branch master updated: * tp/Texinfo/Convert/Converter.pm (converter_options_for_output): rename encode_converter_for_output as converter_options_for_output. Update callers.

[Patrice Dumas]

This is an automated email from the git hooks/post-receive script.

pertusus pushed a commit to branch master
in repository texinfo.

The following commit(s) were added to refs/heads/master by this push:
     new 150d948494 * tp/Texinfo/Convert/Converter.pm 
(converter_options_for_output): rename encode_converter_for_output as 
converter_options_for_output. Update callers.
150d948494 is described below

commit 150d94849494cd9d34da1e50d6bf947302ef1d92
Author: Patrice Dumas <pertusus@free.fr>
AuthorDate: Sun Nov 5 20:31:56 2023 +0100

    * tp/Texinfo/Convert/Converter.pm (converter_options_for_output):
    rename encode_converter_for_output as converter_options_for_output.
    Update callers.
    
    * tp/Texinfo/Convert/HTML.pm (_prepare_conversion_units)
    (_prepare_units_directions_files, _prepare_title_titlepage)
    (_html_convert_output), tp/Texinfo/Convert/Text.pm
    (select_text_options), tp/Texinfo/Translations.pm (_XS_gdt),
    tp/Texinfo/XS/convert/ConvertXS.xs (html_prepare_conversion_units)
    (html_prepare_units_directions_files)
    (html_prepare_title_titlepage, html_convert_output),
    tp/Texinfo/XS/main/TranslationsXS.xs (gettree),
    tp/Texinfo/XS/main/get_perl_info.c (copy_sv_options_for_convert_text):
    do not convert strings to UTF-8 in perl but in XS.  Rename
    encode_text_options as select_text_options in Texinfo::Convert::Text.
    
    * tp/Texinfo/Translations.pm (import),
    tp/Texinfo/XS/main/TranslationsXS.xs (gdt): remove _XS_gettree, have a
    direct override of gdt (in comments, as it does not work).  In XS,
    rename gettree as gdt and adapt call to match gdt call in perl.
---
 ChangeLog                            | 23 +++++++++++++++++
 tp/Texinfo/Convert/Converter.pm      |  5 ++--
 tp/Texinfo/Convert/HTML.pm           | 50 +++++++++++++-----------------------
 tp/Texinfo/Convert/Text.pm           | 44 ++++++++++---------------------
 tp/Texinfo/Translations.pm           | 39 +++++-----------------------
 tp/Texinfo/XS/convert/ConvertXS.xs   | 22 ++++++++--------
 tp/Texinfo/XS/main/TranslationsXS.xs | 24 ++++++++++-------
 tp/Texinfo/XS/main/get_perl_info.c   |  4 +--
 8 files changed, 90 insertions(+), 121 deletions(-)

diff --git a/ChangeLog b/ChangeLog
index 58fb5724f9..b031d5e694 100644
--- a/ChangeLog
+++ b/ChangeLog
@@ -1,3 +1,26 @@
+2023-11-05  Patrice Dumas  <pertusus@free.fr>
+
+       * tp/Texinfo/Convert/Converter.pm (converter_options_for_output):
+       rename encode_converter_for_output as converter_options_for_output.
+       Update callers.
+
+       * tp/Texinfo/Convert/HTML.pm (_prepare_conversion_units)
+       (_prepare_units_directions_files, _prepare_title_titlepage)
+       (_html_convert_output), tp/Texinfo/Convert/Text.pm
+       (select_text_options), tp/Texinfo/Translations.pm (_XS_gdt),
+       tp/Texinfo/XS/convert/ConvertXS.xs (html_prepare_conversion_units)
+       (html_prepare_units_directions_files)
+       (html_prepare_title_titlepage, html_convert_output),
+       tp/Texinfo/XS/main/TranslationsXS.xs (gettree),
+       tp/Texinfo/XS/main/get_perl_info.c (copy_sv_options_for_convert_text):
+       do not convert strings to UTF-8 in perl but in XS.  Rename
+       encode_text_options as select_text_options in Texinfo::Convert::Text.
+
+       * tp/Texinfo/Translations.pm (import),
+       tp/Texinfo/XS/main/TranslationsXS.xs (gdt): remove _XS_gettree, have a
+       direct override of gdt (in comments, as it does not work).  In XS,
+       rename gettree as gdt and adapt call to match gdt call in perl.
+
 2023-11-05  Patrice Dumas  <pertusus@free.fr>
 
        * tp/Texinfo/Common.pm, tp/Texinfo/Config.pm
diff --git a/tp/Texinfo/Convert/Converter.pm b/tp/Texinfo/Convert/Converter.pm
index 808afa7d31..bd27c40b49 100644
--- a/tp/Texinfo/Convert/Converter.pm
+++ b/tp/Texinfo/Convert/Converter.pm
@@ -548,9 +548,8 @@ sub encode_converter_document($)
   return $result;
 }
 
-# to be used before output
-# TODO document?
-sub encode_converter_for_output($)
+# FIXME remove, do in XS
+sub converter_options_for_output($)
 {
   my $self = shift;
 
diff --git a/tp/Texinfo/Convert/HTML.pm b/tp/Texinfo/Convert/HTML.pm
index d2645869bf..0d43e18bd9 100644
--- a/tp/Texinfo/Convert/HTML.pm
+++ b/tp/Texinfo/Convert/HTML.pm
@@ -8524,8 +8524,7 @@ sub convert_tree($$;$)
   my $explanation = shift;
 
   # No XS, convert_tree is not called on trees registered in XS
-  #my $XS_result = _XS_html_convert_tree($self, $tree,
-  #     (defined($explanation) ? Encode::encode('UTf-8', $explanation) : ''));
+  #my $XS_result = _XS_html_convert_tree($self, $tree, $explanation);
   #return $XS_result if (defined($XS_result));
 
   # when formatting accents, goes through xml_accent without
@@ -9378,13 +9377,12 @@ sub _prepare_conversion_units($$$)
   my ($output_units, $special_units, $associated_special_units);
 
   if ($self->{'converter_descriptor'} and $XS_convert) {
-    my $encoded_converter = $self->encode_converter_for_output();
-    my $encoded_document_name = Encode::encode('UTF-8', $document_name);
+    my $converter_info = $self->converter_options_for_output();
     my ($targets, $special_targets, $seen_ids);
     ($output_units, $special_units, $associated_special_units,
      $targets, $special_targets, $seen_ids)
-      = _XS_prepare_conversion_units($encoded_converter,
-                                     $encoded_document_name);
+      = _XS_prepare_conversion_units($converter_info,
+                                     $document_name);
     $self->{'targets'} = $targets;
     $self->{'special_targets'} = $special_targets;
     $self->{'seen_ids'} = $seen_ids;
@@ -9451,20 +9449,15 @@ sub _prepare_units_directions_files($$$$$$$$)
   my $document_name = shift;
 
   if ($self->{'converter_descriptor'} and $XS_convert) {
-    my $encoded_converter = $self->encode_converter_for_output();
-    my $encoded_document_name = Encode::encode('UTF-8', $document_name);
-    my $encoded_output_file = Encode::encode('UTF-8', $output_file);
-    my $encoded_destination_directory
-         = Encode::encode('UTF-8', $destination_directory);
-    my $encoded_output_filename = Encode::encode('UTF-8', $output_filename);
+    my $converter_info = $self->converter_options_for_output();
 
     my ($XS_files_source_info, $global_units_directions,
         $elements_in_file_count, $filenames,
         $file_counters, $out_filepaths)
-      = _XS_prepare_units_directions_files($encoded_converter,
+      = _XS_prepare_units_directions_files($converter_info,
            $output_units, $special_units, $associated_special_units,
-           $encoded_output_file, $encoded_destination_directory,
-           $encoded_output_filename, $encoded_document_name);
+           $output_file, $destination_directory,
+           $output_filename, $document_name);
     $self->{'global_units_directions'} = $global_units_directions;
     $self->{'elements_in_file_count'} = $elements_in_file_count;
 
@@ -11141,10 +11134,8 @@ sub _prepare_title_titlepage($$$$)
   my $output_filename = shift;
 
   if ($self->{'converter_descriptor'} and $XS_convert) {
-    my $encoded_output_filename = Encode::encode('UTF-8', $output_filename);
-    my $encoded_output_file = Encode::encode('UTF-8', $output_file);
     _XS_html_prepare_title_titlepage($self, $output_units,
-                       $encoded_output_file, $encoded_output_filename);
+                            $output_file, $output_filename);
     return;
   }
 
@@ -11177,7 +11168,7 @@ sub convert($$)
   my $self = shift;
   my $document = shift;
 
-  my $encoded_converter;
+  my $converter_info;
   my $root = $document->tree();
 
   my $result = '';
@@ -11207,9 +11198,9 @@ sub convert($$)
   if ($self->{'converter_descriptor'} and $XS_convert) {
     # Do it preferentially in XS, and import to perl, to have data
     # setup in C for XS too.
-    $encoded_converter = $self->encode_converter_for_output();
+    $converter_info = $self->converter_options_for_output();
     my $global_units_directions =
-      _XS_prepare_output_units_global_targets($encoded_converter,
+      _XS_prepare_output_units_global_targets($converter_info,
            $output_units, $special_units, $associated_special_units);
     $self->{'global_units_directions'} = $global_units_directions;
   } else {
@@ -11246,7 +11237,7 @@ sub convert($$)
   $self->{'current_filename'} = '';
 
   if ($self->{'converter_descriptor'} and $XS_convert) {
-    my $XS_result = _XS_html_convert_convert ($encoded_converter, $root,
+    my $XS_result = _XS_html_convert_convert ($converter_info, $root,
                                               $output_units, $special_units);
     $self->_finalize_output_state();
     return $XS_result;
@@ -11533,18 +11524,13 @@ sub _html_convert_output($$$$$$$$)
       $destination_directory, $output_filename, $document_name) = @_;
 
   if ($self->{'converter_descriptor'} and $XS_convert) {
-    my $encoded_converter = $self->encode_converter_for_output();
-    my $encoded_document_name = Encode::encode('UTF-8', $document_name);
-    my $encoded_output_file = Encode::encode('UTF-8', $output_file);
-    my $encoded_destination_directory
-         = Encode::encode('UTF-8', $destination_directory);
-    my $encoded_output_filename = Encode::encode('UTF-8', $output_filename);
+    my $converter_info = $self->converter_options_for_output();
 
     my $XS_text_output
-           = _XS_html_convert_output ($encoded_converter,
-                     $root, $output_units, $special_units, 
$encoded_output_file,
-                     $encoded_destination_directory, $encoded_output_filename,
-                     $encoded_document_name);
+           = _XS_html_convert_output ($converter_info,
+                     $root, $output_units, $special_units, $output_file,
+                     $destination_directory, $output_filename,
+                     $document_name);
     return $XS_text_output;
   }
 
diff --git a/tp/Texinfo/Convert/Text.pm b/tp/Texinfo/Convert/Text.pm
index f00e418d0c..8422b22e48 100644
--- a/tp/Texinfo/Convert/Text.pm
+++ b/tp/Texinfo/Convert/Text.pm
@@ -412,50 +412,32 @@ sub copy_options_for_convert_text($;$)
   return %options;
 }
 
-# encode to UTF-8 bytes before passing to XS code.  Specific
-# text options are in general ASCII strings, but this is still
-# cleaner.  Also encode and select converter options passed.
-sub encode_text_options($)
+# select converter options passed.
+sub select_text_options($)
 {
   my $options = shift;
-  my $encoded_options = {};
-
-  foreach my $option ('enabled_encoding') {
-    if (defined($options->{$option})) {
-      $encoded_options->{$option}
-        = Encode::encode("UTF-8", $options->{$option});
-    }
-  }
+  my $selected_options = {};
 
   foreach my $option (@text_indicator_converter_options,
                       'INCLUDE_DIRECTORIES',
+                      'expanded_formats',
                       # non-converter indicator options
-                      'sc', 'code', 'sort_string') {
+       'enabled_encoding', 'sc', 'code', 'sort_string') {
     if (defined($options->{$option})) {
-      $encoded_options->{$option} = $options->{$option};
-    }
-  }
-
-  if (defined($options->{'expanded_formats'})) {
-    # FIXME may not need to encode, as the formats are ascii strings
-    my $expanded_formats = {};
-    foreach my $format (keys(%{$options->{'expanded_formats'}})) {
-      my $encoded_format = Encode::encode("UTF-8", $format);
-      $expanded_formats->{$encoded_format} = 1;
+      $selected_options->{$option} = $options->{$option};
     }
-    $encoded_options->{'expanded_formats'} = $expanded_formats;
   }
 
   # called through convert_to_text with a converter in text options
   if ($options->{'converter'}
       and $options->{'converter'}->{'conf'}) {
-    $encoded_options->{'other_converter_options'}
+    $selected_options->{'other_converter_options'}
        = $options->{'converter'}->{'conf'};
   }
 
-  $encoded_options->{'self_converter_options'} = $options;
+  $selected_options->{'self_converter_options'} = $options;
 
-  return $encoded_options;
+  return $selected_options;
 }
 
 # This is used if the document is available for XS, but XS is not
@@ -490,8 +472,8 @@ sub convert_to_text($;$)
 
   # Interface with XS converter.
   if ($XS_convert and defined($root->{'tree_document_descriptor'})) {
-    my $encoded_options = encode_text_options($options);
-    my $XS_result = _convert_tree_with_XS($encoded_options, $root, $options);
+    my $selected_options = select_text_options($options);
+    my $XS_result = _convert_tree_with_XS($selected_options, $root, $options);
     if (defined ($XS_result)) {
       return $XS_result;
     } else {
@@ -999,8 +981,8 @@ sub output($$)
   my $result;
   # Interface with XS converter.
   if ($XS_convert and defined($root->{'tree_document_descriptor'})) {
-    my $encoded_options = encode_text_options($self);
-    my $XS_result = _convert_tree_with_XS($encoded_options, $root, $self);
+    my $selected_options = select_text_options($self);
+    my $XS_result = _convert_tree_with_XS($selected_options, $root, $self);
     if (defined ($XS_result)) {
       $result = $XS_result;
     } else {
diff --git a/tp/Texinfo/Translations.pm b/tp/Texinfo/Translations.pm
index 0071934e01..a916af8a8e 100644
--- a/tp/Texinfo/Translations.pm
+++ b/tp/Texinfo/Translations.pm
@@ -52,10 +52,13 @@ sub import {
     Texinfo::XSLoader::override(
       "Texinfo::Translations::_XS_configure",
       "Texinfo::TranslationsXS::configure");
-    # not loaded because it is not usable.  See comments near _XS_gettree
+    # Example of how gdt could be overriden.  Not used because
+    # the approach is flawed as there won't be any substitution if the trees in
+    # $replaced_substrings are not registered in C data, as is the case in
+    # general.
     #Texinfo::XSLoader::override(
-    #  "Texinfo::Translations::_XS_gettree",
-    #  "Texinfo::TranslationsXS::gettree");
+    #  "Texinfo::Translations::gdt",
+    #  "Texinfo::TranslationsXS::gdt");
     $module_loaded = 1;
   }
   # The usual import method
@@ -457,36 +460,6 @@ sub pgdt($$$;$$)
                                          $translation_context, $lang);
 }
 
-
-sub _XS_gettree($;$$$$)
-{
-}
-
-# Example of what a wrapper around XS gdt code could be.  Not used because
-# the approach is flawed as there won't be any substitution if the trees in
-# $replaced_substrings are not registered in C data, as is the case in general.
-# The code is fine, though.
-sub _XS_gdt($$;$$$)
-{
-  my ($customization_information, $string, $replaced_substrings,
-      $translation_context, $lang) = @_;
-
-  my $encoded_string = Encode::encode('utf-8', $string);
-  my $encoded_translation_context;
-  $encoded_translation_context = Encode::encode('utf-8', $translation_context)
-    if (defined($translation_context));
-  my $encoded_lang;
-  $encoded_lang = Encode::encode('utf-8', $lang)
-    if (defined($lang));
-
-  my $tree = _XS_gettree ($encoded_string, $customization_information,
-                          $replaced_substrings,
-                          $encoded_translation_context, $encoded_lang);
-
-  return $tree;
-}
-
-
 if (0) {
   # it is needed to mark the translation as gdt is called like
   # gdt($customization_information, '....')
diff --git a/tp/Texinfo/XS/convert/ConvertXS.xs 
b/tp/Texinfo/XS/convert/ConvertXS.xs
index 17b4f17914..246d321d25 100644
--- a/tp/Texinfo/XS/convert/ConvertXS.xs
+++ b/tp/Texinfo/XS/convert/ConvertXS.xs
@@ -254,7 +254,7 @@ html_prepare_conversion_units (SV *converter_in, ...)
          SV *seen_ids_sv;
       PPCODE:
          if (items > 1 && SvOK(ST(1)))
-           document_name = SvPVbyte_nolen (ST(1));
+           document_name = SvPVutf8_nolen (ST(1));
 
          /* add warn string? */
          self = set_output_converter_sv (converter_in, 0);
@@ -290,10 +290,10 @@ html_prepare_conversion_units (SV *converter_in, ...)
 
 void
 html_prepare_units_directions_files (SV *converter_in, SV *output_units_in, SV 
*special_units_in, SV *associated_special_units_in, output_file, 
destination_directory, output_filename, document_name)
-         char *output_file = (char *)SvPVbyte_nolen($arg);
-         char *destination_directory = (char *)SvPVbyte_nolen($arg);
-         char *output_filename = (char *)SvPVbyte_nolen($arg);
-         char *document_name = (char *)SvPVbyte_nolen($arg);
+         char *output_file = (char *)SvPVutf8_nolen($arg);
+         char *destination_directory = (char *)SvPVutf8_nolen($arg);
+         char *output_filename = (char *)SvPVutf8_nolen($arg);
+         char *document_name = (char *)SvPVutf8_nolen($arg);
   PREINIT:
          CONVERTER *self = 0;
          int output_units_descriptor = 0;
@@ -428,8 +428,8 @@ html_translate_names (SV *converter_in)
 
 void
 html_prepare_title_titlepage (SV *converter_in, SV *output_units_in, 
output_file, output_filename)
-         char *output_file = (char *)SvPVbyte_nolen($arg);
-         char *output_filename = (char *)SvPVbyte_nolen($arg);
+         char *output_file = (char *)SvPVutf8_nolen($arg);
+         char *output_filename = (char *)SvPVutf8_nolen($arg);
   PREINIT:
          CONVERTER *self = 0;
          int output_units_descriptor = 0;
@@ -526,10 +526,10 @@ html_convert_tree (SV *converter_in, SV *tree_in, 
explanation)
 
 SV *
 html_convert_output (SV *converter_in, SV *tree_in, SV *output_units_in, SV 
*special_units_in, output_file, destination_directory, output_filename, 
document_name)
-         char *output_file = (char *)SvPVbyte_nolen($arg);
-         char *destination_directory = (char *)SvPVbyte_nolen($arg);
-         char *output_filename = (char *)SvPVbyte_nolen($arg);
-         char *document_name = (char *)SvPVbyte_nolen($arg);
+         char *output_file = (char *)SvPVutf8_nolen($arg);
+         char *destination_directory = (char *)SvPVutf8_nolen($arg);
+         char *output_filename = (char *)SvPVutf8_nolen($arg);
+         char *document_name = (char *)SvPVutf8_nolen($arg);
   PREINIT:
          CONVERTER *self = 0;
          DOCUMENT *document = 0;
diff --git a/tp/Texinfo/XS/main/TranslationsXS.xs 
b/tp/Texinfo/XS/main/TranslationsXS.xs
index 286e9c2374..789c210a0b 100644
--- a/tp/Texinfo/XS/main/TranslationsXS.xs
+++ b/tp/Texinfo/XS/main/TranslationsXS.xs
@@ -48,11 +48,17 @@ configure (localesdir, 
strings_textdomain="texinfo_document")
        configure (localesdir, strings_textdomain);
 
 
+# TODO not sure that the options_in argument is good to be
+# copy_sv_options argument, may need to retrieve a converter
+# first or Parser configuration.  Does not matter much as
+# the approach does not work because replaced_substrings
+# perl element tree cannot be retrieved in C stored documents.
 # optional:
-# options, replaced_substrings, translation_context, lang
+# replaced_substrings, translation_context, lang
 SV *
-gettree (char *string, ...)
-      PROTOTYPE: $;$$$$
+gdt (SV *options_in, string, ...)
+        char *string = (char *)SvPVutf8_nolen($arg);
+      PROTOTYPE: $$;$$$
       PREINIT:
         char *translation_context = 0;
         char *in_lang = 0;
@@ -63,10 +69,14 @@ gettree (char *string, ...)
         int gdt_document_descriptor;
         DOCUMENT *gdt_document;
       CODE:
+         if (SvOK(options_in))
+           {
+             options = copy_sv_options (options_in);
+           }
         if (items > 4 && SvOK(ST(4)))
-           in_lang = (char *)SvPVbyte_nolen(ST(4));
+           in_lang = (char *)SvPVutf8_nolen(ST(4));
         if (items > 3 && SvOK(ST(3)))
-           translation_context = (char *)SvPVbyte_nolen(ST(3));
+           translation_context = (char *)SvPVutf8_nolen(ST(3));
         if (items > 2 && SvOK(ST(2)))
            {
              /* TODO put in get_perl_info.h */
@@ -92,10 +102,6 @@ gettree (char *string, ...)
                        replaced_substrings, key, document->tree);
                }
            }
-         if (items > 1 && SvOK(ST(1)))
-           {
-             options = copy_sv_options (ST(1));
-           }
 
          gdt_document_descriptor
                      = gdt (string, options, replaced_substrings,
diff --git a/tp/Texinfo/XS/main/get_perl_info.c 
b/tp/Texinfo/XS/main/get_perl_info.c
index d1ceb3d4b9..b8dfe38af5 100644
--- a/tp/Texinfo/XS/main/get_perl_info.c
+++ b/tp/Texinfo/XS/main/get_perl_info.c
@@ -287,7 +287,7 @@ copy_sv_options_for_convert_text (SV *sv_in)
   enabled_encoding_sv = hv_fetch (hv_in, "enabled_encoding",
                                   strlen ("enabled_encoding"), 0);
   if (enabled_encoding_sv)
-    text_options->encoding = strdup (SvPVbyte_nolen (*enabled_encoding_sv));
+    text_options->encoding = strdup (SvPVutf8_nolen (*enabled_encoding_sv));
 
   include_directories_sv = hv_fetch (hv_in, "INCLUDE_DIRECTORIES",
                                      strlen ("INCLUDE_DIRECTORIES"), 0);
@@ -1594,7 +1594,7 @@ set_conf (CONVERTER *converter, const char *conf, SV 
*value)
 {
   if (converter->conf)
     get_sv_option (converter->conf, conf, value);
-   /* Too early to have aoptions set
+   /* Too early to have options set
   else
     fprintf (stderr, "HHH no converter conf %s\n", conf);
     */