branch master updated: * tp/Texinfo/ParserNonXS.pm (parse_texi_file, _setup_text_root) (_parse_texi), tp/Texinfo/Structuring.pm (no_root_command_tree), tp/Texinfo/Transformations.pm: use a function to setup 'text_root' type in tp/Texinfo/ParserNonXS.pm. Prepare recognizing trees with text_root and @bye only as trees without root command in no_root_command_tree(). No functional change, aim is to reduce difference with future code changing the tree in ParserNonXS.pm.

[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 9436ea0  * tp/Texinfo/ParserNonXS.pm (parse_texi_file, 
_setup_text_root) (_parse_texi), tp/Texinfo/Structuring.pm 
(no_root_command_tree), tp/Texinfo/Transformations.pm: use a function to setup 
'text_root' type in tp/Texinfo/ParserNonXS.pm.  Prepare recognizing trees with 
text_root and @bye only as trees without root command in 
no_root_command_tree(). No functional change, aim is to reduce difference with 
future code changing the tree in ParserNonXS.pm.
9436ea0 is described below

commit 9436ea007b15a63ab0066eb0483c488c0fc4f8f0
Author: Patrice Dumas <pertusus@free.fr>
AuthorDate: Tue Sep 7 22:33:12 2021 +0200

    * tp/Texinfo/ParserNonXS.pm (parse_texi_file, _setup_text_root)
    (_parse_texi), tp/Texinfo/Structuring.pm (no_root_command_tree),
    tp/Texinfo/Transformations.pm: use a function to setup 'text_root'
    type in tp/Texinfo/ParserNonXS.pm.  Prepare recognizing trees
    with text_root and @bye only as trees without root command in
    no_root_command_tree().
    No functional change, aim is to reduce difference with future code
    changing the tree in ParserNonXS.pm.
---
 ChangeLog                     | 11 +++++++++++
 tp/Texinfo/ParserNonXS.pm     | 44 +++++++++++++++++++++++++------------------
 tp/Texinfo/Structuring.pm     | 26 +++++++++++++++++++------
 tp/Texinfo/Transformations.pm |  9 +++------
 tp/t/accents.t                | 15 +++++++++++++--
 5 files changed, 73 insertions(+), 32 deletions(-)

diff --git a/ChangeLog b/ChangeLog
index 09b65ba..0d8039e 100644
--- a/ChangeLog
+++ b/ChangeLog
@@ -1,5 +1,16 @@
 2021-09-07  Patrice Dumas  <pertusus@free.fr>
 
+       * tp/Texinfo/ParserNonXS.pm (parse_texi_file, _setup_text_root)
+       (_parse_texi), tp/Texinfo/Structuring.pm (no_root_command_tree),
+       tp/Texinfo/Transformations.pm: use a function to setup 'text_root'
+       type in tp/Texinfo/ParserNonXS.pm.  Prepare recognizing trees
+       with text_root and @bye only as trees without root command in
+       no_root_command_tree().
+       No functional change, aim is to reduce difference with future code
+       changing the tree in ParserNonXS.pm.
+
+2021-09-07  Patrice Dumas  <pertusus@free.fr>
+
        * tp/Texinfo/Convert/Converter.pm (determine_files_and_directory)
        (_set_element_file, _top_node_filename, _set_pages_files)
        (create_destination_directory), tp/DebugTexinfo/DebugTree.pm,
diff --git a/tp/Texinfo/ParserNonXS.pm b/tp/Texinfo/ParserNonXS.pm
index 6fb387e..a72025a 100644
--- a/tp/Texinfo/ParserNonXS.pm
+++ b/tp/Texinfo/ParserNonXS.pm
@@ -933,15 +933,15 @@ sub parse_texi_file($$)
       last;
     }
   }
-  my $root = { 'contents' => [], 'type' => 'text_root' };
+  my ($document_root, $text_root) = _setup_text_root();
   if (@first_lines) {
-    push @{$root->{'contents'}}, { 'type' => 'preamble', 'contents' => [],
-                                   'parent' => $root };
+    push @{$text_root->{'contents'}}, { 'type' => 'preamble', 'contents' => [],
+                                   'parent' => $text_root };
     foreach my $line (@first_lines) {
-      push @{$root->{'contents'}->[-1]->{'contents'}}, 
+      push @{$text_root->{'contents'}->[-1]->{'contents'}},
                                    { 'text' => $line,
                                      'type' => 'preamble_text',
-                                     'parent' => $root->{'contents'}->[-1]
+                                     'parent' => $text_root->{'contents'}->[-1]
                                    };
     }
   }
@@ -958,18 +958,18 @@ sub parse_texi_file($$)
   $self->{'info'}->{'input_file_name'} = $file_name;
   $self->{'info'}->{'input_directory'} = $directories;
   
-  my $tree = $self->_parse_texi($root);
-
-  # Find 'text_root', which contains everything before first node/section.
-  # if there are elements, 'text_root' is the first content, otherwise it
-  # is the root.
-  my $text_root;
-  if ($tree->{'type'} eq 'text_root') {
-    $text_root = $tree;
-  } elsif ($tree->{'contents'} and $tree->{'contents'}->[0]
-           and $tree->{'contents'}->[0]->{'type'} eq 'text_root') {
-    $text_root = $tree->{'contents'}->[0];
-  }
+  my $tree = $self->_parse_texi($document_root);
+
+  ## Find 'text_root', which contains everything before first node/section.
+  ## if there are elements, 'text_root' is the first content, otherwise it
+  ## is the root.
+  #my $text_root;
+  #if ($tree->{'type'} eq 'text_root') {
+  #  $text_root = $tree;
+  #} elsif ($tree->{'contents'} and $tree->{'contents'}->[0]
+  #         and $tree->{'contents'}->[0]->{'type'} eq 'text_root') {
+  #  $text_root = $tree->{'contents'}->[0];
+  #}
 
   # Put everything before @setfilename in a special type.  This allows to
   # ignore everything before @setfilename.
@@ -3618,13 +3618,20 @@ sub _check_valid_nesting {
   }
 }
 
+sub _setup_text_root()
+{
+  my $text_root = { 'contents' => [], 'type' => 'text_root' };
+  my $document_root = $text_root;
+  return ($document_root, $text_root);
+}
 
 # the main subroutine
 sub _parse_texi($;$)
 {
   my ($self, $root) = @_;
 
-  $root = { 'contents' => [], 'type' => 'text_root' } if (!defined($root));
+  ($root) = _setup_text_root() if (!defined($root));
+
   my $current = $root;
 
   my $line_nr;
@@ -5444,6 +5451,7 @@ sub _parse_texi($;$)
         }
       # Misc text except end of line
       } elsif (defined $misc_text) {
+        print STDERR "MISC TEXT: $misc_text\n" if ($self->{'DEBUG'});
         my $new_text = $misc_text;
         substr ($line, 0, length ($misc_text)) = '';
         $current = _merge_text($self, $current, $new_text);
diff --git a/tp/Texinfo/Structuring.pm b/tp/Texinfo/Structuring.pm
index dfce63c..f24beef 100644
--- a/tp/Texinfo/Structuring.pm
+++ b/tp/Texinfo/Structuring.pm
@@ -114,6 +114,23 @@ sub section_level($)
 }
 
 
+sub no_root_command_tree($)
+{
+  my $root = shift;
+  if (!$root->{'type'} or $root->{'type'} ne 'document_root'
+      or !$root->{'contents'}
+      or scalar(@{$root->{'contents'}}) == 0
+      or ($root->{'contents'}->[0]->{'type'}
+          and $root->{'contents'}->[0]->{'type'} eq 'text_root'
+          and (scalar(@{$root->{'contents'}}) == 1
+              or (scalar(@{$root->{'contents'}}) == 2
+                  and $root->{'contents'}->[1]->{'cmdname'}
+                  and $root->{'contents'}->[1]->{'cmdname'} eq 'bye')))) {
+    return 1;
+  }
+  return 0;
+}
+
 # Go through the sectioning commands (e.g. @chapter, not @node), and
 # set:
 # 'number'
@@ -130,8 +147,7 @@ sub sectioning_structure($$$)
   my $configuration_informations = shift;
   my $root = shift;
 
-  if (!$root->{'type'} or $root->{'type'} ne 'document_root'
-      or !$root->{'contents'}) {
+  if (no_root_command_tree($root)) {
     return undef;
   }
 
@@ -819,8 +835,7 @@ sub nodes_tree($$$$$)
 sub split_by_node($)
 {
   my $root = shift;
-  if (!$root->{'type'} or $root->{'type'} ne 'document_root'
-      or !$root->{'contents'} or !@{$root->{'contents'}}) {
+  if (no_root_command_tree($root)) {
     return undef;
   }
   my $elements;
@@ -868,8 +883,7 @@ sub split_by_node($)
 sub split_by_section($)
 {
   my $root = shift;
-  if (!$root->{'type'} or $root->{'type'} ne 'document_root'
-      or !$root->{'contents'} or !@{$root->{'contents'}}) {
+  if (no_root_command_tree($root)) {
     return undef;
   }
   my $elements;
diff --git a/tp/Texinfo/Transformations.pm b/tp/Texinfo/Transformations.pm
index 72b8744..a25a6ea 100644
--- a/tp/Texinfo/Transformations.pm
+++ b/tp/Texinfo/Transformations.pm
@@ -73,8 +73,7 @@ sub _correct_level($$;$)
 sub fill_gaps_in_sectioning($)
 {
   my $root = shift;
-  if (!$root->{'type'} or $root->{'type'} ne 'document_root'
-      or !$root->{'contents'}) {
+  if (Texinfo::Structuring::no_root_command_tree($root)) {
     return undef;
   }
   my @sections_list;
@@ -323,8 +322,7 @@ sub insert_nodes_for_sectioning_commands($$$$)
   my $targets_list = shift;
   my $labels = shift;
 
-  if (!$root->{'type'} or $root->{'type'} ne 'document_root'
-      or !$root->{'contents'}) {
+  if (Texinfo::Structuring::no_root_command_tree($root)) {
     return (undef, undef);
   }
   my @added_nodes;
@@ -457,8 +455,7 @@ sub _get_non_automatic_nodes_with_sections($)
 {
   my $root = shift;
 
-  if (!$root->{'type'} or $root->{'type'} ne 'document_root'
-      or !$root->{'contents'}) {
+  if (Texinfo::Structuring::no_root_command_tree($root)) {
     return undef;
   }
   my @non_automatic_nodes;
diff --git a/tp/t/accents.t b/tp/t/accents.t
index 4fda0c4..d43e9a3 100644
--- a/tp/t/accents.t
+++ b/tp/t/accents.t
@@ -18,6 +18,17 @@ use Texinfo::Convert::HTML;
 
 ok(1, "modules loading");
 
+sub _find_accent($)
+{
+  my $text_root = shift;
+  my $current = $text_root;
+  while ($current->{'type'} and ($current->{'type'} eq 'text_root'
+                                 or $current->{'type'} eq 'document_root'
+                                 or $current->{'type'} eq 'paragraph')) {
+    $current = $current->{'contents'}->[0];
+  }
+  return $current;
+}
 
 sub test_accent_stack ($)
 {
@@ -27,7 +38,7 @@ sub test_accent_stack ($)
   my $reference = $test->[2]; 
   my $parser = Texinfo::Parser::parser();
   my $text_root = $parser->parse_texi_text($texi);
-  my $tree = $text_root->{'contents'}->[0]->{'contents'}->[0];
+  my $tree = _find_accent($text_root);
   my ($contents, $commands_stack) = 
     Texinfo::Convert::Utils::find_innermost_accent_contents($tree);
   my $text = Texinfo::Convert::Text::convert_to_text({'contents' => 
$contents});
@@ -76,7 +87,7 @@ sub test_enable_encoding ($)
   my $reference_unicode = $test->[5];
   my $parser = Texinfo::Parser::parser();
   my $text_root = $parser->parse_texi_text($texi);
-  my $tree = $text_root->{'contents'}->[0]->{'contents'}->[0];
+  my $tree = _find_accent($text_root);
 
   my ($contents, $commands_stack) = 
     Texinfo::Convert::Utils::find_innermost_accent_contents($tree);