On 22 January 2014 12:55, Richard Hipp <[email protected]> wrote: > ... >> http://myhost/myrepo/vdiff?from=2014-01-22T07:30&to=tip&sbs=1 > > The URL you supplied is requesting the aggregate diff of check-ins > [84ef6a7729], [fb73686d55], and [0a6b833945]. Those are the check-ins that > occur after time 2014-01-22T:07:30 through tip. Is it not giving you that?
Well, it is showing the complete diff, but not all checkins (not all checkin-comments if you will): --- $ f timeline children '2014-01-22' -n 4 === 2014-01-22 === 07:34:13 [0a6b833945] added Y (user: michai tags: trunk) 07:33:27 [fb73686d55] changed X (user: michai tags: trunk) 07:32:56 [84ef6a7729] added X (user: michai tags: trunk) === 2014-01-21 === 22:40:48 [959ae30958] bla (user: michai tags: trunk) +++ no more data (4) +++ $ --- i.e. 2 checkins each adding a file, and in between those, a checkin changing the file X that was just added. I'm having a little trouble setting up a fossil on my host atm, so I paste links(1) output here: --- $ links -dump 'http://myhost/myrepo/vdiff?sbs=1&from=2014-01-22T07:30&to=2014-01-22T07:40' logo MyProject Check-in Differences Not logged in Home Timeline Branches Tags Tickets Wiki Login Hide Diff Invert Unified Diff Many hyperlinks are disabled. Use anonymous login to enable hyperlinks. Difference From: [959ae30958] bla (user: michai tags: trunk, date: 2014-01-21 22:40:48) To: [0a6b833945] added Y (user: michai tags: trunk, date: 2014-01-22 07:34:13) ---------------------------------------------------------------------- Added X > 1 changed content Added Y > 1 initial content This page was generated in about 0.029s by Fossil version [bf7524b13f] 2014-01-21 19:26:14 $ --- The diff for checkin fb73686d55 is shown first, although the header (I don't know what to call it) displayed for this diff reads 'Added X', suggesting it is the initial contents of X, from checkin 84ef6a7729. (Analogous to the bottom-most diff displayed here - that is the diff corresponding to checkin 0a6b833945; the header of this diff reads 'Added Y', so that makes sense to me. When I look at this list, I would think that the line "changed content" is the initial content of X (which isn't true; initial content was the line "initial content"), and the initial content of file Y was "initial content" (which is true). To me this is misleading; furthermore, when using 'sbs=0' I would probably miss the fact a file was being modified - not added - altogether. I don't know if this is a flaw or feature or a mistake on my side. Recent fossil-changes in vdiff ('sbs=0' feature now globbing) and and tree-view for files have made me use the web UI a lot more in a relatively short time, so it's likely this behaviour has been present for a long time, and I just happen to notice it for the first time now. Michai _______________________________________________ fossil-users mailing list [email protected] http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
