Re: compilation error on NetBSD

On Mon, Jul 21, 2025 at 11:02:17PM +0200, Bruno Haible via Bug reports for the GNU Texinfo documentation system wrote: > The CI reports a compilation error on NetBSD 10.0: > > gcc -DHAVE_CONFIG_H -I. -I../../info -I.. -I../.. > -I../../gnulib/lib

Re: dependency on libperl?

On Mon, Jul 21, 2025 at 11:05:24PM +0200, Bruno Haible via Bug reports for the GNU Texinfo documentation system wrote: > The CI reports a link error on Ubuntu (both when compiling natively and > when cross-compiling for RISCV), that was not present last week: > > /bin/bash ../../libtool --tag=CC

Re: "make distcheck" failure

On Mon, Jul 21, 2025 at 09:04:09PM +0200, Bruno Haible via Bug reports for the GNU Texinfo documentation system wrote: > Hi, > > The CI reports today that, unlike a week ago, "make distcheck" fails: > > (cd info && make top_distdir=../texinfo-7.2.20250721dev > distdir=../texinfo-7.2.20250721d

Re: [PATCH v1 1/1] Short Sample: show how to avoid warning: node `[...]' unreferenced.

On Sat, Jul 19, 2025 at 02:20:36AM +0200, Denis 'GNUtoo' Carikli wrote: > While the current Short Sample work fine, if we remove the second > section and run run the following command: >makeinfo -D "VERSION 0.1" -D "UPDATED now" shortsample.texi' > we end up with the following warning: > sh

Re: Unable to build Texinfo from git ("SWIG not found")

On Fri, Jul 18, 2025 at 06:46:35PM +0200, Andreas Schwab wrote: > > -- Macro: AC_ARG_WITH (PACKAGE, HELP-STRING, [ACTION-IF-GIVEN], > > [ACTION-IF-NOT-GIVEN]) > > > > The option now looks like it was given, so the third argument to > > AC_ARG_WITH, [ with_swig='yes' ], runs, leading to

Re: Highlight syntax with source-highlight produces error with split html output

On Fri, Jul 18, 2025 at 10:27:34AM -0700, Raymond Toy wrote: > Well, dang. I don't get this error anymore either. I didn't change anything. > > Oh, wait. If I remove src-hl_html/, I get the error. But if I run it again, > I don't. And the files are all there as you show. You figured it out. I'm

Re: Highlight syntax with source-highlight produces error with split html output

On Fri, Jul 18, 2025 at 08:43:21AM -0700, Raymond Toy wrote: > |@include src-hl.c | > > I changed it to > > |#include src-hl.c | > > Now I get the message: > > |texi2any: warning: highlight_syntax.pm: could not open > src-hl_html/src-hl_highlight_C_input.C: No such file or directory | > > It a

Re: Highlight syntax with source-highlight produces error with split html output

On Fri, Jul 18, 2025 at 08:43:21AM -0700, Raymond Toy wrote: > Huh, this is weird. Perhaps I messed up building texinfo 7.2 from sources a > while back? > > Anyway, with your test, I ran it. I get a message that it can't include > src-hl.c. That seems to come from the line > > |@include src-hl.c

Re: Unable to build Texinfo from git ("SWIG not found")

On Thu, Jul 17, 2025 at 02:28:20AM +0100, Gavin Smith wrote: > The responsible code is in tta/configure.ac: > > > - > # SWIG interfaces > > # FIXME if enable_xs=yes and embedded_perl=no, the interface w

Re: Highlight syntax with source-highlight produces error with split html output

On Wed, Jul 16, 2025 at 02:27:25PM -0700, Raymond Toy wrote: > I'm experimenting with HIGHLIGHT_SYNTAX=source-highligit. It works fine with > --no-split and produces pretty nice highlighting. > > However, when I don't use --no-split, I get an error. An example error > message: > > |makeinfo: warn

Unable to build Texinfo from git ("SWIG not found")

I'm now unable to build Texinfo from the git repository. After rerunning autogen.sh (as I couldn't run 'configure' due to an error message about tta/build-aux/py-compile being missing), configure doesn't complete: checking whether iconv converts EUC-CN to UTF-8... yes checking for newlocale... yes

Re: Texinfo 7.2 pdf output for @example lisp

On Wed, Jul 16, 2025 at 02:13:43PM -0700, Raymond Toy wrote: > If I have a texinfo file that has > > |@example lisp (defun foo (x) (+ x x)) @end example | (The line breaks are not present in the plaintext version of your email). > > and do |texinfo --pdf|, (with texinfo 7.2), the pdf output lo

Re: Info hook script for file not found?

On Wed, Jul 16, 2025 at 04:19:53PM +0200, Patrice Dumas wrote: > On Wed, Jul 16, 2025 at 12:57:03AM +0100, Gavin Smith wrote: > > This feature still needs work: > > > > * At what point do we run the hook - before checking for man pages, > > or after? E.g. if "ta

Re: Info hook script for file not found?

On Sun, Feb 23, 2025 at 03:51:47PM +, Gavin Smith wrote: > I've commited code to use a hook program under > "XDG_CONFIG_DIRS/texinfo/info-hooks/manual-not-found", where XDG_CONFIG_DIRS > could be e.g. $HOME/.config or /etc/xdg. Previous discussion: https://lists.

Re: Start on replacement for titlepage in texinfo.tex

On Tue, Jul 15, 2025 at 02:27:05PM +0300, Eli Zaretskii wrote: > > It would be good in my opinion to get rid of the @vskip line as it > > looks ugly and confusing and probably turns people off. > > I think this calls for some research: how many manuals do NOT use the > standard @vskip line, and if

Re: Start on replacement for titlepage in texinfo.tex

On Mon, Jul 14, 2025 at 09:13:36PM +0100, Gavin Smith wrote: > On Mon, Jun 30, 2025 at 11:30:30PM +0100, Gavin Smith wrote: > > So just using @maketitle to produce the copyright page too is > > a real possibility. > > So I've been seeing what I can implement in texinf

Start on replacement for titlepage in texinfo.tex

On Mon, Jun 30, 2025 at 11:30:30PM +0100, Gavin Smith wrote: > So just using @maketitle to produce the copyright page too is > a real possibility. So I've been seeing what I can implement in texinfo.tex as a prototype for this feature area. The idea is that the user specifies inf

Re: Could texinfo.tex skip re-reading defaults from txi-en.tex ?

On Fri, Jul 11, 2025 at 06:35:04PM +0100, Gavin Smith wrote: > On Fri, Jul 11, 2025 at 07:14:13PM +0200, Sebastian Meyer wrote: > > Would it make sense for texinfo.tex to skip over @documentlanguage en when > > no other language has been set before? > > It seems like it is a

Re: Could texinfo.tex skip re-reading defaults from txi-en.tex ?

On Fri, Jul 11, 2025 at 07:14:13PM +0200, Sebastian Meyer wrote: > Now that CTAN no longer contains the translation files (following a recent > thread on this list, > ), the > texinfo bundle from CTAN is no longer sufficient to p

Re: Texinfo::Common::debug_print_tree doesn't work well with Texinfo::TreeElement

On Fri, Jul 11, 2025 at 11:26:19AM +0200, Patrice Dumas wrote: > On Fri, Jul 11, 2025 at 12:45:05AM +0100, Gavin Smith wrote: > > As Texinfo tree elements are now blessed into the Texinfo::TreeElement > > class, this class is output by Data::Dumper when used by >

Texinfo::Common::debug_print_tree doesn't work well with Texinfo::TreeElement

As Texinfo tree elements are now blessed into the Texinfo::TreeElement class, this class is output by Data::Dumper when used by Texinfo::Common::debug_print_tree. I use this fairly often when developing and/or debugging. Sample output: $VAR1 = bless( { 'cmdname' => 'node', 'contents' => [

Re: why 'verbatim' output for menu comments in HTML?

On Wed, Jul 09, 2025 at 07:23:53AM +, Werner LEMBERG wrote: > > >> OK, so please document this. Looking at section `Writing a menu` > >> in the Texinfo info file (of current git) I don't see anything > >> about HTML output of `@menu` blocks and how it is formatted (and > >> the description of

Re: how to get HTML navigation panel in footer with `-c FORMAT_MENU=menu`

On Wed, Jul 09, 2025 at 05:04:34AM +, Werner LEMBERG wrote: > Ouch! I simply searched for the string 'footer' in the Texinfo info > pages and I got no hit. I now see that the description of > `WORDS_IN_PAGE` is in the 'Customization of Navigation and Headers' > section (as it should be), whic

Master menu output for Info with explicit menus

Currently running "info texinfo menu" with the current output from texinfo.texi does not go to the node you expect ('Menus'), but to the documentation of the Info format instead ('Info Format Menu'). Likewise, "info texinfo image" goes to 'Info Format Image', not 'Images', and "info texinfo printi

Re: script for updating CSS entities

On Tue, Jul 08, 2025 at 12:19:58PM +, Werner LEMBERG wrote: > > Comparing the HTML of version 6.8 with version 7.2, I see that many > classes got new names. This means that we essentially have to rewrite > our custom CSS file, which is not fun. > > What about providing a script > > ``` > te

Re: why 'verbatim' output for menu comments in HTML?

On Tue, Jul 08, 2025 at 12:07:11PM +, Werner LEMBERG wrote: > > [texi2any 7.2] > > > Consider the following example file `menu.texinfo`. > > ``` > \input texinfo > > @node Top > @top Top > > @menu > * Foo:: Foo foo foo. > > > XXX @var{yyy} ZZZ > > > * Bar:: Bar bar bar. > @end menu >

Re: how to get HTML navigation panel in footer with `-c FORMAT_MENU=menu`

On Tue, Jul 08, 2025 at 07:42:13PM +, Werner LEMBERG wrote: > There is both a header and a footer navigation panel. However, the > files `Foo.html` and `Bar.html` look like the following: > > ``` > Previous: Foo, Up: Top > --- > > ... > > ``` > > There is only a

Re: ugly page break in ToC

On Mon, Jul 07, 2025 at 06:08:37PM +, Werner LEMBERG wrote: > > I am not really sure which is better. There is more in chapter 17 > > and one argument is that it is better to start a new page for > > chapter 17's contents listing. > > Both are OK to me. However, what I have a problem with ar

Re: wrong to use Index match in Info reader for the presence of index?

On Sun, Jun 29, 2025 at 11:55:11PM +0200, Patrice Dumas wrote: > On Sat, Jun 28, 2025 at 07:54:36PM +0100, Gavin Smith wrote: > > > I checked that a tag that is not 'Node:' or 'Ref:' has the Info reader > > > (but not the Emacs Info reader) stop going throu

Re: ugly page break in ToC

On Mon, Jul 07, 2025 at 06:10:52AM +, Werner LEMBERG wrote: > > [texinfo.tex 2025-03-22.08] > [luatex Version 1.22.0 (TeX Live 2025), Development id: 7673] > > > While playing around with adding parts to LilyPond's Notation > Reference, I got the attached layout in the ToC. As can be seen,

Re: New structure warnings and deeply nested nodes in libc manual

On Sat, Jul 05, 2025 at 10:29:00AM +0200, Patrice Dumas wrote: > On Sat, Jul 05, 2025 at 01:06:29AM +0100, Gavin Smith wrote: > > On Fri, Jul 04, 2025 at 10:18:05PM +0100, Gavin Smith wrote: > > > All these nodes (in the Sphinx output) have explicit pointers on the > > >

Re: New structure warnings and deeply nested nodes in libc manual

On Fri, Jul 04, 2025 at 10:18:05PM +0100, Gavin Smith wrote: > All these nodes (in the Sphinx output) have explicit pointers on the > node lines, just like the affected part of the libc manual. > > So I expect we will have to take some account of node pointers > when generating

Re: New structure warnings and deeply nested nodes in libc manual

On Mon, Jun 23, 2025 at 12:20:05AM +0200, Patrice Dumas wrote: > > > I do not really understand the comment explaining why Getopt and argp > > > start at the @section level, I do not get what "enough room for their > > > internal hierarchy" means, so it is not clear to me why the Glibc people > > >

Re: replacement for titlepage and more metadata commands

On Wed, Jul 02, 2025 at 11:23:14PM +0200, Patrice Dumas wrote: > > I've been learning about features in "KOMA-Script" (a LaTeX system) > > that allow for the creation of copyright pages: > > > > https://tex.stackexchange.com/questions/12570/how-to-add-a-copyright-page-to-the-report-class > > https

Re: replacement for titlepage and more metadata commands

On Sun, Jun 29, 2025 at 06:53:20PM +0200, Patrice Dumas wrote: > > The user uses @maketitle to produce the title page, and the second page > > is created by the contents of the @copyrightpage block. > > > > We could also possibly provide a @makecopyrightpage command > > for a default copyright pag

Re: wrong to use Index match in Info reader for the presence of index?

On Mon, Jun 30, 2025 at 04:04:12PM +0300, Eli Zaretskii wrote: > > Date: Mon, 30 Jun 2025 14:48:40 +0200 > > From: pertu...@free.fr > > Cc: gavinsmith0...@gmail.com, bug-texinfo@gnu.org > > > > On Mon, Jun 30, 2025 at 02:47:01PM +0300, Eli Zaretskii wrote: > > > > Date: Sun, 29 Jun 2025 18:59:22 +

Re: replacement for titlepage and more metadata commands

On Sun, Jun 29, 2025 at 06:53:20PM +0200, Patrice Dumas wrote: > On Sun, Jun 29, 2025 at 01:31:31AM +0100, Gavin Smith wrote: > > > > In the default case, when using @documentinfo, the copying text would > > > > be output > > > > on the verso of the title p

Re: New structure warnings and deeply nested nodes in libc manual

On Sun, Jun 29, 2025 at 02:42:08PM +0300, Eli Zaretskii wrote: > > It's because there are three menu entries for the same node. The > > first menu entry is expected because that node is subordinate in > > the sectioning structure (a @subsection underneath a @section), > > but the other two are une

Re: New structure warnings and deeply nested nodes in libc manual

On Sun, Jun 29, 2025 at 07:44:09AM +0300, Eli Zaretskii wrote: > > The only irregularities I could find in this manual are multiple menu > > entries > > for the same node (in "Help-Inv" and "Help-M"). > > That manual's structure is not a tree. It is pretty close to being a tree, as far as I coul

Re: New structure warnings and deeply nested nodes in libc manual

On Sun, Jun 29, 2025 at 07:50:47AM +0300, Eli Zaretskii wrote: > > From: Gavin Smith > > Date: Sat, 28 Jun 2025 21:45:03 +0100 > > > > And now I have warnings with this manual (patch at the end of this mail): > > > > $ TEXINFO_XS_STRUCTURE=0 ../tta/perl/tex

Re: New structure warnings and deeply nested nodes in libc manual

On Sun, Jun 29, 2025 at 02:15:05AM +0200, pertu...@free.fr wrote: > > Patrice: the > > "$menu_node_relations->{'associated_section'}->{'section_directions'}->{'next'}->{'associated_node'}->{'element'}" > > stuff is very verbose. All this is doing is finding the next > > node based on sectioning.

Re: replacement for titlepage and more metadata commands

> > In the default case, when using @documentinfo, the copying text would be > > output > > on the verso of the title page in PDF output. No separate command would be > > necessary. > > It is necessary if the user wants a specific formatting different than > the default. Could we support input

Re: New structure warnings and deeply nested nodes in libc manual

On Sat, Jun 28, 2025 at 08:47:33PM +0100, Gavin Smith wrote: > > As an example of a manual whose structure is not a tree, but is > > well-thought out, see info.texi. I'd expect to see such warnings > > there as well (unless the few explicit node pointers it has will avoid

Re: wrong to use Index match in Info reader for the presence of index?

On Sat, Jun 28, 2025 at 08:53:22PM +0100, Gavin Smith wrote: > Yes, I believe I said this in another message. I'll try again to explain > how I think Info should find Info nodes: D'oh, I keep on saying "Info nodes", I meant "index nodes".

Re: wrong to use Index match in Info reader for the presence of index?

On Sat, Jun 28, 2025 at 10:23:59PM +0300, Eli Zaretskii wrote: > > From: Gavin Smith > > Date: Sat, 28 Jun 2025 19:54:36 +0100 > > > > I also don't think it is a huge restriction to have "Index" in the > > node name. Non-English Info manuals could ha

Re: New structure warnings and deeply nested nodes in libc manual

On Tue, Jun 24, 2025 at 02:28:44PM +0300, Eli Zaretskii wrote: > > So while the Info output is basically fine, output that doesn't use > > menus, including the default HTML output and output with TeX, has > > problems. Hence I believe we should keep the new warnings with > > CHECK_NORMAL_MENU_STRU

Re: wrong to use Index match in Info reader for the presence of index?

On Sat, Jun 28, 2025 at 07:32:04PM +0200, Patrice Dumas wrote: > On Sat, Jun 28, 2025 at 04:39:14PM +0100, Gavin Smith wrote: > > As I remember, the marker for an index node (^@^H[index^@^H]) was not > > added by older versions of texi2any/makeinfo. So there are Info files > &

Re: wrong to use Index match in Info reader for the presence of index?

On Sat, Jun 28, 2025 at 05:28:07PM +0200, Patrice Dumas wrote: > Hello, > > In the Info reader, the node name for "Index" is checked in addition to > the presence of an index (with (node->flags & N_IsIndex), and possibly > the presence of non empty menu). > > I think that it is wrong to do that

Re: replacement for titlepage and more metadata commands

Coming back to this thread as I was prompted by Patrice: https://lists.gnu.org/archive/html/bug-texinfo/2025-01/msg00066.html I don't have a complete proposal but will post some comments to help get this conversation going again. I like the proposed @documentinfo command to contain metadata such

Re: specifying directory in htmlxref.conf file to include more files

> It appears to be possible to upload to subdirectories of > ftp.gnu.org/gnu/texinfo, pursuant to Info node "(maintain)FTP Upload > Directory Trees". > > I have done this so the new files are at > > https://ftp.gnu.org/gnu/texinfo/htmlxref.d/ > > We could upload a stub file to replace htmlxref.

Re: specifying directory in htmlxref.conf file to include more files

On Tue, Feb 18, 2025 at 12:10:09AM +0100, Patrice Dumas wrote: > On Mon, Feb 17, 2025 at 07:00:37PM +0000, Gavin Smith wrote: > > On Mon, Feb 17, 2025 at 07:44:53PM +0100, Patrice Dumas wrote: > > > On Mon, Feb 17, 2025 at 06:33:30PM +, Gavin Smith wrote: > > > >

Re: Incomplete and outdated packaging on CTAN

On Tue, Jun 24, 2025 at 03:03:51PM -0600, Karl Berry wrote: > For many years (dating back to when I maintained Texinfo), the way > Texinfo gets to CTAN is that I download the distribution tarball(s), > unpack it in ftp://tug.org/tex/texinfo*, and CTAN mirrors from there. > The CTAN metadata does no

Re: Incomplete and outdated packaging on CTAN

On Tue, Jun 24, 2025 at 08:58:50PM +0800, Rivulet Cedar wrote: > Hello everyone! We are having problems localising the Guix manual: > > https://codeberg.org/guix/translations/issues/1#issuecomment-5537951 > > The CTAN package is still at version 6.8, missing files like texinfo-zh.tex. > Rivulet C

Re: New structure warnings and deeply nested nodes in libc manual

On Mon, Jun 23, 2025 at 02:13:39PM +0300, Eli Zaretskii wrote: > The automatic calculation of Prev/Next/Up node pointers assume a > strict tree-like structure of a manual, and it also assumes that the > @menu item for a node appears in its parent node. If these conditions > are not fulfilled, then

Re: New structure warnings and deeply nested nodes in libc manual

On Sun, Jun 22, 2025 at 10:59:07PM +0200, Patrice Dumas wrote: > On Sun, Jun 22, 2025 at 06:53:42PM +0100, Gavin Smith wrote: > > @c Getopt and argp start at the @section level so that there's > > @c enough room for their internal hierarchy (mostly a problem with > >

New structure warnings and deeply nested nodes in libc manual

On Fri, May 30, 2025 at 05:38:11PM +0100, Gavin Smith wrote: > > In the next two or three days, I'll try to do the corresponding change > > in the C code. Then I'll look at splitting out the structure-checking > > code into its own function... > &g

Re: DocBook.pm "Not a HASH reference at ../tta/perl//../perl/Texinfo/Convert/DocBook.pm line 730.

On Sat, Jun 21, 2025 at 05:19:06PM +0200, Patrice Dumas wrote: > On Mon, Jun 09, 2025 at 10:39:32PM +0100, Gavin Smith wrote: > > texi2any --docbook currently does not work on some manuals. > > > > Here is an input file (cut down from the libc manual) producing > > the

Re: doc/texinfo.tex 2025-02-07.12 yields too much inter-word spacing in PDF

; and "Appendix A". Thanks for tracking down the commit where the bug became apparent. In fact, it is possible this bug is 10 years old. It is a problem with the \requireauxfile macro, which is used to read the .aux file when required, rather than in @setfilename which was fo

Re: texinfo 7.2 does not utilize libinfow

On Wed, Jun 18, 2025 at 09:28:25AM +0300, Дилян Палаузов wrote: > Hello, > > libtinfo and libtinfow are created from the source code of ncurses. When > there ./configure is called with --enable-widec then the w libraries are > generated - libticw, libformw, libncursesw, libtinfow. > > These l

Re: doc/texinfo.tex 2025-02-07.12 yields too much inter-word spacing in PDF

On Tue, Jun 17, 2025 at 01:35:53PM +0200, Vincent Lefevre wrote: > Hi, > > The latest doc/texinfo.tex file (2025-03-22) yields too much > inter-word spacing in PDF for the "Permission is granted to copy..." > paragraph (at least) of the GNU MPFR manual, which contains > > @copying > This manual d

Re: texinfo 7.2 does not utilize libinfow

On Tue, Jun 17, 2025 at 08:03:56PM +0300, Дилян Палаузов wrote: > Hello, > > I thought that compiling everything only for wide characters is a very good > idea. But ./configure for texinfo 7.2 prints > > config.status: creating po/Makefile > config.status: creating po_document/POTFILES > config

Re: @lisp vs @example lisp in HTMLL

On Sun, Jun 15, 2025 at 04:16:23PM -0700, Raymond Toy wrote: > In > https://www.gnu.org/software/texinfo/manual/texinfo/html_node/_0040lisp.html, > it says |@lisp| is the same as |@examle lisp|. In the generated HTML file, > visually, they produce the same thing. However, the html code is quite > d

Test failures with TEXINFO_XS_STRUCTURE=0

Commit a2be60e0154934 (today's date). There are two test failures with TEXINFO_XS_STRUCTURE=0. Under tta/perl: $ TEXINFO_XS_STRUCTURE=0 make check FAIL: t/transformations.t 145 - test_insert_nodes_nodedescription converted html FAIL: t/transformations.t 160 - test_insert_nodes_reallocated_node

Re: tta/perl tests regeneration is broken

On Sat, Jun 14, 2025 at 11:07:53PM +0100, Gavin Smith wrote: > Commit e9a68e4d6a95f3a0f62 (today's date). > > I cannot regenerate the tta/perl tests (to test changes I am working on). If > under tta/perl I run > > $ ../maintain/all_tests.sh generate > > I g

tta/perl tests regeneration is broken

Commit e9a68e4d6a95f3a0f62 (today's date). I cannot regenerate the tta/perl tests (to test changes I am working on). If under tta/perl I run $ ../maintain/all_tests.sh generate I get an error message: ./t/02coverage.t ok 1 ReadDocbook::convert_tree: undef reader at t//../../perl/Texinfo/Conve

DocBook.pm "Not a HASH reference at ../tta/perl//../perl/Texinfo/Convert/DocBook.pm line 730.

texi2any --docbook currently does not work on some manuals. Here is an input file (cut down from the libc manual) producing the error: \input texinfo @titlepage @center @titlefont{The GNU C Library} @sp 1 @center @titlefont{Reference Manual} @sp 2 @center Sandra Loosemore @center with @center Ri

Re: HV and SV problem

On Sat, Jun 07, 2025 at 09:51:50AM +0200, Patrice Dumas wrote: > On Fri, Jun 06, 2025 at 07:27:55PM +0100, Gavin Smith wrote: > Indeed, that's the issue, thanks for the analysis. What confused me is > that in Perl code, it does not happens that way, but I think that it is >

Re: HV and SV problem

On Fri, Jun 06, 2025 at 07:27:56PM +0100, Gavin Smith wrote: > Then I suspected something was strange with the way the data structures > were constructed. I suspect it would be hard to replicate the loss without > using XS. After reading through "man perlref" again, I fou

Re: HV and SV problem

On Fri, Jun 06, 2025 at 07:27:56PM +0100, Gavin Smith wrote: > To explain the problem another way, I believe that if you had XS code like: > > HV *hv = newHV (); > SV *sv = newSViv(7); > hv_store (hv, "foo", strlen ("foo"), sv, 0); > hv_store (hv, "bar&

Re: HV and SV problem

On Fri, Jun 06, 2025 at 07:27:56PM +0100, Gavin Smith wrote: > I had a breakthrough when I realised that the referenced parent element, > the $previous_ending_container->{'contents'}->[$i]->{'parent'}, was the > before_item > element that was being overwri

HV and SV problem

On Thu, Jun 05, 2025 at 10:45:17PM +0200, Patrice Dumas wrote: > I do not have any good idea on what more could be done to understand > what is going on. I based my investigations on commit 08c59dc0cbc36 (2025-06-05) as you said that your later commits did not fix the problem. Summary: I think I

Re: Duplicate variable use in Texinfo/ManipulateTree.pm

On Thu, Jun 05, 2025 at 01:07:51AM +0200, Patrice Dumas wrote: > On Wed, Jun 04, 2025 at 09:37:20PM +0100, Gavin Smith wrote: > > When I run the texi2any test suite with TEXINFO_XS=omit, there is > > a warning printed: > > > > "my" variable $current_nr masks

Duplicate variable use in Texinfo/ManipulateTree.pm

When I run the texi2any test suite with TEXINFO_XS=omit, there is a warning printed: "my" variable $current_nr masks earlier declaration in same scope at t//../../perl/Texinfo/ManipulateTree.pm line 839. This is presumably a bug. Here is the relevant part of the code: sub print_element_detail

TEXINFO_XS_STRUCTURE=0 test divergence

Today, I ran $ TEXINFO_XS_STRUCTURE=0 ../maintain/all_tests.sh generate under tta/perl to test some changes to the Perl structuring code. However, there are again test divergences from the XS code. The diff begins: diff --git a/tta/perl/t/results/converters_tests/indices_in_begin_tables_lists

Re: Improved diagnostic messages for ill-formed menus?

On Wed, Jun 04, 2025 at 12:21:52AM +0200, Patrice Dumas wrote: > On Fri, May 30, 2025 at 05:38:10PM +0100, Gavin Smith wrote: > > I've done this. I noticed that complete_node_tree_with_menus it was > > not run from C/texi2any.c, but some of this looks quite different to >

gawk compatibility mode build error?

On Wed, Jun 04, 2025 at 04:07:43AM -0500, NR wrote: > On Wed, Jun 4, 2025 at 4:05 AM NR wrote: > > > > b92d3c94189a7cbf8ca7fba5a4ccdec03459386c > > ``` > > gawk -f ./jrtangle ./ti.twjr || rm -f texindex.awk > > gawk: ./jrtangle:28:match(PROCINFO["version"], /^[0-9]+\./, v); > > gawk: .

Re: Usefulness of menu directions

On Tue, Jun 03, 2025 at 10:44:47PM +0200, Patrice Dumas wrote: > > It would be better to report an error on the menu directly rather > > indirectly through the menu directions. > > On the whole menu, or on the menu entry? It should not be too difficult > to do on the menu entry or menu entry node

Re: info tests reading $ginfo_output started failing

On Tue, Jun 03, 2025 at 08:23:44AM +0200, Patrice Dumas wrote: > On Thu, May 29, 2025 at 04:04:14PM +0100, Gavin Smith wrote: > > On Thu, May 29, 2025 at 04:33:23PM +0200, Patrice Dumas wrote: > > You could try reverting to the previous version to see if the test suites > > pa

Menu structure check reform proposal

On Fri, May 30, 2025 at 06:16:30PM +0100, Gavin Smith wrote: > On Wed, May 28, 2025 at 09:23:15AM +0200, Patrice Dumas wrote: > > > It could be possible to write completely new code just checking for > > > faults, which doesn't rely on "menu directions" at all

Usefulness of menu directions

On Wed, May 28, 2025 at 09:23:15AM +0200, Patrice Dumas wrote: > > It could be possible to write completely new code just checking for > > faults, which doesn't rely on "menu directions" at all. (It may not be > > clear to the user what "menu directions" are as these are an abstraction > > created

Re: Improved diagnostic messages for ill-formed menus?

On Wed, May 28, 2025 at 09:33:39PM +0100, Gavin Smith wrote: > > > What I would like to change, as a preliminary step before trying to > > > find better ways of reporting on problems with the node structure, > > > is to completely separate: > > > > &g

Re: Many differences in test results with TEXINFO_XS_STRUCTURE=0

On Thu, May 29, 2025 at 10:04:46AM +0200, Patrice Dumas wrote: > On Tue, May 27, 2025 at 08:23:30PM +0100, Gavin Smith wrote: > > However, running the test suite with TEXINFO_XS_STRUCTURE=0 produces > > many other changes too: > > I fixed all the issues, I think. One issue

Re: info tests reading $ginfo_output started failing

On Thu, May 29, 2025 at 04:33:23PM +0200, Patrice Dumas wrote: > Hello, > > Many info tests (38 over 87) started failing for me. The tests that > fail are those that look at the $ginfo_output content. I tried to > comment out some code modified by the last commit on the info test > suite, but it

Re: Improved diagnostic messages for ill-formed menus?

On Wed, May 28, 2025 at 09:23:15AM +0200, Patrice Dumas wrote: > On Tue, May 27, 2025 at 08:54:39PM +0100, Gavin Smith wrote: > > > > 'complete_node_tree_with_menus' is not an easy function to read. > > This function, as it is named (complete_node_tree_with_menus

Improved diagnostic messages for ill-formed menus?

A mistake in a @menu block can produce many errors that are hard to understand. For example, one missing menu entry: @menu * One:: * Three:: @end menu instead of @menu * One:: * Two:: * Three:: @end menu produces 5 separate errors: test.texi:13: warning: node next pointer for `One' is `Two' b

Many differences in test results with TEXINFO_XS_STRUCTURE=0

I wanted to try some changes to the code checking the structure of sections and menus. I changed the code in Texinfo/Structuring.pm first, and wanted to know how this affected the output of the test suite. However, running the test suite with TEXINFO_XS_STRUCTURE=0 produces many other changes too:

Re: automake-history in htmlxref.d/Texinfo_GNU.cnf

On Sun, May 25, 2025 at 11:45:27AM -0600, Karl Berry wrote: > Can we add automake-history to the new Texinfo_GNU.cnf please? > To avoid the warning from makeinfo. I think it's this: Thanks I have commited this. > > diff --git a/util/htmlxref.d/Texinfo_GNU.cnf b/util/htmlxref.d/Texinfo_GNU.cnf

Re: Build - possible bug for "multiple outputs"

On Sat, Mar 01, 2025 at 10:51:44AM +, Gavin Smith wrote: > I came across a section in the Automake manual describing problems with > Makefile rules with multiple outputs. > > https://www.gnu.org/software/automake/manual/html_node/Multiple-Outputs.html > > We have such rules

Re: makeinfo --latex and long names in cross references

On Sun, May 25, 2025 at 10:57:03AM +0100, Gavin Smith wrote: > I've just checked that the "@xrefautomaticsectiontitle on" command > works as expected for LaTeX output and uses the section name for > the text of a cross-reference. Place this on a line early in your file

Re: makeinfo --latex and long names in cross references

On Sun, May 25, 2025 at 03:08:35AM -0600, arn...@skeeve.com wrote: > Hi. > > I'm using `makeinfo --latex' to prepare a book for publication. > With texinfo.tex, cross references automatically use the "long" > name of the referred to chapter, section etc. > > However, in the generated LaTeX, the t

Re: blank lines behavior and documentation

On Tue, May 20, 2025 at 05:17:31PM +0200, Vincent Lefevre wrote: > Hi, > > There is a lack of documentation concerning blank lines in the > texinfo source, in particular concerning their effect with the > various output formats. > > In the GNU Texinfo 7.1.1 manual, it is said that blank lines are

Re: EPUB still non-conformance

On Fri, May 23, 2025 at 11:27:29AM +0100, Daniel Cerqueira wrote: > Hi. > > No surprises here. The `epubcheck` with 7.2 gives me no errors. > Thanks for your work. That's great!

Re: "texi2any --html --no-split" should not generate accesskey attributes

On Wed, May 21, 2025 at 09:36:38AM +0200, Patrice Dumas wrote: > > Thanks for the report. I suppose we should avoid outputting the accesskey > > attribute for non-split output. > > Actually, the issue happens for other split options that end up with > more than one node/section in an HTML page, s

Re: [PATCH] Fix an uninitialized variable.

On Tue, May 20, 2025 at 01:06:38PM -0700, Collin Funk wrote: > Is there any benefit to doing this over just doing: > > $ git am 0001-example-commit.patch > > That command should set the author correctly. > > Collin It's because we keep a ChangeLog file changes to which need to be included i

Re: [PATCH] Fix an uninitialized variable.

On Tue, May 20, 2025 at 10:06:17AM +0200, Patrice Dumas wrote: > Hello, > > The analysis and patch look good to me. Gavin, I think that you can > apply it (I know that it is a kind of patch practical for git to get > both the patch and commiter right, but I don't know how to apply that > kind of

Re: "texi2any --html --no-split" should not generate accesskey attributes

> > The following seems to work for the non-XS code: > > > > diff --git a/tta/perl/Texinfo/Convert/HTML.pm > > b/tta/perl/Texinfo/Convert/HTML.pm > > index c68559c7eb..3435405254 100644 > > --- a/tta/perl/Texinfo/Convert/HTML.pm > > +++ b/tta/perl/Texinfo/Convert/HTML.pm > > @@ -9247,6 +9247,13 @

Re: "texi2any --html --no-split" should not generate accesskey attributes

On Tue, May 13, 2025 at 04:28:28PM +0200, Vincent Lefevre wrote: > "texi2any --html --no-split" generates many accesskey attributes > with the same value. This does not make sense. In particular, the > behavior in such a case depends on the web browser: > > https://www.w3schools.com/TAGs/att_acc

Re: info selects wrong language in multiple language manual

On Sat, May 17, 2025 at 01:11:40AM +0200, pelzflorian (Florian Pelz) wrote: > Gavin Smith writes: > > Yes, that is expected. If the entry in 'dir' > > > > * Guix: (guix.es). Gestión del software instalado y la > >

Re: capitalization inconsistency in top of info tree.

On Fri, May 16, 2025 at 07:45:23PM +, keenban--- via Bug reports for the GNU Texinfo documentation system wrote: > Hello GNU texinfo team. > This is my first bug report, so hopefully I am sending to the correct > location and with the correct information. > In the Node: "Top," there are menu

Re: info selects wrong language in multiple language manual

On Fri, May 16, 2025 at 02:18:36PM +0200, pelzflorian (Florian Pelz) wrote: > Hello Texinfo. Guix has a Texinfo manual translated to multiple > languages: > > share/info/guix.info.gz leads to the English manual. > > There are also guix.de.info.gz, guix.fr.info.gz, guix.pt_BR.info.gz and > more l

  1   2   3   4   5   6   7   8   9   10   >