char *mysql_query_digest_and_first_comment(char *s, int _len, char **first_comment){
int i = 0;
char cur_comment[FIRST_COMMENT_MAX_LENGTH];
cur_comment[0]=0;
int ccl=0;
int cmd=0;
int len = _len;
if (_len > QUERY_DIGEST_MAX_LENGTH) {
len = QUERY_DIGEST_MAX_LENGTH;
}
char *r = (char *) malloc(len + SIZECHAR);
char *p_r = r;
char *p_r_t = r;
char prev_char = 0;
char qutr_char = 0;
char flag = 0;
char fc=0;
int fc_len=0;
char fns=0;
bool lowercase=0;
lowercase=mysql_thread___query_digests_lowercase;
while(i < len)
{
// =================================================
// START - read token char and set flag what's going on.
// =================================================
if(flag == 0)
{
// store current position
p_r_t = p_r;
// comment type 1 - start with '/*'
if(prev_char == '/' && *s == '*')
{
ccl=0;
flag = 1;
if (*(s+1)=='!')
cmd=1;
}
// comment type 2 - start with '#'
else if(*s == '#')
{
flag = 2;
}
// string - start with '
else if(*s == '\'' || *s == '"')
{
flag = 3;
qutr_char = *s;
}
// may be digit - start with digit
else if(is_token_char(prev_char) && is_digit_char(*s))
{
flag = 4;
if(len == i+1)
continue;
}
// not above case - remove duplicated space char
else
{
flag = 0;
if (fns==0 && is_space_char(*s)) {
s++;
i++;
continue;
}
if (fns==0) fns=1;
if(is_space_char(prev_char) && is_space_char(*s)){
prev_char = ' ';
*p_r = ' ';
s++;
i++;
continue;
}
}
}
// =================================================
// PROCESS and FINISH - do something on each case
// =================================================
else
{
// --------
// comment
// --------
if (flag == 1) {
if (cmd) {
if (ccl<FIRST_COMMENT_MAX_LENGTH-1) {
cur_comment[ccl]=*s;
ccl++;
}
}
if (fc==0) {
fc=1;
}
if (fc==1) {
if (fc_len<FIRST_COMMENT_MAX_LENGTH-1) {
if (*first_comment==NULL) {
*first_comment=(char *)malloc(FIRST_COMMENT_MAX_LENGTH);
}
char *c=*first_comment+fc_len;
*c = !is_space_char(*s) ? *s : ' ';
fc_len++;
}
if (prev_char == '*' && *s == '/') {
if (fc_len>=2) fc_len-=2;
char *c=*first_comment+fc_len;
*c=0;
//*first_comment[fc_len]=0;
fc=2;
}
}
}
if(
// comment type 1 - /* .. */
(flag == 1 && prev_char == '*' && *s == '/') ||
// comment type 2 - # ... \n
(flag == 2 && (*s == '\n' || *s == '\r'))
)
{
p_r = flag == 1 ? p_r_t - SIZECHAR : p_r_t;
if (cmd) {
cur_comment[ccl]=0;
if (ccl>=2) {
ccl-=2;
cur_comment[ccl]=0;
char el=0;
int fcc=0;
while (el==0 && fcc<ccl ) {
switch (cur_comment[fcc]) {
case '/':
case '*':
case '!':
case '0':
case '1':
case '2':
case '3':
case '4':
case '5':
case '6':
case '7':
case '8':
case '9':
case ' ':
fcc++;
break;
default:
el=1;
break;
}
}
if (el) {
memcpy(p_r,cur_comment+fcc,ccl-fcc);
p_r+=(ccl-fcc);
*p_r++=' ';
}
}
cmd=0;
}
prev_char = ' ';
flag = 0;
s++;
i++;
continue;
}
// --------
// string
// --------
else if(flag == 3)
{
// Last char process
if(len == i + 1)
{
p_r = p_r_t;
*p_r++ = '?';
flag = 0;
break;
}
// need to be ignored case
if(p_r > p_r_t + SIZECHAR)
{
if(
(prev_char == '\\' && *s == '\\') || // to process '\\\\', '\\'
(prev_char == '\\' && *s == qutr_char) || // to process '\''
(prev_char == qutr_char && *s == qutr_char) // to process ''''
)
{
prev_char = 'X';
s++;
i++;
continue;
}
}
// satisfied closing string - swap string to ?
if(*s == qutr_char && (len == i+1 || *(s + SIZECHAR) != qutr_char))
{
p_r = p_r_t;
*p_r++ = '?';
flag = 0;
if(i < len)
s++;
i++;
continue;
}
}
// --------
// digit
// --------
else if(flag == 4)
{
// last single char
if(p_r_t == p_r)
{
*p_r++ = '?';
i++;
continue;
}
// token char or last char
if(is_token_char(*s) || len == i+1)
{
if(is_digit_string(p_r_t, p_r))
{
p_r = p_r_t;
*p_r++ = '?';
if(len == i+1)
{
if(is_token_char(*s))
*p_r++ = *s;
i++;
continue;
}
}
flag = 0;
}
}
}
// =================================================
// COPY CHAR
// =================================================
// convert every space char to ' '
if (lowercase==0) {
*p_r++ = !is_space_char(*s) ? *s : ' ';
} else {
*p_r++ = !is_space_char(*s) ? (tolower(*s)) : ' ';
}
prev_char = *s++;
i++;
}
// remove a trailing space
if (p_r>r) {
char *e=p_r;
e--;
if (*e==' ') {
*e=0;
}
}
*p_r = 0;
// process query stats
return r;
}
char *mysql_query_digest_and_first_comment(char *s, int len, char *first_comment){
int i = 0;
char *r = (char *) malloc(len + SIZECHAR);
char *p_r = r;
char *p_r_t = r;
char prev_char = 0;
char qutr_char = 0;
char flag = 0;
char fc=0;
int fc_len=0;
char fns=0;
while(i < len)
{
// =================================================
// START - read token char and set flag what's going on.
// =================================================
if(flag == 0)
{
// store current position
p_r_t = p_r;
// comment type 1 - start with '/*'
if(prev_char == '/' && *s == '*')
{
flag = 1;
}
// comment type 2 - start with '#'
else if(*s == '#')
{
flag = 2;
}
// string - start with '
else if(*s == '\'' || *s == '"')
{
flag = 3;
qutr_char = *s;
}
// may be digit - start with digit
else if(is_token_char(prev_char) && is_digit_char(*s))
{
flag = 4;
if(len == i+1)
continue;
}
// not above case - remove duplicated space char
else
{
flag = 0;
if (fns==0 && is_space_char(*s)) {
s++;
i++;
continue;
}
if (fns==0) fns=1;
if(is_space_char(prev_char) && is_space_char(*s)){
prev_char = ' ';
*p_r = ' ';
s++;
i++;
continue;
}
}
}
// =================================================
// PROCESS and FINISH - do something on each case
// =================================================
else
{
// --------
// comment
// --------
if (flag == 1) {
if (fc==0) {
fc=1;
}
if (fc==1) {
if (fc_len<FIRST_COMMENT_MAX_LENGTH-1) {
first_comment[fc_len]= !is_space_char(*s) ? *s : ' ';
fc_len++;
}
if (prev_char == '*' && *s == '/') {
if (fc_len>=2) fc_len-=2;
first_comment[fc_len]=0;
fc=2;
}
}
}
if(
// comment type 1 - /* .. */
(flag == 1 && prev_char == '*' && *s == '/') ||
// comment type 2 - # ... \n
(flag == 2 && (*s == '\n' || *s == '\r'))
)
{
p_r = flag == 1 ? p_r_t - SIZECHAR : p_r_t;
prev_char = ' ';
flag = 0;
s++;
i++;
continue;
}
// --------
// string
// --------
else if(flag == 3)
{
// Last char process
if(len == i + 1)
{
p_r = p_r_t;
*p_r++ = '?';
flag = 0;
break;
}
// need to be ignored case
if(p_r > p_r_t + SIZECHAR)
{
if(
(prev_char == '\\' && *s == '\\') || // to process '\\\\', '\\'
(prev_char == '\\' && *s == qutr_char) || // to process '\''
(prev_char == qutr_char && *s == qutr_char) // to process ''''
)
{
prev_char = 'X';
s++;
i++;
continue;
}
}
// satisfied closing string - swap string to ?
if(*s == qutr_char && (len == i+1 || *(s + SIZECHAR) != qutr_char))
{
p_r = p_r_t;
*p_r++ = '?';
flag = 0;
if(i < len)
s++;
i++;
continue;
}
}
// --------
// digit
// --------
else if(flag == 4)
{
// last single char
if(p_r_t == p_r)
{
*p_r++ = '?';
i++;
continue;
}
// token char or last char
if(is_token_char(*s) || len == i+1)
{
if(is_digit_string(p_r_t, p_r))
{
p_r = p_r_t;
*p_r++ = '?';
if(len == i+1)
{
if(is_token_char(*s))
*p_r++ = *s;
i++;
continue;
}
}
flag = 0;
}
}
}
// =================================================
// COPY CHAR
// =================================================
// convert every space char to ' '
*p_r++ = !is_space_char(*s) ? *s : ' ';
prev_char = *s++;
i++;
}
// remove a trailing space
if (p_r>r) {
char *e=p_r;
e--;
if (*e==' ') {
*e=0;
}
}
*p_r = 0;
// process query stats
return r;
}
