initial

src/filters/rclzip

@@ -0,0 +1,133 @@
+#!/usr/bin/env python
+
+###########################################
+## Generic recoll multifilter communication code
+import sys
+import os
+
+myname = os.path.basename(sys.argv[0])
+
+def ermsg(s, fatal=0, exitvalue=1):
+    print >> sys.stderr, "RCLMFILT:", myname, ":", s
+    if fatal:
+        exit(exitvalue)
+
+def readparam():
+    s = sys.stdin.readline()
+    if s == '':
+        ermsg(": EOF on input", 1, 0)
+
+    s = s.rstrip("\n")
+
+    if s == "":
+        return ("","")
+    l = s.split()
+    if len(l) != 2:
+        ermsg("bad line: [" + s + "]", 1, 1)
+
+    paramname = l[0].lower()
+    paramsize = int(l[1])
+    if paramsize > 0:
+        paramdata = sys.stdin.read(paramsize)
+        if len(paramdata) != paramsize:
+            ermsg("Bad read: wanted %d, got %d" % (paramsize, len(paramdata)),
+                  1,1)
+    else:
+        paramdata = ""
+    
+    ermsg("paramname [%s] paramsize %d value [%s]" %
+          (paramname, paramsize, paramdata))
+    return (paramname, paramdata)
+
+# Send answer: document, ipath, possible eof.
+def answer(docdata, ipath, iseof):
+
+    print "Document:", len(docdata)
+    sys.stdout.write(docdata)
+
+    if len(ipath):
+        print "Ipath:", len(ipath)
+        sys.stdout.write(ipath)
+
+    # If we're at the end of the contents, say so
+    if iseof:
+        print "Eof: 0"
+    # End of message
+    print
+    sys.stdout.flush()
+    ermsg("done writing data")
+
+def mainloop(processor):
+    while 1:
+        ermsg("waiting for command")
+
+        params = dict()
+
+        # Read at most 10 parameters (we only actually use one), stop
+        # at empty line
+        for i in range(10):
+            paramname,paramdata = readparam()
+            if paramname == "":
+                break
+            params[paramname] = paramdata
+
+        # Call worker
+        processor.extractone(params)
+    
+###################################################################
+# Code specific to the zip file filter from here
+
+class ZipExtractor:
+    def __init__(self):
+        self.currentindex = 0
+        self.contents = []
+        self.zipfile = ""
+        
+    # Open zipfile and retrieve TOC
+    def listzip(self, fname):
+        cmd = "unzip -Z -1 " + fname
+        f = os.popen(cmd, "r")
+        self.contents = f.readlines()
+        f.close()
+        return self.contents
+
+    # Extract given file
+    def extractzipentry(self, fname, name):
+        cmd = " ".join(["unzip -p ", fname, name])
+        f = os.popen(cmd, "r")
+        data = f.read()
+        # ermsg("data: %s" % data)
+        f.close
+        return data
+
+    def extractone(self, params):
+        # See what's asked of us: open new zip file, next entry or specific one
+        if not params.has_key("filename:"):
+            ermsg("no filename ??", 1, 1)
+
+        # If we're given a file name, open it. Else increment position in
+        # current file (but wait for possible ipath)
+        if len(params["filename:"]) != 0:
+            self.zipfile = params["filename:"]
+            self.currentindex = 0
+            ermsg("opening [%s]" % (self.zipfile,))
+            self.listzip(self.zipfile)
+        else:
+            self.currentindex += 1
+
+        # If we have an ipath, that's what we look for, else process next entry
+        if params.has_key("ipath:") and len(params["ipath:"]):
+            entryname = params["ipath:"]
+        else:
+            if self.currentindex >= len(self.contents):
+                # EOF: "Document: 0\n\n"
+                answer("", "", 0)
+                return True
+            else:
+                entryname = self.contents[self.currentindex].rstrip("\n")
+        
+        docdata = self.extractzipentry(self.zipfile, entryname)
+        answer(docdata,entryname, (self.currentindex >= len(self.contents) -1))
+        return True
+
+mainloop(ZipExtractor())