Make \regex_show:n show printable characters (fixes #590)

Previously only the charcode was shown, which made it harder to debug things.
latex3 · Apr 27, 2021 · ab92696 · ab92696
1 parent aecd3a5
commit ab92696
Show file tree

Hide file tree

Showing 4 changed files with 126 additions and 112 deletions.
diff --git a/l3kernel/CHANGELOG.md b/l3kernel/CHANGELOG.md
@@ -20,6 +20,7 @@ this project uses date-based 'snapshot' version identifiers.
 - `\pdf_version_gset:n` for `dvips`.
 - Improve handling of `\exp_not:n` in `\text_expand:n` (issue #875)
 - `\file_full_name:n` now avoids calling `\pdffilesize` multiple times on the same file.
+- Show printable characters explicitly in `\regex_show:n`
 
 ### Fixed
 - Evalutate integer constants only once (issue#861)

diff --git a/l3kernel/l3regex.dtx b/l3kernel/l3regex.dtx
@@ -509,7 +509,7 @@
 %   which never change.
 % \end{function}
 %
-% \begin{function}[added = 2017-05-26]{\regex_show:n, \regex_show:N}
+% \begin{function}[added = 2017-05-26, updated = 2021-04-15]{\regex_show:n, \regex_show:N}
 %   \begin{syntax}
 %     \cs{regex_show:n} \Arg{regex}
 %   \end{syntax}
@@ -3554,18 +3554,18 @@
       \cs_set:Npn \@@_A_test: { anchor~at~start~(\iow_char:N\\A) }
       \cs_set:Npn \@@_G_test: { anchor~at~start~of~match~(\iow_char:N\\G) }
       \cs_set_protected:Npn \@@_item_caseful_equal:n ##1
-        { \@@_show_one:n { char~code~\int_eval:n{##1} } }
+        { \@@_show_one:n { char~code~\@@_show_char:n{##1} } }
       \cs_set_protected:Npn \@@_item_caseful_range:nn ##1##2
         {
           \@@_show_one:n
-            { range~[\int_eval:n{##1}, \int_eval:n{##2}] }
+            { range~[\@@_show_char:n{##1}, \@@_show_char:n{##2}] }
         }
       \cs_set_protected:Npn \@@_item_caseless_equal:n ##1
-        { \@@_show_one:n { char~code~\int_eval:n{##1}~(caseless) } }
+        { \@@_show_one:n { char~code~\@@_show_char:n{##1}~(caseless) } }
       \cs_set_protected:Npn \@@_item_caseless_range:nn ##1##2
         {
           \@@_show_one:n
-            { Range~[\int_eval:n{##1}, \int_eval:n{##2}]~(caseless) }
+            { Range~[\@@_show_char:n{##1}, \@@_show_char:n{##2}]~(caseless) }
         }
       \cs_set_protected:Npn \@@_item_catcode:nT
         { \@@_show_item_catcode:NnT \c_true_bool }
@@ -3574,7 +3574,7 @@
       \cs_set_protected:Npn \@@_item_reverse:n
         { \@@_show_scope:nn { Reversed~match } }
       \cs_set_protected:Npn \@@_item_exact:nn ##1##2
-        { \@@_show_one:n { char~##2,~catcode~##1 } }
+        { \@@_show_one:n { char~\@@_show_char:n{##2},~catcode~##1 } }
       \cs_set_eq:NN \@@_item_exact_cs:n \@@_show_item_exact_cs:n
       \cs_set_protected:Npn \@@_item_cs:n
         { \@@_show_scope:nn { control~sequence } }
@@ -3590,6 +3590,19 @@
 %    \end{macrocode}
 % \end{macro}
 %
+% \begin{macro}[EXP]{\@@_show_char:n}
+%   Show a single character, together with its ascii representation if available.
+%   This could be extended to beyond ascii.  It is not ideal for parentheses themselves.
+%    \begin{macrocode}
+\cs_new:Npn \@@_show_char:n #1
+  {
+    \int_eval:n {#1}
+    \int_compare:nT { 32 <= #1 <= 126 }
+      { ~ ( \char_generate:nn {#1} {12} ) }
+  }
+%    \end{macrocode}
+% \end{macro}
+%
 % \begin{macro}{\@@_show_one:n}
 %   Every part of the final message go through this function, which adds
 %   one line to the output, with the appropriate prefix.