Changes in uspace/lib/c/generic/str.c [933cadf:c3d19ac] in mainline

File:

• uspace/lib/c/generic/str.c (modified) (5 diffs)

uspace/lib/c/generic/str.c

@@ -2 +2 @@
  * Copyright (c) 2005 Martin Decky
  * Copyright (c) 2008 Jiri Svoboda
+ * Copyright (c) 2011 Martin Sucha
+ * Copyright (c) 2011 Oleg Romanenko
  * All rights reserved.
  *
@@ -540 +542 @@
 
         dstr_size = str_size(dest);
+        if (dstr_size >= size)
+                return;
+        
         str_cpy(dest + dstr_size, size - dstr_size, src);
+}
+
+/** Convert space-padded ASCII to string.
+ *
+ * Common legacy text encoding in hardware is 7-bit ASCII fitted into
+ * a fixed-width byte buffer (bit 7 always zero), right-padded with spaces
+ * (ASCII 0x20). Convert space-padded ascii to string representation.
+ *
+ * If the text does not fit into the destination buffer, the function converts
+ * as many characters as possible and returns EOVERFLOW.
+ *
+ * If the text contains non-ASCII bytes (with bit 7 set), the whole string is
+ * converted anyway and invalid characters are replaced with question marks
+ * (U_SPECIAL) and the function returns EIO.
+ *
+ * Regardless of return value upon return @a dest will always be well-formed.
+ *
+ * @param dest          Destination buffer
+ * @param size          Size of destination buffer
+ * @param src           Space-padded ASCII.
+ * @param n             Size of the source buffer in bytes.
+ *
+ * @return              EOK on success, EOVERFLOW if the text does not fit
+ *                      destination buffer, EIO if the text contains
+ *                      non-ASCII bytes.
+ */
+int spascii_to_str(char *dest, size_t size, const uint8_t *src, size_t n)
+{
+        size_t sidx;
+        size_t didx;
+        size_t dlast;
+        uint8_t byte;
+        int rc;
+        int result;
+
+        /* There must be space for a null terminator in the buffer. */
+        assert(size > 0);
+        result = EOK;
+
+        didx = 0;
+        dlast = 0;
+        for (sidx = 0; sidx < n; ++sidx) {
+                byte = src[sidx];
+                if (!ascii_check(byte)) {
+                        byte = U_SPECIAL;
+                        result = EIO;
+                }
+
+                rc = chr_encode(byte, dest, &didx, size - 1);
+                if (rc != EOK) {
+                        assert(rc == EOVERFLOW);
+                        dest[didx] = '\0';
+                        return rc;
+                }
+
+                /* Remember dest index after last non-empty character */
+                if (byte != 0x20)
+                        dlast = didx;
+        }
+
+        /* Terminate string after last non-empty character */
+        dest[dlast] = '\0';
+        return result;
 }
 
@@ -572 +640 @@
         dest[dest_off] = '\0';
 }
+
+/** Convert UTF16 string to string.
+ *
+ * Convert utf16 string @a src to string. The output is written to the buffer
+ * specified by @a dest and @a size. @a size must be non-zero and the string
+ * written will always be well-formed. Surrogate pairs also supported.
+ *
+ * @param dest  Destination buffer.
+ * @param size  Size of the destination buffer.
+ * @param src   Source utf16 string.
+ *
+ * @return EOK, if success, negative otherwise.
+ */
+int utf16_to_str(char *dest, size_t size, const uint16_t *src)
+{
+        size_t idx = 0, dest_off = 0;
+        wchar_t ch;
+        int rc = EOK;
+
+        /* There must be space for a null terminator in the buffer. */
+        assert(size > 0);
+
+        while (src[idx]) {
+                if ((src[idx] & 0xfc00) == 0xd800) {
+                        if (src[idx + 1] && (src[idx + 1] & 0xfc00) == 0xdc00) {
+                                ch = 0x10000;
+                                ch += (src[idx] & 0x03FF) << 10;
+                                ch += (src[idx + 1] & 0x03FF);
+                                idx += 2;
+                        }
+                        else
+                                break;
+                } else {
+                        ch = src[idx];
+                        idx++;
+                }
+                rc = chr_encode(ch, dest, &dest_off, size - 1);
+                if (rc != EOK)
+                        break;
+        }
+        dest[dest_off] = '\0';
+        return rc;
+}
+
+int str_to_utf16(uint16_t *dest, size_t size, const char *src)
+{
+        int rc = EOK;
+        size_t offset = 0;
+        size_t idx = 0;
+        wchar_t c;
+
+        assert(size > 0);
+        
+        while ((c = str_decode(src, &offset, STR_NO_LIMIT)) != 0) {
+                if (c > 0x10000) {
+                        if (idx + 2 >= size - 1) {
+                                rc = EOVERFLOW;
+                                break;
+                        }
+                        c = (c - 0x10000);
+                        dest[idx] = 0xD800 | (c >> 10);
+                        dest[idx + 1] = 0xDC00 | (c & 0x3FF);
+                        idx++;
+                } else {
+                         dest[idx] = c;
+                }
+
+                idx++;
+                if (idx >= size - 1) {
+                        rc = EOVERFLOW;
+                        break;
+                }
+        }
+
+        dest[idx] = '\0';
+        return rc;
+}
+
 
 /** Convert wide string to new string.
@@ -652 +798 @@
 
         dest[dlen - 1] = '\0';
+}
+
+/** Convert string to wide string.
+ *
+ * Convert string @a src to wide string. A new wide NULL-terminated
+ * string will be allocated on the heap.
+ *
+ * @param src   Source string.
+ */
+wchar_t *str_to_awstr(const char *str)
+{
+        size_t len = str_length(str);
+        
+        wchar_t *wstr = calloc(len+1, sizeof(wchar_t));
+        if (wstr == NULL)
+                return NULL;
+        
+        str_to_wstr(wstr, len + 1, str);
+        return wstr;
 }
 
@@ -950 +1115 @@
         return dest;
 }
-
 
 /** Convert initial part of string to unsigned long according to given base.