Re: [PATCH] {yacc-work} tests: remove redundancy from silent lex/yacc tests

[Ralf Wildenhues]

* Stefano Lattarini wrote on Fri, Jan 28, 2011 at 05:06:51PM CET:
> Finally a testsuite patch that offers a simplification rather
> than a complication (at least IMHO).  The ChangeLog entry below
> should explain the details.
> 
> OK for the 'yacc-work' branch?  I will push in 72 hours if
> there's no objection.

I must object to this sort of patch ping pong, first splitting
silent5.test into these two tests and then back.  This steals
developer time from work on bugs.

Can we at least be openly honest and replace this patch with a patch
that simply reverts v1.11-246-g22ee3bd and reinstates silent5.test?
Alternatively, please describe in what way your proposed patch isn't
equivalent to that reversal.

When changing code, especially in the testsuite, I almost always look at
the history of a test.  Does it make sense to go this way?  Changes such
as this one don't make sense.  Either we made an error in the earlier
split: then we should be openly honest about it and admit it in the log
entry or a code comment, or we didn't, then the new patch is likely
faulty.  Or some infrastructure changed that now allows us to reunify
now that didn't before: then the patch /might/ be acceptable, but it
tells us that we should have developed things in a different order.

Thanks,
Ralf

> tests: remove redundancy from silent lex/yacc tests
> 
> * tests/silent-yacc-gcc.test, tests/silent-yacc-generic.test: Merge
> these two testcases into ...
> * tests/silent-yacc.test: ... this new one, which doesn't fiddle
> which dependency tracking, as that shouldn't impact on generation
> of yacc-derived C source and header files (and, for what concerns
> compilation of C files, is already tested in other testcases).
> * tests/silent-lex-gcc.test, tests/silent-lex-generic.test: Merge
> these two testcases into ...
> * tests/silent-lex.test: ... this new test, for similar reasons.
> * tests/Makefile.am (TESTS): Update.