def pdftk_merge(out_fn, inputs):
cmd = ([PDFTK] + [to_filename(inp, opened_ok=False) for inp in inputs] +
['cat', 'output', out_fn])
LOG.info('pdftk_merge: %r', cmd)
subprocess.check_call(cmd)
unlink(cmd[1])
return out_fn
def imagick_exec(todo, args, inp=None, executable=None):
if not executable:
executable = EXECUTABLE
LOG.debug('EXE=%r cmd_end=%r', executable, args)
if os.path.basename(executable) in ('gm', 'gm.exe'):
cmd = [executable, todo]
else:
executable = replace_bbase(executable, todo)
cmd = [executable]
try:
inpfn, outfn = None, None
data = None
kwds = {}
if USE_TEMPFILES and (inp or args[1].endswith('-')):
if inp: # stdin
inp = MMStringIO(inp).getvalue()
ext = '.' + args[0].split(':', 1)[0] if ':' in args[0] else ''
inpfd, inpfn = tempfile.mkstemp('-IM' + ext)
os.write(inpfd, inp)
os.close(inpfd)
args[0] = inpfn
assert os.path.exists(inpfn), inpfn
assert os.path.getsize(inpfn) == len(inp), \
'inp=%d <> %s=%d' % (len(inp), inpfn,
os.path.getsize(inpfn))
if args[1].endswith('-'): # stdout
ext = '.' + args[1].split(':', 1)[0] if ':' in args[1] else ''
outfd, outfn = tempfile.mkstemp('-IM' + ext)
os.close(outfd)
args[1] = outfn
assert os.path.exists(outfn), outfn
else:
data = MMStringIO(inp).getvalue() if inp else ''
if args[0].endswith('-'):
assert data, args
LOG.debug('data=%r...%r', data[:100], data[-100:])
kwds.update({'stdin': subprocess.PIPE, 'stdout': subprocess.PIPE})
LOG.debug('cmd=%r args=%r kwds=%r => %s', cmd, args, kwds,
' '.join(cmd + args))
p = subprocess.Popen(cmd + args)
out, err = p.communicate(data)
if inpfn:
unlink(inpfn)
assert not err and p.returncode in (None, 0), (p.returncode, err)
res = (MMFile(outfn, 'rb', delete_on_close=True) if outfn
else StringIO(out))
LOG.debug("imagick_exec(%s, %r): %r", todo, args, res)
if outfn:
unlink(outfn)
return res
except OSError:
if not executable or EXECUTABLE == executable:
executable2 = replace_bbase(executable, todo)
if executable2 == executable:
raise
return convert(inp, args[1].split(':', 1)[0],
executable=executable2)
else:
raise
def pdftk_split(stream):
pref = 'page_'
with mktempdir('pdf', delete_asap=True) as out_dir:
cmd = [PDFTK, to_filename(stream), 'burst', 'output',
out_dir + '/' + pref + '%03d.pdf']
LOG.debug('pdftk_split: %r', cmd)
subprocess.check_call(cmd)
files = sorted(os.listdir(out_dir))
LOG.debug('out_dir=%s: %s', out_dir, files)
for fn in files:
if fn.startswith(pref) and fn.endswith('.pdf'):
yield MMFile(os.path.join(out_dir, fn), 'rb',
delete_on_close=True)
unlink(cmd[1])
def _request_files(files=None):
if files:
for elt in files:
yield elt
return
fn = None
if SAVE_REQ_DIR:
with _SRD_lock:
fns = os.listdir(SAVE_REQ_DIR)
if len(fns) > SAVE_REQ_DIR_THRESHOLD:
for x in sorted(fns)[:SAVE_REQ_DIR_THRESHOLD // 2]:
unlink(os.path.join(SAVE_REQ_DIR, x))
fn = os.path.join(SAVE_REQ_DIR,
str(int(time.time() * 1000)) + '--' +
('%06d' % REQ_COUNTER) +
urllib.quote_plus(request.url, safe=''))
import gzip
files = ((stor.stream, stor.mimetype, stor.name)
for _, stor in request.files.iteritems(multi=True))
try:
first = files.next()
files = itertools.chain([first], files)
except StopIteration: # nothing there
LOG.warn('empty "files"')
if request.data:
import re
rgx = re.compile('''filename=['"](.*?)["']''')
files = [(StringIO(request.data), request.headers['Content-Type'],
rgx.search(request.headers.get('Content-Disposition',
'filename=""')).group(1))]
LOG.info('files=%r', files)
i = 0
for stream, mimetype, name in files:
if len(stream.read(1)) == 1: # skip empty files
stream.seek(0)
if fn:
with gzip.open(fn + ('--file%02d' % i) + '.dmp.gz',
'wb') as fh:
while 1:
chunk = stream.read(65536)
if not chunk:
break
fh.write(chunk)
stream.seek(0)
i += 1
yield (stream, mimetype, basename(name))
except Exception, e:
LOG.error('pdftk_split cannot split %s: %s [decrypt? %s]',
stream_s, e, decrypt)
if not decrypt:
raise
#pdffn = to_filename(stream)
try:
pdffn = clean_pdf(pdffn)
except:
LOG.exception('error with PDF->PS->PDF of %s', pdffn)
raise
LOG.info('trying again with %r', pdffn)
try:
splitted = split_pdf(to_stream(pdffn), decrypt=False)
finally:
unlink(pdffn)
LOG.debug('splitted files=%r', splitted)
return splitted
try:
from pyPdf import PdfFileWriter as PFW
PdfFileWriter = PFW
del PFW
except ImportError:
PdfFileWriter = None
def simple_split(stream):
stream_s = repr(stream)[:100]
pages = list(get_pages(stream))
