util, utf8: make ellipsize take multi-byte characters into account

rename old versions to ascii_*

Do not take into account zerowidth characters, but do consider double-wide characters.
Import needed utf8 helper code from glib.

v3: rebase ontop of utf8 restructuring work

[zj: tweak the algorithm a bit, move new code to separate file]

Makefile.am

@@ -683,6 +683,8 @@ libsystemd_shared_la_SOURCES = \
 	src/shared/exit-status.h \
 	src/shared/utf8.c \
 	src/shared/utf8.h \
+	src/shared/gunicode.c \
+	src/shared/gunicode.h \
 	src/shared/pager.c \
 	src/shared/pager.h \
 	src/shared/ioprio.h \

TODO

@@ -17,10 +17,6 @@ Bugfixes:
 
 * properly handle .mount unit state tracking when two mount points are stacked one on top of another on the exact same mount point.
 
-* ellipsize_mem must take into account multi-byte unicode characters, and
-  - make the resulting line the requested number of *characters*, not *bytes*,
-  - avoid truncuating multi-byte sequences in the middle.
-
 * When we detect invalid UTF-8, we cant't use it in an error message:
   log...("Path is not UTF-8 clean, ignoring assignment: %s", rvalue);
 

src/shared/gunicode.c

@@ -0,0 +1,108 @@
+/* gunicode.c - Unicode manipulation functions
+ *
+ *  Copyright (C) 1999, 2000 Tom Tromey
+ *  Copyright 2000, 2005 Red Hat, Inc.
+ */
+
+#include "gunicode.h"
+
+#define unichar uint32_t
+
+/**
+ * g_utf8_prev_char:
+ * @p: a pointer to a position within a UTF-8 encoded string
+ *
+ * Finds the previous UTF-8 character in the string before @p.
+ *
+ * @p does not have to be at the beginning of a UTF-8 character. No check
+ * is made to see if the character found is actually valid other than
+ * it starts with an appropriate byte. If @p might be the first
+ * character of the string, you must use g_utf8_find_prev_char() instead.
+ *
+ * Return value: a pointer to the found character.
+ **/
+char *
+utf8_prev_char (const char *p)
+{
+  while (1)
+    {
+      p--;
+      if ((*p & 0xc0) != 0x80)
+        return (char *)p;
+    }
+}
+
+struct Interval
+{
+  unichar start, end;
+};
+
+static int
+interval_compare (const void *key, const void *elt)
+{
+  unichar c = (unichar) (long) (key);
+  struct Interval *interval = (struct Interval *)elt;
+
+  if (c < interval->start)
+    return -1;
+  if (c > interval->end)
+    return +1;
+
+  return 0;
+}
+
+/*
+ * NOTE:
+ *
+ * The tables for g_unichar_iswide() and g_unichar_iswide_cjk() are
+ * generated from the Unicode Character Database's file
+ * extracted/DerivedEastAsianWidth.txt using the gen-iswide-table.py
+ * in this way:
+ *
+ *   ./gen-iswide-table.py < path/to/ucd/extracted/DerivedEastAsianWidth.txt | fmt
+ *
+ * Last update for Unicode 6.0.
+ */
+
+/**
+ * g_unichar_iswide:
+ * @c: a Unicode character
+ *
+ * Determines if a character is typically rendered in a double-width
+ * cell.
+ *
+ * Return value: %TRUE if the character is wide
+ **/
+bool
+unichar_iswide (unichar c)
+{
+  /* See NOTE earlier for how to update this table. */
+  static const struct Interval wide[] = {
+    {0x1100, 0x115F}, {0x2329, 0x232A}, {0x2E80, 0x2E99}, {0x2E9B, 0x2EF3},
+    {0x2F00, 0x2FD5}, {0x2FF0, 0x2FFB}, {0x3000, 0x303E}, {0x3041, 0x3096},
+    {0x3099, 0x30FF}, {0x3105, 0x312D}, {0x3131, 0x318E}, {0x3190, 0x31BA},
+    {0x31C0, 0x31E3}, {0x31F0, 0x321E}, {0x3220, 0x3247}, {0x3250, 0x32FE},
+    {0x3300, 0x4DBF}, {0x4E00, 0xA48C}, {0xA490, 0xA4C6}, {0xA960, 0xA97C},
+    {0xAC00, 0xD7A3}, {0xF900, 0xFAFF}, {0xFE10, 0xFE19}, {0xFE30, 0xFE52},
+    {0xFE54, 0xFE66}, {0xFE68, 0xFE6B}, {0xFF01, 0xFF60}, {0xFFE0, 0xFFE6},
+    {0x1B000, 0x1B001}, {0x1F200, 0x1F202}, {0x1F210, 0x1F23A}, {0x1F240,
+    0x1F248}, {0x1F250, 0x1F251}, {0x20000, 0x2FFFD}, {0x30000, 0x3FFFD}
+  };
+
+  if (bsearch ((void *)(uintptr_t)c, wide, (sizeof (wide) / sizeof ((wide)[0])), sizeof wide[0],
+               interval_compare))
+    return true;
+
+  return false;
+}
+
+const char utf8_skip_data[256] = {
+  1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,
+  1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,
+  1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,
+  1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,
+  1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,
+  1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,
+  2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,
+  3,3,3,3,3,3,3,3,3,3,3,3,3,3,3,3,4,4,4,4,4,4,4,4,5,5,5,5,6,6,1,1
+};

src/shared/gunicode.h

@@ -0,0 +1,28 @@
+/* gunicode.h - Unicode manipulation functions
+ *
+ *  Copyright (C) 1999, 2000 Tom Tromey
+ *  Copyright 2000, 2005 Red Hat, Inc.
+ */
+
+#include <stdint.h>
+#include <stdbool.h>
+#include <stdlib.h>
+
+char *utf8_prev_char (const char *p);
+
+extern const char utf8_skip_data[256];
+
+/**
+ * g_utf8_next_char:
+ * @p: Pointer to the start of a valid UTF-8 character
+ *
+ * Skips to the next character in a UTF-8 string. The string must be
+ * valid; this macro is as fast as possible, and has no error-checking.
+ * You would use this macro to iterate over a string character by
+ * character. The macro returns the start of the next UTF-8 character.
+ * Before using this macro, use g_utf8_validate() to validate strings
+ * that may contain invalid UTF-8.
+ */
+#define utf8_next_char(p) (char *)((p) + utf8_skip_data[*(const unsigned char *)(p)])
+
+bool unichar_iswide (uint32_t c);

src/shared/utf8.c

@@ -98,7 +98,7 @@ static int utf8_encoded_expected_len(const char *str) {
 }
 
 /* decode one unicode char */
-static int utf8_encoded_to_unichar(const char *str) {
+int utf8_encoded_to_unichar(const char *str) {
         int unichar;
         int len;
         int i;

src/shared/utf8.h

@@ -35,3 +35,4 @@ char *ascii_filter(const char *s);
 char *utf16_to_utf8(const void *s, size_t length);
 
 int utf8_encoded_valid_unichar(const char *str);
+int utf8_encoded_to_unichar(const char *str);

src/shared/util.c

@@ -74,6 +74,8 @@
 #include "env-util.h"
 #include "fileio.h"
 #include "device-nodes.h"
+#include "utf8.h"
+#include "gunicode.h"
 
 int saved_argc = 0;
 char **saved_argv = NULL;
@@ -3288,7 +3290,7 @@ int running_in_chroot(void) {
                 a.st_ino != b.st_ino;
 }
 
-char *ellipsize_mem(const char *s, size_t old_length, size_t new_length, unsigned percent) {
+static char *ascii_ellipsize_mem(const char *s, size_t old_length, size_t new_length, unsigned percent) {
         size_t x;
         char *r;
 
@@ -3319,6 +3321,80 @@ char *ellipsize_mem(const char *s, size_t old_length, size_t new_length, unsigne
         return r;
 }
 
+char *ellipsize_mem(const char *s, size_t old_length, size_t new_length, unsigned percent) {
+        size_t x;
+        char *e;
+        const char *i, *j;
+        unsigned k, len, len2;
+
+        assert(s);
+        assert(percent <= 100);
+        assert(new_length >= 3);
+
+        /* if no multibyte characters use ascii_ellipsize_mem for speed */
+        if (ascii_is_valid(s))
+                return ascii_ellipsize_mem(s, old_length, new_length, percent);
+
+        if (old_length <= 3 || old_length <= new_length)
+                return strndup(s, old_length);
+
+        x = (new_length * percent) / 100;
+
+        if (x > new_length - 3)
+                x = new_length - 3;
+
+        k = 0;
+        for (i = s; k < x && i < s + old_length; i = utf8_next_char(i)) {
+                int c;
+
+                c = utf8_encoded_to_unichar(i);
+                if (c < 0)
+                        return NULL;
+                k += unichar_iswide(c) ? 2 : 1;
+        }
+
+        if (k > x) /* last character was wide and went over quota */
+                x ++;
+
+        for (j = s + old_length; k < new_length && j > i; ) {
+                int c;
+
+                j = utf8_prev_char(j);
+                c = utf8_encoded_to_unichar(j);
+                if (c < 0)
+                        return NULL;
+                k += unichar_iswide(c) ? 2 : 1;
+        }
+        assert(i <= j);
+
+        /* we don't actually need to ellipsize */
+        if (i == j)
+                return memdup(s, old_length + 1);
+
+        /* make space for ellipsis */
+        j = utf8_next_char(j);
+
+        len = i - s;
+        len2 = s + old_length - j;
+        e = new(char, len + 3 + len2 + 1);
+        if (!e)
+                return NULL;
+
+        /*
+        printf("old_length=%zu new_length=%zu x=%zu len=%u len2=%u k=%u\n",
+               old_length, new_length, x, len, len2, k);
+        */
+
+        memcpy(e, s, len);
+        e[len]   = 0xe2; /* tri-dot ellipsis: … */
+        e[len + 1] = 0x80;
+        e[len + 2] = 0xa6;
+
+        memcpy(e + len + 3, j, len2 + 1);
+
+        return e;
+}
+
 char *ellipsize(const char *s, size_t length, unsigned percent) {
         return ellipsize_mem(s, strlen(s), length, percent);
 }

src/shared/util.h

@@ -405,6 +405,7 @@ static inline const char *ansi_highlight_off(void) {
 int running_in_chroot(void);
 
 char *ellipsize(const char *s, size_t length, unsigned percent);
+                                   /* bytes                 columns */
 char *ellipsize_mem(const char *s, size_t old_length, size_t new_length, unsigned percent);
 
 int touch(const char *path);