def __init__(self, pdf, page_obj, page_number=None, initial_doctop=0, clean_unicode=True):
self.pdf = pdf
self.page_obj = page_obj
self.page_number = page_number
self.rotation = self.page_obj.attrs.get("Rotate", 0) % 360
self.page_obj.rotate = self.rotation
self.initial_doctop = self.decimalize(initial_doctop)
cropbox = page_obj.attrs.get("CropBox")
mediabox = page_obj.attrs.get("MediaBox")
self.cropbox = self.decimalize(resolve_all(cropbox)) if cropbox is not None else None
self.mediabox = self.decimalize(resolve_all(mediabox) or self.cropbox)
m = self.mediabox
if self.rotation in [ 90, 270 ]:
self.bbox = self.decimalize((
min(m[1], m[3]),
min(m[0], m[2]),
max(m[1], m[3]),
max(m[0], m[2]),
))
else:
self.bbox = self.decimalize((
min(m[0], m[2]),
min(m[1], m[3]),
max(m[0], m[2]),
max(m[1], m[3]),
))
if clean_unicode:
has_unicode = any(ord(c['text']) < 128 for c in self.chars)
if has_unicode:
self.chars = [i.update({'text':unicodedata.normalize(i['text'])}) for i in self.chars]
def __init__(self, pdf, page_obj, page_number=None, initial_doctop=0):
self.pdf = pdf
self.page_obj = page_obj
self.page_number = page_number
self.rotation = self.page_obj.attrs.get("Rotate", 0) % 360
self.page_obj.rotate = self.rotation
self.initial_doctop = self.decimalize(initial_doctop)
cropbox = page_obj.attrs.get("CropBox")
mediabox = page_obj.attrs.get("MediaBox")
self.cropbox = self.decimalize(resolve_all(cropbox)) if cropbox is not None else None
self.mediabox = self.decimalize(resolve_all(mediabox) or self.cropbox)
m = self.mediabox
if self.rotation in [ 90, 270 ]:
self.bbox = self.decimalize((
min(m[1], m[3]),
min(m[0], m[2]),
max(m[1], m[3]),
max(m[0], m[2]),
))
else:
self.bbox = self.decimalize((
min(m[0], m[2]),
min(m[1], m[3]),
max(m[0], m[2]),
max(m[1], m[3]),
))
def _decimalize(v, q=None):
# If already a decimal, just return itself
if type(v) == Decimal:
return v
# If tuple/list passed, bulk-convert
elif isinstance(v, (tuple, list)):
return type(v)(decimalize(x, q) for x in v)
# If PDFObjRef passed, resolve it
elif isinstance(v, PDFObjRef):
return decimalize(resolve_all(v), q)
# Convert int-like
elif isinstance(v, numbers.Integral):
return Decimal(int(v))
# Convert float-like
elif isinstance(v, numbers.Real):
if q != None:
return Decimal(repr(v)).quantize(Decimal(repr(q)),
rounding=ROUND_HALF_UP)
else:
return Decimal(repr(v))
else:
raise ValueError("Cannot convert {0} to Decimal.".format(v))
def process_object(obj):
attr = dict((k, CONVERSIONS[k](resolve_all(v)))
for k, v in obj.__dict__.items() if k not in IGNORE)
kind = re.sub(lt_pat, "", obj.__class__.__name__).lower()
attr["object_type"] = kind
attr["page_number"] = pno
if hasattr(obj, "get_text"):
attr["text"] = obj.get_text()
if kind == "curve":
attr["points"] = list(map(point2coord, obj.pts))
if attr.get("y0") != None:
attr["top"] = h - attr["y1"]
attr["bottom"] = h - attr["y0"]
attr["doctop"] = idc + attr["top"]
if objects.get(kind) == None:
objects[kind] = []
objects[kind].append(attr)
if hasattr(obj, "_objs"):
for child in obj._objs:
process_object(child)
def _parse_info(doc):
xref = doc.xrefs[0]
info_ref=xref.trailer.get('Info')
if info_ref:
info=resolve_all(info_ref)
return info
