bool CSV_Parser::parse(const STR& input_line, CSV_FIELDS& output_fields)
{
/*
A private method which handles the parsing logic used by both the overloaded public methods
*/
STR field;
int i, j;
if(input_line.length() == 0)
{
return false;
}
i = 0;
do
{
if(i < input_line.length() && input_line[i] == CSV_QUOTE)
{
j = parse_quoted_fields(input_line, field, ++i);
}
else
{
j = parse_normal_fields(input_line, field, i);
}
output_fields.push_back(field);
i = j + 1;
}while(j < input_line.length());
return true;
}
int CSV_Parser::parse_quoted_fields(const STR& input_line, STR& field, int& i)
{
/*
Quoted fields are the ones which are enclosed within quotes
For instance - Consider that input_line is - 1997,Ford,E350,"Super, luxurious truck"
An example for a quoted field would be - Super luxurious truck
Another instance being - 1997,Ford,E350,"Super, ""luxurious"" truck"
*/
int j;
field = "";
for(j=i; j<input_line.length(); j++)
{
if(input_line[j] == '"' && input_line[++j] != '"')
{
int k = input_line.find_first_of(CSV_DELIMITER, j);
if(k > input_line.length())
{
k = input_line.length();
}
for(k -= j; k-- > 0; )
{
field += input_line[j++];
}
break;
}
else
{
field += input_line[j];
}
}
return j;
}
bool EndsWithT(const STR& str, const STR& search, bool case_sensitive) {
size_t str_length = str.length();
size_t search_length = search.length();
if (search_length > str_length)
return false;
if (case_sensitive)
return str.compare(str_length - search_length, search_length, search) == 0;
return std::equal(search.begin(), search.end(),
str.begin() + (str_length - search_length),
base::CaseInsensitiveCompare<typename STR::value_type>());
}
TrimPositions TrimStringT(const STR& input,
const STR& trim_chars,
TrimPositions positions,
STR* output) {
// Find the edges of leading/trailing whitespace as desired.
const size_t last_char = input.length() - 1;
const size_t first_good_char = (positions & TRIM_LEADING) ?
input.find_first_not_of(trim_chars) : 0;
const size_t last_good_char = (positions & TRIM_TRAILING) ?
input.find_last_not_of(trim_chars) : last_char;
// When the string was all whitespace, report that we stripped off whitespace
// from whichever position the caller was interested in. For empty input, we
// stripped no whitespace, but we still need to clear |output|.
if (input.empty() ||
(first_good_char == STR::npos) || (last_good_char == STR::npos)) {
bool input_was_empty = input.empty(); // in case output == &input
output->clear();
return input_was_empty ? TRIM_NONE : positions;
}
// Trim the whitespace.
*output =
input.substr(first_good_char, last_good_char - first_good_char + 1);
// Return where we trimmed from.
return static_cast<TrimPositions>(
((first_good_char == 0) ? TRIM_NONE : TRIM_LEADING) |
((last_good_char == last_char) ? TRIM_NONE : TRIM_TRAILING));
}
TrimPositions TrimStringT(const STR& input,
const typename STR::value_type trim_chars[],
TrimPositions positions,
STR* output)
{
// 根据移除选项positions查找两端边界.
const typename STR::size_type last_char = input.length() - 1;
const typename STR::size_type first_good_char = (positions&TRIM_LEADING) ?
input.find_first_not_of(trim_chars) : 0;
const typename STR::size_type last_good_char = (positions&TRIM_TRAILING) ?
input.find_last_not_of(trim_chars) : last_char;
// 当字符串所有字符都是空白, 根据调用传入的positions返回TrimPositions.
// 对于空输入没有去除任何空白, 但仍需要对output串调用clear.
if(input.empty() || (first_good_char==STR::npos) || (last_good_char==STR::npos))
{
bool input_was_empty = input.empty();
output->clear();
return input_was_empty ? TRIM_NONE : positions;
}
// 移除空白.
*output = input.substr(first_good_char, last_good_char-first_good_char+1);
// 返回两端哪边移除过.
return static_cast<TrimPositions>(
((first_good_char==0)?TRIM_NONE:TRIM_LEADING) |
((last_good_char==last_char)?TRIM_NONE:TRIM_TRAILING));
}
int CSV_Parser::parse_normal_fields(const STR& input_line, STR& field, int& i)
{
/*
Normal fields are the ones which contain no escaped or quoted characters
For instance - Consider that input_line is - 1997,Ford,E350,"Super, luxurious truck"
An example for a normal field would be - Ford
*/
int j;
j = input_line.find_first_of(CSV_DELIMITER, i);
if(j > input_line.length())
{
j = input_line.length();
}
field = std :: string(input_line, i, j-i);
return j;
}
static bool DoIsStringASCII(const STR& str) {
for (size_t i = 0; i < str.length(); i++) {
typename ToUnsigned<typename STR::value_type>::Unsigned c = str[i];
if (c > 0x7F)
return false;
}
return true;
}
bool StartsWithT(const STR& str, const STR& search, bool case_sensitive) {
if (case_sensitive) {
return str.compare(0, search.length(), search) == 0;
}
else {
if (search.size() > str.size())
return false;
return std::equal(search.begin(), search.end(), str.begin(),
base::CaseInsensitiveCompare<typename STR::value_type>());
}
}
STR CCrypt::md5(STR value){
MD5_CTX ctx;
unsigned char buff[MD5_DIGEST_LENGTH];
MD5_Init(&ctx);
MD5_Update(&ctx, value.c_str(), value.length());
MD5_Final(buff, &ctx);
char res[33];
for(int i = 0; i < 16; i++){
sprintf(res+i*2, "%02x", buff[i]);
}
return STR(res);
}
void SplitStringAlongWhitespaceT(const STR& str, std::vector<STR>* result)
{
const size_t length = str.length();
if(!length)
{
return;
}
bool last_was_ws = false;
size_t last_non_ws_start = 0;
for(size_t i=0; i<length; ++i)
{
switch(str[i])
{
// HTML 5定义的空白: space, tab, LF, line tab, FF, or CR.
case L' ':
case L'\t':
case L'\xA':
case L'\xB':
case L'\xC':
case L'\xD':
if(!last_was_ws)
{
if(i > 0)
{
result->push_back(str.substr(last_non_ws_start,
i-last_non_ws_start));
}
last_was_ws = true;
}
break;
default: // 不是空白字符.
if(last_was_ws)
{
last_was_ws = false;
last_non_ws_start = i;
}
break;
}
}
if(!last_was_ws)
{
result->push_back(str.substr(last_non_ws_start,
length-last_non_ws_start));
}
}
bool ReplaceCharsT(const STR& input,
const STR& replace_chars,
const STR& replace_with,
STR* output) {
bool removed = false;
size_t replace_length = replace_with.length();
*output = input;
size_t found = output->find_first_of(replace_chars);
while (found != STR::npos) {
removed = true;
output->replace(found, 1, replace_with);
found = output->find_first_of(replace_chars, found + replace_length);
}
return removed;
}