lmi
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [lmi] PATCH: Submodule changes


From: Vadim Zeitlin
Subject: Re: [lmi] PATCH: Submodule changes
Date: Fri, 9 Oct 2020 17:30:41 +0200

On Fri, 9 Oct 2020 15:12:03 +0000 Greg Chicares <gchicares@sbcglobal.net> wrote:

GC> I have only one important question: the wx{W,P}* SHA1s are embedded in
GC> the submodule stuff now AIUI, so should the following 'wx_commit*_sha'
GC> variables...
GC> 
GC> $ git --no-pager grep _sha install_wx* 
GC> install_wx.sh:vendor=${LMI_TRIPLET}-$gcc_version-$wx_commit_sha
GC> 
install_wxpdfdoc.sh:build_dir="$exec_prefix/wxpdfdoc-ad_hoc/wxpdfdoc-$wxpdfdoc_commit_sha"
GC> 
GC> ...still be used? AFAICT without actually rebuilding, those variables
GC> ought to be empty now, right?

 Oops, you're absolutely right and I just haven't noticed this. AFAICS
there are just two solutions:

1. Remove the SHA suffixes from the vendor string.
2. Set these suffixes to the actual commits used in the submodule.

 I guess you prefer the solution (2), so I'm going to do this, but how
should I do it: wait until you push your changes and then make another PR
or update the existing PR with this change, so that you could cherry-pick
the new commit(s?)? What would you prefer?

GC> Here, for the record, is how I applied PR 160:
GC> 
GC> $git fetch xanadu                                 
GC> remote: Enumerating objects: 23, done.
GC> remote: Counting objects: 100% (23/23), done.
GC> remote: Total 39 (delta 23), reused 23 (delta 23), pack-reused 16
GC> Unpacking objects: 100% (39/39), done.
GC> From https://github.com/vadz/lmi
GC>    61f55594..14d582d4  master             -> xanadu/master
GC>  * [new branch]        wx-only-submodules -> xanadu/wx-only-submodules
GC>  * [new branch]        wx-submodules      -> xanadu/wx-submodules
GC>    d66edf01..b808e3a3  xmlwrapp-submodule -> xanadu/xmlwrapp-submodule
GC> 
GC> $git cherry-pick ..xanadu/wx-submodules           

 As usual, I'd have preferred "git merge xanadu/wx-submodules" but if you
prefer linear history, this works, of course. An alternative to cherry
picking would be to create a local branch and then rebase it on master, but
this basically boils down to the same thing.

 Anyhow, thanks for applying this and please let me know how should I fix
the problem you've found above.

VZ

Attachment: pgprqAVnYOxnz.pgp
Description: PGP signature


reply via email to

[Prev in Thread] Current Thread [Next in Thread]