--- htmltextslave.c 2003-06-12 17:02:18.000000000 +0800
+++ htmltextslave.c.1.1 2003-06-12 17:18:21.000000000 +0800
@@ -35,9 +35,147 @@
#include "htmlplainpainter.h"
#include "htmlgdkpainter.h"
-
/* #define HTML_TEXT_SLAVE_DEBUG */
+/*
+ * Added by zhancb 2003-6-6,char of puncutation should be avoid breaking
+ */
+
+// Define a struct for UTF8 code
+struct _BreakUTF8Code
+{
+ gint xCode ;
+ gint yCode ;
+ gint zCode ;
+} ;
+typedef struct _BreakUTF8Code BreakUTF8Code ;
+
+// The below is the puntuation should be broken after(UTF-8),total 18 puncs
+// £¬¡£¡¢£¡£©£º£»¡±¡¯£¿¡µ¡»¡·¡¹¡½¡¿£Ý£ý
+static BreakUTF8Code cBreakAfter[] =
+ {
+ { 0Xffffffef, 0Xffffffbc, 0Xffffff8c } ,
+ { 0Xffffffe3, 0Xffffff80, 0Xffffff82 } ,
+ { 0Xffffffe3, 0Xffffff80, 0Xffffff81 } ,
+ { 0Xffffffef, 0Xffffffbc, 0Xffffff81 } ,
+ { 0Xffffffef, 0Xffffffbc, 0Xffffff89 } ,
+ { 0Xffffffef, 0Xffffffbc, 0Xffffff9a } ,
+ { 0Xffffffef, 0Xffffffbc, 0Xffffff9b } ,
+ { 0Xffffffe2, 0Xffffff80, 0Xffffff9d } ,
+ { 0Xffffffe2, 0Xffffff80, 0Xffffff99 } ,
+ { 0Xffffffef, 0Xffffffbc, 0Xffffff9f } ,
+ { 0Xffffffe3, 0Xffffff80, 0Xffffff89 } ,
+ { 0Xffffffe3, 0Xffffff80, 0Xffffff8f } ,
+ { 0Xffffffe3, 0Xffffff80, 0Xffffff8b } ,
+ { 0Xffffffe3, 0Xffffff80, 0Xffffff8d } ,
+ { 0Xffffffe3, 0Xffffff80, 0Xffffff97 } ,
+ { 0Xffffffe3, 0Xffffff80, 0Xffffff91 } ,
+ { 0Xffffffef, 0Xffffffbc, 0Xffffffbd } ,
+ { 0Xffffffef, 0Xffffffbd, 0Xffffff9d }
+ };
+
+// The below is the puntuation should be broken before(UTF-8),totally 11 puncs
+// ¡°¡®£¨¡¶¡´¡¸¡º¡¼¡¾£Û£û
+static BreakUTF8Code cBreakBefore[] =
+ {
+ { 0Xffffffe2, 0Xffffff80, 0Xffffff9c } ,
+ { 0Xffffffe2, 0Xffffff80, 0Xffffff98 } ,
+ { 0Xffffffef, 0Xffffffbc, 0Xffffff88 } ,
+ { 0Xffffffe3, 0Xffffff80, 0Xffffff8a } ,
+ { 0Xffffffe3, 0Xffffff80, 0Xffffff88 } ,
+ { 0Xffffffe3, 0Xffffff80, 0Xffffff8c } ,
+ { 0Xffffffe3, 0Xffffff80, 0Xffffff8e } ,
+ { 0Xffffffe3, 0Xffffff80, 0Xffffff96 } ,
+ { 0Xffffffe3, 0Xffffff80, 0Xffffff90 } ,
+ { 0Xffffffef, 0Xffffffbc, 0Xffffffbb } ,
+ { 0Xffffffef, 0Xffffffbd, 0Xffffff9b }
+ };
+
+// break type of the char
+typedef enum
+{
+ BREAK_NONE ,
+ BREAK_FREE ,
+ BREAK_AFTER ,
+ BREAK_BEFORE
+} BreakType ;
+
+// The function to check if a Chinese word could be broken.
+// Param: if xByte < 0,yByte and zByte is the last two code of a Chinese word;
+// if xByte >= 0,yByte is the prev char of xByte and zByte is next char
+// Return: break type of the char
+BreakType ChkBreakType( gchar xByte, gchar yByte, gchar zByte )
+{
+
+ if ( xByte < 0 ) // In UTF-8 and under Linux,Chinese text code less than 0
+ {
+ int i = 0 ;
+ for ( ; i < 18; ++i )
+ {
+ if ( cBreakAfter[i].xCode == xByte && cBreakAfter[i].yCode == yByte
+ && cBreakAfter[i].zCode == zByte )
+ {
+ return BREAK_AFTER ;
+ }
+ }
+
+ for ( i = 0; i < 11; ++i )
+ {
+ if ( cBreakBefore[i].xCode == xByte && cBreakBefore[i].yCode == yByte
+ && cBreakBefore[i].zCode == zByte )
+ {
+ return BREAK_BEFORE ;
+ }
+ }
+
+ // Other Chinese words could be broken free
+ return BREAK_FREE ;
+ }
+ else
+ {
+ if ( xByte == 0 || xByte == -1 ) return BREAK_FREE ;
+
+ if ( ( xByte >='a' && xByte <= 'z' ) || ( xByte >='A' || xByte <='Z' ) || ( xByte >='0' && xByte <= '9' ) )
+ {
+ if ( ( yByte == ' ' || yByte == '\t' )
+ && ( zByte == ' ' || zByte == '\t' ) )
+ {
+ return BREAK_FREE ;
+ }
+ else if ( yByte == ' ' || yByte == '\t' )
+ {
+ return BREAK_BEFORE ;
+ }
+ else if ( zByte == ' ' || zByte == '\t' )
+ {
+ return BREAK_AFTER ;
+ }
+ else
+ {
+ return BREAK_NONE ;
+ }
+ }
+ else if ( '!' == xByte || '%' == xByte || ',' == xByte || '.' == xByte
+ || ']' == xByte || '}' == xByte || ';' == xByte || ':' == xByte
+ || '?' == xByte || ')' == xByte)
+ {
+ return BREAK_AFTER ;
+ }
+ else if ( '[' == xByte || '{' == xByte || '(' == xByte )
+ {
+ return BREAK_BEFORE ;
+ }
+ else
+ {
+ return BREAK_FREE ;
+ }
+ }
+}
+
+/*
+ * Added End
+ */
+
HTMLTextSlaveClass html_text_slave_class;
static HTMLObjectClass *parent_class = NULL;
@@ -112,7 +250,7 @@
space = str;
while (words && *space && len < slave->posLen) {
- if (*space == ' ')
+ if (*space == ' ' || *space < 0/*Added by zhancb,new condition*/)
words --;
if (words) {
@@ -123,7 +261,7 @@
/* printf ("width %d --> ", width); */
width += html_painter_get_space_width (p, html_text_get_font_style (text), text->face)*(html_text_text_line_length (str, &line_offset, len, &tabs) - len);
- /* printf ("%d\n", width); */
+ /* printf ("In get_words_width(),second width is%d\n", width); */
}
return width;
@@ -224,7 +362,7 @@
str = sep = buffer;
while (off < len && *sep) {
- if (*sep == ' ')
+ if (*sep == ' ' || *sep < 0/*Added by zhancb,a new condiftion*/)
(*words) ++;
sep = g_utf8_next_char (sep);
@@ -330,7 +468,6 @@
for (i = 0; i < len; i++)
putchar (text [i]);
- printf ("'\n");
}
#endif
@@ -371,11 +508,12 @@
gint width = 0;
if (HTML_TEXT (slave->owner)->text_len == 0
- || html_text_get_char (HTML_TEXT (slave->owner), slave->posStart + slave->posLen - 1) != ' ') {
+ || (html_text_get_char (HTML_TEXT (slave->owner), slave->posStart + slave->posLen - 1) != ' '
+ && html_text_get_char (HTML_TEXT (slave->owner), slave->posStart + slave->posLen - 1) >= 0)/*Added by zhancb,a new and condition*/) {
HTMLObject *obj;
obj = html_object_next_not_slave (HTML_OBJECT (slave));
if (obj && html_object_is_text (obj)
- && html_text_get_char (HTML_TEXT (obj), 0) != ' ')
+ && html_text_get_char (HTML_TEXT (obj), 0) != ' ' && html_text_get_char (HTML_TEXT (obj), 0) > 0/*Added by zhancb,a new and condition*/ )
width = html_text_get_nb_width (HTML_TEXT (obj), painter, TRUE);
}
@@ -457,7 +595,6 @@
begin = html_text_slave_remove_leading_space (slave, painter, lineBegin);
- /* printf ("fit_line %d left: %d lspacetext: \"%s\"\n", firstRun, widthLeft, begin); */
prev = html_object_prev_not_slave (HTML_OBJECT (text));
forceFit = orig_start_word == slave->start_word
@@ -473,15 +610,27 @@
}
words ++;
- while (*sep && *sep != ' ') {
+
+ /*
+ *Modifed by zhancb 2003-6-1,For line break of Chinese Text,now I limited the line width to 60,but didn't work.
+ *The original code is: while (*sep && *sep != ' ')
+ */
+ while (*sep && *sep != ' ' && *sep >= 0)
+ /*
+ *Modified End
+ */
+ {
sep = g_utf8_next_char (sep);
pos++;
}
if (words + slave->start_word >= text->words)
+ {
break;
}
+ }
+
if (words + slave->start_word == text->words)
rv = HTML_FIT_COMPLETE;
else if (words == 0 || get_words_width (slave, painter, words) == 0) {
@@ -506,7 +655,8 @@
words ++;
- while (*sep && *sep != ' ') {
+ while (*sep && *sep != ' ')
+ {
sep = g_utf8_next_char (sep);
pos ++;
}
@@ -514,9 +664,58 @@
}
if (rv == HTML_FIT_PARTIAL)
+ {
+ // Added by zhancb 2003-6-8
+ // Check if current char could be broken after and next char could be broken before
+ if ( *sep < 0 )
+ {
+ gchar xByte, yByte, zByte ;
+ gchar xPrevByte, yPrevByte, zPrevByte ;
+
+ zPrevByte = -1 ;
+
+ while (1)
+ {
+
+ xByte = *sep ;
+ xPrevByte = *g_utf8_prev_char( sep ) ;
+ if ( *sep < 0 )
+ {
+ yByte = *(sep + 1 ) ;
+ zByte = *(sep + 2 ) ;
+ }
+ else
+ {
+ yByte = *g_utf8_prev_char( sep ) ;
+ zByte = *g_utf8_next_char( sep ) ;
+ }
+ if ( xPrevByte < 0 )
+ {
+ yPrevByte = *(g_utf8_prev_char( sep ) + 1) ;
+ zPrevByte = *(g_utf8_prev_char( sep ) + 2 ) ;
+ }
+ else
+ {
+ yPrevByte = *g_utf8_prev_char(g_utf8_prev_char( sep ) ) ;
+ zPrevByte = *sep ;
+ }
+
+ if ( ( (ChkBreakType( xByte, yByte, zByte ) == BREAK_BEFORE || ChkBreakType( xByte, yByte, zByte ) == BREAK_FREE ) &&
+ (ChkBreakType( xPrevByte, yPrevByte, zPrevByte ) == BREAK_AFTER || ChkBreakType( xPrevByte, yPrevByte, zPrevByte ) == BREAK_FREE) )
+ || 0 == pos )
+ break ;
+
+ sep = g_utf8_prev_char( sep ) ;
+ pos -- ;
+ }
+ }
+ // Added End
+
if (pos < slave->posLen) {
split (slave, pos, slave->start_word + words, *sep ? sep : NULL);
o->width = get_words_width (slave, painter, words);
+
+ }
}
#ifdef HTML_TEXT_SLAVE_DEBUG