index: support webarchive (.war) and mimehtml (.mhtml) formats

src/filters/rclwar

@@ -0,0 +1,54 @@
+#!/usr/bin/env python
+
+# WAR web archive filter for recoll. War file are gzipped tar files
+
+import rclexecm
+import tarfile
+
+class WarExtractor:
+    def __init__(self, em):
+        self.em = em
+
+    def extractone(self, tarinfo):
+        docdata = ""
+        try:
+            member = self.tar.extractfile(tarinfo)
+            docdata = member.read()
+            ok = True
+        except Exception, err:
+            self.em.rclog("extractone: failed: [%s]" % err)
+            ok = False
+        return (ok, docdata, tarinfo.name, rclexecm.RclExecM.noteof)
+
+    ###### File type handler api, used by rclexecm ---------->
+    def openfile(self, params):
+        self.currentindex = 0
+        try:
+            self.tar = tarfile.open(params["filename:"])
+            return True
+        except Exception, err:
+            self.em.rclog(str(err))
+            return False
+
+    def getipath(self, params):
+        ipath = params["ipath:"]
+        try:
+            tarinfo = self.tar.getmember(ipath)
+        except Exception, err:
+            self.em.rclog(str(err))
+            return (False, "", ipath, rclexecm.RclExecM.noteof)
+        return self.extractone(tarinfo)
+
+    def getnext(self, params):
+        tarinfo = self.tar.next()
+        if tarinfo is None:
+            #self.em.rclog("getnext: EOF hit")
+            return (False, "", "", rclexecm.RclExecM.eofnow)
+        else:
+            ret = self.extractone(tarinfo)
+            return ret
+
+# Main program: create protocol handler and extractor and run them
+proto = rclexecm.RclExecM()
+extract = WarExtractor(proto)
+rclexecm.main(proto, extract)