branch master updated: * tp/Texinfo/Convert/HTML.pm (url_protect_url_text) (url_protect_file_text), tp/Texinfo/XS/convert/convert_html.c (url_protect_url_text, url_protect_file_text): replace end of lines by spaces before percent encoding.

[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 d0eab9ad38 * tp/Texinfo/Convert/HTML.pm (url_protect_url_text) 
(url_protect_file_text), tp/Texinfo/XS/convert/convert_html.c 
(url_protect_url_text, url_protect_file_text): replace end of lines by spaces 
before percent encoding.
d0eab9ad38 is described below

commit d0eab9ad38d7bd7f9e34b7f9eb5580c9728ee3a4
Author: Patrice Dumas <pertusus@free.fr>
AuthorDate: Sun Feb 18 16:39:18 2024 +0100

    * tp/Texinfo/Convert/HTML.pm (url_protect_url_text)
    (url_protect_file_text), tp/Texinfo/XS/convert/convert_html.c
    (url_protect_url_text, url_protect_file_text): replace end of lines by
    spaces before percent encoding.
---
 ChangeLog                                      |  7 ++++
 tp/Texinfo/Convert/HTML.pm                     |  8 ++--
 tp/Texinfo/XS/convert/convert_html.c           | 54 +++++++++++++++++---------
 tp/t/results/html_tests/end_of_line_in_uref.pl |  2 +-
 4 files changed, 49 insertions(+), 22 deletions(-)

diff --git a/ChangeLog b/ChangeLog
index 59d3d8a06e..54642d4b74 100644
--- a/ChangeLog
+++ b/ChangeLog
@@ -1,3 +1,10 @@
+2024-02-18  Patrice Dumas  <pertusus@free.fr>
+
+       * tp/Texinfo/Convert/HTML.pm (url_protect_url_text)
+       (url_protect_file_text), tp/Texinfo/XS/convert/convert_html.c
+       (url_protect_url_text, url_protect_file_text): replace end of lines by
+       spaces before percent encoding.
+
 2024-02-18  Patrice Dumas  <pertusus@free.fr>
 
        * tp/Texinfo/Convert/HTML.pm (_convert_heading_command),
diff --git a/tp/Texinfo/Convert/HTML.pm b/tp/Texinfo/Convert/HTML.pm
index f533f70375..89010b1163 100644
--- a/tp/Texinfo/Convert/HTML.pm
+++ b/tp/Texinfo/Convert/HTML.pm
@@ -9066,13 +9066,13 @@ sub convert_tree($$;$)
 
 # Protect an url, in which characters with specific meaning in url are
 # considered to have their specific meaning.
-# TODO turn end of lines to spaces?  Currently, an end of line is percent
-# protected, it is most likely not what the author intended.  Tested in
-# html_tests.t end_of_line_in_uref
 sub url_protect_url_text($$)
 {
   my $self = shift;
   my $input_string = shift;
+  # turn end of lines to spaces, as it is most likely what is expected
+  # rather than a percent encoded end of line.
+  $input_string =~ s/[\n\r]+/ /g;
   # percent encode character string.  It is better use UTF-8 irrespective
   # of the actual charset of the HTML output file, according to the tests done.
   my $href = encode("UTF-8", $input_string);
@@ -9088,6 +9088,8 @@ sub url_protect_file_text($$)
 {
   my $self = shift;
   my $input_string = shift;
+  # turn end of lines to spaces, as it is most likely what is expected.
+  $input_string =~ s/[\n\r]+/ /g;
   # percent encode character string.  It is better use UTF-8 irrespective
   # of the actual charset of the HTML output file, according to the tests done.
   my $href = encode("UTF-8", $input_string);
diff --git a/tp/Texinfo/XS/convert/convert_html.c 
b/tp/Texinfo/XS/convert/convert_html.c
index 4fe3e7656f..dab4658b83 100644
--- a/tp/Texinfo/XS/convert/convert_html.c
+++ b/tp/Texinfo/XS/convert/convert_html.c
@@ -2583,21 +2583,30 @@ url_protect_url_text (CONVERTER *self, const char 
*input_string)
             }
           else
             {
-              int i;
-              int char_len = 1;
-              if (!isascii (*p))
+              int n = strspn (p, "\r\n");
+              if (n)
                 {
-                  /* Protect UTF-8 with continuation bytes. */
-                  while ((p[char_len] & 0xC0) == 0x80)
-                    char_len++;
+                  text_printf (&text, "%%%02x", (unsigned char)' ');
+                  p += n;
                 }
-              for (i = 0; i < char_len; i++)
+              else
                 {
+                  int i;
+                  int char_len = 1;
+                  if (!isascii (*p))
+                    {
+                      /* Protect UTF-8 with continuation bytes. */
+                      while ((p[char_len] & 0xC0) == 0x80)
+                        char_len++;
+                    }
+                  for (i = 0; i < char_len; i++)
+                    {
             /* the reason for forcing (unsigned char) is that the %x modifier
                expects an unsigned int parameter and a char will usually be
                promoted to an int when passed to a varargs function */
-                  text_printf (&text, "%%%02x", (unsigned char)*p);
-                  p += 1;
+                      text_printf (&text, "%%%02x", (unsigned char)*p);
+                      p += 1;
+                    }
                 }
             }
         }
@@ -2642,21 +2651,30 @@ url_protect_file_text (CONVERTER *self, const char 
*input_string)
             }
           else
             {
-              int i;
-              int char_len = 1;
-              if (!isascii (*p))
+              int n = strspn (p, "\r\n");
+              if (n)
                 {
-                  /* Protect UTF-8 with continuation bytes. */
-                  while ((p[char_len] & 0xC0) == 0x80)
-                    char_len++;
+                  text_printf (&text, "%%%02x", (unsigned char)' ');
+                  p += n;
                 }
-              for (i = 0; i < char_len; i++)
+              else
                 {
+                  int i;
+                  int char_len = 1;
+                  if (!isascii (*p))
+                    {
+                      /* Protect UTF-8 with continuation bytes. */
+                      while ((p[char_len] & 0xC0) == 0x80)
+                        char_len++;
+                    }
+                  for (i = 0; i < char_len; i++)
+                    {
             /* the reason for forcing (unsigned char) is that the %x modifier
                expects an unsigned int parameter and a char will usually be
                promoted to an int when passed to a varargs function */
-                  text_printf (&text, "%%%02x", (unsigned char)*p);
-                  p += 1;
+                      text_printf (&text, "%%%02x", (unsigned char)*p);
+                      p += 1;
+                    }
                 }
             }
         }
diff --git a/tp/t/results/html_tests/end_of_line_in_uref.pl 
b/tp/t/results/html_tests/end_of_line_in_uref.pl
index 5ddc9f74b0..97bbf79f53 100644
--- a/tp/t/results/html_tests/end_of_line_in_uref.pl
+++ b/tp/t/results/html_tests/end_of_line_in_uref.pl
@@ -92,7 +92,7 @@ $result_converted{'html'}->{'end_of_line_in_uref'} = 
'<!DOCTYPE html>
 </head>
 
 <body lang="en">
-<p>See the <a class="uref" 
href="https://gcc.gnu.org/codingconventions.html#Spelling%0aSpelling";>terminology
 and markup</a> section.</p>
+<p>See the <a class="uref" 
href="https://gcc.gnu.org/codingconventions.html#Spelling%20Spelling";>terminology
 and markup</a> section.</p>
 
 
 </body>