Merge my 'autoupdate' branch #289

cooljeanius · 2023-09-26T17:58:33Z

~~This isn't actually ready to merge yet (see cooljeanius#1), but I thought I'd open this as a draft anyways just so I can have something to look at while I fix the merge conflicts...~~

OK, so, basically what I did here was to:

Run autoupdate
Notice that it broke some things, so fix the things it broke
Get a testsuite started, to try to catch the sorts of things that autoupdate might break
Fix the remaining breakages that the testsuite caught
Do a few other misc. style tweaks that I noticed while I was at it

…ravis is using also check that variable for gnulib-tool is empty before setting it; this allows the user to override it by setting it to the path to their own gnulib-tool also print more bootstrap messages the gitignore files got updating by my re-running the bootstrap script while testing it

also change the test for empty variable in the bootstrap script to accommodate Travis and add comment about bug in doc/Makefile.am also deal with issue with `echo -n`

…or a C compiler and remove the manual call to AC_PROG_CC in configure.ac (gl_EARLY still seems to call it, though...) Also enable maintainer mode, and some other things to get the bootstrap script to work on Travis.

…commit

also try to fix docs build failure

…ild version.texi when needed

… are parse-able still need to try actually calling them all...

…his branch for in the first place that is, run `autoupdate` on all the m4 files, and see what the fallout is. (note: this will probably cause a lot of errors)

there is still plenty more remaining to clean up though also bump up various serial numbers and replace instances of configure.in with configure.ac and misc. other edits noticed while doing the aforementioned cleanup (I fully expect this build to fail btw)

should be good now

try marking TODO as org-mode

Check for path to texi2dvi, to make it easier to override. I was trying to work around the following error: ``` /usr/bin/texi2dvi: Can't use option `--output' with more than one argument. ``` ...but it seems that doing `make distcheck AM_DISTCHECK_DVI_TARGET=html` works better.

…pdate

serial bump caught by CI

cooljeanius · 2023-09-28T13:44:24Z

ok merge conflicts fixed; next step is getting CI passing again...

ok, 1 more...

good grief, how many serial bumps is this going to take?!

not complete yet, though; failure expected

still need to do everything after that...

still got the rest to go...

(locally, at least)

serial bump caught on CI

bump serial some more

GitHub Actions needs lzip now, too [travis skip]

try to skip the dvi step [travis skip]

I *say* I actually like autotools, but sometimes the haters can actually be right: autotools can actually be really annoying sometimes!

minor testsuite tweak

let's fix Travis next

maybe the difference between CIs is their distro version?

cooljeanius · 2023-09-28T23:12:40Z

OK it now passes both CIs

cooljeanius · 2023-10-23T09:48:32Z

@peti so I realize that this might be a bit too much to review all at once; do you have any tips on how I could make this easier to review?

peti · 2023-10-23T13:04:54Z

Makefile.am


-dist_doc_DATA           = AUTHORS COPYING COPYING.EXCEPTION README
+dist_doc_DATA           = AUTHORS COPYING COPYING.EXCEPTION README.md


We need a file called README in the release tar all or Automake will complain. Can you please revert this change?

@peti the plain-text README already gets included by default without needing to be specified explicitly because it's one of the files that automake puts under its am__DIST_COMMON variable. If I change README.md to just README, then make distcheck fails with:

Making all in test /Applications/Xcode.app/Contents/Developer/usr/bin/make all-am make[3]: Nothing to be done for `all-am'. make[2]: *** No rule to make target `../../README.md', needed by `../../README'. Stop. make[1]: *** [all-recursive] Error 1 make: *** [distcheck] Error 1

If I change it back to README.md, however, it gets past that step successfully (it then proceeds to fail with an error from texi2dvi making autoconf-archive.dvi in doc, but that's probably why the CI runs distcheck with AM_DISTCHECK_DVI_TARGET="", and in any case, it's unrelated to the issue of what the README is named)

bootstrap.sh

peti · 2023-10-23T13:07:43Z

bootstrap.sh


-set -eu
+# Wait until after we finish testing a possibly empty variable to do `set -u`


I'm not sure whether I understand this change. What problem does this solve?

@peti: The -u flag causes errors whenever the script encounters an undefined variable. This defeats the point of testing to see if certain variables are empty or not. For example, the test on line 11 of:

if [ "x${gnulibtool}" = "x" ]; then

...will cause the entire script to error out if the script is run with set -u, like this:

$ sh ./bootstrap.sh ./bootstrap.sh: line 11: gnulibtool: unbound variable

revert change to shebang line as requested in PR

replace some tabs with spaces, and disable some shellcheck warnings

cooljeanius · 2023-11-02T15:58:14Z

@peti I changed bootstrap.sh's shebang back to #!/bin/sh in deddd9a, and silenced a few additional shellcheck warnings for bootstrap.sh in 300ae70.

cooljeanius · 2024-03-30T04:08:32Z

Hi @peti, may I please gently ping this? I think the recent xz backdoor has shown that testing to ensure that autoconf macros actually do what they say they do can be important for security, so in that light, I think that makes the beginnings of a testsuite that I've started here to be a bit more important now... Please let me know if there's anything I can do to help move this along.

bump suggested by CI run of prev. commit

annoying

there's gotta be some sort of bug with the serial-bumping check; applying the exact diff that it suggests still fails to get it to pass...

how many serial bumps is this going to take?! ``` --- m4/ax_check_compile_flag.m4 2024-06-07 04:05:57.348392358 +0000 +++ stage/ax_check_compile_flag.m4 2024-06-07 04:06:46.236406376 +0000 @@ -34,7 +34,7 @@ # and this notice are preserved. This file is offered as-is, without any # warranty. -#serial 14 +#serial 15 AC_DEFUN([AX_CHECK_COMPILE_FLAG], [AC_PREREQ([2.64])dnl for _AC_LANG_PREFIX and AS_VAR_IF make: *** [cfg.mk:36: stage/ax_check_compile_flag.m4] Error 1 Error: Process completed with exit code 2. ```

--- m4/ax_check_compile_flag.m4 2024-06-07 04:09:26.127066767 +0000 +++ stage/ax_check_compile_flag.m4 2024-06-07 04:10:15.695371330 +0000 @@ -34,7 +34,7 @@ # and this notice are preserved. This file is offered as-is, without any # warranty. -#serial 15 +#serial 16 AC_DEFUN([AX_CHECK_COMPILE_FLAG], [AC_PREREQ([2.64])dnl for _AC_LANG_PREFIX and AS_VAR_IF make: *** [cfg.mk:36: stage/ax_check_compile_flag.m4] Error 1

cooljeanius · 2024-06-07T04:24:35Z

OK CI is passing again; I'd recommend squashing when merging, to compress all these intermediary commits...

cooljeanius added 20 commits October 16, 2014 11:58

autoupdate the main configure.ac

b5777c8

update configury by following directions from gnulib-tool and autoscan

f1ad5c8

also change the test for empty variable in the bootstrap script to accommodate Travis and add comment about bug in doc/Makefile.am also deal with issue with `echo -n`

Travis can't get bash right now, so revert that part of the previous …

9ca4439

…commit

lower AC_PREREQ from 2.69 to 2.68 for Travis

85dd775

trying to avoid C compiler fails, so add it back

9bb0e7b

also try to fix docs build failure

try to debug Travis's automake failing to generate proper rules to bu…

3627bde

…ild version.texi when needed

maybe it needs maintainer mode?

57fc05d

cfg.mk needs a hack for VPATH builds

d276b02

I now have a testsuite that at least tests to see if all the m4 files…

6f52945

… are parse-able still need to try actually calling them all...

now that I have a proper testsuite, I can finally do what I created t…

c560445

…his branch for in the first place that is, run `autoupdate` on all the m4 files, and see what the fallout is. (note: this will probably cause a lot of errors)

more cleaning up after autoupdate

660e690

should be good now

oops, missed a few serial bumps locally that Travis caught...

b31ebb6

Create .gitattributes

26fabde

try marking TODO as org-mode

minor build tweak

1d22e5e

Merge https://github.com/autoconf-archive/autoconf-archive into autou…

94d832b

…pdate

Update ax_append_flag.m4

b1fad20

serial bump caught by CI

cooljeanius added 9 commits September 28, 2023 09:47

Update ax_append_flag.m4

4c7554a

ok, 1 more...

Update ax_append_flag.m4

1bd6ea6

good grief, how many serial bumps is this going to take?!

continuing CI fixing

3891353

forgot I could just check these locally

78a3381

not complete yet, though; failure expected

serial bumps for all of m4/ax_cxx_*.m4

0869d65

still need to do everything after that...

through the 1st half of the alphabet

b838317

still got the rest to go...

passes make and make check now

c196c1a

(locally, at least)

Update ax_func_accept_argtypes.m4

2c55fb1

serial bump caught on CI

Update ax_func_accept_argtypes.m4

c0d70b2

bump serial some more

cooljeanius added 10 commits September 28, 2023 17:26

let's see if this will help

ee30e9e

Update autoconf-archive.yml

c5a1086

GitHub Actions needs lzip now, too [travis skip]

Update autoconf-archive.yml

b8b782e

try to skip the dvi step [travis skip]

grrrr

f1edd81

I *say* I actually like autotools, but sometimes the haters can actually be right: autotools can actually be really annoying sometimes!

ok, maybe I've figured it out now?

6b3051a

Update test/Makefile.am

95920a9

minor testsuite tweak

CI tweak

5d1f24c

darn CI systems

97676fd

it works on GitHub Actions now

d8c8b14

let's fix Travis next

I dunno what I'm doing

85da969

maybe the difference between CIs is their distro version?

cooljeanius marked this pull request as ready for review September 28, 2023 23:12

cooljeanius mentioned this pull request Sep 28, 2023

Rebase autoupdate branch on upstream cooljeanius/autoconf-archive#1

Closed

rm some lines that I guess I don't need anymore

22ff658

peti reviewed Oct 23, 2023

View reviewed changes

cooljeanius added 2 commits October 29, 2023 08:17

Update bootstrap.sh

deddd9a

revert change to shebang line as requested in PR

Update bootstrap.sh

300ae70

replace some tabs with spaces, and disable some shellcheck warnings

cooljeanius requested a review from peti November 8, 2023 17:33

cooljeanius added 8 commits April 1, 2024 19:17

a few things noticed upon review

fd76a70

Merge branch 'master' into autoupdate

406253d

Update ax_check_compile_flag.m4

89f87ca

bump suggested by CI run of prev. commit

Update ax_check_compile_flag.m4

2db6896

annoying

Update ax_check_compile_flag.m4

d72fe5c

there's gotta be some sort of bug with the serial-bumping check; applying the exact diff that it suggests still fails to get it to pass...

ok now it at least succeeds locally

ed24c89

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Merge my 'autoupdate' branch #289

Merge my 'autoupdate' branch #289

cooljeanius commented Sep 26, 2023 •

edited

cooljeanius commented Sep 28, 2023

cooljeanius commented Sep 28, 2023

cooljeanius commented Oct 23, 2023

peti Oct 23, 2023

cooljeanius Oct 29, 2023 •

edited

peti Oct 23, 2023

cooljeanius Oct 29, 2023 •

edited

cooljeanius commented Nov 2, 2023

cooljeanius commented Mar 30, 2024

cooljeanius commented Jun 7, 2024


		dist_doc_DATA = AUTHORS COPYING COPYING.EXCEPTION README
		dist_doc_DATA = AUTHORS COPYING COPYING.EXCEPTION README.md


		set -eu
		# Wait until after we finish testing a possibly empty variable to do `set -u`

Merge my 'autoupdate' branch #289

Are you sure you want to change the base?

Merge my 'autoupdate' branch #289

Conversation

cooljeanius commented Sep 26, 2023 • edited

cooljeanius commented Sep 28, 2023

cooljeanius commented Sep 28, 2023

cooljeanius commented Oct 23, 2023

peti Oct 23, 2023

Choose a reason for hiding this comment

cooljeanius Oct 29, 2023 • edited

Choose a reason for hiding this comment

peti Oct 23, 2023

Choose a reason for hiding this comment

cooljeanius Oct 29, 2023 • edited

Choose a reason for hiding this comment

cooljeanius commented Nov 2, 2023

cooljeanius commented Mar 30, 2024

cooljeanius commented Jun 7, 2024

cooljeanius commented Sep 26, 2023 •

edited

cooljeanius Oct 29, 2023 •

edited

cooljeanius Oct 29, 2023 •

edited