Kornel Benko wrote: > > > > how difficult it would be to create regression test where: > > > > 1. we load beamer example, export to .tex > > > > 2. we pick some (first) frame, apply 20x lfuns for moving slide down, > > > > 20x lfun for moving up > > > > 3. export again to .tex and compare that both versions did not change > > > > > > > > I guess the second part is the hard one, but might be possible via lyx > > > > --batch? > > > > > > We _have_ similar tests. If you could _exactly_ tell which keystrokes to > > > do > > > and what to expect (for instance in a log-file), it would be one of the > > > keytests. > > > > I know we have keystroke test; but these are very fragile and not even > > needed here. > > My idea was sending sequence of lfuns to lyx via batch commandline > > parameter. > > I believe we have all lfuns needed (open file,repat lfuns for moving > > section and export lfun). > > Output wouldn't be log file but output .tex file. > > > > Pavel > > OK. If you are able to create the batchfile, I should be able to use it.
I will try to come up with it once we sort out the current problems. P
