def __init__(self, source):
name = source.BaseFont[1:]
self.name = self.lookup.get(name, name)
self.remap = chr
self.twobyte = False
info = source.ToUnicode
if not info:
return
info = info.stream.split('beginbfchar')[1].split('endbfchar')[0]
info = list(PdfTokens(info))
assert not len(info) & 1
info2 = []
for x in info:
assert x[0] == '<' and x[-1] == '>' and len(x) in (4, 6), x
i = int(x[1:-1], 16)
info2.append(i)
self.remap = dict(
(x, chr(y)) for (x, y) in zip(info2[::2], info2[1::2])).get
self.twobyte = len(info[0]) > 4
from music21 import *
from pdfrw import PdfReader, PdfWriter, PdfTokens
from pdfrw.findobjs import page_per_xobj
CLEF_MAPPING = {"(&)": clef.TrebleClef,
"(V)": clef.Treble8vbClef,
"(?)": clef.BassClef}
inpfn, = sys.argv[1:]
outfn = 'extract.' + os.path.basename(inpfn)
doc = PdfReader(inpfn)
page = doc.pages[0]
# page.Contents.stream = page.Contents.stream[:21000]
tokens = PdfTokens(page.Contents.stream)
indent = 0
commands = ["q", "Q", "ET", "BT", "cm", "Tm", "Tf", "s", "m", "l", "S", "TJ", "f", "Tj", "k", "re", "W", "n", "K", "w", "c"]
params = []
items = []
subcommands = []
for token in tokens:
if token == "q":
indent += 1
if subcommands:
items.append(subcommands)
subcommands = []
elif token == "Q":
indent -= 1
items.append(subcommands)
#print(subcommands)