|Title:||make wiki (code?, bugs?) searchable|
|Last Modified:||2015-02-14 15:24:22|
|Version Found In:|
Search is a very useful feature to have <hr><i>anonymous added on 2008-11-26 22:55:35:</i><br> For what it's worth, you can do source code searches with a lot of editors out there, particularly the ones that support [http://cscope.sourceforge.net/ | Cscope]. <hr><i>anonymous added on 2009-12-08 19:37:06:</i><br> Yes, I can use cscope et al, but sometimes you need to do a search (grep) on older, checked-in, versions. <hr><i>anonymous added on 2009-12-23 15:43:35:</i><br> See http://trac.edgewall.org/search for example of search feature request. <hr><i>anonymous added on 2009-12-23 17:28:48:</i><br> Since I am living in huge text files (todos, notes, etc), a grep over different versions of these files would be perfect. Something like hg's grep subcommand over all files and all revisions. First, I thought this was an easy change, but having seen the internal diff format and the sources, I found that this might not be so easy -- at least for me... <hr><i>anonymous added on 2009-12-24 14:12:55:</i><br> full text search like fts library inside fossile.exe? <hr><i>anonymous added on 2009-12-28 21:50:06:</i><br> Thanks for the hint. Yup, something like that will most probably do the job. Will check this one out. <hr><i>ron added on 2010-01-25 10:02:45:</i><br> Checkin [07bd8796dc91cc6cd631481f5b1712809becf1e6] add search to wiki, although it only searches wiki page titles. <hr><i>jeremy_c added on 2010-01-25 14:14:06:</i><br> Why not have it search the body as well? <hr><i>ron added on 2010-01-25 14:28:02:</i><br> only because I wasn't sure how to do that ... <hr><i>ron added on 2010-01-25 20:33:49:</i><br> Well, it's not straightforward to search wiki page contents, since the pages are stored as BLOBs and I don't see any easy way to search a BLOB. It would be really nice if there were a helper function in SQL like "blobsearch(text)"... but it would still have to perform the uncompressing of the BLOB before doing the search through it. <hr><i>anonymous added on 2010-01-25 22:00:06:</i><br> The same applies to search in code (see above remarks on that) :-( <hr><i>anonymous added on 2010-02-04 13:54:52:</i><br> Back in April 2009 there has been some drh notes on fossil mailing list about searching/using FTS on repositories, for details see e.g. [http://lists.fossil-scm.org:8080/cgi-bin/mailman/private/fossil-users/2009-April/000695.html] where: <pre> It has been "considered". But little action has been taken yet.... I did modify the makefile so that the Full Text Search engine of SQLite is complied in. I was thinking that might be useful for implementing search. But so far it is unused. An interesting research project would be to try to figure out how to do search across all historical versions of a project. </pre> In march 2009 there was an offer to help with "free advice on how to proceed" with integrating SQLite CLI to fossil by drh. Would it be possible to provide some hints, e.g. for searching content of artifact in compressed/deltified blobs? Hopefully this would allow more interested peole to focus on implementation of this needed functionality. Many thanks. <hr><i>anonymous added on 2010-02-04 17:13:49:</i><br> Actually, this is the only thing holding me back from using fossil :-( I was trying to understand the data format used in fossil and just try and hack a simple way through. Idea was to decode all artefacts and just try to find simple strings. But I'm already sucking here. Or will fts take care of that all? Seems that the efficient data format does not make it easy to do such things as text searches... <hr><i>anonymous added on 2010-03-02 21:22:16:</i><br> It seems that recently implemented features in fossil (e.g. ticket search, datetime string used as hyperlink to timeline, graph of changes in the timeline) could be a part of general effort to bring interfaced artefact search into fossil. If so or not are there plans (which could be revealed) about implementing search in fossil? Maybe implementing this request on higly compressed and efficiently stored artifacts pose some prohibitive computational penalty on this otherwise low resource application. Even it that case it would be really nice to hear some remarks. Is there any way how can casual user of fossil help with this much needed feature?
nobody added on 2014-09-19 15:50:03:
Ticket search is very important for project management. I'm not sure if this can be done by customizing the ticket reports.
nobody added on 2014-12-10 15:12:39: (text/x-fossil-wiki)
as a new user of fossil (attracted by the HN discussion), I like it so far, thanks! But the lack of full text search is killing me, how do you guys find related wikis you wrote long time ago? I support the "fossil grep pattern" idea discussed above... thanks