/**
* xmlEncodeEntitiesReentrant:
* @doc: the document containing the string
* @input: A string to convert to XML.
*
* Do a global encoding of a string, replacing the predefined entities
* and non ASCII values with their entities and CharRef counterparts.
* Contrary to xmlEncodeEntities, this routine is reentrant, and result
* must be deallocated.
*
* Returns A newly allocated string with the substitution done.
*/
xmlChar *
xmlEncodeEntitiesReentrant(xmlDocPtr doc, const xmlChar *input) {
const xmlChar *cur = input;
xmlChar *buffer = NULL;
xmlChar *out = NULL;
int buffer_size = 0;
int html = 0;
if (input == NULL) return(NULL);
if (doc != NULL)
html = (doc->type == XML_HTML_DOCUMENT_NODE);
/*
* allocate an translation buffer.
*/
buffer_size = 1000;
buffer = (xmlChar *) xmlMalloc(buffer_size * sizeof(xmlChar));
if (buffer == NULL) {
xmlEntitiesErrMemory("xmlEncodeEntitiesReentrant: malloc failed");
return(NULL);
}
out = buffer;
while (*cur != '\0') {
if (out - buffer > buffer_size - 100) {
int indx = out - buffer;
growBufferReentrant();
out = &buffer[indx];
}
/*
* By default one have to encode at least '<', '>', '"' and '&' !
*/
if (*cur == '<') {
*out++ = '&';
*out++ = 'l';
*out++ = 't';
*out++ = ';';
} else if (*cur == '>') {
*out++ = '&';
*out++ = 'g';
*out++ = 't';
*out++ = ';';
} else if (*cur == '&') {
*out++ = '&';
*out++ = 'a';
*out++ = 'm';
*out++ = 'p';
*out++ = ';';
} else if (((*cur >= 0x20) && (*cur < 0x80)) ||
(*cur == '\n') || (*cur == '\t') || ((html) && (*cur == '\r'))) {
/*
* default case, just copy !
*/
*out++ = *cur;
} else if (*cur >= 0x80) {
if (((doc != NULL) && (doc->encoding != NULL)) || (html)) {
/*
* Bjørn Reese <*****@*****.**> provided the patch
xmlChar xc;
xc = (*cur & 0x3F) << 6;
if (cur[1] != 0) {
xc += *(++cur) & 0x3F;
*out++ = xc;
} else
*/
*out++ = *cur;
} else {
/*
* We assume we have UTF-8 input.
*/
char buf[11], *ptr;
int val = 0, l = 1;
if (*cur < 0xC0) {
xmlEntitiesErr(XML_CHECK_NOT_UTF8,
"xmlEncodeEntitiesReentrant : input not UTF-8");
if (doc != NULL)
doc->encoding = xmlStrdup(BAD_CAST "ISO-8859-1");
snprintf(buf, sizeof(buf), "&#%d;", *cur);
buf[sizeof(buf) - 1] = 0;
ptr = buf;
while (*ptr != 0) *out++ = *ptr++;
cur++;
continue;
} else if (*cur < 0xE0) {
val = (cur[0]) & 0x1F;
val <<= 6;
val |= (cur[1]) & 0x3F;
l = 2;
} else if (*cur < 0xF0) {
val = (cur[0]) & 0x0F;
val <<= 6;
val |= (cur[1]) & 0x3F;
val <<= 6;
val |= (cur[2]) & 0x3F;
l = 3;
} else if (*cur < 0xF8) {
val = (cur[0]) & 0x07;
val <<= 6;
val |= (cur[1]) & 0x3F;
val <<= 6;
val |= (cur[2]) & 0x3F;
val <<= 6;
val |= (cur[3]) & 0x3F;
l = 4;
}
if ((l == 1) || (!IS_CHAR(val))) {
xmlEntitiesErr(XML_ERR_INVALID_CHAR,
"xmlEncodeEntitiesReentrant : char out of range\n");
if (doc != NULL)
doc->encoding = xmlStrdup(BAD_CAST "ISO-8859-1");
snprintf(buf, sizeof(buf), "&#%d;", *cur);
buf[sizeof(buf) - 1] = 0;
ptr = buf;
while (*ptr != 0) *out++ = *ptr++;
cur++;
continue;
}
/*
* We could do multiple things here. Just save as a char ref
*/
if (html)
snprintf(buf, sizeof(buf), "&#%d;", val);
else
snprintf(buf, sizeof(buf), "&#x%X;", val);
buf[sizeof(buf) - 1] = 0;
ptr = buf;
while (*ptr != 0) *out++ = *ptr++;
cur += l;
continue;
}
} else if (IS_BYTE_CHAR(*cur)) {
char buf[11], *ptr;
snprintf(buf, sizeof(buf), "&#%d;", *cur);
buf[sizeof(buf) - 1] = 0;
ptr = buf;
while (*ptr != 0) *out++ = *ptr++;
}
cur++;
}
*out++ = 0;
return(buffer);
}
/**
* xmlEncodeEntitiesReentrant:
* @param doc the document containing the string
* @param input A string to convert to XML.
*
* Do a global encoding of a string, replacing the predefined entities
* and non ASCII values with their entities and CharRef counterparts.
* Contrary to xmlEncodeEntities, this routine is reentrant, and result
* must be deallocated.
*
* Returns A newly allocated string with the substitution done.
*
* OOM: possible --> returns NULL (for input!=NULL), sets OOM flag
*/
XMLPUBFUNEXPORT xmlChar*
xmlEncodeEntitiesReentrant(xmlDocPtr doc, const xmlChar *input) {
const xmlChar* cur = input;
xmlChar* buffer = NULL;
xmlChar* out = NULL;
int buffer_size;
int html;
LOAD_GS_SAFE_DOC(doc)
if (input == NULL)
return(NULL);
html = doc && (doc->type == XML_HTML_DOCUMENT_NODE);
/*
* allocate an translation buffer.
*/
buffer_size = 1000;
buffer = (xmlChar *) xmlMalloc(buffer_size * sizeof(xmlChar));
if (buffer == NULL) {
xmlGenericError(xmlGenericErrorContext, EMBED_ERRTXT("malloc failed\n"));
return(NULL);
}
out = buffer;
while (*cur != '\0') {
if (out - buffer > buffer_size - 100) {
xmlChar* newbuf;
int indx = out - buffer;
newbuf = (xmlChar*)xmlGrowBufferReentrant(&buffer_size, buffer); // on OOM returns NULL (buffer is not freed)
if(!buffer)
{
xmlFree(buffer);
return NULL;
}
buffer = newbuf;
out = &buffer[indx];
}
/*
* By default one have to encode at least '<', '>', '"' and '&' !
*/
if (*cur == '<') {
*out++ = '&';
*out++ = 'l';
*out++ = 't';
*out++ = ';';
} else if (*cur == '>') {
*out++ = '&';
*out++ = 'g';
*out++ = 't';
*out++ = ';';
} else if (*cur == '&') {
*out++ = '&';
*out++ = 'a';
*out++ = 'm';
*out++ = 'p';
*out++ = ';';
} else if (((*cur >= 0x20) && (*cur < 0x80)) ||
(*cur == '\n') || (*cur == '\t') || ((html) && (*cur == '\r'))) {
/*
* default case, just copy !
*/
*out++ = *cur;
} else if (*cur >= 0x80) {
if (((doc != NULL) && (doc->encoding != NULL)) || (html)) {
/*
* Bjorn Reese <*****@*****.**> provided the patch
xmlChar xc;
xc = (*cur & 0x3F) << 6;
if (cur[1] != 0) {
xc += *(++cur) & 0x3F;
*out++ = xc;
} else
*/
*out++ = *cur;
} else {
/*
* We assume we have UTF-8 input.
*/
char buf[11], *ptr;
// DONE: rename 'l' variable -- hard to understand and error-prone otherwise (looks like '1')
int val = 0, len = 1; // 'l' became 'len'
if (*cur < 0xC0) {
xmlGenericError(xmlGenericErrorContext, EMBED_ERRTXT("xmlEncodeEntitiesReentrant : input not UTF-8\n"));
if (doc != NULL){
doc->encoding = xmlStrdup(BAD_CAST "ISO-8859-1");
if(OOM_FLAG)
goto OOM;
}
snprintf(buf, sizeof(buf), "&#%d;", *cur);
buf[sizeof(buf) - 1] = 0;
ptr = buf;
while (*ptr != 0)
{
*out++ = *ptr++;
}
cur++;
continue;
} else if (*cur < 0xE0) {
val = (cur[0]) & 0x1F;
val <<= 6;
val |= (cur[1]) & 0x3F;
len = 2;
} else if (*cur < 0xF0) {
val = (cur[0]) & 0x0F;
val <<= 6;
val |= (cur[1]) & 0x3F;
val <<= 6;
val |= (cur[2]) & 0x3F;
len = 3;
} else if (*cur < 0xF8) {
val = (cur[0]) & 0x07;
val <<= 6;
val |= (cur[1]) & 0x3F;
val <<= 6;
val |= (cur[2]) & 0x3F;
val <<= 6;
val |= (cur[3]) & 0x3F;
len = 4;
}
if ((len == 1) || (!IS_CHAR(val))) {
xmlGenericError(xmlGenericErrorContext, EMBED_ERRTXT("xmlEncodeEntitiesReentrant : char out of range\n"));
if (doc != NULL){
doc->encoding = xmlStrdup(BAD_CAST "ISO-8859-1");
if(OOM_FLAG)
goto OOM;
}
// 2-->
snprintf(buf, sizeof(buf), "&#%d;", *cur);
buf[sizeof(buf) - 1] = 0;
ptr = buf;
while (*ptr != 0)
{
*out++ = *ptr++;
}
cur++;
continue;
// <--2
}
/*
* We could do multiple things here. Just save as a char ref
*/
// 3-->
if (html)
snprintf(buf, sizeof(buf), "&#%d;", val);
else
snprintf(buf, sizeof(buf), "&#x%X;", val);
buf[sizeof(buf) - 1] = 0;
ptr = buf;
while (*ptr != 0)
{
*out++ = *ptr++;
}
cur += len;
continue;
// <--3
}
} else if (IS_BYTE_CHAR(*cur)) {
char buf[11], *ptr;
// 4-->
snprintf(buf, sizeof(buf), "&#%d;", *cur);
buf[sizeof(buf) - 1] = 0;
ptr = buf;
while (*ptr != 0)
{
*out++ = *ptr++;
}
}
cur++;
// continue; is implied here
// <--4
} // while (*cur != '\0')
*out++ = 0;
return(buffer);
OOM:
xmlFree(buffer);
return NULL;
}
/**
* xmlEncodeEntitiesReentrant:
* @doc: the document containing the string
* @input: A string to convert to XML.
*
* Do a global encoding of a string, replacing the predefined entities
* and non ASCII values with their entities and CharRef counterparts.
* Contrary to xmlEncodeEntities, this routine is reentrant, and result
* must be deallocated.
*
* Returns A newly allocated string with the substitution done.
*
* OOM: possible --> returns NULL (for input!=NULL), sets OOM flag
*/
xmlChar*
xmlEncodeEntitiesReentrant(xmlDocPtr doc, const xmlChar *input) {
const xmlChar* cur = input;
xmlChar* buffer = NULL;
xmlChar* out = NULL;
int buffer_size;
int html;
if (input == NULL)
return(NULL);
html = doc && (doc->type == XML_HTML_DOCUMENT_NODE);
/*
* allocate an translation buffer.
*/
// TODO: "Magic number" for buffer size - make a controlled parameter
// TODO: OPTIMIZE: Select most appropriate buffer size by default / reuse some preallocated buffer
buffer_size = 1000;
buffer = (xmlChar *) xmlMalloc(buffer_size * sizeof(xmlChar));
if (buffer == NULL) {
SET_OOM_FLAG;
xmlGenericError(xmlGenericErrorContext, EMBED_ERRTXT("malloc failed\n"));
return(NULL);
}
out = buffer;
while (*cur != '\0') {
// TODO: Another "magic" number -- select optimal for current buffer_size
if (out - buffer > buffer_size - 100) {
xmlChar* newbuf;
int indx = out - buffer;
newbuf = (xmlChar*)xmlGrowBufferReentrant(&buffer_size, buffer); // on OOM returns NULL (buffer is not freed)
if(!buffer)
{
xmlFree(buffer);
return NULL;
}
buffer = newbuf;
out = &buffer[indx];
}
/*
* By default one have to encode at least '<', '>', '"' and '&' !
*/
// TODO: OPTIMIZE: Rearrange IFs according expectations of conditions
// Obviously, the default case is the most expected
if (*cur == '<') {
*out++ = '&';
*out++ = 'l';
*out++ = 't';
*out++ = ';';
} else if (*cur == '>') {
*out++ = '&';
*out++ = 'g';
*out++ = 't';
*out++ = ';';
} else if (*cur == '&') {
*out++ = '&';
*out++ = 'a';
*out++ = 'm';
*out++ = 'p';
*out++ = ';';
} else if (((*cur >= 0x20) && (*cur < 0x80)) ||
(*cur == '\n') || (*cur == '\t') || ((html) && (*cur == '\r'))) {
/*
* default case, just copy !
*/
*out++ = *cur;
} else if (*cur >= 0x80) {
if (((doc != NULL) && (doc->encoding != NULL)) || (html)) {
/*
* Bj?rn Reese <*****@*****.**> provided the patch
xmlChar xc;
xc = (*cur & 0x3F) << 6;
if (cur[1] != 0) {
xc += *(++cur) & 0x3F;
*out++ = xc;
} else
*/
*out++ = *cur;
} else {
/*
* We assume we have UTF-8 input.
*/
char buf[11], *ptr;
// TODO: rename 'l' variable -- hard to understand and error-prone otherwise (looks like '1')
int val = 0, l = 1;
if (*cur < 0xC0) {
xmlGenericError(xmlGenericErrorContext, EMBED_ERRTXT("xmlEncodeEntitiesReentrant : input not UTF-8\n"));
if (doc != NULL){
doc->encoding = xmlStrdup(BAD_CAST "ISO-8859-1");
if(OOM_FLAG)
goto OOM;
}
// TODO: Detect OOM..
// TODO: these several lines repeat four times in WHILE loop -- try to combine and use GOTO
snprintf(buf, sizeof(buf), "&#%d;", *cur);
buf[sizeof(buf) - 1] = 0;
ptr = buf;
while (*ptr != 0)
{
*out++ = *ptr++;
}
cur++;
continue;
// ENDTODO:
} else if (*cur < 0xE0) {
val = (cur[0]) & 0x1F;
val <<= 6;
val |= (cur[1]) & 0x3F;
l = 2;
} else if (*cur < 0xF0) {
val = (cur[0]) & 0x0F;
val <<= 6;
val |= (cur[1]) & 0x3F;
val <<= 6;
val |= (cur[2]) & 0x3F;
l = 3;
} else if (*cur < 0xF8) {
val = (cur[0]) & 0x07;
val <<= 6;
val |= (cur[1]) & 0x3F;
val <<= 6;
val |= (cur[2]) & 0x3F;
val <<= 6;
val |= (cur[3]) & 0x3F;
l = 4;
}
if ((l == 1) || (!IS_CHAR(val))) {
xmlGenericError(xmlGenericErrorContext, EMBED_ERRTXT("xmlEncodeEntitiesReentrant : char out of range\n"));
if (doc != NULL){
doc->encoding = xmlStrdup(BAD_CAST "ISO-8859-1");
if(OOM_FLAG)
goto OOM;
}
// 2-->
snprintf(buf, sizeof(buf), "&#%d;", *cur);
buf[sizeof(buf) - 1] = 0;
ptr = buf;
while (*ptr != 0)
{
*out++ = *ptr++;
}
cur++;
continue;
// <--2
}
/*
* We could do multiple things here. Just save as a char ref
*/
// 3-->
if (html)
snprintf(buf, sizeof(buf), "&#%d;", val);
else
snprintf(buf, sizeof(buf), "&#x%X;", val);
buf[sizeof(buf) - 1] = 0;
ptr = buf;
while (*ptr != 0)
{
*out++ = *ptr++;
}
cur += l;
continue;
// <--3
}
} else if (IS_BYTE_CHAR(*cur)) {
char buf[11], *ptr;
// 4-->
snprintf(buf, sizeof(buf), "&#%d;", *cur);
buf[sizeof(buf) - 1] = 0;
ptr = buf;
while (*ptr != 0)
{
*out++ = *ptr++;
}
}
cur++;
// continue; is implied here
// <--4
} // while (*cur != '\0')
*out++ = 0;
return(buffer);
OOM:
xmlFree(buffer);
return NULL;
}
/**
* xmlEncodeEntitiesInternal:
* @doc: the document containing the string
* @input: A string to convert to XML.
* @attr: are we handling an atrbute value
*
* Do a global encoding of a string, replacing the predefined entities
* and non ASCII values with their entities and CharRef counterparts.
* Contrary to xmlEncodeEntities, this routine is reentrant, and result
* must be deallocated.
*
* Returns A newly allocated string with the substitution done.
*/
static xmlChar *
xmlEncodeEntitiesInternal(xmlDocPtr doc, const xmlChar *input, int attr) {
const xmlChar *cur = input;
xmlChar *buffer = NULL;
xmlChar *out = NULL;
size_t buffer_size = 0;
int html = 0;
if (input == NULL) return(NULL);
if (doc != NULL)
html = (doc->type == XML_HTML_DOCUMENT_NODE);
/*
* allocate an translation buffer.
*/
buffer_size = 1000;
buffer = (xmlChar *) xmlMalloc(buffer_size * sizeof(xmlChar));
if (buffer == NULL) {
xmlEntitiesErrMemory("xmlEncodeEntities: malloc failed");
return(NULL);
}
out = buffer;
while (*cur != '\0') {
size_t indx = out - buffer;
if (indx + 100 > buffer_size) {
growBufferReentrant();
out = &buffer[indx];
}
/*
* By default one have to encode at least '<', '>', '"' and '&' !
*/
if (*cur == '<') {
const xmlChar *end;
/*
* Special handling of server side include in HTML attributes
*/
if (html && attr &&
(cur[1] == '!') && (cur[2] == '-') && (cur[3] == '-') &&
((end = xmlStrstr(cur, BAD_CAST "-->")) != NULL)) {
while (cur != end) {
*out++ = *cur++;
indx = out - buffer;
if (indx + 100 > buffer_size) {
growBufferReentrant();
out = &buffer[indx];
}
}
*out++ = *cur++;
*out++ = *cur++;
*out++ = *cur++;
continue;
}
*out++ = '&';
*out++ = 'l';
*out++ = 't';
*out++ = ';';
} else if (*cur == '>') {
*out++ = '&';
*out++ = 'g';
*out++ = 't';
*out++ = ';';
} else if (*cur == '&') {
/*
* Special handling of &{...} construct from HTML 4, see
* http://www.w3.org/TR/html401/appendix/notes.html#h-B.7.1
*/
if (html && attr && (cur[1] == '{') &&
(strchr((const char *) cur, '}'))) {
while (*cur != '}') {
*out++ = *cur++;
indx = out - buffer;
if (indx + 100 > buffer_size) {
growBufferReentrant();
out = &buffer[indx];
}
}
*out++ = *cur++;
continue;
}
*out++ = '&';
*out++ = 'a';
*out++ = 'm';
*out++ = 'p';
*out++ = ';';
} else if (((*cur >= 0x20) && (*cur < 0x80)) ||
(*cur == '\n') || (*cur == '\t') || ((html) && (*cur == '\r'))) {
/*
* default case, just copy !
*/
*out++ = *cur;
} else if (*cur >= 0x80) {
if (((doc != NULL) && (doc->encoding != NULL)) || (html)) {
/*
* Bjørn Reese <*****@*****.**> provided the patch
xmlChar xc;
xc = (*cur & 0x3F) << 6;
if (cur[1] != 0) {
xc += *(++cur) & 0x3F;
*out++ = xc;
} else
*/
*out++ = *cur;
} else {
/*
* We assume we have UTF-8 input.
*/
char buf[11], *ptr;
int val = 0, l = 1;
if (*cur < 0xC0) {
xmlEntitiesErr(XML_CHECK_NOT_UTF8,
"xmlEncodeEntities: input not UTF-8");
if (doc != NULL)
doc->encoding = xmlStrdup(BAD_CAST "ISO-8859-1");
snprintf(buf, sizeof(buf), "&#%d;", *cur);
buf[sizeof(buf) - 1] = 0;
ptr = buf;
while (*ptr != 0) *out++ = *ptr++;
cur++;
continue;
} else if (*cur < 0xE0) {
val = (cur[0]) & 0x1F;
val <<= 6;
val |= (cur[1]) & 0x3F;
l = 2;
} else if (*cur < 0xF0) {
val = (cur[0]) & 0x0F;
val <<= 6;
val |= (cur[1]) & 0x3F;
val <<= 6;
val |= (cur[2]) & 0x3F;
l = 3;
} else if (*cur < 0xF8) {
val = (cur[0]) & 0x07;
val <<= 6;
val |= (cur[1]) & 0x3F;
val <<= 6;
val |= (cur[2]) & 0x3F;
val <<= 6;
val |= (cur[3]) & 0x3F;
l = 4;
}
if ((l == 1) || (!IS_CHAR(val))) {
xmlEntitiesErr(XML_ERR_INVALID_CHAR,
"xmlEncodeEntities: char out of range\n");
if (doc != NULL)
doc->encoding = xmlStrdup(BAD_CAST "ISO-8859-1");
snprintf(buf, sizeof(buf), "&#%d;", *cur);
buf[sizeof(buf) - 1] = 0;
ptr = buf;
while (*ptr != 0) *out++ = *ptr++;
cur++;
continue;
}
/*
* We could do multiple things here. Just save as a char ref
*/
snprintf(buf, sizeof(buf), "&#x%X;", val);
buf[sizeof(buf) - 1] = 0;
ptr = buf;
while (*ptr != 0) *out++ = *ptr++;
cur += l;
continue;
}
} else if (IS_BYTE_CHAR(*cur)) {
char buf[11], *ptr;
snprintf(buf, sizeof(buf), "&#%d;", *cur);
buf[sizeof(buf) - 1] = 0;
ptr = buf;
while (*ptr != 0) *out++ = *ptr++;
}
cur++;
}
*out = 0;
return(buffer);
mem_error:
xmlEntitiesErrMemory("xmlEncodeEntities: realloc failed");
xmlFree(buffer);
return(NULL);
}
