def begin_page(self, page, ctm):
(x0, y0, x1, y1) = page.mediabox
(x0, y0) = apply_matrix_pt(ctm, (x0, y0))
(x1, y1) = apply_matrix_pt(ctm, (x1, y1))
mediabox = (0, 0, abs(x0-x1), abs(y0-y1))
self.cur_item = LTPage(self.pageno, mediabox)
return
def begin_page(self, page: PDFPage, ctm: Tuple[int, int, int, int, int,
int]) -> None:
(x0, y0, x1, y1) = page.mediabox
(x0, y0) = apply_matrix_pt(ctm, (x0, y0))
(x1, y1) = apply_matrix_pt(ctm, (x1, y1))
mediabox = (0, 0, abs(x0 - x1), abs(y0 - y1))
self.cur_item = LTPageExtended(self.pageno, mediabox)
return
def paint_single_path(self, gstate, stroke, fill, evenodd, path):
'''
Converting a single path draw command into lines and curves objects
'''
if len(path) < 2:
return
shape = ''.join(x[0] for x in path)
pts = []
for p in path:
for i in xrange(1, len(p), 2):
pts.append(apply_matrix_pt(self.ctm, (p[i], p[i + 1])))
# Line mode
if self.line_only_shape.match(shape):
# check for sloped lines first
has_slope = False
for i in xrange(len(pts) - 1):
if pts[i][0] != pts[i + 1][0] and pts[i][1] != pts[i + 1][1]:
has_slope = True
break
if not has_slope:
for i in xrange(len(pts) - 1):
self.cur_item.add(
LTLine(gstate.linewidth, pts[i], pts[i + 1]))
# Adding the closing line for a polygon, especially rectangles
if shape.endswith('h'):
self.cur_item.add(LTLine(gstate.linewidth, pts[0],
pts[-1]))
return
# Add the curve as an arbitrary polyline (belzier curve info is lost here)
self.cur_item.add(LTCurve(gstate.linewidth, pts))
def paint_single_path(self, gstate, stroke, fill, evenodd, path):
'''
Converting a single path draw command into lines and curves objects
'''
if len(path) < 2:
return
shape = ''.join(x[0] for x in path)
pts = []
for p in path:
for i in xrange(1, len(p), 2):
pts.append(apply_matrix_pt(self.ctm, (p[i], p[i+1])))
# Line mode
if self.line_only_shape.match(shape):
# check for sloped lines first
has_slope = False
for i in xrange(len(pts)-1):
if pts[i][0] != pts[i+1][0] and pts[i][1] != pts[i+1][1]:
has_slope = True
break
if not has_slope:
for i in xrange(len(pts) - 1):
self.cur_item.add(LTLine(gstate.linewidth, pts[i], pts[i + 1]))
# Adding the closing line for a polygon, especially rectangles
if shape.endswith('h'):
self.cur_item.add(LTLine(gstate.linewidth, pts[0], pts[-1]))
return
# Add the curve as an arbitrary polyline (belzier curve info is lost here)
self.cur_item.add(LTCurve(gstate.linewidth, pts))
def _paint_network(self, path):
network = LTNetwork()
previous_element = None
for tuple in path:
if len(tuple) == 3:
state, x, y = tuple
element = (x, y)
element = apply_matrix_pt(self.ctm, element)
element = int_round(element[0]), int_round(element[1])
element = Point(element)
elif len(tuple) == 1 and tuple[0] == 'h':
element = previous_element
state = 'l'
else:
raise IndexError
network.add_point(element)
# state 'l' is to draw a line, which means that we have link.
# Otherwise, we only have the point.
if state == 'l':
network.add_link(element, previous_element)
previous_element = element
self.cur_item.add(network)
def paint_path(self, gstate, stroke, fill, evenodd, path):
shape = ''.join(x[0] for x in path)
if shape == 'ml':
# horizontal/vertical line
(_, x0, y0) = path[0]
(_, x1, y1) = path[1]
(x0, y0) = apply_matrix_pt(self.ctm, (x0, y0))
(x1, y1) = apply_matrix_pt(self.ctm, (x1, y1))
if x0 == x1 or y0 == y1:
self.cur_item.add(LTLine(gstate.linewidth, (x0, y0), (x1, y1)))
return
if shape == 'mlllh':
# rectangle
(_, x0, y0) = path[0]
(_, x1, y1) = path[1]
(_, x2, y2) = path[2]
(_, x3, y3) = path[3]
(x0, y0) = apply_matrix_pt(self.ctm, (x0, y0))
(x1, y1) = apply_matrix_pt(self.ctm, (x1, y1))
(x2, y2) = apply_matrix_pt(self.ctm, (x2, y2))
(x3, y3) = apply_matrix_pt(self.ctm, (x3, y3))
if ((x0 == x1 and y1 == y2 and x2 == x3 and y3 == y0)
or (y0 == y1 and x1 == x2 and y2 == y3 and x3 == x0)):
self.cur_item.add(LLTRect(gstate.linewidth, (x0, y0, x2, y2)))
print self.interpreter.scs, self.interpreter.ncs, gstate, stroke, fill, (
x0, y0, x2, y2)
return
# other shapes
pts = []
for p in path:
for i in xrange(1, len(p), 2):
pts.append(apply_matrix_pt(self.ctm, (p[i], p[i + 1])))
self.cur_item.add(LTCurve(gstate.linewidth, pts))
return
def newLTCharInit(self, matrix, font, fontsize, scaling, rise, text, textwidth, textdisp): LTText.__init__(self) # Patched in line self.font = font self.fontsize = fontsize self._text = text self.matrix = matrix self.fontname = font.fontname self.adv = textwidth * fontsize * scaling # compute the boundary rectangle. if font.is_vertical(): # vertical width = font.get_width() * fontsize (vx, vy) = textdisp if vx is None: vx = width//2 else: vx = vx * fontsize * .001 vy = (1000 - vy) * fontsize * .001 tx = -vx ty = vy + rise bll = (tx, ty+self.adv) bur = (tx+width, ty) else: # horizontal height = font.get_height() * fontsize descent = font.get_descent() * fontsize ty = descent + rise bll = (0, ty) bur = (self.adv, ty+height) (a, b, c, d, e, f) = self.matrix self.upright = (0 < a*d*scaling and b*c <= 0) (x0, y0) = apply_matrix_pt(self.matrix, bll) (x1, y1) = apply_matrix_pt(self.matrix, bur) if x1 < x0: (x0, x1) = (x1, x0) if y1 < y0: (y0, y1) = (y1, y0) LTComponent.__init__(self, (x0, y0, x1, y1)) if font.is_vertical(): self.size = self.width else: self.size = self.height return
def paint_path(self, gstate, stroke, fill, evenodd, path):
shape = ''.join(x[0] for x in path)
if shape == 'ml': # horizontal/vertical line
(_,x0,y0) = path[0]
(_,x1,y1) = path[1]
(x0,y0) = apply_matrix_pt(self.ctm, (x0,y0))
(x1,y1) = apply_matrix_pt(self.ctm, (x1,y1))
if y0 == y1:
# horizontal ruler
self.cur_item.add(LTLine(gstate.linewidth, 'H', (x0,y0,x1,y1)))
elif x0 == x1:
# vertical ruler
self.cur_item.add(LTLine(gstate.linewidth, 'V', (x0,y0,x1,y1)))
elif shape == 'mlllh':
# rectangle
(_,x0,y0) = path[0]
(_,x1,y1) = path[1]
(_,x2,y2) = path[2]
(_,x3,y3) = path[3]
(x0,y0) = apply_matrix_pt(self.ctm, (x0,y0))
(x1,y1) = apply_matrix_pt(self.ctm, (x1,y1))
(x2,y2) = apply_matrix_pt(self.ctm, (x2,y2))
(x3,y3) = apply_matrix_pt(self.ctm, (x3,y2))
if ((x0 == x1 and y1 == y2 and x2 == x3 and y3 == y0) or
(y0 == y1 and x1 == x2 and y2 == y3 and x3 == x0)):
self.cur_item.add(LTRect(gstate.linewidth, (x0,y0,x2,y2)))
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 paint_path(self, gstate, stroke, fill, evenodd, path):
shape = ''.join(x[0] for x in path)
if shape == 'ml':
# horizontal/vertical line
(_, x0, y0) = path[0]
(_, x1, y1) = path[1]
(x0, y0) = utils.apply_matrix_pt(self.ctm, (x0, y0))
(x1, y1) = utils.apply_matrix_pt(self.ctm, (x1, y1))
if x0 == x1 or y0 == y1:
self.cur_item.add(LTLine(gstate.linewidth, (x0, y0), (x1, y1),
stroke, fill, evenodd, gstate.scolor, gstate.ncolor))
return
if shape == 'mlllh':
#print("Painting rectangle!")
# rectangle
(_, x0, y0) = path[0]
(_, x1, y1) = path[1]
(_, x2, y2) = path[2]
(_, x3, y3) = path[3]
(x0, y0) = utils.apply_matrix_pt(self.ctm, (x0, y0))
(x1, y1) = utils.apply_matrix_pt(self.ctm, (x1, y1))
(x2, y2) = utils.apply_matrix_pt(self.ctm, (x2, y2))
(x3, y3) = utils.apply_matrix_pt(self.ctm, (x3, y3))
if ((x0 == x1 and y1 == y2 and x2 == x3 and y3 == y0) or
(y0 == y1 and x1 == x2 and y2 == y3 and x3 == x0)):
xlist = [x0,x1,x2,x3]
ylist = [y0,y1,y2,y3]
minx = min(xlist)
maxx = max(xlist)
miny = min(ylist)
maxy = max(ylist)
self.rectangles.append(Rectangle(minx, miny, maxx-minx, maxy-miny))
self.cur_item.add(LTRect(gstate.linewidth, (x0, y0, x2, y2),
stroke, fill, evenodd, gstate.scolor, gstate.ncolor))
return
# other shapes
pts = []
for p in path:
for i in range(1, len(p), 2):
pts.append(utils.apply_matrix_pt(self.ctm, (p[i], p[i+1])))
self.cur_item.add(LTCurve(gstate.linewidth, pts, stroke, fill,
evenodd, gstate.scolor, gstate.ncolor))
return
def push_textbox(self, p, h, font, matrix):
if self.temp_text is not None:
(xt, yt) = utils.apply_matrix_pt(matrix, self.init_p)
self.text_boxes.append(TextBox(self.temp_text, xt,
yt, (p[0]-self.init_p[0])*matrix[0], h, font))
self.temp_text = None
def paint_single_path(self, gstate, stroke, fill, evenodd, path):
'''
Converting a single path draw command into lines and curves objects
'''
shape = ''.join(x[0] for x in path)
if shape == 'mh':
return
if shape == 'ml':
# horizontal/vertical line
(_, x0, y0) = path[0]
(_, x1, y1) = path[1]
(x0, y0) = apply_matrix_pt(self.ctm, (x0, y0))
(x1, y1) = apply_matrix_pt(self.ctm, (x1, y1))
if x0 == x1 or y0 == y1:
self.cur_item.add(LTLine(gstate.linewidth, (x0, y0), (x1, y1)))
return
if shape == 'mlllh':
# rectangle
(_, x0, y0) = path[0]
(_, x1, y1) = path[1]
(_, x2, y2) = path[2]
(_, x3, y3) = path[3]
(x0, y0) = apply_matrix_pt(self.ctm, (x0, y0))
(x1, y1) = apply_matrix_pt(self.ctm, (x1, y1))
(x2, y2) = apply_matrix_pt(self.ctm, (x2, y2))
(x3, y3) = apply_matrix_pt(self.ctm, (x3, y3))
if ((x0 == x1 and y1 == y2 and x2 == x3 and y3 == y0) or
(y0 == y1 and x1 == x2 and y2 == y3 and x3 == x0)):
self.cur_item.add(LTRect(gstate.linewidth, (x0, y0, x2, y2)))
return
# other shapes
pts = []
for p in path:
for i in xrange(1, len(p), 2):
pts.append(apply_matrix_pt(self.ctm, (p[i], p[i+1])))
direction = None
for i in xrange(len(pts)-1):
if pts[i][0]==pts[i+1][0]:
cur_dir = 'v'
elif pts[i][1]==pts[i+1][1]:
cur_dir = 'h'
else:
direction = None
break
if direction is None:
direction = cur_dir
elif direction != cur_dir:
direction = None
break
# implicitly if direction == cur_dir, continue
# If the entire segment has a consistent vertical/horizontal
# direction, we split all draw commands into segments instead
# of dumping the entire path as a curve
if direction is not None:
for i in xrange(len(pts)-1):
self.cur_item.add(LTLine(gstate.linewidth,pts[i],pts[i+1]))
else:# Add the curve as an arbitrary polyline (belzier curve info is lost here)
self.cur_item.add(LTCurve(gstate.linewidth, pts))
def paint_path(self, gstate, stroke, fill, evenodd, path):
shape = ''.join(x[0] for x in path)
if shape == 'ml':
# horizontal/vertical line
(_, x0, y0) = path[0]
(_, x1, y1) = path[1]
(x0, y0) = apply_matrix_pt(self.ctm, (x0, y0))
(x1, y1) = apply_matrix_pt(self.ctm, (x1, y1))
if x0 == x1 or y0 == y1:
x0, y0, x1, y1 = get_bound([(x0, y0), (x1, y1)])
item = add_wh({
"x0": x0,
"y0": y0,
"x1": x1,
"y1": y1,
"object_type": "line",
})
if self.interpreter.parse_styles:
item.update({
"stroke_width":
0 if stroke is False else gstate.linewidth,
"stroke":
None if stroke is False else
(gstate.scolor_hex or "#000000"),
"fill":
None if fill is False else
(gstate.ncolor_hex or "#000000"),
"dash":
gstate.dash,
"evenodd":
evenodd,
})
self.cur_item.add(item)
return
if shape == 'mlllh':
# rectangle
(_, x0, y0) = path[0]
(_, x1, y1) = path[1]
(_, x2, y2) = path[2]
(_, x3, y3) = path[3]
(x0, y0) = apply_matrix_pt(self.ctm, (x0, y0))
(x1, y1) = apply_matrix_pt(self.ctm, (x1, y1))
(x2, y2) = apply_matrix_pt(self.ctm, (x2, y2))
(x3, y3) = apply_matrix_pt(self.ctm, (x3, y3))
if ((x0 == x1 and y1 == y2 and x2 == x3 and y3 == y0)
or (y0 == y1 and x1 == x2 and y2 == y3 and x3 == x0)):
y_min = min(y0, y1, y2, y3)
y_max = max(y0, y1, y2, y3)
x_min = min(x0, x1, x2, x3)
x_max = max(x0, x1, x2, x3)
item = add_wh({
"x0": x_min,
"y0": y_min,
"x1": x_max,
"y1": y_max,
"object_type": "rect",
})
if self.interpreter.parse_styles:
item.update({
"stroke_width":
0 if stroke is False else gstate.linewidth,
"stroke":
None if stroke is False else
(gstate.scolor_hex or "#000000"),
"fill":
None if fill is False else
(gstate.ncolor_hex or "#000000"),
"dash":
gstate.dash,
"evenodd":
evenodd,
})
self.cur_item.add(item)
return
# other shapes
pts = []
for p in path:
for i in range(1, len(p), 2):
pts.append(apply_matrix_pt(self.ctm, (p[i], p[i + 1])))
x0, y0, x1, y1 = get_bound(pts)
item = add_wh({
"x0": x0,
"y0": y0,
"x1": x1,
"y1": y1,
"points": pts,
"path": path,
"object_type": "curve"
})
if self.interpreter.parse_styles:
item.update({
"stroke_width":
0 if stroke is False else gstate.linewidth,
"stroke":
None if stroke is False else (gstate.scolor_hex or "#000000"),
"fill":
None if fill is False else (gstate.ncolor_hex or "#000000"),
"dash":
gstate.dash,
"evenodd":
evenodd,
})
self.cur_item.add(item)
return
def render_char(self, matrix, font, fontsize, scaling, rise, cid):
try:
text = font.to_unichr(cid)
assert isinstance(text, six.text_type), text
except PDFUnicodeNotDefined:
text = self.handle_undefined_char(font, cid)
textwidth = font.char_width(cid)
textdisp = font.char_disp(cid)
adv = textwidth * fontsize * scaling
# compute the boundary rectangle.
if font.is_vertical():
# vertical
width = font.get_width() * fontsize
(vx, vy) = textdisp
if vx is None:
vx = width * 0.5
else:
vx = vx * fontsize * .001
vy = (1000 - vy) * fontsize * .001
tx = -vx
ty = vy + rise
bll = (tx, ty + self.adv)
bur = (tx + width, ty)
else:
# horizontal
height = font.get_height() * fontsize
descent = font.get_descent() * fontsize
ty = descent + rise
bll = (0, ty)
bur = (adv, ty + height)
(a, b, c, d, e, f) = matrix
upright = (0 < a * d * scaling and b * c <= 0)
(x0, y0) = apply_matrix_pt(matrix, bll)
(x1, y1) = apply_matrix_pt(matrix, bur)
if x1 < x0:
(x0, x1) = (x1, x0)
if y1 < y0:
(y0, y1) = (y1, y0)
if font.is_vertical():
size = x1 - x0
else:
size = y1 - y0
# cf. "Table 106 – Text rendering modes" of PDF 1.7 spec
gs = self.interpreter.graphicstate
r = self.interpreter.textstate.render
if r in (0, 2, 4, 6):
fill = gs.ncolor_hex
else:
fill = None
if r in (1, 2, 5, 6):
stroke = gs.scolor_hex
else:
stroke = None
item = add_wh({
"x0": x0,
"y0": y0,
"x1": x1,
"y1": y1,
"fontname": font.fontname,
"size": size,
"text": text,
"object_type": "char",
})
if self.interpreter.parse_styles:
item.update({
"fill": fill,
"stroke": stroke,
})
self.cur_item.add(item)
return adv