rclzip
112 lines (102 with data), 4.2 kB
#!/usr/bin/env python
# Zip file filter for Recoll
import rclexecm
from zipfile import ZipFile
# Note about file names (python 2.6. 2.7, don't know about 3.)
#
# There is a bit in zip entries to indicate if the filename is encoded
# as utf-8 or not. If the bit is set, zipfile decodes the file name
# and stores it in the catalog as an unicode object. Else it uses a
# binary string.
#
# When reading the file, the input file name is used directly as an
# index into the catalog.
#
# When we send the file name data to the indexer, we have to serialize
# it as byte string, we can't pass unicode objects to and fro. This
# means that we have to test if the name is unicode. If it is, we send
# the string encoded as utf-8. When reading, if the input is utf-8, we
# turn it to unicode and use this to access the zip member, else we
# use the binary string.
#
# In the case where an archive member name is a valid non-ascii utf-8
# string, but the flag is not set (which could probably happen if the
# archiver did not try to detect utf-8 file names), this will fail,
# because we'll convert back the utf-8 string to unicode and pass this
# to zipfile, but an utf-8 string, not a unicode object is actually in
# the catalog in this case, so the access will fail (will be seen at
# preview or open time). This does not affect ascii file names because
# the representation is the same anyway.
#
# To avoid this problem, we'd need to pass a separate bit of
# information indicating that encoding was performed, not just rely on
# the utf-8 validity test (ie have a 1st char switch), but this would be
# incompatible with existing indexes. Instead we try both ways...
#
class ZipExtractor:
def __init__(self, em):
self.currentindex = 0
self.em = em
def extractone(self, ipath):
#self.em.rclog("extractone: [%s]" % ipath)
docdata = ""
try:
info = self.zip.getinfo(ipath)
# There could be a 4GB Iso in the zip. We have to set a limit
if info.file_size > self.em.maxmembersize:
self.em.rclog("extractone: entry %s size %d too big" %
(ipath, info.file_size))
docdata = ""
#raise BadZipfile()
else:
docdata = self.zip.read(ipath)
ok = True
except Exception, err:
self.em.rclog("extractone: failed: [%s]" % err)
ok = False
iseof = rclexecm.RclExecM.noteof
if self.currentindex >= len(self.zip.namelist()) -1:
iseof = rclexecm.RclExecM.eofnext
if isinstance(ipath, unicode):
ipath = ipath.encode("utf-8")
return (ok, docdata, ipath, iseof)
###### File type handler api, used by rclexecm ---------->
def openfile(self, params):
self.currentindex = -1
try:
self.zip = ZipFile(params["filename:"])
return True
except:
return False
def getipath(self, params):
ipath = params["ipath:"]
ok, data, ipath, eof = self.extractone(ipath)
if ok:
return (ok, data, ipath, eof)
# Not found. Maybe we need to decode the path?
try:
ipath = ipath.decode("utf-8")
return self.extractone(ipath)
except Exception, err:
return (ok, data, ipath, eof)
def getnext(self, params):
if self.currentindex == -1:
# Return "self" doc
self.currentindex = 0
self.em.setmimetype('text/plain')
if len(self.zip.namelist()) == 0:
eof = rclexecm.RclExecM.eofnext
else:
eof = rclexecm.RclExecM.noteof
return (True, "", "", eof)
if self.currentindex >= len(self.zip.namelist()):
#self.em.rclog("getnext: EOF hit")
return (False, "", "", rclexecm.RclExecM.eofnow)
else:
ret= self.extractone(self.zip.namelist()[self.currentindex])
self.currentindex += 1
return ret
# Main program: create protocol handler and extractor and run them
proto = rclexecm.RclExecM()
extract = ZipExtractor(proto)
rclexecm.main(proto, extract)