Hugh Myers wrote:
Thanks for the tip on 2.odd versus 2.stable as well. Hadn't realized that a policy of collateral damage was allowed.
It isn't allowed, AFAICS, but can you prognose if you do collateral damage every time you write a patch? The testing is not as thorough as for stable releases. You noticed yourself that you /need/ the unstable version, since it has extra features not available with the older version. Which, in return, means that those features might not yet integrate seamlessly with all existing stuff.
In your case, the new TabStaff kinda seems to break vertical spacing / page layout. Which is bad, and probably also a regression against 2.12., but it's a inevitable risk when implementing new features. However, it does not seem to break anything "globally", just what is incident to TabStaff. I'm not surprised, though, if even everything using the old TabStaff features works as expected, since probably some regression test case would have shown the converse, and the patches had not been applied as they are.
Given your approach to testing (and your experience) this should work as well as more conservative approaches. [...]
Note, by the way, that I can't write you an "official" version - I'm not a developer, I just read the mailing lists every now and then.
Cheers, Alexander _______________________________________________ lilypond-user mailing list lilypond-user@gnu.org http://lists.gnu.org/mailman/listinfo/lilypond-user
