diff --git a/src/filters/rcldia b/src/filters/rcldia index 185a4dc3..937204f5 100755 --- a/src/filters/rcldia +++ b/src/filters/rcldia @@ -4,7 +4,7 @@ # 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/html;charset=utf-8 +# 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]: @@ -15,8 +15,6 @@ # .dia = application/x-dia-diagram # Small fixes from jfd: dia files are sometimes not compressed. -# And a note: this file actually has no reason to return HTML as there is -# no metadata. We could just as well and more simply return text/plain import rclexecm import re from gzip import GzipFile @@ -24,20 +22,10 @@ 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+]$') +rhs = re.compile(r'^#\s*(.*)') +rhe = re.compile(r'(.*)\s*#$') rempty = re.compile(r'^#?\s*#?$') -htmltemplate = ''' - - - - -{0} - - -''' - # xml parser for dia xml file class Parser: def __init__(self,rclem): @@ -58,11 +46,9 @@ class Parser: def chardata(self,data): if self.handlethis: # check if line is not empty and replace hashes/spaces - # tricky: after htmlescape check also for umlauts if not rempty.search(data): - self.string.append(self.rclem.htmlescape( - rhe.sub(r'\1',rhs.sub(r'\1',data))).encode('ascii', 'xmlcharrefreplace')) - + self.string.append(rhe.sub(r'\1',rhs.sub(r'\1',data))) + def endelement(self,name): self.handlethis = False @@ -83,7 +69,7 @@ class DiaExtractor: except Exception, err: ok = False iseof = rclexecm.RclExecM.eofnext - self.em.setmimetype("text/html") + self.em.setmimetype("text/plain") return (ok, docdata, ipath, iseof) ###### File type handler api, used by rclexecm ----------> @@ -116,7 +102,7 @@ class DiaExtractor: def ExtractDiaText(self): diap = Parser(self.em) diap.feed(self.dia) - return htmltemplate.format('\n'.join(diap.string)) + return '\n'.join(diap.string) # Main program: create protocol handler and extractor and run them proto = rclexecm.RclExecM() diff --git a/src/sampleconf/mimeconf b/src/sampleconf/mimeconf index 37774f47..76d90b33 100644 --- a/src/sampleconf/mimeconf +++ b/src/sampleconf/mimeconf @@ -71,7 +71,7 @@ application/vnd.wordperfect = exec wpd2html;mimetype=text/html application/x-abiword = exec rclabw application/x-awk = internal text/plain application/x-chm = execm rclchm -application/x-dia-diagram = execm rcldia;mimetype=text/html;charset=utf-8 +application/x-dia-diagram = execm rcldia;mimetype=text/plain;charset=utf-8 application/x-dvi = exec rcldvi application/x-flac = execm rclaudio application/x-gnuinfo = execm rclinfo