Collin Funk wrote:
> sed -e 's,/,\\/,g' -e 's,^,/^,' -e 's,$,\$/d,' < "$tmp"/ignore-removed
> if test -n "$anchor"; then sed -e 's,/,\\/,g' -e
> "s,^,/^${doubly_escaped_anchor}," -e 's,$,$/d,' < "$tmp"/ignore-removed; fi
>
> Which uses sed expressions to create sed expressions right? Like this:
>
In the output of test-create-testdir4, the list of modules in the output
is not correctly indented:
--- test-create-testdir-4.out 2024-03-19 17:48:35.022049580 +0100
+++ tmp2229544-out 2024-03-24 01:42:30.769446029 +0100
@@ -1,3262 +1,3342 @@
Module list with included dependencies (indente
On 3/23/24 5:01 PM, Bruno Haible wrote:
> * Is os.walk() as efficient as the 'find' command? On some file systems,
> determining the type (file vs. directory) of a directory entry requires
> an extra stat() call; on others, it is no extra overhead. We can assume
> that the 'find' command is w
Hi Bruno,
On 3/23/24 3:49 AM, Bruno Haible wrote:
> What's still missing, is the handling of 'files_removed'.
> GLImport.py lines 804..820 ought to match gnulib-tool.sh lines 6253..6273.
Sorry I haven't fixed this yet. I'm pretty sure I have been
overthinking this issue for the past hour or two h
Here it's gnulib-tool.sh which produces bad output:
$ diff -u <(./gnulib-tool.sh --extract-automake-snippet relocatable-prog)
<(./gnulib-tool.py --extract-automake-snippet relocatable-prog)
--- /dev/fd/63 2024-03-24 01:30:55.783950235 +0100
+++ /dev/fd/62 2024-03-24 01:30:55.783950235 +0100
@@
I see this test failure:
$ GNULIB_TOOL_IMPL=py ./test-create-testdir-4.sh
Files ./test-create-testdir-4.result/gllib/Makefile.am and
tmp2146793-result/gllib/Makefile.am differ
FAIL: gnulib-tool's result has unexpected differences.
$ diff -u ./test-create-testdir-4.result/gllib/Makefile.am
tmp214
Hi Simon and Collin,
> > Could putting the following into bootstrap.conf be a method that
> > we could recommend? Then developers can override it with
> > GNULIB_TOOL_IMPL=sh ./bootstrap if they want.
> >
> > GNULIB_TOOL_IMPL=${GNULIB_TOOL_IMPL:-py}
>
> I'd like to hear what Bruno thinks about
Hi Collin,
> Since one bullet point is inlining 'sed' invocations, it might also be
> worth considering doing the same for 'find'.
Two considerations:
* Is os.walk() as efficient as the 'find' command? On some file systems,
determining the type (file vs. directory) of a directory entry require
Collin Funk wrote:
> I've attached an updated patch.
Thanks! Applied.
Bruno
Hi Simon,
On 3/23/24 2:24 AM, Simon Josefsson wrote:
> Is there a way to opt-in inetutils to prefer python gnulib-tool, before
> gnulib as a whole changes its default behaviour? I think doing that
> will allow better testing of gnulib-tool.py in the wild until gnulib as
> a whole can change. Thi
Hi Bruno,
On 3/23/24 8:30 AM, Bruno Haible wrote:
> So, I've trimmed and reorganized the TODO file.
Thanks for checking these. I was going to mention a few of them but
you beat me to it. :)
Since one bullet point is inlining 'sed' invocations, it might also be
worth considering doing the same fo
On 3/23/24 4:49 AM, Bruno Haible wrote:
> If the variable has 3 possible values (None, False, True), then writing
> 'if value == True:' is the simpler way. Why would one use a conversion to
> bool, i.e. 3-values to 2-values conversion? It would only make things
> more complicated.
Yes, I agree now
Running
$ GNULIB_TOOL_IMPL=py ./test-create-testdir-1.sh
invokes configure, make, make distclean
in the generated directory, although this is not needed.
It's due to a typo ('cleaned_files' instead of 'tests_cleaned_files').
This patch fixes it.
2024-03-23 Bruno Haible
gnulib-tool.py
With the option --single-configure, gnulib-tool.py omits many
autoconf snippets from the generated configure.ac:
$ GNULIB_TOOL_IMPL=py ./test-create-testdir-1.sh
Files ./test-create-testdir-1.result/configure.ac and
tmp1933038-result/configure.ac differ
FAIL: gnulib-tool's result has unexpected
With Automake 1.16.3 in $PATH, I see this test failure:
$ GNULIB_TOOL_IMPL=py ./test-create-testdir-2.sh
./test-create-testdir-2.out tmp2119378-out differ: byte 16366, line 886
--- ./test-create-testdir-2.out 2024-03-19 17:28:37.195424671 +0100
+++ tmp2119378-out 2024-03-23 18:05:58.725173290
This patch reduces the mismatch from
GNULIB_TOOL_IMPL=py ./test-create-testdir-2.sh
2024-03-23 Bruno Haible
gnulib-tool.py: Print "executing touch config.h.in".
* pygnulib/GLTestDir.py (GLTestDir.execute): Print
"executing touch config.h.in", like gnulib-tool.sh does.
I've checked the remaining items in the TODO file:
- "regex-quote: \] -> ] in EREs and BREs"
Does not apply to regexes in Python syntax. => Nothing to do.
- "Fix bug in func_symlink_if_changed, from 2006-11-13."
The code that I committed in 2022-08 appears to handle this
already.
Collin Funk wrote:
> I'm used to writing this:
>
> if value:
> print('1')
> else:
> print('2')
>
> instead of this:
>
> if value == True:
> print('1')
> elif value == False:
> print('2')
If the variable has 2 possible values, then writing 'if value:' is the
simpler way of writin
On 3/23/24 3:54 AM, Bruno Haible wrote:
> But this part of your patch is a no-op, since
>
> >>> None == True
> False
> >>> None == False
> False
>
> No?
Oops... Yes you are right.
I'm used to writing this:
if value:
print('1')
else:
print('2')
instead of this:
if value == Tru
Hi Collin,
> This patch handles some issues with --vc-files/--no--vc-files.
> gnulib-tool.py would always print this in the actioncmd message at the
> top of files. The proper behavior would be to check if vc_files ==
> None before treating it as a bool.
But this part of your patch is a no-op, si
Hi Collin,
> This patch should hopefully stop gnulib-tool.py from mangling
> .gitignore files like it did previously. I think I did a decent job
> translating the join calls into Python sets, but any criticism is
> welcome.
I've applied the patch, although it is incomplete.
Next, I applied the a
(moving to bug-gnulib)
Collin Funk writes:
> On 3/22/24 2:18 PM, Simon Josefsson wrote:
>> Upgrading inetutils to use gnulib-tool.py would be nice. As a start, I
>> bumped the gnulib submodule.
>
> Bruno and I are still working on it with a test suite. We want the
> file output and stdout outpu
This patch handles some issues with --vc-files/--no--vc-files.
gnulib-tool.py would always print this in the actioncmd message at the
top of files. The proper behavior would be to check if vc_files ==
None before treating it as a bool. In gnulib-tool.sh it is empty else
true/false.
The second issu
23 matches
Mail list logo
