char *xml_string(char *s)
{
static char *buf = NULL;
static int bufsize = 0;
char *p, *sub;
int len, pos = 0;
if (!buf) {
bufsize = 64;
buf = gmalloc(bufsize);
}
p = buf;
while (*s) {
if (pos > (bufsize - 8)) {
bufsize *= 2;
buf = grealloc(buf, bufsize);
p = buf + pos;
}
/* these are safe even if string is already UTF-8 coded
* since UTF-8 strings won't contain '<' or '>' */
if (*s == '<') {
sub = "<";
len = 4;
} else if (*s == '>') {
sub = ">";
len = 4;
} else if (*s == '"') {
sub = """;
len = 6;
} else if (*s == '-') { /* can't be used in xml comment strings */
sub = "-";
len = 5;
} else if (*s == '\'') {
sub = "'";
len = 5;
}
/* escape '&' only if not part of a legal entity sequence */
else if (*s == '&' && !(xml_isentity(s))) {
sub = "&";
len = 5;
} else {
sub = s;
len = 1;
}
while (len--) {
*p++ = *sub++;
pos++;
}
s++;
}
*p = '\0';
return buf;
}
/* a variant of xml_string for urls in hrefs */
char *xml_url_string(char *s)
{
static char *buf = NULL;
static int bufsize = 0;
char *p, *sub;
#if 0
char *prev = NULL;
#endif
int len, pos = 0;
if (!buf) {
bufsize = 64;
buf = gmalloc(bufsize);
}
p = buf;
while (s && *s) {
if (pos > (bufsize - 8)) {
bufsize *= 2;
buf = grealloc(buf, bufsize);
p = buf + pos;
}
/* escape '&' only if not part of a legal entity sequence */
if (*s == '&' && !(xml_isentity(s))) {
sub = "&";
len = 5;
}
/* '<' '>' are safe to substitute even if string is already UTF-8 coded
* since UTF-8 strings won't contain '<' or '>' */
else if (*s == '<') {
sub = "<";
len = 4;
}
else if (*s == '>') {
sub = ">";
len = 4;
}
#if 0
else if (*s == '-') { /* can't be used in xml comment strings */
sub = "-";
len = 5;
}
else if (*s == ' ' && prev && *prev == ' ') {
/* substitute 2nd and subsequent spaces with required_spaces */
sub = " "; /* inkscape doesn't recognise */
len = 6;
}
#endif
else if (*s == '"') {
sub = """;
len = 6;
}
else if (*s == '\'') {
sub = "'";
len = 5;
}
else {
sub = s;
len = 1;
}
while (len--) {
*p++ = *sub++;
pos++;
}
#if 0
prev = s;
#endif
s++;
}
*p = '\0';
return buf;
}
/* html_string is a modified version of xml_string */
char *html_string(char *s)
{
static char *buf = NULL;
static int bufsize = 0;
char *p, *sub, *prev = NULL;
int len, pos = 0;
int temp,cnt,remaining=0;
char workstr[16];
long unsigned int charnum=0;
unsigned char byte;
unsigned char mask;
if (!buf) {
bufsize = 64;
buf = gmalloc(bufsize);
}
p = buf;
while (s && *s) {
if (pos > (bufsize - 8)) {
bufsize *= 2;
buf = grealloc(buf, bufsize);
p = buf + pos;
}
/* escape '&' only if not part of a legal entity sequence */
if (*s == '&' && !(xml_isentity(s))) {
sub = "&";
len = 5;
}
/* '<' '>' are safe to substitute even if string is already UTF-8 coded
* since UTF-8 strings won't contain '<' or '>' */
else if (*s == '<') {
sub = "<";
len = 4;
}
else if (*s == '>') {
sub = ">";
len = 4;
}
else if (*s == '-') { /* can't be used in xml comment strings */
sub = "-";
len = 5;
}
else if (*s == ' ' && prev && *prev == ' ') {
/* substitute 2nd and subsequent spaces with required_spaces */
sub = " "; /* inkscape doesn't recognise */
len = 6;
}
else if (*s == '"') {
sub = """;
len = 6;
}
else if (*s == '\'') {
sub = "'";
len = 5;
}
else if ((unsigned char)*s > 127) {
byte=(unsigned char)*s;
cnt=0;
for (mask=127; mask < byte; mask=mask >>1){
cnt++;
byte=byte & mask;
}
if (cnt>1){
charnum=byte;
remaining=cnt-1;
}else{
charnum=charnum<<6;
charnum+=byte;
remaining--;
}
if (remaining>0){
s++;
continue;
}
/* we will build the html value right-to-left
* (least significant-to-most) */
workstr[15]=';';
sub=&workstr[14];
len=3; /* &# + ; */
do {
temp=charnum%10;
*(sub--)=(char)((int)'0'+ temp);
charnum/=10;
len++;
if (len>12){ /* 12 is arbitrary, but clearly in error */
fprintf(stderr, "Error during conversion to \"UTF-8\". Quiting.\n");
exit(1);
}
} while (charnum>0);
*(sub--)='#';
*(sub)='&';
}
else {
