Excerpts from Michael Stapelberg's message of 2012-01-08 12:50:02 -0800: > I think you meant --sup-label-file, or you did not push some commits yet :).
Yes, you are right. Thanks. > The total time for importing was: > ; scanned 133872, indexed 122314, skipped 9485 bad and 2073 seen messages in > 12484.5s = 10.7 m/s > > Which is awfully long. I took a look at the process list and saw lots of > html2text process running. Why are they run at all? Are they also run for > plain-text emails? Most of my emails surely are plain-text, without any need > to > convert HTML. They should only be run for text/html mime parts that are not part of a multipart/alternative mime part with a text component (in that case the text part is just used directly.) I believe that is the case, but if you can find a counterexample, please file it as a bug: http://github.com/wmorgan/heliotrope/issues. > ; reindexed 120451 messages, skipped 0 spam and 1863 deleted in 9509.3s = > 12.7 m/s > > Also, the numbers don’t add up. 120451 + 1863 = 122314, so they do add up. Perhaps what's confusing is that import happily preserves your deleted emails, but reindex by default skips over them? Glad to hear you're giving it a try. Keep the bug reports coming. -- William <wmor...@masanjin.net> _______________________________________________ Sup-devel mailing list Sup-devel@rubyforge.org http://rubyforge.org/mailman/listinfo/sup-devel
