Read the input field with unescaping if TAG_OUTPUT_MODE is "u-ctags" and TAG_OUTPUT_FILESEP is "slash"

.gitignore

@@ -44,3 +44,5 @@ tests/test-fix-unescaping
 tests/test-fix-unescaping-input-fields
 tests/test-fix-unescaping-input-fields-exuberant
 tests/test-fix-unescaping-input-fields-no-mode
+tests/test-fix-unescaping-input-fields-backslash
+tests/test-fix-unescaping-input-fields-no-filesep

NEWS.md

@@ -1,7 +1,8 @@
 # Version ???
 
 - read input fields, values at the second column in a tag file, with
-  unescaping if !_TAG_OUTPUT_MODE in the tag file is "u-ctags".
+  unescaping if !_TAG_OUTPUT_MODE is "u-ctags" and
+  !_TAG_OUTPUT_FILESEP is "slash" in the tag file.
 
 - LT_VERSION ?:?:?
 

readtags.c

@@ -46,7 +46,8 @@ struct sTagFile {
 		/* format of tag file */
 	unsigned char format;
 		/* 1 "u-ctags" is set to !_TAG_OUTPUT_MODE pseudo tag
-		 * in the tags file. */
+		 * and "slash" is set to !_TAG_OUTPUT_FILESEP
+		 * pseudo tag in the tags file. */
 	unsigned char inputUCtagsMode;
 		/* how is the tag file sorted? */
 	tagSortType sortMethod;
@@ -691,6 +692,8 @@ static tagResult readPseudoTags (tagFile *const file, tagFileInfo *const info)
 	int err = 0;
 	tagResult result = TagSuccess;
 	const size_t prefixLength = strlen (PseudoTagPrefix);
+	int tag_output_mode_u_ctags = 0;
+	int tag_output_filesep_slash = 0;
 
 	info->file.format     = 1;
 	info->file.sort       = TAG_UNSORTED;
@@ -779,7 +782,12 @@ static tagResult readPseudoTags (tagFile *const file, tagFileInfo *const info)
 			else if (strcmp (key, "TAG_OUTPUT_MODE") == 0)
 			{
 				if (strcmp (value, "u-ctags") == 0)
-					file->inputUCtagsMode = 1;
+					tag_output_mode_u_ctags = 1;
+			}
+			else if (strcmp (key, "TAG_OUTPUT_FILESEP") == 0)
+			{
+				if (strcmp (value, "slash") == 0)
+					tag_output_filesep_slash = 1;
 			}
 
 			info->file.format     = file->format;
@@ -790,6 +798,10 @@ static tagResult readPseudoTags (tagFile *const file, tagFileInfo *const info)
 			info->program.version = file->program.version;
 		}
 	}
+
+	if (tag_output_mode_u_ctags && tag_output_filesep_slash)
+		file->inputUCtagsMode = 1;
+
 	if (fsetpos (file->fp, &startOfLine) < 0)
 		err = errno;
 

tests/Makefile.am

@@ -14,6 +14,8 @@ TESTS = \
 	test-fix-unescaping-input-fields \
 	test-fix-unescaping-input-fields-exuberant \
 	test-fix-unescaping-input-fields-no-mode \
+	test-fix-unescaping-input-fields-backslash \
+	test-fix-unescaping-input-fields-no-filesep \
 	\
 	$(NULL)
 
@@ -33,6 +35,8 @@ check_PROGRAMS = \
 	test-fix-unescaping-input-fields \
 	test-fix-unescaping-input-fields-exuberant \
 	test-fix-unescaping-input-fields-no-mode \
+	test-fix-unescaping-input-fields-backslash \
+	test-fix-unescaping-input-fields-no-filesep \
 	\
 	$(NULL)
 
@@ -108,3 +112,11 @@ EXTRA_DIST += unescaping-input-fields-exuberant.tags
 test_fix_unescaping_input_fields_no_mode = test-fix-unescaping-input-fields-no-mode.c
 test_fix_unescaping_input_fields_no_mode_DEPENDENCIES = $(DEPS)
 EXTRA_DIST += unescaping-input-fields-no-mode.tags
+
+test_fix_unescaping_input_fields_backslash = test-fix-unescaping-input-fields-backslash.c
+test_fix_unescaping_input_fields_backslash_DEPENDENCIES = $(DEPS)
+EXTRA_DIST += unescaping-input-fields-backslash.tags
+
+test_fix_unescaping_input_fields_no_filesep = test-fix-unescaping-input-fields-no-filesep.c
+test_fix_unescaping_input_fields_no_filesep_DEPENDENCIES = $(DEPS)
+EXTRA_DIST += unescaping-input-fields-no-filesep.tags

tests/test-fix-unescaping-input-fields-backslash.c

@@ -0,0 +1,94 @@
+/*
+*   Copyright (c) 2022, Masatake YAMATO
+*
+*   This source code is released into the public domain.
+*
+*   Testing the fix for handling unescaping
+*/
+
+#include "readtags.h"
+
+#include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
+#include <unistd.h>
+
+#include "test-fields.h"
+
+
+#define CHECK3(NAME,FILE,PAT)					\
+	CHECK ((NAME), name);						\
+	CHECK ((FILE), file);						\
+	CHECK ((PAT),  address.pattern)
+
+#define NEXT_CHECK3(NAME,FILE,PAT)				\
+	NEXT ();									\
+	CHECK3 (NAME,FILE,PAT)
+
+int
+main (void)
+{
+	char *srcdir = getenv ("srcdir");
+	if (srcdir)
+	{
+		if (chdir (srcdir) == -1)
+		{
+			perror ("chdir");
+			return 99;
+		}
+	}
+
+	tagFile *t;
+	tagFileInfo info;
+
+	const char *tags0 = "./unescaping-input-fields-backslash.tags";
+	t = tagsOpen (tags0, &info);
+	if (t == NULL
+		|| info.status.opened == 0)
+	{
+		fprintf (stderr, "unexpected result (t: %p, opened: %d)\n",
+				 t, info.status.opened);
+		return 1;
+	}
+	fprintf (stderr, "ok\n");
+
+	tagEntry e;
+	tagResult r;
+
+	r = tagsFirst (t, &e);
+	if (r != TagSuccess)
+	{
+		fprintf (stderr, "error in tagsFirst\n");
+		return 1;
+	}
+
+	CHECK3 ("tab0", "\\tabc", "/^tab0$/");
+	NEXT_CHECK3 ("tab1", "a\\tbc", "/^tab1$/");
+	NEXT_CHECK3 ("tab2", "ab\\tc", "/^tab2$/");
+	NEXT_CHECK3 ("tab3", "abc\\t", "/^tab3$/");
+	NEXT_CHECK3 ("tab4", "\\\\abc", "/^tab4$/");
+	NEXT_CHECK3 ("tab5", "a\\\\bc", "/^tab5$/");
+	NEXT_CHECK3 ("tab6", "ab\\\\c", "/^tab6$/");
+	NEXT_CHECK3 ("tab7", "abc\\\\", "/^tab7$/");
+	NEXT_CHECK3 ("tab8", "\\nabc", "/^tab8$/");
+	NEXT_CHECK3 ("tab9", "a\\nbc", "/^tab9$/");
+	NEXT_CHECK3 ("taba", "ab\\nc", "/^taba$/");
+	NEXT_CHECK3 ("tabb", "abc\\n", "/^tabb$/");
+	NEXT_CHECK3 ("tabc", "\\n\\\\abc", "/^tabc$/");
+	NEXT_CHECK3 ("tabd", "\\\\\\nabc", "/^tabd$/");
+	NEXT_CHECK3 ("tabe", "a\\n\\\\bc", "/^tabe$/");
+	NEXT_CHECK3 ("tabf", "a\\\\\\nbc", "/^tabf$/");
+	NEXT_CHECK3 ("tabg", "abc\\n\\\\", "/^tabg$/");
+	NEXT_CHECK3 ("tabh", "abc\\\\\\n", "/^tabh$/");
+	NEXT_CHECK3 ("tabi", "\\t\\\\\\n",  "/^tabi$/");
+	NEXT_CHECK3 ("tabj", "ab\\t\\\\\\nc",  "/^tabj$/");
+
+	r = tagsClose(t);
+	if (r != TagSuccess)
+	{
+		fprintf (stderr, "error in tagsClose\n");
+		return 1;
+	}
+
+	return 0;
+}

tests/test-fix-unescaping-input-fields-no-filesep.c

@@ -0,0 +1,94 @@
+/*
+*   Copyright (c) 2022, Masatake YAMATO
+*
+*   This source code is released into the public domain.
+*
+*   Testing the fix for handling unescaping
+*/
+
+#include "readtags.h"
+
+#include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
+#include <unistd.h>
+
+#include "test-fields.h"
+
+
+#define CHECK3(NAME,FILE,PAT)					\
+	CHECK ((NAME), name);						\
+	CHECK ((FILE), file);						\
+	CHECK ((PAT),  address.pattern)
+
+#define NEXT_CHECK3(NAME,FILE,PAT)				\
+	NEXT ();									\
+	CHECK3 (NAME,FILE,PAT)
+
+int
+main (void)
+{
+	char *srcdir = getenv ("srcdir");
+	if (srcdir)
+	{
+		if (chdir (srcdir) == -1)
+		{
+			perror ("chdir");
+			return 99;
+		}
+	}
+
+	tagFile *t;
+	tagFileInfo info;
+
+	const char *tags0 = "./unescaping-input-fields-no-filesep.tags";
+	t = tagsOpen (tags0, &info);
+	if (t == NULL
+		|| info.status.opened == 0)
+	{
+		fprintf (stderr, "unexpected result (t: %p, opened: %d)\n",
+				 t, info.status.opened);
+		return 1;
+	}
+	fprintf (stderr, "ok\n");
+
+	tagEntry e;
+	tagResult r;
+
+	r = tagsFirst (t, &e);
+	if (r != TagSuccess)
+	{
+		fprintf (stderr, "error in tagsFirst\n");
+		return 1;
+	}
+
+	CHECK3 ("tab0", "\\tabc", "/^tab0$/");
+	NEXT_CHECK3 ("tab1", "a\\tbc", "/^tab1$/");
+	NEXT_CHECK3 ("tab2", "ab\\tc", "/^tab2$/");
+	NEXT_CHECK3 ("tab3", "abc\\t", "/^tab3$/");
+	NEXT_CHECK3 ("tab4", "\\\\abc", "/^tab4$/");
+	NEXT_CHECK3 ("tab5", "a\\\\bc", "/^tab5$/");
+	NEXT_CHECK3 ("tab6", "ab\\\\c", "/^tab6$/");
+	NEXT_CHECK3 ("tab7", "abc\\\\", "/^tab7$/");
+	NEXT_CHECK3 ("tab8", "\\nabc", "/^tab8$/");
+	NEXT_CHECK3 ("tab9", "a\\nbc", "/^tab9$/");
+	NEXT_CHECK3 ("taba", "ab\\nc", "/^taba$/");
+	NEXT_CHECK3 ("tabb", "abc\\n", "/^tabb$/");
+	NEXT_CHECK3 ("tabc", "\\n\\\\abc", "/^tabc$/");
+	NEXT_CHECK3 ("tabd", "\\\\\\nabc", "/^tabd$/");
+	NEXT_CHECK3 ("tabe", "a\\n\\\\bc", "/^tabe$/");
+	NEXT_CHECK3 ("tabf", "a\\\\\\nbc", "/^tabf$/");
+	NEXT_CHECK3 ("tabg", "abc\\n\\\\", "/^tabg$/");
+	NEXT_CHECK3 ("tabh", "abc\\\\\\n", "/^tabh$/");
+	NEXT_CHECK3 ("tabi", "\\t\\\\\\n",  "/^tabi$/");
+	NEXT_CHECK3 ("tabj", "ab\\t\\\\\\nc",  "/^tabj$/");
+
+	r = tagsClose(t);
+	if (r != TagSuccess)
+	{
+		fprintf (stderr, "error in tagsClose\n");
+		return 1;
+	}
+
+	return 0;
+}

tests/unescaping-input-fields-backslash.tags

@@ -0,0 +1,29 @@
+!_TAG_FILE_FORMAT	2	/extended format; --format=1 will not append ;" to lines/
+!_TAG_FILE_SORTED	0	/0=unsorted, 1=sorted, 2=foldcase/
+!_TAG_OUTPUT_FILESEP	backslash	/slash or backslash/
+!_TAG_OUTPUT_MODE	u-ctags	/u-ctags or e-ctags/
+!_TAG_PATTERN_LENGTH_LIMIT	96	/0 for no limit/
+!_TAG_PROGRAM_AUTHOR	Universal Ctags Team	//
+!_TAG_PROGRAM_NAME	Universal Ctags	/Derived from Exuberant Ctags/
+!_TAG_PROGRAM_URL	https://ctags.io/	/official site/
+!_TAG_PROGRAM_VERSION	0.0.0	/8d952eba/
+tab0	\tabc	/^tab0$/;"	n
+tab1	a\tbc	/^tab1$/;"	n
+tab2	ab\tc	/^tab2$/;"	n
+tab3	abc\t	/^tab3$/;"	n
+tab4	\\abc	/^tab4$/;"	n
+tab5	a\\bc	/^tab5$/;"	n
+tab6	ab\\c	/^tab6$/;"	n
+tab7	abc\\	/^tab7$/;"	n
+tab8	\nabc	/^tab8$/;"	n
+tab9	a\nbc	/^tab9$/;"	n
+taba	ab\nc	/^taba$/;"	n
+tabb	abc\n	/^tabb$/;"	n
+tabc	\n\\abc	/^tabc$/;"	n
+tabd	\\\nabc	/^tabd$/;"	n
+tabe	a\n\\bc	/^tabe$/;"	n
+tabf	a\\\nbc	/^tabf$/;"	n
+tabg	abc\n\\	/^tabg$/;"	n
+tabh	abc\\\n	/^tabh$/;"	n
+tabi	\t\\\n	/^tabi$/;"	n
+tabj	ab\t\\\nc	/^tabj$/;"	n

tests/unescaping-input-fields-no-filesep.tags

@@ -0,0 +1,28 @@
+!_TAG_FILE_FORMAT	2	/extended format; --format=1 will not append ;" to lines/
+!_TAG_FILE_SORTED	0	/0=unsorted, 1=sorted, 2=foldcase/
+!_TAG_OUTPUT_MODE	u-ctags	/u-ctags or e-ctags/
+!_TAG_PATTERN_LENGTH_LIMIT	96	/0 for no limit/
+!_TAG_PROGRAM_AUTHOR	Universal Ctags Team	//
+!_TAG_PROGRAM_NAME	Universal Ctags	/Derived from Exuberant Ctags/
+!_TAG_PROGRAM_URL	https://ctags.io/	/official site/
+!_TAG_PROGRAM_VERSION	0.0.0	/8d952eba/
+tab0	\tabc	/^tab0$/;"	n
+tab1	a\tbc	/^tab1$/;"	n
+tab2	ab\tc	/^tab2$/;"	n
+tab3	abc\t	/^tab3$/;"	n
+tab4	\\abc	/^tab4$/;"	n
+tab5	a\\bc	/^tab5$/;"	n
+tab6	ab\\c	/^tab6$/;"	n
+tab7	abc\\	/^tab7$/;"	n
+tab8	\nabc	/^tab8$/;"	n
+tab9	a\nbc	/^tab9$/;"	n
+taba	ab\nc	/^taba$/;"	n
+tabb	abc\n	/^tabb$/;"	n
+tabc	\n\\abc	/^tabc$/;"	n
+tabd	\\\nabc	/^tabd$/;"	n
+tabe	a\n\\bc	/^tabe$/;"	n
+tabf	a\\\nbc	/^tabf$/;"	n
+tabg	abc\n\\	/^tabg$/;"	n
+tabh	abc\\\n	/^tabh$/;"	n
+tabi	\t\\\n	/^tabi$/;"	n
+tabj	ab\t\\\nc	/^tabj$/;"	n