Re: [lmi] Tests PR 174

[Vadim Zeitlin]

On Wed, 24 Mar 2021 19:43:41 +0000 Greg Chicares <gchicares@sbcglobal.net> 
wrote:

GC> On 3/24/21 1:58 AM, Vadim Zeitlin wrote:
GC> > On Wed, 24 Mar 2021 00:45:07 +0000 Greg Chicares 
<gchicares@sbcglobal.net> wrote:
GC> > 
GC> > GC> On 3/23/21 10:00 PM, Vadim Zeitlin wrote:
GC> > [...]
GC> > GC> > I'd like to ask you once again if you could please merge the PR 
fixing the
GC> > GC> > tests with autotools and adding them to the CI builds from
GC> > GC> > https://github.com/let-me-illustrate/lmi/pull/174
GC> > GC> Maybe I can look at it tomorrow.
GC> 
GC> I haven't finished testing it, but I'm pushing it now,

 Thank you!

GC> Please take a look at commit 593a7121f737, which may be
GC> erroneous.

 Yes, it is, and should be reverted -- or, alternatively, 41bd0a36d (Update
wx submodule to avoid clang warning in wx/hashmap.h, 2021-03-22) should be
re-applied again.

GC> As the commit message explains, I did this:
GC> 
GC> $git cherry-pick e571a6e11..2769d69ba xanadu/automake-libtest
GC> $git status                                                  
GC> On branch master
GC> Your branch is ahead of 'origin/master' by 28 commits.
GC>   (use "git push" to publish your local commits)
GC> 
GC> Changes not staged for commit:
GC>   (use "git add <file>..." to update what will be committed)
GC>   (use "git restore <file>..." to discard changes in working directory)
GC>         modified:   third_party/wx (new commits)
GC> 
GC> and blindly followed the "git add" advice.

 It tells you to add the files you want to commit, but you don't want to
commit your local submodule in this situation because it hasn't been
updated to the version in the previous commit.

 One of the confusing things about git submodules is that they, unlike
normal files, are not updated automatically. This makes sense because
updating a submodule is much, much more expensive than updating a single
file, so you wouldn't want this to happen automatically when using many
and/or big submodules (and wx is big enough already), but, again, it is
confusing.

 Instead you need to update the submodules manually whenever they are
modified by a commit not done by you (if you had committed the submodule
yourself, it is already at the right version). This is done using
"git submodule update third_party/wx" command or, if you want to save on
typing, just "git submodule update", which updates all submodules.

 Again, I know that the submodules behaviour is confusing, and I should
have warned you about it in advance, sorry for not thinking about it. But
at least I feel vindicated in my idea that it would be good to try a
submodule update on this trivial change, as at least it didn't actually
break anything for you, which wouldn't have been the case if we upgraded
directly to wx 3.1.5.

GC> Now it looks
GC> like I should have ignored it...and instead have used
GC> "git restore" to discard it? Is that the proper way to
GC> get git-status to report that everything's clean?

 Yes, please see above.

GC> Here are the two commits of concern, and it appears
GC> that the second undoes the first:

 Yes, exactly. The initial version in master was 6cdaedd42, I've updated it
to c9486f9ce, but your local version remained at 6cdaedd42 even after
applying my commits and you've committed it on top of mine, returning the
submodule to 6cdaedd42.

 So could you please revert this commit using "git revert 593a7121f" and
push it to fix the current CI failure in clang build?

 Thanks in advance!
VZ

pgp70WJ4XLKL6.pgp
Description: PGP signature