ls: --color: honor separate sequences for extension cases

Following on from commit v8.29-45-g24053fbd8 which unconditionally
used case insensitive extension matching, support selective
case sensitive matching when there are separate extension cases
defined with different display sequences.

* src/dircolors.hin: Document how file name suffixes are matched.
Note this is displayed with `dircolors --print-database` which
the texi info recommends to use for details.
* src/ls.c (parse_ls_color): Postprocess the list to
mark entries for case sensitive matching,
and also adjust so that unmatchable entries are more quickly ignored.
(get_color_indicator): Use exact matching rather than
case insensitive matching if so marked.
* tests/ls/color-ext.sh: Add test cases.
* NEWS: Mention the change in behavior.
Addresses https://bugs.gnu.org/33123

4 files changed, 114 insertions, 9 deletions

diff --git a/NEWS b/NEWS
index 7bc8f9109..f46b16b6f 100644
--- a/NEWS
+++ b/NEWS
@@ -103,6 +103,9 @@ GNU coreutils NEWS                                    -*- outline -*-
   reverting to the behavior in coreutils-9.0 and earlier.
   This behavior is now documented.
 
+  ls --color now matches a file extension case sensitively
+  if there are different sequences defined for separate cases.
+
   printf unicode \uNNNN, \UNNNNNNNN syntax, now supports all valid
   unicode code points.  Previously is was restricted to the C
   universal character subset, which restricted most points <= 0x9F.
diff --git a/src/dircolors.hin b/src/dircolors.hin
index e5d9eab94..24a90cf94 100644
--- a/src/dircolors.hin
+++ b/src/dircolors.hin
@@ -74,8 +74,10 @@ STICKY 37;44	# dir with the sticky bit set (+t) and not other-writable
 EXEC 01;32
 
 # List any file extensions like '.gz' or '.tar' that you would like ls
-# to color below. Put the extension, a space, and the color init string.
-# (and any comments you want to add after a '#')
+# to color below. Put the suffix, a space, and the color init string.
+# (and any comments you want to add after a '#').
+# Suffixes are matched case insensitively, but if you define different
+# init strings for separate cases, those will be honored.
 
 # If you use DOS-style suffixes, you may want to uncomment the following:
 #.cmd 01;32 # executables (bright green)
@@ -88,7 +90,7 @@ EXEC 01;32
 #.sh  01;32
 #.csh 01;32
 
- # archives or compressed (bright red)
+# archives or compressed (bright red)
 .tar 01;31
 .tgz 01;31
 .arc 01;31
diff --git a/src/ls.c b/src/ls.c
index bd9130ccb..71d94fd6a 100644
--- a/src/ls.c
+++ b/src/ls.c
@@ -612,6 +612,7 @@ struct color_ext_type
   {
     struct bin_str ext;		/* The extension we're looking for */
     struct bin_str seq;		/* The sequence to output when we do */
+    bool   exact_match;		/* Whether to compare case insensitively */
     struct color_ext_type *next;	/* Next in list */
   };
 
@@ -643,7 +644,7 @@ static struct bin_str color_indicator[] =
     { LEN_STR_PAIR ("\033[K") },	/* cl: clear to end of line */
   };
 
-/* FIXME: comment  */
+/* A list mapping file extensions to corresponding display sequence.  */
 static struct color_ext_type *color_ext_list = NULL;
 
 /* Buffer for color sequences */
@@ -2775,6 +2776,7 @@ parse_ls_color (void)
               ext = xmalloc (sizeof *ext);
               ext->next = color_ext_list;
               color_ext_list = ext;
+              ext->exact_match = false;
 
               ++p;
               ext->ext.string = buf;
@@ -2860,6 +2862,49 @@ parse_ls_color (void)
         }
       print_with_color = false;
     }
+  else
+    {
+      /* Postprocess list to set EXACT_MATCH on entries where there are
+         different cased extensions with separate sequences defined.
+         Also set ext.len to SIZE_MAX on any entries that can't
+         match due to precedence, to avoid redundant string compares.  */
+      struct color_ext_type *e1;
+
+      for (e1 = color_ext_list; e1 != NULL; e1 = e1->next)
+        {
+          struct color_ext_type *e2;
+          bool case_ignored = false;
+
+          for (e2 = e1->next; e2 != NULL; e2 = e2->next)
+            {
+              if (e2->ext.len < SIZE_MAX && e1->ext.len == e2->ext.len)
+                {
+                  if (memcmp (e1->ext.string, e2->ext.string, e1->ext.len) == 0)
+                    e2->ext.len = SIZE_MAX; /* Ignore */
+                  else if (c_strncasecmp (e1->ext.string, e2->ext.string,
+                                          e1->ext.len) == 0)
+                    {
+                      if (case_ignored)
+                        {
+                          e2->ext.len = SIZE_MAX; /* Ignore */
+                        }
+                      else if (e1->seq.len == e2->seq.len
+                               && memcmp (e1->seq.string, e2->seq.string,
+                                          e1->seq.len) == 0)
+                        {
+                          e2->ext.len = SIZE_MAX; /* Ignore */
+                          case_ignored = true;    /* Ignore all subsequent */
+                        }
+                      else
+                        {
+                          e1->exact_match = true;
+                          e2->exact_match = true;
+                        }
+                    }
+                }
+            }
+        }
+    }
 
   if (color_indicator[C_LINK].len == 6
       && !STRNCMP_LIT (color_indicator[C_LINK].string, "target"))
@@ -5040,10 +5085,21 @@ get_color_indicator (const struct fileinfo *f, bool symlink_target)
       name += len;		/* Pointer to final \0.  */
       for (ext = color_ext_list; ext != NULL; ext = ext->next)
         {
-          if (ext->ext.len <= len
-              && c_strncasecmp (name - ext->ext.len, ext->ext.string,
-                                ext->ext.len) == 0)
-            break;
+          if (ext->ext.len <= len)
+            {
+              if (ext->exact_match)
+                {
+                  if (STREQ_LEN (name - ext->ext.len, ext->ext.string,
+                                 ext->ext.len))
+                    break;
+                }
+              else
+                {
+                  if (c_strncasecmp (name - ext->ext.len, ext->ext.string,
+                                     ext->ext.len) == 0)
+                    break;
+                }
+            }
         }
     }
 
diff --git a/tests/ls/color-ext.sh b/tests/ls/color-ext.sh
index 091895c9f..ab045ec6a 100755
--- a/tests/ls/color-ext.sh
+++ b/tests/ls/color-ext.sh
@@ -20,13 +20,16 @@
 print_ver_ ls
 working_umask_or_skip_
 
-touch img1.jpg IMG2.JPG file1.z file2.Z || framework_failure_
+touch img1.jpg IMG2.JPG img3.JpG file1.z file2.Z || framework_failure_
 code_jpg='01;35'
+code_JPG='01;35;46'
 code_z='01;31'
 c0=$(printf '\033[0m')
 c_jpg=$(printf '\033[%sm' $code_jpg)
+c_JPG=$(printf '\033[%sm' $code_JPG)
 c_z=$(printf '\033[%sm' $code_z)
 
+# Case insenitive extensions
 LS_COLORS="*.jpg=$code_jpg:*.Z=$code_z" ls -U1 --color=always \
   img1.jpg IMG2.JPG file1.z file2.Z > out || fail=1
 printf "$c0\
@@ -37,5 +40,46 @@ ${c_z}file2.Z$c0
 " > out_ok || framework_failure_
 compare out out_ok || fail=1
 
+# Case sensitive extensions
+LS_COLORS="*.jpg=$code_jpg:*.JPG=$code_JPG" ls -U1 --color=always \
+  img1.jpg IMG2.JPG img3.JpG > out || fail=1
+printf "$c0\
+${c_jpg}img1.jpg$c0
+${c_JPG}IMG2.JPG$c0
+img3.JpG
+" > out_ok || framework_failure_
+compare out out_ok || fail=1
+
+# Case insensitive extensions (due to same sequences)
+LS_COLORS="*.jpg=$code_jpg:*.JPG=$code_jpg" ls -U1 --color=always \
+  img1.jpg IMG2.JPG img3.JpG > out || fail=1
+printf "$c0\
+${c_jpg}img1.jpg$c0
+${c_jpg}IMG2.JPG$c0
+${c_jpg}img3.JpG$c0
+" > out_ok || framework_failure_
+compare out out_ok || fail=1
+
+# Case insensitive extensions (due to same sequences (after ignored sequences))
+# Note later entries in LS_COLORS take precedence.
+LS_COLORS="*.jpg=$code_jpg:*.jpg=$code_JPG:*.JPG=$code_JPG" \
+  ls -U1 --color=always img1.jpg IMG2.JPG img3.JpG > out || fail=1
+printf "$c0\
+${c_JPG}img1.jpg$c0
+${c_JPG}IMG2.JPG$c0
+${c_JPG}img3.JpG$c0
+" > out_ok || framework_failure_
+compare out out_ok || fail=1
+
+# Case sensitive extensions (due to diff sequences (after ignored sequences))
+# Note later entries in LS_COLORS take precedence.
+LS_COLORS="*.jpg=$code_JPG:*.jpg=$code_jpg:*.JPG=$code_JPG" \
+  ls -U1 --color=always img1.jpg IMG2.JPG img3.JpG > out || fail=1
+printf "$c0\
+${c_jpg}img1.jpg$c0
+${c_JPG}IMG2.JPG$c0
+img3.JpG
+" > out_ok || framework_failure_
+compare out out_ok || fail=1
 
 Exit $fail