Changeset 82bb9c1 in mainline

Timestamp:

2009-04-02T20:38:51Z (16 years ago)

Author:

Jiri Svoboda <jirik.svoboda@…>

Branches:

lfn, master, serial, ticket/834-toolchain-update, topic/msim-upgrade, topic/simplify-dev-export

Children:

58d5a7e7

Parents:

d09f84e6

Message:

Consider character display width somewhat. Explain naming scheme.

Location:

kernel/generic

Files:

• include/string.h (modified) (1 diff)
• src/lib/string.c (modified) (6 diffs)
• src/printf/printf_core.c (modified) (2 diffs)

kernel/generic/include/string.h

@@ -43 +43 @@
 
 extern wchar_t chr_decode(const char *, size_t *, size_t);
-extern int chr_encode(const wchar_t, char *, size_t *, size_t);
+extern int chr_encode(wchar_t, char *, size_t *, size_t);
+extern count_t chr_width(wchar_t);
 
-extern size_t str_size(const char *str);
+extern size_t str_size(const char *);
 extern size_t str_lsize(const char *, count_t);
-extern count_t str_length(const char *str);
-extern count_t wstr_length(const wchar_t *str);
+extern size_t str_wsize(const char *, count_t);
+
+extern count_t str_length(const char *);
+extern count_t wstr_length(const wchar_t *);
+extern count_t wstr_wlength(const wchar_t *, count_t);
 
 extern bool ascii_check(const wchar_t ch);

kernel/generic/src/lib/string.c

@@ -33 +33 @@
 /**
  * @file
- * @brief Miscellaneous functions.
+ * @brief String functions.
+ *
+ * Strings and characters use the Universal Character Set (UCS). The standard
+ * strings, called just strings are encoded in UTF-8. Wide strings (encoded
+ * in UTF-32) are supported to a limited degree. A single character is
+ * represented as wchar_t.
+ *
+ * Strings have the following metrics:
+ *
+ *      Metric  Abbrev. Meaning
+ *      ------  ------  -------
+ *      size    n       Number of bytes the string is encoded into, excluding
+ *                      the null terminator.
+ *      length  l       The number of characters in the string, excluding
+ *                      the null terminator.
+ *      width   w       The number of character cells the string takes up on a
+ *                      monospace display.
+ *
+ * Naming scheme:
+ *
+ *      chr_xxx         operate on characters
+ *      str_xxx         operate on strings
+ *      wstr_xxx        operate on wide strings
+ *
+ *      [w]str_[n|l|w]xxx       operate on a prefix limited by size, length
+ *                              or width.
  */
 
@@ -146 +171 @@
  *         code was invalid.
  */
-int chr_encode(const wchar_t ch, char *str, size_t *offset, size_t sz)
+int chr_encode(wchar_t ch, char *str, size_t *offset, size_t sz)
 {
         uint32_t cc;            /* Unsigned version of ch. */
@@ -200 +225 @@
 }
 
+/** Get display width of character.
+ *
+ * @param ch    The character.
+ * @return      Character width in display cells.
+ */
+count_t chr_width(wchar_t ch)
+{
+        return 1;
+}
+
 /** Get size of string, with length limit.
  *
@@ -206 +241 @@
  * the length of @a str, the entire string is measured.
  *
- * @param str   String to consider.
- * @param count Maximum number of characters to measure.
- *
- * @return Number of bytes used by the characters.
+ * @param str   String to consider.
+ * @param count Maximum number of characters to measure.
+ *
+ * @return      Number of bytes used by the characters.
  */
 size_t str_lsize(const char *str, count_t max_len)
@@ -231 +266 @@
 }
 
+/** Get size of string, with width limit.
+ *
+ * Get the number of bytes which are used by the longest prefix of @a str
+ * that can fit into @a max_width display cells.
+ *
+ * @param str   String to consider.
+ * @param count Maximum number of display cells.
+ *
+ * @return      Number of bytes used by the characters that fit.
+ */
+size_t str_wsize(const char *str, count_t max_width)
+{
+        count_t width = 0;
+        size_t cur = 0;
+        size_t prev;
+        wchar_t ch;
+
+        while (true) {
+                prev = cur;
+                if (width >= max_width)
+                        break;
+                ch = chr_decode(str, &cur, UTF8_NO_LIMIT);
+                if (ch == '\0') break;
+
+                width += chr_width(ch);
+        }
+
+        return prev;
+}
+
+
+/** Get length of wide string, with width limit.
+ *
+ * Get the number of characters in a wide string that can fit into @a max_width
+ * display cells.
+ *
+ * @param wstr   Wide string to consider.
+ * @param count Maximum number of display cells.
+ *
+ * @return      Number of bytes used by the characters that fit.
+ */
+count_t wstr_wlength(const wchar_t *wstr, count_t max_width)
+{
+        count_t width = 0;
+        index_t cur = 0;
+
+        while (true) {
+                if (width >= max_width)
+                        break;
+                if (wstr[cur] == '\0') break;
+
+                width += chr_width(wstr[cur]);
+                ++cur;
+        }
+
+        return (count_t) cur;
+}
+
 /** Check whether character is plain ASCII.
  *
@@ -291 +384 @@
 /** Return number of characters in a wide string.
  *
- * @param str NULL-terminated wide string.
- * @return Number of characters in @a str.
+ * @param       str NULL-terminated wide string.
+ * @return      Number of characters in @a str.
  */
 count_t wstr_length(const wchar_t *wstr)

kernel/generic/src/printf/printf_core.c

@@ -269 +269 @@
 
         int retval;
-        size_t size = str_lsize(str, precision);
+        size_t size = str_wsize(str, precision);
         if ((retval = printf_putnchars_utf8(str, size, ps)) < 0)
                 return -counter;
@@ -314 +314 @@
 
         int retval;
-        size_t size = min(strw, precision) * sizeof(wchar_t);
+        size_t size = wstr_wlength(wstr, precision) * sizeof(wchar_t);
         if ((retval = printf_putnchars_utf32(wstr, size, ps)) < 0)
                 return -counter;