Open Source Projects Europe

rcldia    113 lines (97 with data), 3.5 kB

#!/usr/bin/env python2
# -*- coding: utf-8 -*-
from __future__ import print_function

# dia (http://live.gnome.org/Dia) file filter for recoll
# stefan.friedel@iwr.uni-heidelberg.de 2012
#
# add the following to ~/.recoll/mimeconf into the [index] section:
# application/x-dia-diagram = execm rcldia;mimetype=text/plain;charset=utf-8
# and into the [icons] section:
# application/x-dia-diagram = drawing
# and finally under [categories]:
# other = ...\
#       application/x-dia-diagram
#
# in ~/.recoll/mimemap:
# .dia = application/x-dia-diagram

# Small fixes from jfd: dia files are sometimes not compressed.  
import rclexecm
import re
from gzip import GzipFile
import xml.parsers.expat

# some regexps to parse/format the xml data: delete #/spaces at the b/eol and
# ignore empty lines
rhs = re.compile(r'^#\s*(.*)')
rhe = re.compile(r'(.*)\s*#$')
rempty = re.compile(r'^#?\s*#?$')

# xml parser for dia xml file
class Parser:
    def __init__(self,rclem):
        self._parser = xml.parsers.expat.ParserCreate(encoding='UTF-8')
        self._parser.StartElementHandler = self.startelement 
        self._parser.EndElementHandler = self.endelement
        self._parser.CharacterDataHandler = self.chardata
        self.string = []
        self.handlethis = False
        self.rclem = rclem

    def startelement(self, name, attrs):
        if name == 'dia:string':
            self.handlethis = True
        else:
            self.handlethis = False
    
    def chardata(self,data):
        if self.handlethis:
            # check if line is not empty and replace hashes/spaces
            if not rempty.search(data):
                self.string.append(rhe.sub(r'\1',rhs.sub(r'\1',data)))
    
    def endelement(self,name):
        self.handlethis = False

    def feed(self, fh):
        self._parser.ParseFile(fh)
        del self._parser

class DiaExtractor:
    def __init__(self, em):
        self.em = em

    def extractdia(self):
        docdata = ""
        ipath = ""
        try:
            docdata = self.ExtractDiaText()
            ok = True
        except Exception as err:
            ok = False
        iseof = rclexecm.RclExecM.eofnext
        self.em.setmimetype("text/plain")
        return (ok, docdata, ipath, iseof)

    ###### File type handler api, used by rclexecm ---------->
    def openfile(self, params):
        try:
            self.dia = GzipFile(params["filename:"], 'r')
            # Dia files are sometimes not compressed. Quite weirdly,
            # GzipFile does not complain until we try to read. Have to do it
            # here to be able to retry an uncompressed open.
            data = self.dia.readline()
            self.dia.seek(0)
            return True
        except:
            # File not compressed ?
            try:
                self.dia = open(params["filename:"], 'r')
            except:
                return False
            return True

    def getipath(self, params):
        ok, data, ipath, eof = self.extractdia()
        return (ok, data, ipath, eof)
        
    def getnext(self, params):
        ok, data, ipath, eof = self.extractdia()
        return (ok, data, ipath, eof)

    ###### read data
    def ExtractDiaText(self):
        diap = Parser(self.em)
        diap.feed(self.dia)
        return '\n'.join(diap.string)

# Main program: create protocol handler and extractor and run them
proto = rclexecm.RclExecM()
extract = DiaExtractor(proto)
rclexecm.main(proto, extract)