DO NOT REPLY TO THIS MESSAGE. INSTEAD, POST ANY RESPONSES TO THE LINK BELOW.
[STR New]
Link: http://www.fltk.org/str.php?L2348
Version: 1.3-current
Link: http://www.fltk.org/str.php?L2348
Version: 1.3-current
Index: src/Fl_Text_Buffer.cxx
===================================================================
--- src/Fl_Text_Buffer.cxx (revision 7860)
+++ src/Fl_Text_Buffer.cxx (working copy)
@@ -1024,7 +1024,7 @@
*foundPos = startPos;
return 1;
}
- int l = fl_utf8len(c);
+ int l = fl_utf8len(c); // TODO: replace?
if (memcmp(sp, address(bp), l))
break;
sp += l; bp += l;
@@ -1076,7 +1076,7 @@
*foundPos = startPos;
return 1;
}
- int l = fl_utf8len(c);
+ int l = fl_utf8len(c); // TODO: replace?
if (memcmp(sp, address(bp), l))
break;
sp += l; bp += l;
@@ -1560,6 +1560,29 @@
/*
+ * temporary functions for debug purposes
+ */
+void printByte(const char byte) {
+ if (' ' <= byte && byte <= '~')
+ printf(" %c", byte);
+ else
+ printf("0x%02hhx", byte);
+}
+void printBytes(int pos, const char byte) {
+ printf("pos = %d byte = [", pos);
+ printByte(byte);
+ printf("]\n");
+}
+void printBytes(int pos, const char* bytes, int n) {
+ printf("pos = %d bytes = [", pos);
+ for (int i = 0; i < n && bytes[i] != '\0'; i++) {
+ printByte(bytes[i]);
+ printf(",");
+ }
+ printf("]\n");
+}
+
+/*
Return the previous character position.
Uncode safe.
*/
@@ -1570,6 +1593,15 @@
IS_UTF8_ALIGNED2(this, (pos))
+ char bytes[10]; memset(bytes, '\0', 10); // longest UTF-8 sequence is 6
bytes
+ int b = pos;
+ for (int i = 0; i < 10 && b > 0; i++, b--)
+ bytes[10-1-i] = byte_at(pos-i);
+ const char* p = &bytes[10-1];
+ const char* q = fl_utf8back(p-1, &bytes[b], p);
+ int d = p - q;
+ pos = pos - d;
+/*
char c;
do {
pos--;
@@ -1577,7 +1609,7 @@
return 0;
c = byte_at(pos);
} while ( (c&0xc0) == 0x80);
-
+*/
IS_UTF8_ALIGNED2(this, (pos))
return pos;
}
@@ -1593,13 +1625,23 @@
return prev_char_clipped(pos);
}
-
/*
Return the next character position.
Returns length() if the end of the buffer is reached.
*/
int Fl_Text_Buffer::next_char(int pos) const
{
+ char bytes[10]; memset(bytes, '\0', 10); // longest UTF-8 sequence is 6
bytes
+ for (int i = 0; i < 10 && pos+i < mLength; i++)
+ bytes[i+0] = byte_at(pos+i);
+ const char *p = &bytes[0];
+ const char *q = fl_utf8fwd(p+1, p, p+6);
+ int d = q - p;
+ pos += d;
+ if (pos >= mLength)
+ return mLength;
+ return pos;
+ /*
IS_UTF8_ALIGNED2(this, (pos))
int n = fl_utf8len(byte_at(pos));
pos += n;
@@ -1607,6 +1649,7 @@
return mLength;
IS_UTF8_ALIGNED2(this, (pos))
return pos;
+ */
}
@@ -1624,14 +1667,53 @@
*/
int Fl_Text_Buffer::utf8_align(int pos) const
{
+ char bytes[10]; memset(bytes, '\0', 10); // longest UTF-8 sequence is 6
bytes
+ int b = pos;
+ for (int i = 0; i < 10 && b > 0; i++, b--)
+ bytes[10-1-i] = byte_at(pos-i);
+ const char* p = &bytes[10-1];
+ const char* q = fl_utf8back(p, &bytes[b], p);
+ int d = p - q;
+ return (d == 0) ? pos : pos-d;
+/*
char c = byte_at(pos);
while ( (c&0xc0) == 0x80) {
pos--;
c = byte_at(pos);
}
return pos;
+*/
}
+/*
+ * temporary routines for use in IS_UTF8_ALIGN macros
+ */
+int isUtf8Aligned(const char* s) {
+ if (s && *s && fl_utf8len(*s)<=0) {
+ const char* p = fl_utf8fwd(s, s, s+6);
+ if (p == s)
+ return 1;
+ return 0;
+ }
+ return 1;
+}
+int isUtf8Aligned(const Fl_Text_Buffer* tb, int pos) {
+ if (pos < 0 || pos >= tb->length())
+ return 1;
+
+ char bytes[10]; memset(bytes, '\0', 10); // longest UTF-8 sequence is 6
bytes
+ int len = min(10-1, tb->length() - pos);
+ for (int i = 0; i < len; i++)
+ bytes[i] = tb->byte_at(pos+i);
+ const char* p = &bytes[0];
+ const char* q = fl_utf8fwd(p, p, p+6);
+ if (p == q)
+ return 1;
+ if (pos>=0 && pos<tb->length() && fl_utf8len(tb->byte_at(pos))<=0)
+ return 0;
+ return 1;
+}
+
//
// End of "$Id$".
//
Index: src/Fl_Text_Display.cxx
===================================================================
--- src/Fl_Text_Display.cxx (revision 7860)
+++ src/Fl_Text_Display.cxx (working copy)
@@ -84,7 +84,35 @@
// CET - FIXME
#define TMPFONTWIDTH 6
+/* fl_utf8len(char b) replacement that handles CP1252 C1 control chars, etc.
+ * temporary name to aid finding and replacing the originals
+ * temporary location to minimize file changes
+ */
+int fl_drg8len(const char* s)
+{
+ int n = 0;
+ unsigned ucs = fl_utf8decode(s, 0, &n);
+ ucs = ucs + 1; // keep g++ quiet
+ return n;
+}
+/* even more temporary version with added debug
+ */
+int fl_drg8len(const char* s, const char* f)
+{
+ int m = fl_utf8len(*s);
+ int n = fl_drg8len(s);
+ if (m != n) {
+ // printf("%s: utf8len=%d drg8len=%d ", f, m, n);
+ // for (int i = 0; i < max(m,n); i++) {
+ // printByte(s[i]);
+ // printf(",");
+ // }
+ // printf("\n");
+ }
+ return n;
+}
+
/**
\brief Creates a new text display widget.
@@ -757,10 +785,20 @@
/* determine how many displayed character positions are covered */
startIndent = mBuffer->count_displayed_characters( lineStart, startPos );
indent = startIndent;
+
for ( c = text; *c != '\0'; c += fl_utf8len(*c) )
indent++;
endIndent = indent;
+ // the following trumps the block above
+ int drgIndent = startIndent;
+ for ( c = text; *c != '\0'; c += fl_drg8len(c, "overstrike") )
+ drgIndent++;
+ // if (drgIndent != indent)
+ // printf("drgIndent = %d indent = %d\n", drgIndent, indent);
+ indent = drgIndent;
+ endIndent = indent;
+
/* find which characters to remove, and if necessary generate additional
padding to make up for removed control characters at the end */
indent = startIndent;
@@ -1742,6 +1780,9 @@
style = position_style(lineStartPos, lineLen, 0);
for (i=0; i<lineLen; ) {
int len = fl_utf8len(lineStr[i]);
+ int drgLen = fl_drg8len(&lineStr[i], "handle_vline");
+ len = drgLen;
+
charStyle = position_style(lineStartPos, lineLen, i);
// FIXME: if the character is a tab, we need to do the correct indenting
// FIXME: if the character is an optional hyphen, we need to ignore it
unless we wrap the text
@@ -1809,6 +1850,8 @@
int i = 0;
while (i<len) {
int cl = fl_utf8len(s[i]);
+ int drgLen = fl_drg8len(&s[i], "find_x");
+ cl = drgLen;
int w = int( string_width(s, i+cl, style) );
if (w>x)
return i;
@@ -1867,6 +1910,23 @@
}
+/*
+ * expand CP1252(?) src string to full UTF-8 representation in dst
+ * temporary name and location
+ */
+int drgExpand(const char* src, int srcBytes, char* dst) {
+ int len = 0, dstBytes = 0;
+ unsigned ucs;
+ char buffer[10]; // max utf-8 sequence is 6 bytes
+ char* p = (char*)src;
+ for (int i = 0; i < srcBytes; i += len, p += len) { // TODO: assert(len !=
1) ?
+ ucs = fl_utf8decode(p, &src[srcBytes], &len);
+ int utf8len = fl_utf8encode(ucs, buffer);
+ for (int j = 0; j < utf8len; j++)
+ dst[dstBytes++] = buffer[j];
+ }
+ return dstBytes;
+}
/**
\brief Draw a text segment in a single style.
@@ -1944,7 +2004,13 @@
if (!(style & BG_ONLY_MASK)) {
fl_color( foreground );
fl_font( font, fsize );
- fl_draw( string, nChars, X, Y + mMaxsize - fl_descent());
+ // fl_draw( string, nChars, X, Y + mMaxsize - fl_descent());
+
+ // need to expand to full UTF-8 or fl_draw() won't work as expected
+ char* expanded = (char*)malloc(nChars*6*sizeof(char));
+ int numBytes = drgExpand(string, nChars, expanded);
+ fl_draw(expanded, numBytes, X, Y + mMaxsize - fl_descent());
+ free((void*)expanded);
}
// CET - FIXME
@@ -2158,7 +2224,13 @@
}
fl_font( font, fsize );
- return fl_width( string, length );
+ // need to expand to full UTF-8 or fl_width() won't work as expected
+ char* expanded = (char*)malloc(length*6*sizeof(char));
+ int numBytes = drgExpand(string, length, expanded);
+ double width = fl_width(expanded, numBytes);
+ free((void*)expanded);
+
+ return width;
}
@@ -3175,6 +3247,8 @@
IS_UTF8_ALIGNED(s)
int charLen = fl_utf8len(*s), style = 0;
+ int drgLen = fl_drg8len(s, "measure_prop_char");
+ charLen = drgLen;
if (mStyleBuffer) {
style = mStyleBuffer->byte_at(pos);
}
@@ -3253,6 +3327,7 @@
return 1;
c = buffer()->char_at(lineEndPos);
+ // TODO: if c is ucs, then fl_utf8len(c) is just plain wrong! no?
return c == '\n' || ((c == '\t' || c == ' ') &&
lineEndPos + fl_utf8len(c) < buffer()->length());
}
Index: FL/Fl_Text_Buffer.H
===================================================================
--- FL/Fl_Text_Buffer.H (revision 7860)
+++ FL/Fl_Text_Buffer.H (working copy)
@@ -38,8 +38,18 @@
#ifdef ASSERT_UTF8
# include <assert.h>
-# define IS_UTF8_ALIGNED(a) if (a && *a) assert(fl_utf8len(*(a))>0);
-# define IS_UTF8_ALIGNED2(a, b) if (b>=0 && b<a->length())
assert(fl_utf8len(a->byte_at(b))>0);
+// define IS_UTF8_ALIGNED(a) if (a && *a) assert(fl_utf8len(*(a))>0);
+// define IS_UTF8_ALIGNED2(a, b) if (b>=0 && b<a->length())
assert(fl_utf8len(a->byte_at(b))>0);
+ // replace body of macros with functions to allow extra debugging.
+ // temporary location for forward declaration and externs.
+ class Fl_Text_Buffer;
+ extern void printByte(const char byte);
+ extern void printBytes(int pos, const char byte);
+ extern void printBytes(int pos, const char* bytes, int n);
+ extern int isUtf8Aligned(const char* s);
+ extern int isUtf8Aligned(const Fl_Text_Buffer* tb, int pos);
+# define IS_UTF8_ALIGNED(a) assert(isUtf8Aligned(a));
+# define IS_UTF8_ALIGNED2(a, b) assert(isUtf8Aligned(a,b));
#else
# define IS_UTF8_ALIGNED(a)
# define IS_UTF8_ALIGNED2(a, b)
_______________________________________________
fltk-bugs mailing list
[email protected]
http://lists.easysw.com/mailman/listinfo/fltk-bugs
