def all(self,firstpage=1):
doc = PDFDocument(self.filename)
pages = doc.count_pages()
chosen_pages = [i for i in xrange(firstpage,pages+1)]
self._l.info("%s - %d pages. All chosen",self.filename,pages)
def pi(n): return Page(self.filename,n,self.c)
return [ pi(p) for p in chosen_pages ]
def randompages(self,count,firstpage=1):
doc = PDFDocument(self.filename)
pages = doc.count_pages()
chosen_pages = random.sample([i for i in xrange(firstpage,pages+1)],min(pages-firstpage+1,count))
chosen_pages.sort()
self._l.info("%s - %d pages. %s chosen",self.filename,pages,chosen_pages)
def pi(n): return Page(self.filename,n,self.c)
return [ pi(p) for p in chosen_pages ]
sys.path.append("..")
from sheetmusic import *
from gamera.core import *
init_gamera()
gspath = '"c:/Program Files/gs/gs8.70/bin/gswin32c.exe"'.replace("/", "\\")
FORMAT = "%(asctime)-15s %(levelname)s [%(name)s.%(funcName)s] %(message)s"
logging.basicConfig(level=logging.DEBUG, format=FORMAT, filename="convert.log")
for file in sys.argv[1:]:
print "Converting %s" % file
start = time.time()
try:
doc = PDFDocument(file)
pages = doc.count_pages()
chosen_pages = random.sample([i for i in xrange(1, pages + 1)], min(pages, 10))
chosen_pages.sort()
logging.info("%s - %d pages. %s chosen", file, pages, chosen_pages)
dirname = file[:-4]
if not os.path.exists(dirname):
os.mkdir(dirname)
for page in chosen_pages:
sys.stdout.flush()
outfile = "%s/%s-page%d.tif" % (dirname, file[:-4], page)
outfile_staves = "%s/%s-page%d-staves.tif" % (dirname, file[:-4], page)
outfile_gamscript = "%s-page%d-class.py" % (file[:-4], page)
cmd = " ".join(
[
gspath,
from pdftools.pdffile import PDFDocument
from pdftools.pdftext import Text
def contents_to_text(contents):
for item in contents:
if isinstance(item, type([])):
for i in contents_to_text(item):
yield i
elif isinstance(item, Text):
yield item.text
doc = PDFDocument("declaration.pdf")
n_pages = doc.count_pages()
text = []
for n_page in range(1, (n_pages + 1)):
print "Page", n_page
page = doc.read_page(n_page)
contents = page.read_contents().contents
text.extend(contents_to_text(contents))
print "".join(text)
f = open("ok.txt", "w")
f.write("".join(text))
f.close()
sys.path.append("..")
from sheetmusic import *
from gamera.core import *
init_gamera()
gspath = "\"c:/Program Files/gs/gs8.70/bin/gswin32c.exe\"".replace("/", "\\")
FORMAT = "%(asctime)-15s %(levelname)s [%(name)s.%(funcName)s] %(message)s"
logging.basicConfig(level=logging.DEBUG, format=FORMAT, filename="convert.log")
for file in sys.argv[1:]:
print "Converting %s" % file
start = time.time()
try:
doc = PDFDocument(file)
pages = doc.count_pages()
chosen_pages = random.sample([i for i in xrange(1, pages + 1)],
min(pages, 10))
chosen_pages.sort()
logging.info("%s - %d pages. %s chosen", file, pages, chosen_pages)
dirname = file[:-4]
if not os.path.exists(dirname):
os.mkdir(dirname)
for page in chosen_pages:
sys.stdout.flush()
outfile = "%s/%s-page%d.tif" % (dirname, file[:-4], page)
outfile_staves = "%s/%s-page%d-staves.tif" % (dirname, file[:-4],
page)
outfile_gamscript = "%s-page%d-class.py" % (file[:-4], page)
cmd = ' '.join([