New command-line options for 'etags'

This adds '--no-fallback-lang' and '--no-empty-file-entries'
options, and their opposites '--fallback-lang' and
'--empty-file-entries'.
* lib-src/etags.c (fallback_lang, empty_files): New toggles.
(main): Initialize them to 'true'.
(longopts) [!CTAGS]: Add the '--(no-)fallback-lang' and
'--(no-)empty-file-entries' options.
(find_entries): If 'fallback_lang' is false, don't attempt Fortran
and C/C++ fallbacks.
(print_help): Add help for new options.
(main): If 'empty_files' is false, don't output file entries for
files that have no tags.  (Bug#73484)

* doc/emacs/maintaining.texi (Create Tags Table):
* etc/NEWS:
* doc/man/etags.1: Document the new options.

* test/manual/etags/Makefile (check): Add test for new options.
* test/manual/etags/ETAGS.good_7: New file.

doc/emacs/maintaining.texi

@@ -3048,6 +3048,10 @@ language not yet supported by @command{etags}, and you want to avoid
 having @command{etags} fall back on Fortran and C as the default
 languages.
 
+  You can also prevent @command{etags} from falling back on Fortran and
+C if you specify the @samp{--no-fallback-lang} option.  The option
+@samp{--fallback-lang} countermands that.
+
   The option @samp{--parse-stdin=@var{file}} is mostly useful when
 calling @command{etags} from programs.  It can be used (only once) in
 place of a file name on the command line.  @command{etags} will read from
@@ -3068,6 +3072,15 @@ explanation.  If followed by one or more @samp{--language=@var{lang}}
 options, it outputs detailed information about how tags are generated for
 @var{lang}.
 
+  By default, @command{etags} includes in the tags table it produces all
+the files it scans, including files where it found no tags at all.
+Specify @samp{--no-empty-file-entries} to prevent that; then files with
+no tags will not be mentioned in the tags table.  However, note that
+commands which process files mentioned in the tags table, such as
+@code{tags-search} (@pxref{Identifier Search}), will process files which
+were thus omitted from the tags table.  The option
+@samp{--empty-file-entries} countermands that.
+
 @findex etags-regen-mode
   Instead of creating and updating the tags table by manually invoking
 @command{etags}, you can ask Emacs to do it for you automatically.

doc/man/etags.1

@@ -1,5 +1,5 @@
 .\" See section COPYING for copyright and redistribution information.
-.TH ETAGS 1 "2022-06-10" "GNU Tools" "GNU"
+.TH ETAGS 1 "2024-10-06" "GNU Tools" "GNU"
 .de BP
 .sp
 .ti -.2i
@@ -22,6 +22,8 @@ etags, ctags \- generate tag file for Emacs, vi
 [\|\-\-members\|] [\|\-\-no\-members\|] [\|\-\-output=\fItagfile\fP\|]
 [\|\-\-class\-qualify\|]
 [\|\-\-regex=\fIregexp\fP\|] [\|\-\-no\-regex\|]
+[\|\-\-no\-fallback\-lang\|] [\|\-\-fallback\-lang\|]
+[\|\-\-no\-empty\-file\-entries\|] [\|\-\-empty\-file\-entries\|]
 [\|\-\-help\|] [\|\-\-version\|]
 \fIfile\fP .\|.\|.
 
@@ -256,6 +258,20 @@ reads the regexes contained in the file regex.file.
 Don't do any more regexp matching on the following files.  May be
 freely intermixed with filenames and the \fB\-\-regex\fP option.
 .TP
+\fB\-\-fallback\-lang
+If a file's language cannot be determined, attempt to parse it as
+Fortran and C/C++.  This is the default.
+.TP
+\fB\-\-no\-fallback\-lang
+Do not fall back to Fortran and C/C++ for files whose language is
+could not be determined.
+.TP
+\fB\-\-empty\-file\-entries
+Produce file entries for files with no tags.  This is the default.
+.TP
+\fB\-\-no\-empty\-file\-entries
+Do not output file entries for files with no tags.
+.TP
 .B \-u, \-\-update
 Update tag entries for \fIfiles\fP specified on command line, leaving
 tag entries for other files in place.  Currently, this is implemented

etc/NEWS

@@ -44,6 +44,16 @@ why it the mark trace buffer is enabled by default.
 
 * Changes in Emacs 31.1
 
+** Etags
++++
+New command-line options for handling unrecognized programming languages.
+The new command-line option '--no-fallback-lang' disables attempts to
+parse as Fortran or C/C++ files whose programming language 'etags' could
+not determine.  This allows to avoid false positives and reduce the time
+required to scan directories with many such files.  Another new option
+'--no-empty-file-entries' disables generation of file entries in tags
+tables for files in which no tags were found.
+
 ---
 ** find-func.el commands now have history enabled.
 The 'find-function', 'find-library', 'find-face-definition', and

lib-src/etags.c

@@ -480,6 +480,8 @@ static bool ignoreindent;	/* -I: ignore indentation in C */
 static int packages_only;	/* --packages-only: in Ada, only tag packages*/
 static int class_qualify;	/* -Q: produce class-qualified tags in C++/Java */
 static int debug;		/* --debug */
+static int fallback_lang;	/* --(no-)fallback-lang: Fortran/C fallbacks */
+static int empty_files;		/* --(no-)empty-file-entries */
 
 /* STDIN is defined in LynxOS system headers */
 #ifdef STDIN
@@ -530,6 +532,10 @@ static struct option longopts[] =
   { "no-defines",         no_argument,       NULL,               'D'   },
   { "no-globals",         no_argument,       &globals,           0     },
   { "include",            required_argument, NULL,               'i'   },
+  { "no-fallback-lang",   no_argument,       &fallback_lang,     0     },
+  { "fallback-lang",      no_argument,       &fallback_lang,     1     },
+  { "no-empty-file-entries", no_argument,    &empty_files,       0     },
+  { "empty-file-entries", no_argument,       &empty_files,       1     },
 #endif
   { NULL }
 };
@@ -1039,6 +1045,20 @@ Relative ones are stored relative to the output file's directory.\n");
 	Do not create tag entries for members of structures\n\
 	in some languages.");
 
+  if (!CTAGS)
+    {
+      puts ("--fallback-lang\n\
+	If a file's language could not be determined, try to parse\n\
+	it as Fortran and C/C++.");
+      puts ("--no-fallback-lang\n\
+	Do not fall back to Fortran and C/C++ if a file's language\n\
+	could not be determined.");
+      puts ("--empty-file-entries\n\
+	Produce file entries for files with no tags.");
+      puts ("--no-empty-file-entries\n\
+	Do not output file entries for files with no tags.");
+    }
+
   puts ("-Q, --class-qualify\n\
         Qualify tag names with their class name in C++, ObjC, Java, and Perl.\n\
         This produces tag names of the form \"class::member\" for C++,\n\
@@ -1161,6 +1181,15 @@ main (int argc, char **argv)
   typedefs = typedefs_or_cplusplus = constantypedefs = true;
   globals = members = true;
 
+  /* By default, fall back to Fortran/C/C++ if no language is detected by the
+     file's name.  This could be reversed in a future version, but only for
+     ETAGS.  */
+  fallback_lang = true;
+
+  /* By default, output file entries for files that have no tags.  This affects
+     only ETAGS. */
+  empty_files = true;
+
   /* When the optstring begins with a '-' getopt_long does not rearrange the
      non-options arguments to be at the end, but leaves them alone. */
   optstring = concat ("-ac:Cf:Il:o:Qr:RSVhH",
@@ -1388,10 +1417,13 @@ main (int argc, char **argv)
 	{
 	  fdesc *fdp;
 
-	  /* Output file entries that have no tags. */
-	  for (fdp = fdhead; fdp != NULL; fdp = fdp->next)
-	    if (!fdp->written)
-	      fprintf (tagf, "\f\n%s,0\n", fdp->taggedfname);
+	  /* Output file entries that have no tags, unless disabled. */
+	  if (empty_files)
+	    {
+	      for (fdp = fdhead; fdp != NULL; fdp = fdp->next)
+		if (!fdp->written)
+		  fprintf (tagf, "\f\n%s,0\n", fdp->taggedfname);
+	    }
 
 	  while (nincluded_files-- > 0)
 	    fprintf (tagf, "\f\n%s,include\n", *included_files++);
@@ -1951,22 +1983,30 @@ find_entries (FILE *inf)
 	}
     }
 
-  /* Else try Fortran or C. */
+  /* Else try Fortran or C if that fallback is not disabled. */
   if (parser == NULL)
     {
-      node *old_last_node = last_node;
-
-      curfdp->lang = get_language_from_langname ("fortran");
-      find_entries (inf);
-
-      if (old_last_node == last_node)
-	/* No Fortran entries found.  Try C. */
+      if (fallback_lang)
 	{
-	  reset_input (inf);
-	  curfdp->lang = get_language_from_langname (cplusplus ? "c++" : "c");
+	  node *old_last_node = last_node;
+
+	  curfdp->lang = get_language_from_langname ("fortran");
 	  find_entries (inf);
+
+	  if (old_last_node == last_node)
+	    /* No Fortran entries found.  Try C. */
+	    {
+	      reset_input (inf);
+	      curfdp->lang = get_language_from_langname (cplusplus
+							 ? "c++" : "c");
+	      find_entries (inf);
+	    }
+	  return;
 	}
-      return;
+      /* If fallbacks are disabled, treat files without a language as if
+	 '--language=none' was specified for them. */
+      curfdp->lang = get_language_from_langname ("none");
+      parser = curfdp->lang->function;
     }
 
   if (!no_line_directive

test/manual/etags/Makefile

@@ -59,6 +59,7 @@ check:
 	@$(MAKE) OPTIONS='--regex=@regexfile --no-members' ediff_4
 	@$(MAKE) OPTIONS='nonexistent --members --declarations --regex=@regexfile' ediff_5
 	@$(MAKE) OPTIONS='--class-qualify --members --declarations --regex=@regexfile' ediff_6
+	@$(MAKE) OPTIONS='--no-members --no-fallback --no-empty' ediff_7
 	@$(MAKE) cdiff
 	@$(MAKE) ctags_update