diff --git a/src/filters/rclimg.py b/src/filters/rclimg.py
new file mode 100755
index 00000000..ac21d130
--- /dev/null
+++ b/src/filters/rclimg.py
@@ -0,0 +1,123 @@
+#!/usr/bin/env python
+
+# Python-based Image Tag extractor for Recoll. This is less thorough than the
+# Perl-based rclimg script, but useful if you don't want to have to install Perl
+# (e.g. on Windows).
+#
+# Uses pyexiv2. Also tried Pillow, found it useless for tags.
+#
+
+import sys
+import os
+import rclexecm
+import re
+
+try:
+ import pyexiv2
+except:
+ print "RECFILTERROR HELPERNOTFOUND python:pyexiv2"
+ sys.exit(1);
+
+khexre = re.compile('.*\.0[xX][0-9a-fA-F]+$')
+
+pyexiv2_titles = {
+ 'Xmp.dc.subject',
+ 'Xmp.lr.hierarchicalSubject',
+ 'Xmp.MicrosoftPhoto.LastKeywordXMP',
+ }
+
+# Keys for which we set meta tags
+meta_pyexiv2_keys = {
+ 'Xmp.dc.subject',
+ 'Xmp.lr.hierarchicalSubject',
+ 'Xmp.MicrosoftPhoto.LastKeywordXMP',
+ 'Xmp.digiKam.TagsList',
+ 'Exif.Photo.DateTimeDigitized',
+ 'Exif.Photo.DateTimeOriginal',
+ 'Exif.Image.DateTime',
+ }
+
+exiv2_dates = ['Exif.Photo.DateTimeOriginal',
+ 'Exif.Image.DateTime', 'Exif.Photo.DateTimeDigitized']
+
+class ImgTagExtractor:
+ def __init__(self, em):
+ self.em = em
+ self.currentindex = 0
+
+ def extractone(self, params):
+ #self.em.rclog("extractone %s" % params["filename:"])
+ ok = False
+ if not params.has_key("filename:"):
+ self.em.rclog("extractone: no file name")
+ return (ok, docdata, "", rclexecm.RclExecM.eofnow)
+ filename = params["filename:"]
+
+ try:
+ metadata = pyexiv2.ImageMetadata(filename)
+ metadata.read()
+ keys = metadata.exif_keys + metadata.iptc_keys + metadata.xmp_keys
+ mdic = {}
+ for k in keys:
+ # we skip numeric keys and undecoded makernote data
+ if k != 'Exif.Photo.MakerNote' and not khexre.match(k):
+ mdic[k] = str(metadata[k].raw_value)
+ except Exception, err:
+ self.em.rclog("extractone: extract failed: [%s]" % err)
+ return (ok, "", "", rclexecm.RclExecM.eofnow)
+
+ docdata = "\n"
+
+ ttdata = set()
+ for k in pyexiv2_titles:
+ if k in mdic:
+ ttdata.add(self.em.htmlescape(mdic[k]))
+ if ttdata:
+ title = ""
+ for v in ttdata:
+ v = v.replace('[', '').replace(']', '').replace("'", "")
+ title += v + " "
+ docdata += '' + title + '\n'
+
+ for k in exiv2_dates:
+ if k in mdic:
+ # Recoll wants: %Y-%m-%d %H:%M:%S.
+ # We get 2014:06:27 14:58:47
+ dt = mdic[k].replace(':', '-', 2)
+ docdata += '\n'
+ break
+
+ for k,v in mdic.iteritems():
+ if k == 'Xmp.digiKam.TagsList':
+ docdata += '\n'
+
+ docdata += "\n"
+ for k,v in mdic.iteritems():
+ docdata += k + " : " + self.em.htmlescape(mdic[k]) + "
\n"
+ docdata += ""
+
+ self.em.setmimetype("text/html")
+
+ return (True, docdata, "", rclexecm.RclExecM.eofnext)
+
+ ###### File type handler api, used by rclexecm ---------->
+ def openfile(self, params):
+ self.currentindex = 0
+ return True
+
+ def getipath(self, params):
+ return self.extractone(params)
+
+ def getnext(self, params):
+ if self.currentindex >= 1:
+ return (False, "", "", rclexecm.RclExecM.eofnow)
+ else:
+ ret= self.extractone(params)
+ self.currentindex += 1
+ return ret
+
+if __name__ == '__main__':
+ proto = rclexecm.RclExecM()
+ extract = ImgTagExtractor(proto)
+ rclexecm.main(proto, extract)