2014-05-14 18:07 GMT+02:00 Richard Shann <[email protected]>: > On Wed, 2014-05-14 at 14:09 +0200, Éloi Rivard wrote: > > The first is integration-data/hemiola.denemo and not > > integration-tmp/hemiola.denemo > > ok, I have checked in hemiola.denemo and loaded and saved it as > hemiola.denemo.denemo, but there is no directory in the master git > called integration-tmp and I am sure you don't mean one should be > created. Where do I put this reference file for checking against? > I tried the make update-samples but this generated differences for the > input files thus: > > integration-tmp is automatically created and deleted during the test, you won't see it The reference file should be in integration-data. Comparison against reference files is done for .denemo files in example directory, and scm / mxml files in integration-data
> # modified: actions/init.denemo > > # modified: actions/musicgames/ReadingNoteNames.denemo > > # modified: samples/AllDenemoDirectiveOutputs.denemo > > # modified: samples/DurationError.denemo > > # modified: templates/Accordion/accordion.denemo > > # modified: templates/EarlyMusic/BreveLonga.denemo > > # modified: templates/Tablature.denemo > > # modified: templates/band/BrassBand.denemo > > # modified: templates/band/BrassTrio.denemo > > # modified: templates/chamber/StringQuartet.denemo > > # modified: templates/choral/SATB-Piano.denemo > > # modified: templates/choral/SATB.denemo > > # modified: templates/choral/TTBB.denemo > > # modified: templates/drummap.denemo > > # modified: templates/jazz/ChordSymbols.denemo > > # modified: templates/jazz/JazzCombo.denemo > > # modified: tests/Interactive/Playback/CursorToEnd.denemo > > # modified: tests/Interactive/Typesetting/testDynamicStaff.denemo > > > > I'm puzzled ... > What do you mean ? > But I think I have now a good basic test for a new command of this sort > (commands that generate standalone directives). The hemiola.denemo has a > script to be run on loading which deletes the object at the cursor and > then runs the new command, which should re-instate it. So if the command > fails to restore the file to its original state there is an error. > > Something similar can be done for other sorts of directive creating > commands, which is many of the new scripted commands. > > Richard > > > > > > Comparision between the two files is automatically done. > > > > > > > > 2014-05-14 13:15 GMT+02:00 Richard Shann <[email protected]>: > > On Wed, 2014-05-14 at 13:03 +0200, Éloi Rivard wrote: > > > integration-data/hemiola.denemo will be opened and saved at > > > integration-tmp/hemiola.denemo.denemo > > > > > > Then integration-tmp/hemiola.denemo.denemo will be compared > > against > > > integration-data/hemiola.denemo > > > > So I should check in two files, > > > > integration-tmp/hemiola.denemo > > integration-tmp/hemiola.denemo.denemo > > > > > > where the second is the result of opening and saving the first > > on a > > clean machine? > > > > Richard > > > > > > > > > > Test fails if the files are different > > > > > > > > > > > > 2014-05-14 12:48 GMT+02:00 Richard Shann > > <[email protected]>: > > > On Wed, 2014-05-14 at 12:31 +0200, Éloi Rivard > > wrote: > > > > I just added a small thing. > > > > > > > > mxml files in tests/integration-data were just > > opened and > > > saved. > > > > > > > > Now, if a .denemo file has the same name, it will > > be > > > compared to the > > > > saved file. > > > > > > > > For example "integration-data/foobar.mxml" will be > > opened > > > and saved, > > > > and compared against > > "integration-data/foobar.denemo" if it > > > exists. > > > > > > > > > > > > At this point, does the test system seem good to > > you about > > > file > > > > opening and saving ? > > > > > > > > > It sounds good, I need to give it some testing I > > guess. I have > > > created a > > > file illustrating the newest feature (support for > > hemiola), if > > > I check > > > it in (to integration-data/hemiola.denemo ?) what > > will happen? > > > > > > Richard > > > > > > > > > > > > > > > > > > > > > > 2014-05-14 12:09 GMT+02:00 Éloi Rivard > > <[email protected]>: > > > > make update-samples should be good to use > > now. > > > > > > > > > > > > I also set up the test that compares > > opened and > > > saved files > > > > with the original ones. > > > > > > > > > > > > > > > > 2014-05-13 19:07 GMT+02:00 Éloi Rivard > > > <[email protected]>: > > > > > > > > I did not noticed this when I ran > > git diff. > > > > > > > > > > > > > > > > You should not use make > > update-samples at > > > the moment > > > > however, since non interactive > > file saving > > > is not > > > > working completely yet. I just > > fixed issues > > > with > > > > lyrics, there is the same problem > > with > > > <scheme> tag. I > > > > let you know when it is fixed. > > > > > > > > > > > > 2014-05-13 18:36 GMT+02:00 Richard > > Shann > > > > <[email protected]>: > > > > > > > > On Tue, 2014-05-13 at > > 13:47 +0100, > > > Richard > > > > Shann wrote: > > > > > Hmm, I get it telling me > > that all > > > > the .denemo files are > > changed when I > > > > > run <<make > > update-samples>>, and > > > when I try > > > > git diff it says that they > > > > > are binary files and > > that they > > > differ, > > > > > > > > > > > > actually, not all of them > > say they > > > are binary > > > > files. Is it perhaps the > > > > case that you didn't check > > in the > > > initial > > > > versions of the files? (We > > > > used to generate > > compressed files > > > as .denemo, > > > > so some could well still > > > > be binary). > > > > > > > > Richard > > > > > > > > > but they are not binary > > files ... > > > > > any ideas? > > > > > > > > > > Richard > > > > > > > > > > > > > > > > > > > > > > > > > On Tue, 2014-05-13 at > > 11:36 +0200, > > > Éloi > > > > Rivard wrote: > > > > > > This is pushed, you > > can do it > > > with « make > > > > update-samples ». At the > > > > > > moment however, saving > > files > > > does not save > > > > lyrics for examples > > > > > > (because lyrics > > depends on > > > GtkTextBuffer, > > > > that cannot be used in non > > > > > > interactive mode). > > > > > > > > > > > > > > > > > > 2014-05-11 16:18 GMT > > +02:00 > > > Richard Shann > > > > <[email protected]>: > > > > > > Eloi, > > > > > > I introduced a > > new field > > > in > > > > the .denemo format and the > > > > > > automatic > > build's > > > > > > test started > > failing > > > because it > > > > wasn't present in the > > > > > > reference file > > > > > > blank.denemo. > > > > > > I edited this > > by hand, > > > but we need > > > > a make target which > > behaves > > > > > > like the > > > > > > test target > > except that > > > instead of > > > > diffing the outputs it > > > > > > copies them as > > > > > > the new > > reference files. > > > > > > > > > > > > Richard > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > _______________________________________________ > > > > > > Denemo-devel > > mailing > > > list > > > > > > > > [email protected] > > > > > > > > > > > > > https://lists.gnu.org/mailman/listinfo/denemo-devel > > > > > > > > > > > > > > > > > > > > > > > > -- > > > > > > Éloi Rivard - > > [email protected] > > > > > > > > > > > > « On perd plus à être > > indécis > > > qu'à se > > > > tromper. » > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > _______________________________________________ > > > > > Denemo-devel mailing > > list > > > > > [email protected] > > > > > > > > > > > > https://lists.gnu.org/mailman/listinfo/denemo-devel > > > > > > > > > > > > > > > > > > > > > > > > > > > > -- > > > > Éloi Rivard - [email protected] > > > > > > > > « On perd plus à être indécis qu'à > > se > > > tromper. » > > > > > > > > > > > > > > > > > > > > -- > > > > Éloi Rivard - [email protected] > > > > > > > > « On perd plus à être indécis qu'à se > > tromper. » > > > > > > > > > > > > > > > > > > > > -- > > > > Éloi Rivard - [email protected] > > > > > > > > « On perd plus à être indécis qu'à se tromper. » > > > > > > > > > > > > > > > > > > > > > > > > > -- > > > Éloi Rivard - [email protected] > > > > > > « On perd plus à être indécis qu'à se tromper. » > > > > > > > > > > > > > > > > > -- > > Éloi Rivard - [email protected] > > > > « On perd plus à être indécis qu'à se tromper. » > > > > > -- Éloi Rivard - [email protected] « On perd plus à être indécis qu'à se tromper. »
_______________________________________________ Denemo-devel mailing list [email protected] https://lists.gnu.org/mailman/listinfo/denemo-devel
