TeX: use charsets, and nicer error message when input file is not found

This fixes #421 in a lazy way:
it would be better to not even try to parse \input commands that are
located within a verbatim environment, but it is incredibly hard while
this trick is almost trivial: I'm just pointing to the right piece of
code to allow users to circumvent the issue.

MANIFEST

@@ -947,6 +947,12 @@ t/fmt/tex/basic.pot
 t/fmt/tex/basic.tex
 t/fmt/tex/basic.trans
 t/fmt/tex/basic.trans.stderr
+t/fmt/tex/input-in-verbatim.norm
+t/fmt/tex/input-in-verbatim.po
+t/fmt/tex/input-in-verbatim.pot
+t/fmt/tex/input-in-verbatim.stderr-without-ignore
+t/fmt/tex/input-in-verbatim.tex
+t/fmt/tex/input-in-verbatim.trans
 t/fmt/texinfo/comments.norm
 t/fmt/texinfo/comments.po
 t/fmt/texinfo/comments.pot

NEWS

@@ -38,6 +38,8 @@ Asciidoc:
 Tex:
  * Allow verbatim environment declaration without trailing space
    (GitHub's #407). Thanks Mayeul Cantan for the report & the fix.
+ * Informative error message when file not found for \input, indicating
+   how to skip this include (GitHub's #421) [mquinson].
 
 Texinfo:
  * Translate partial menu node names (GitHub's #437) [Apteryks].

lib/Locale/Po4a/TeX.pm

@@ -86,7 +86,7 @@ use vars qw(@ISA @EXPORT);
 use Locale::Po4a::TransTractor;
 use Locale::Po4a::Common;
 use File::Basename qw(dirname);
-use Carp qw(croak);
+use Carp           qw(croak);
 
 use Encode;
 use Encode::Guess;
@@ -926,15 +926,16 @@ Overloads Transtractor's read().
 
 =cut
 
-sub read {
+sub read ($$$$) {
     my $self     = shift;
     my $filename = shift;
     my $refname  = shift;
+    my $charset  = shift;
 
     # keep the directory name of the main file.
     $my_dirname = dirname($filename);
 
-    push @{ $self->{TT}{doc_in} }, read_file( $self, $filename, $refname );
+    push @{ $self->{TT}{doc_in} }, read_file( $self, $filename, $refname, $charset );
 }
 
 =item B<read_file>
@@ -954,10 +955,11 @@ sub read_file {
     my $filename = shift
       or croak wrap_mod( "po4a::tex", dgettext( "po4a", "Cannot read from file without having a filename" ) );
     my $refname = shift // $filename;
+    my $charset = shift || 'UTF-8';
     my $linenum = 0;
     my @entries = ();
 
-    open( my $in, $filename )
+    open( my $in, "<:encoding($charset)", $filename )
       or croak wrap_mod( "po4a::tex", dgettext( "po4a", "Cannot read from %s: %s" ), $filename, $! );
     while ( defined( my $textline = <$in> ) ) {
         $linenum++;
@@ -995,11 +997,19 @@ sub read_file {
                 open( KPSEA, "kpsewhich " . $newfilename . " |" );
                 my $newfilepath = <KPSEA>;
 
-                if ( $newfilename ne "" and $newfilepath eq "" ) {
-                    die wrap_mod( "po4a::tex", dgettext( "po4a", "Cannot find %s with kpsewhich" ), $filename );
+                if ( $newfilename ne "" and ( $newfilepath // '' ) eq '' ) {
+                    die wrap_mod(
+                        "po4a::tex",
+                        dgettext(
+                            "po4a",
+                            "Cannot find '%s' with kpsewhich. To prevent this file to be included, add '-o exclude_include=%s' to the options, either on the command line or in your po4a.conf file."
+                        ),
+                        $newfilename,
+                        $newfilename
+                    );
                 }
 
-                push @entries, read_file( $self, $newfilepath, $newfilename );
+                push @entries, read_file( $self, $newfilepath, $newfilename, $charset );
                 if ( $tag eq "include" ) {
                     $textline = "\\clearpage" . $end;
                 } else {

t/Testhelper.pm

@@ -429,6 +429,12 @@ sub run_one_format {
     } elsif ( $error != 0 && $exit_status != 0 ) {
         pass("Expected error detected in $doc");
         note("  Failing as expected: $cmd (retcode: $exit_status)");
+        note("Produced output:");
+        open FH, $real_stderr || die "Cannot open output file that I just created, I'm puzzled";
+        while (<FH>) {
+            note("  $_");
+        }
+        note("(end of command output)\n");
     } else {
         fail("Normalizing $doc: $exit_status");
         note("  FAIL: $cmd");

t/fmt-tex.t

@@ -18,5 +18,20 @@ foreach my $t (qw(basic theorem)) {
       };
 }
 
+push @tests,
+  {
+    'doc'         => 'invalid input command, without ignore',
+    'format'      => 'latex',
+    'input'       => "fmt/tex/input-in-verbatim.tex",
+    'error'       => 1,
+    'norm_stderr' => 'fmt/tex/input-in-verbatim.stderr-without-ignore',
+  },
+  {
+    'doc'     => 'invalid input command, with exclude_include',
+    'format'  => 'latex',
+    'input'   => "fmt/tex/input-in-verbatim.tex",
+    'options' => '-o exclude_include=main.tex'
+  };
+
 run_all_tests(@tests);
 0;

t/fmt/tex/input-in-verbatim.norm

@@ -0,0 +1,11 @@
+% This file was generated with po4a. Translate the source file.
+%
+\documentclass[12pt]{article}
+\begin{document}
+
+\begin{verbatim}
+\documentclass[ignorenonframetext]{beamer}
+\input{main.tex}
+\end{verbatim}
+
+\end{document}

t/fmt/tex/input-in-verbatim.po

@@ -0,0 +1,27 @@
+# SOME DESCRIPTIVE TITLE
+# Copyright (C) YEAR Free Software Foundation, Inc.
+# This file is distributed under the same license as the PACKAGE package.
+# FIRST AUTHOR <EMAIL@ADDRESS>, YEAR.
+#
+#, fuzzy
+msgid ""
+msgstr ""
+"Project-Id-Version: PACKAGE VERSION\n"
+"POT-Creation-Date: 2024-01-04 01:52+0100\n"
+"PO-Revision-Date: YEAR-MO-DA HO:MI+ZONE\n"
+"Last-Translator: FULL NAME <EMAIL@ADDRESS>\n"
+"Language-Team: LANGUAGE <[email protected]>\n"
+"Language: \n"
+"MIME-Version: 1.0\n"
+"Content-Type: text/plain; charset=UTF-8\n"
+"Content-Transfer-Encoding: 8bit\n"
+
+#. type: verbatim
+#: input-in-verbatim.tex:7
+#, no-wrap
+msgid ""
+"\\documentclass[ignorenonframetext]{beamer}\n"
+"\\input{main.tex}"
+msgstr ""
+"\\documentclass[ignorenonframetext]{beamer}\n"
+"\\input{main.tex}"

t/fmt/tex/input-in-verbatim.pot

@@ -0,0 +1,25 @@
+# SOME DESCRIPTIVE TITLE
+# Copyright (C) YEAR Free Software Foundation, Inc.
+# This file is distributed under the same license as the PACKAGE package.
+# FIRST AUTHOR <EMAIL@ADDRESS>, YEAR.
+#
+#, fuzzy
+msgid ""
+msgstr ""
+"Project-Id-Version: PACKAGE VERSION\n"
+"POT-Creation-Date: 2024-01-04 01:50+0100\n"
+"PO-Revision-Date: YEAR-MO-DA HO:MI+ZONE\n"
+"Last-Translator: FULL NAME <EMAIL@ADDRESS>\n"
+"Language-Team: LANGUAGE <[email protected]>\n"
+"Language: \n"
+"MIME-Version: 1.0\n"
+"Content-Type: text/plain; charset=UTF-8\n"
+"Content-Transfer-Encoding: 8bit\n"
+
+#. type: verbatim
+#: input-in-verbatim.tex:7
+#, no-wrap
+msgid ""
+"\\documentclass[ignorenonframetext]{beamer}\n"
+"\\input{main.tex}"
+msgstr ""

t/fmt/tex/input-in-verbatim.stderr-without-ignore

@@ -0,0 +1 @@
+po4a::tex: Cannot find 'main.tex' with kpsewhich. To prevent this file to be included, add '-o exclude_include=main.tex' to the options, either on the command line or in your po4a.conf file.

t/fmt/tex/input-in-verbatim.tex

@@ -0,0 +1,9 @@
+\documentclass[12pt]{article}
+\begin{document}
+
+\begin{verbatim}
+\documentclass[ignorenonframetext]{beamer}
+\input{main.tex}
+\end{verbatim}
+
+\end{document}

t/fmt/tex/input-in-verbatim.trans

@@ -0,0 +1,11 @@
+% This file was generated with po4a. Translate the source file.
+%
+\documentclass[12pt]{article}
+\begin{document}
+
+\begin{verbatim}
+\documentclass[ignorenonframetext]{beamer}
+\input{main.tex}
+\end{verbatim}
+
+\end{document}