branch master updated: * tp/Texinfo/Convert/Unicode.pm (%unicode_to_eight_bit) (_eight_bit_and_unicode_point, encoded_accents) (unicode_point_decoded_in_encoding): use directly the input encodings names. Separate mappings of koi8-u and koi8-r.

[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 d57ee64a19 * tp/Texinfo/Convert/Unicode.pm (%unicode_to_eight_bit) 
(_eight_bit_and_unicode_point, encoded_accents) 
(unicode_point_decoded_in_encoding): use directly the input encodings names.  
Separate mappings of koi8-u and koi8-r.
d57ee64a19 is described below

commit d57ee64a194e3043dd35e16616891f653c7f2785
Author: Patrice Dumas <pertusus@free.fr>
AuthorDate: Fri Feb 18 22:26:00 2022 +0100

    * tp/Texinfo/Convert/Unicode.pm (%unicode_to_eight_bit)
    (_eight_bit_and_unicode_point, encoded_accents)
    (unicode_point_decoded_in_encoding): use directly the input encodings
    names.  Separate mappings of koi8-u and koi8-r.
    
    * tp/Texinfo/Encoding.pm (encoding_alias): remove
    %eight_bit_encoding_aliases.  Use my, not our for variables not used
    elsewhere.
    
    * util/txixml2texi.pl: use directly Encode::resolve_alias() instead of
    Texinfo::Encoding::encoding_alias().
    
    * tp/Texinfo/XS/parsetexi/end_line.c (end_line_misc_line): add latin-1
    mapping.  Remove impossible or incorrect mappings from encoding_map.
    
    * doc/tp_api/Makefile.am: do not document Texinfo::Encoding.
---
 ChangeLog                          | 19 +++++++++++++++++
 doc/tp_api/Makefile.am             |  1 -
 tp/Texinfo/Convert/Unicode.pm      | 42 ++++++++++++++++++++++----------------
 tp/Texinfo/Encoding.pm             | 32 ++++++++---------------------
 tp/Texinfo/XS/parsetexi/end_line.c | 15 +++++++++-----
 util/txixml2texi.pl                |  5 +++--
 6 files changed, 65 insertions(+), 49 deletions(-)

diff --git a/ChangeLog b/ChangeLog
index ffd684842f..b10e737064 100644
--- a/ChangeLog
+++ b/ChangeLog
@@ -1,3 +1,22 @@
+2022-02-18  Patrice Dumas  <pertusus@free.fr>
+
+       * tp/Texinfo/Convert/Unicode.pm (%unicode_to_eight_bit)
+       (_eight_bit_and_unicode_point, encoded_accents)
+       (unicode_point_decoded_in_encoding): use directly the input encodings
+       names.  Separate mappings of koi8-u and koi8-r.
+
+       * tp/Texinfo/Encoding.pm (encoding_alias): remove
+       %eight_bit_encoding_aliases.  Use my, not our for variables not used
+       elsewhere.
+
+       * util/txixml2texi.pl: use directly Encode::resolve_alias() instead of
+       Texinfo::Encoding::encoding_alias().
+
+       * tp/Texinfo/XS/parsetexi/end_line.c (end_line_misc_line): add latin-1
+       mapping.  Remove impossible or incorrect mappings from encoding_map.
+
+       * doc/tp_api/Makefile.am: do not document Texinfo::Encoding.
+
 2022-02-18  Patrice Dumas  <pertusus@free.fr>
 
        * Pod-Simple-Texinfo/pod2texi.pl: do not ignore an input file that
diff --git a/doc/tp_api/Makefile.am b/doc/tp_api/Makefile.am
index ca4fd0e2fd..46e0e87795 100644
--- a/doc/tp_api/Makefile.am
+++ b/doc/tp_api/Makefile.am
@@ -56,7 +56,6 @@ BUILT_SOURCES = tp_api.texi
 # The order matters for the result.
 tp_api_dependencies = \
   $(top_srcdir)/tp/Texinfo/Common.pm \
-  $(top_srcdir)/tp/Texinfo/Encoding.pm \
   $(top_srcdir)/tp/Texinfo/ParserNonXS.pm \
   $(top_srcdir)/tp/Texinfo/Structuring.pm \
   $(top_srcdir)/tp/Texinfo/Report.pm \
diff --git a/tp/Texinfo/Convert/Unicode.pm b/tp/Texinfo/Convert/Unicode.pm
index d43b3c7671..7b1b7b3d91 100644
--- a/tp/Texinfo/Convert/Unicode.pm
+++ b/tp/Texinfo/Convert/Unicode.pm
@@ -652,7 +652,7 @@ foreach my $command (keys(%unicode_accented_letters)) {
 }
 
 my %unicode_to_eight_bit = (
-   'iso8859_1' => {
+   'iso-8859-1' => {
       '00A0' => 'A0',
       '00A1' => 'A1',
       '00A2' => 'A2',
@@ -751,7 +751,7 @@ my %unicode_to_eight_bit = (
       '00FE' => 'FE',
       '00FF' => 'FF',
    },
-   'iso8859_15' => {
+   'iso-8859-15' => {
       '00A0' => 'A0',
       '00A1' => 'A1',
       '00A2' => 'A2',
@@ -849,7 +849,7 @@ my %unicode_to_eight_bit = (
       '00FE' => 'FE',
       '00FF' => 'FF',
    },
-   'iso8859_2' => {
+   'iso-8859-2' => {
       '00A0' => 'A0',
       '0104' => 'A1',
       '02D8' => 'A2',
@@ -946,15 +946,9 @@ my %unicode_to_eight_bit = (
       '0163' => 'FE',
       '02D9' => 'FF',
    },
-   'koi8' => {
+   'koi8-r' => {
       '0415' => 'A3',
-      '0454' => 'A4',
-      '0456' => 'A6',
-      '0457' => 'A7',
       '04D7' => 'B3',
-      '0404' => 'B4',
-      '0406' => 'B6',
-      '0407' => 'B7',
       '042E' => 'C0',
       '0430' => 'C1',
       '0431' => 'C2',
@@ -1020,8 +1014,23 @@ my %unicode_to_eight_bit = (
       '0427' => 'FE',
       '042A' => 'FF',
    },
+   'koi8-u' => {
+      '0454' => 'A4',
+      '0404' => 'B4',
+      '0456' => 'A6',
+      '0406' => 'B6',
+      '0457' => 'A7',
+      '0407' => 'B7',
+      '0491' => 'AD',
+      '0490' => 'BD',
+   }
 );
 
+foreach my $unicode_point (keys(%{$unicode_to_eight_bit{'koi8-r'}})) {
+  $unicode_to_eight_bit{'koi8-u'}->{$unicode_point}
+    = $unicode_to_eight_bit{'koi8-r'}->{$unicode_point};
+}
+
 # currently unused
 my %makeinfo_transliterate_map = (
   '0416' => 'ZH',
@@ -1277,8 +1286,6 @@ sub _eight_bit_and_unicode_point($$)
   my $char = shift;
   my $encoding = shift;
 
-  my $encoding_map_name
-   = $Texinfo::Encoding::eight_bit_encoding_aliases{$encoding};
   my ($eight_bit, $codepoint);
   if (ord($char) <= 128) {
     # 7bit ascii characters, the same in every 8bit encodings
@@ -1286,9 +1293,8 @@ sub _eight_bit_and_unicode_point($$)
     $codepoint = uc(sprintf("%04x",ord($char)));
   } elsif (ord($char) <= hex(0xFFFF)) {
     $codepoint = uc(sprintf("%04x",ord($char)));
-    if (exists($unicode_to_eight_bit{$encoding_map_name}->{$codepoint})) {
-     $eight_bit
-         = $unicode_to_eight_bit{$encoding_map_name}->{$codepoint};
+    if (exists($unicode_to_eight_bit{$encoding}->{$codepoint})) {
+     $eight_bit = $unicode_to_eight_bit{$encoding}->{$codepoint};
     }
   }
   return ($eight_bit, $codepoint);
@@ -1458,7 +1464,7 @@ sub encoded_accents($$$$$;$)
     if ($encoding eq 'utf-8') {
       return _format_unicode_accents_stack($converter, $text, $stack,
                                             $format_accent, $set_case);
-    } elsif ($Texinfo::Encoding::eight_bit_encoding_aliases{$encoding}) {
+    } elsif ($unicode_to_eight_bit{$encoding}) {
       return _format_eight_bit_accents_stack($converter, $text, $stack, 
$encoding,
                                $format_accent, $set_case);
     }
@@ -1474,8 +1480,8 @@ sub unicode_point_decoded_in_encoding($$) {
 
   return 1 if ($encoding
                and ($encoding eq 'utf-8'
-                    or 
($Texinfo::Encoding::eight_bit_encoding_aliases{$encoding}
-                        and 
$unicode_to_eight_bit{$Texinfo::Encoding::eight_bit_encoding_aliases{$encoding}}->{$unicode_point})));
+                    or ($unicode_to_eight_bit{$encoding}
+                        and 
$unicode_to_eight_bit{$encoding}->{$unicode_point})));
   return 0;
 }
 
diff --git a/tp/Texinfo/Encoding.pm b/tp/Texinfo/Encoding.pm
index 67e33d410f..f70e4da229 100644
--- a/tp/Texinfo/Encoding.pm
+++ b/tp/Texinfo/Encoding.pm
@@ -33,7 +33,7 @@ use vars qw(@ISA @EXPORT_OK);
 
 # charset related definitions.
 
-our %perl_charset_to_html = (
+my %perl_charset_to_html = (
               'utf8'       => 'utf-8',
               'utf-8-strict'       => 'utf-8',
               'ascii'      => 'us-ascii',
@@ -41,47 +41,32 @@ our %perl_charset_to_html = (
 );
 
 # encoding name normalization to html-compatible encoding names
-our %encoding_aliases = (
-              'latin1' => 'iso-8859-1',
-);
+my %encoding_aliases;
 
 foreach my $perl_charset (keys(%perl_charset_to_html)) {
    $encoding_aliases{$perl_charset} = $perl_charset_to_html{$perl_charset};
    $encoding_aliases{$perl_charset_to_html{$perl_charset}}
         = $perl_charset_to_html{$perl_charset};
 }
-our %eight_bit_encoding_aliases = (
-  "iso-8859-1",  'iso8859_1',
-  "iso-8859-2",  'iso8859_2',
-  "iso-8859-15", 'iso8859_15',
-  "koi8-r",      'koi8',
-  "koi8-u",      'koi8',
-);
-
-foreach my $encoding (keys(%eight_bit_encoding_aliases)) {
-  $encoding_aliases{$encoding} = $encoding;
-  $encoding_aliases{$eight_bit_encoding_aliases{$encoding}} = $encoding;
-}
 
-our %canonical_texinfo_encodings;
+my %canonical_texinfo_encodings;
 # These are the encodings from the texinfo manual
-foreach my $canonical_encoding('us-ascii', 'utf-8', 'iso-8859-1',
-  'iso-8859-15','iso-8859-2','koi8-r', 'koi8-u') {
+foreach my $canonical_encoding ('us-ascii', 'utf-8', 'iso-8859-1',
+                  'iso-8859-15', 'iso-8859-2', 'koi8-r', 'koi8-u') {
   $canonical_texinfo_encodings{$canonical_encoding} = 1;
+  $encoding_aliases{$canonical_encoding} = $canonical_encoding;
 }
 
 sub encoding_alias($)
 {
   my $encoding = shift;
-  my $canonical_texinfo_encoding;
-  $canonical_texinfo_encoding
-    = $encoding if ($canonical_texinfo_encodings{lc($encoding)});
   my $perl_encoding = Encode::resolve_alias($encoding);
   my $canonical_output_encoding;
   if ($perl_encoding) {
     $canonical_output_encoding = $encoding_aliases{$perl_encoding};
   }
-  foreach my $possible_encoding($encoding, $canonical_output_encoding,
+  my $canonical_texinfo_encoding;
+  foreach my $possible_encoding ($encoding, $canonical_output_encoding,
                                             $perl_encoding) {
     if (defined($possible_encoding)
         and $canonical_texinfo_encodings{lc($possible_encoding)}) {
@@ -120,6 +105,7 @@ Texinfo::Encoding takes care of encoding definition and 
aliasing.
 =over
 
 =item ($canonical_texinfo_encoding, $perl_encoding, 
$canonical_output_encoding) = encoding_alias($encoding)
+X<C<encoding_alias>>
 
 Taking an encoding name as argument, the function returns the
 corresponding canonical Texinfo encoding I<$canonical_texinfo_encoding>
diff --git a/tp/Texinfo/XS/parsetexi/end_line.c 
b/tp/Texinfo/XS/parsetexi/end_line.c
index b67c147247..625d51dc99 100644
--- a/tp/Texinfo/XS/parsetexi/end_line.c
+++ b/tp/Texinfo/XS/parsetexi/end_line.c
@@ -1503,6 +1503,8 @@ end_line_misc_line (ELEMENT *current)
                   static char *known_encodings[] = {
                       "shift_jis",
                       "latin1",
+                      "latin-1",
+                      "utf8",
                       0
                   };
                   for (i = 0; (known_encodings[i]); i++)
@@ -1521,11 +1523,15 @@ end_line_misc_line (ELEMENT *current)
                   struct encoding_map {
                       char *from; char *to;
                   };
+                  /* The map mimics Encode::resolve_alias() result.  Even when
+                     the alias is not good, such as 'utf-8-strict' for 'utf-8'
+                     use the same mapping for consistency with the perl Parser 
*/
                   static struct encoding_map map[] = {
                       "utf-8", "utf-8-strict",
                       "us-ascii", "ascii",
                       "shift_jis", "shiftjis",
-                      "latin1", "iso-8859-1"
+                      "latin1", "iso-8859-1",
+                      "latin-1", "iso-8859-1"
                   };
                   for (i = 0; i < sizeof map / sizeof *map; i++)
                     {
@@ -1544,7 +1550,7 @@ end_line_misc_line (ELEMENT *current)
                 {
                   command_warn (current, "unrecognized encoding name `%s'",
                                 text);
-                  /* Texinfo::Encoding calls Encode::Alias, so knows
+                  /* Texinfo::Encoding calls Encode::resolve_alias, so knows
                      about more encodings than what we know about here.
                      TODO: Check when perl_encoding could be defined when 
                      texinfo_encoding isn't.
@@ -1565,12 +1571,11 @@ end_line_misc_line (ELEMENT *current)
                       "utf-8-strict","utf-8",
                       "ascii",       "us-ascii",
                       "shiftjis",    "shift_jis",
-                      "latin-1",     "iso-8859-1",
                       "iso-8859-1",  "iso-8859-1",
                       "iso-8859-2",  "iso-8859-2",
                       "iso-8859-15", "iso-8859-15",
-                      "koi8-r",      "koi8",
-                      "koi8-u",      "koi8",
+                      "koi8-r",      "koi8-r",
+                      "koi8-u",      "koi8-u",
                   };
                   input_encoding = perl_encoding;
                   for (i = 0; i < sizeof map / sizeof *map; i++)
diff --git a/util/txixml2texi.pl b/util/txixml2texi.pl
index ab585c4823..074a247e78 100755
--- a/util/txixml2texi.pl
+++ b/util/txixml2texi.pl
@@ -24,6 +24,7 @@ use Getopt::Long qw(GetOptions);
 # for dirname.
 use File::Basename;
 use File::Spec;
+use Encode;
 
 Getopt::Long::Configure("gnu_getopt");
 
@@ -294,8 +295,8 @@ while ($reader->read) {
       }
       if ($name eq 'documentencoding' and $reader->hasAttributes() 
           and defined($reader->getAttribute('encoding'))) {
-        my ($texinfo_encoding, $perl_encoding, $output_encoding)
-         = 
Texinfo::Encoding::encoding_alias($reader->getAttribute('encoding'));
+        my $perl_encoding
+          = Encode::resolve_alias($reader->getAttribute('encoding'));
 
         if (defined($perl_encoding)) {
           if ($debug) {