tris/recoll
1
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity

New python-based msword filter + basic arch to convert the others

Browse Source 
--HG--
branch : WINDOWSPORT
This commit is contained in:
Jean-Francois Dockes 2015-09-07 11:16:20 +02:00
parent 5d0e8330fb
commit e40cf64e66
2 changed files with 15 additions and 10 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

22
src/filters/rcldoc.py
View File

@ -6,7 +6,7 @@ import sys
import os import os
# Processing the output from antiword: create html header and tail, process # Processing the output from antiword: create html header and tail, process
# continuation lines escape HTML special characters, accumulate the data # continuation lines escape, HTML special characters, accumulate the data.
class WordProcessData: class WordProcessData:
def __init__(self, em): def __init__(self, em):
self.em = em self.em = em
@ -74,14 +74,18 @@ class WordPassData:
self.em.setmimetype("text/html") self.em.setmimetype("text/html")
return self.out return self.out
# Filter for msword docs. Try antiword, and if this fails, check for # Filter for msword docs. Try antiword, and if this fails, check for
# an rtf or text document (.doc are sometimes like this). Also try # an rtf or text document (.doc are sometimes like this...). Also try
# vwWare if the doc is actually a word doc # vwWare if the doc is actually a word doc
class WordFilter: class WordFilter:
def __init__(self, em, td): def __init__(self, em, td):
self.em = em self.em = em
self.ntry = 0 self.ntry = 0
self.thisdir = td self.execdir = td
def reset(self):
self.ntry = 0
def hasControlChars(self, data): def hasControlChars(self, data):
for c in data: for c in data:
@ -108,7 +112,7 @@ class WordFilter:
return "text/plain" return "text/plain"
def getCmd(self, fn): def getCmd(self, fn):
'''Return command to execute and postprocessor according to '''Return command to execute, and postprocessor, according to
our state: first try antiword, then others depending on mime our state: first try antiword, then others depending on mime
identification. Do 2 tries at most''' identification. Do 2 tries at most'''
if self.ntry == 0: if self.ntry == 0:
@ -116,16 +120,16 @@ class WordFilter:
return (["antiword", "-t", "-i", "1", "-m", "UTF-8"], return (["antiword", "-t", "-i", "1", "-m", "UTF-8"],
WordProcessData(self.em)) WordProcessData(self.em))
elif self.ntry == 1: elif self.ntry == 1:
ntry = 2 self.ntry = 2
# antiword failed. Check for an rtf file, or text and # antiword failed. Check for an rtf file, or text and
# process accordingly. It the doc is actually msword, try # process accordingly. It the doc is actually msword, try
# wvWare. # wvWare.
mt = self.mimetype(fn) mt = self.mimetype(fn)
if mt == "text/plain": if mt == "text/plain":
return ([os.path.join(self.thisdir,"rcltext")], return ([os.path.join(self.execdir,"rcltext")],
WordPassData(self.em)) WordPassData(self.em))
elif mt == "text/rtf": elif mt == "text/rtf":
return ([os.path.join(self.thisdir, "rclrtf")], return ([os.path.join(self.execdir, "rclrtf")],
WordPassData(self.em)) WordPassData(self.em))
elif mt == "application/msword": elif mt == "application/msword":
return (["wvWare", "--nographics", "--charset=utf-8"], return (["wvWare", "--nographics", "--charset=utf-8"],
@ -136,8 +140,8 @@ class WordFilter:
return ([],None) return ([],None)
if __name__ == '__main__': if __name__ == '__main__':
thisdir = os.path.dirname(sys.argv[0]) execdir = os.path.dirname(sys.argv[0])
proto = rclexecm.RclExecM() proto = rclexecm.RclExecM()
filter = WordFilter(proto, thisdir) filter = WordFilter(proto, execdir)
extract = rclexecm.Executor(proto, filter) extract = rclexecm.Executor(proto, filter)
rclexecm.main(proto, extract) rclexecm.main(proto, extract)

1
src/filters/rclexecm.py
View File

@ -233,6 +233,7 @@ class Executor:
def extractone(self, params): def extractone(self, params):
#self.em.rclog("extractone %s %s" % (params["filename:"], \ #self.em.rclog("extractone %s %s" % (params["filename:"], \
# params["mimetype:"])) # params["mimetype:"]))
self.flt.reset()
ok = False ok = False
if not params.has_key("filename:"): if not params.has_key("filename:"):
self.em.rclog("extractone: no mime or file name") self.em.rclog("extractone: no mime or file name")