> On Oct. 12, 2011, 12:36 p.m., Vishesh Handa wrote: > > Ship It! > > Vishesh Handa wrote: > 2 min seems like a lot. Maybe 1 min? But what if one has a huge file > which is being indexed and some other stuff is going on at the same time? Eg > - Compiling some code
Yes, that is why I used 2 minutes. I tried a big movie file which starts the rather slow ffmpeg analyzer and calculates the checksum and it took 2:44 on my i5. So 2min might even not be enough... - Sebastian ----------------------------------------------------------- This is an automatically generated e-mail. To reply, visit: http://git.reviewboard.kde.org/r/102835/#review7256 ----------------------------------------------------------- On Oct. 12, 2011, 10:06 a.m., Sebastian Trueg wrote: > > ----------------------------------------------------------- > This is an automatically generated e-mail. To reply, visit: > http://git.reviewboard.kde.org/r/102835/ > ----------------------------------------------------------- > > (Updated Oct. 12, 2011, 10:06 a.m.) > > > Review request for Nepomuk. > > > Description > ------- > > We already put the indexing into its own process since strigi's indexers tend > to crash every once in a while. But there is another problem: indexers that > run into loops or have other problems that result in them never finishing. > The "solution" is to kill the indexer process after a certain time. In this > patch I chose 2 minutes which I feel is already way too much time than an > indexer should spend on a file. > > > Diffs > ----- > > services/fileindexer/nepomukindexer.h 5b14787 > services/fileindexer/nepomukindexer.cpp d89a6b8 > > Diff: http://git.reviewboard.kde.org/r/102835/diff/diff > > > Testing > ------- > > Tested with a short timeout. The process is killed and the indexing continues > without problems. > > > Thanks, > > Sebastian Trueg > >
_______________________________________________ Nepomuk mailing list [email protected] https://mail.kde.org/mailman/listinfo/nepomuk
