
textstate.linematrix = self.render_string_horizontal(
seq, matrix, textstate.linematrix, font, fontsize,
scaling, charspace, wordspace, rise, dxscale)
return
def do_cm(self, a1, b1, c1, d1, e1, f1): self.ctm = mult_matrix((a1,b1,c1,d1,e1,f1), self.ctm) self.device.set_ctm(self.ctm) return
def begin_figure(self, name, bbox, matrix):
self._stack.append(self.cur_item)
self.cur_item = LTFigure(name, bbox, mult_matrix(matrix, self.ctm))
return
def show_string(self, ts, array):
verbose(ts)
sentence = []
word = []
m = (ts.Tfs * ts.Th, 0, 0, ts.Tfs, 0, ts.Trise)
applicable_Tm = utils.mult_matrix(m, ts.Tm)
(sx, _, _, sy, tx, ty) = applicable_Tm
current_state = (sx, sy, tx, ty)
if self.last_state == None:
self.paragraph = []
verbose('current paragraph becomes=', self.paragraph)
elif current_state[0] == self.last_state[0]:
verbose('DECISION: grouping the text object to last')
else:
verbose('DECISION: finalizing the paragraph')
key = self.last_state[0]
item = self.paragraph_map.get(key, '')
if len(item) > 0:
item = item = ' '
new_item = ' '.join(self.paragraph)
self.paragraph_map[key] = item + new_item
self.paragraph = []
verbose('current paragraph becomes=', self.paragraph)
self.last_state = current_state
for obj in array:
verbose("processing obj=", obj)
if utils.isnumber(obj):
Tj = obj
if Tj < WITHIN_WORD_MOVE_LIMIT:
verbose("DECISION: new word")
sentence.append(''.join(word))
verbose('current sentence becomes=', sentence)
word = []
verbose('current word becomes=', word)
else:
verbose("DECISION: move inside the current word")
if ts.Tf.is_vertical():
tx = 0
ty = ((Tj / 1000) * ts.Tfs)
else:
tx = ((Tj / 1000) * ts.Tfs) * ts.Th
ty = 0
ts.Tm = utils.mult_matrix((1, 0, 0, 1, tx, ty), ts.Tm)
else:
for cid in ts.Tf.decode(obj):
verbose("processing cid=", cid)
m = (ts.Tfs * ts.Th, 0, 0, ts.Tfs, 0, ts.Trise)
applicable_Tm = utils.mult_matrix(m, ts.Tm)
if cid == 32:
applicable_Tw = ts.Tw
sentence.append(''.join(word))
verbose('current sentence becomes=', sentence)
word = []
else:
try:
text = ts.Tf.to_unichr(cid)
except PDFUnicodeNotDefined:
if MISSING_CHAR:
text = MISSING_CHAR
else:
raise
word.append(text)
verbose('current word becomes=', word)
applicable_Tw = 0
w = ts.Tf.char_width(cid)
if ts.Tf.is_vertical():
tx = 0
ty = ((w - 0) * ts.Tfs + ts.Tc + applicable_Tw)
else:
tx = ((w - 0) * ts.Tfs + ts.Tc + applicable_Tw) * ts.Th
ty = 0
ts.Tm = utils.mult_matrix((1, 0, 0, 1, tx, ty), ts.Tm)
if len(word) > 0:
sentence.append(''.join(word))
verbose('current sentence becomes=', sentence)
word = []
verbose('current word becomes=', word)
self.paragraph.append(' '.join(sentence))
verbose('current paragraph becomes=', self.paragraph)
return
def do_Td(self, tx, ty):
verbose_operator("PDF OPERATOR Td: tx=", tx, ", ty=", ty)
m = (1, 0, 0, 1, tx, ty)
self.mpts.Tlm = utils.mult_matrix(m, self.mpts.Tlm)
self.mpts.Tm = self.mpts.Tlm
return
def begin_figure(self, name, bbox, matrix):
super(PDFLocPageAnalyzer, self).begin_figure(name, bbox, matrix)
self.cur_item = PDFLocFigure(name, bbox, mult_matrix(matrix, self.ctm))