Suppose I have this kind of string format:
"<RGB:255,0,0>this text is colored RED.<RGB:0,255,0> While this text is colored GREEN";
I want to extract the values inside the <RGB> i.e 255,0,0 and put it on other variables then delete the chars from '<' to '>'.
My code so far:
//this function is called after the loop that checks for the existence of '<'
void RGB_ExtractAndDelete(std::string& RGBformat, int index, RGB& rgb)
{
int i = index + 5; //we are now next to character ':'
std::string value;
int toNumber;
while (RGBformat[i] != ',')
{
value += RGBformat[i++];
}
++i;
std::stringstream(value) >> toNumber;
rgb.R = toNumber;
value = "";
while (RGBformat[i] != ',')
{
value += RGBformat[i++];
}
++i;
std::stringstream(value) >> toNumber;
value = "";
rgb.G = toNumber;
while (RGBformat[i] != '>')
{
value += RGBformat[i++];
}
++i;
std::stringstream(value) >> toNumber;
value = "";
rgb.B = toNumber;
//I got the right result here which is
//start: <, end: >
printf("start: %c, end: %c\n", RGBformat[index], RGBformat[i]);
//but fail in this one
//this one should erase from '<' until it finds '>'
RGBformat.erase(index, i);
}
If I put the <RGB:?,?,?> on the start of the string, it works but it fails when it finds it next to a non '<' character. Or can you suggest much better approach how to do this?
Use std::str::find to locate the <RGB, :, , and >.
Use std::str::substr to "cut out" the string.
Add if (!std::strinstream(value)>> toNumber) ... to check that the number was actually accepted.
Something like this:
std::string::size_type index = std::RGBformat.find("<RGB");
if (index == std::string::npos)
{
... no "<RGB" found
}
std::string::size_type endAngle = std::RGBformat::find(">", index);
if (endAngle == std::string::npos)
{
... no ">" found...
}
std::string::size_type comma = std::RGBformat.find(",", index);
if (comma == std::string::npos && comma < endAngle)
{
... no "," found ...
}
std::string value = RGBformat.substr(index, comma-index-1);
std::stringstream(value) >> toNumber;
value = "";
rgb.R = toNumber;
std::string::size_type comma2 = std::RGBformat.find(",", comma+1);
if (comma2 == std::string::npos && comma2 < endAngle)
...
Note that this may look a bit clumsier than your current code, but it has the advantage of being a lot safer. If someone passed in "<RGB:55> .... " to your existing code, it would break, because it just keeps going until either you get bored and press a key to stop it, or it crashes, whichever comes first...
If you can use Boost or C++11, this is really the perfect place for regular expressions.
You can match your color specifiers with "\\<RGB:(\\d{1,3}),(\\d{1,3}),(\\d{1,3})\\>" - or if you have C++11 raw string literals, you can write this more readably as R"rx(\<RGB:(\d{1,3}),(\d{1,3}),(\d{1,3})\>)rx".
Parse it with
std::getline
http://en.cppreference.com/w/cpp/string/basic_string/getline
This function accepts a delimiter (e.g. '<' or '>') as third argument.
For an example look at:
Basic C++ program, getline()/parsing a file
Here a modified code I use to extract text from html and retrieve data from html tag when I can't use regexp. Otherwise I advice you to use regular expressions they are much more easier to setup.
In my code I ended my tags with "</>" for the color "<RGB:255,0,0>My text</>".
Hope it would help!
#include <iostream>
#include <string>
#include <vector>
#include <cstdlib>
using namespace std;
typedef struct{
string text;
uint8_t r;
uint8_t g;
uint8_t b;
}TextInfo;
vector<TextInfo> vect;
const vector<TextInfo> & extractInfos(const string & str){
string newStr = str;
vect.clear();
do{
TextInfo info;
int index = newStr.find('>');
if(index != -1 && newStr.find('<') == 0){
// We get "<RGB:r,g,b>"
string color = newStr.substr(0,index+1);
// We extract red color
string red = color.substr(color.find(':') + 1, color.find(',') - color.find(':') - 1);
// We get "g,b>"
color = color.substr(color.find(',') + 1, color.length() - color.find(','));
// We extract green color
string green = color.substr(0,color.find(','));
// We extract "b>"
color = color.substr(color.find(',') + 1, color.length() - color.find('>'));
// We get blue color;
string blue = color.substr(0,color.find('>'));
// string to int into a uint8_t
info.r = atoi(red.c_str());
info.g = atoi(green.c_str());
info.b = atoi(blue.c_str());
// We remove the "<RGB:r,g,b>" part from the string
newStr = newStr.substr(index+1,newStr.length()-index);
index = newStr.find("</>");
// We get the text associated to the color just extracted
info.text = newStr.substr(0,index);
// We remove the "</>" that ends the color
newStr = newStr.substr(index+3,newStr.length()-(index+2));
}else{
// We extract the string to the next '<' or to the end if no other color is set
int i = newStr.find('<');
if(i == -1){
i=newStr.length();
}
info.text = newStr.substr(0,i);
info.r = 0;
info.g = 0;
info.b = 0; // No color then we put default to black
// We get the new part of the string without the one we just exctacted
newStr = newStr.substr(i, newStr.length() - i);
}
// We put the data into a vector
vect.push_back(info);
}while(newStr.length() != 0); // We do it while there is something to extract
return vect;
}
int main(void){
vector<TextInfo> myInfos = extractInfos("<RGB:255,0,0>String to red</><RGB:0,255,0>Green string</>Default color string");
for(vector<TextInfo>::iterator itr = myInfos.begin();itr != myInfos.end();itr++){
cout << (*itr).text << endl;
}
return 0;
}
Related
I have a comma-delimited string that I want to store in a string vector. The string and vectors are:
string s = "1, 10, 'abc', 'test, 1'";
vector<string> v;
Ideally I want the strings 'abc' and 'test, 1' to be stored without the single quotes as below, but I can live with storing them with single quotes:
v[0] = "1";
v[1] = "10";
v[2] = "abc";
v[3] = "test, 1";
bool nextToken(const string &s, string::size_type &start, string &token)
{
token.clear();
start = s.find_first_not_of(" \t", start);
if (start == string::npos)
return false;
string::size_type end;
if (s[start] == '\'')
{
++start;
end = s.find('\'', start);
}
else
end = s.find_first_of(" \t,", start);
if (end == string::npos)
{
token = s.substr(start);
start = s.size();
}
else
{
token = s.substr(start, end-start);
if ((s[end] != ',') && ((end = s.find(',', end + 1)) == string::npos))
start = s.size();
else
start = end + 1;
}
return true;
}
string s = "1, 10, 'abc', 'test, 1'", token;
vector<string> v;
string::size_type start = 0;
while (nextToken(s, start, token))
v.push_back(token);
Demo
What you need to do here, is make yourself a parser that parses as you want it to. Here I have made a parsing function for you:
#include <string>
#include <vector>
using namespace std;
vector<string> parse_string(string master) {
char temp; //the current character
bool encountered = false; //for checking if there is a single quote
string curr_parse; //the current string
vector<string>result; //the return vector
for (int i = 0; i < master.size(); ++i) { //while still in the string
temp = master[i]; //current character
switch (temp) { //switch depending on the character
case '\'': //if the character is a single quote
if (encountered) encountered = false; //if we already found a single quote, reset encountered
else encountered = true; //if we haven't found a single quote, set encountered to true
[[fallthrough]];
case ',': //if it is a comma
if (!encountered) { //if we have not found a single quote
result.push_back(curr_parse); //put our current string into our vector
curr_parse = ""; //reset the current string
break; //go to next character
}//if we did find a single quote, go to the default, and push_back the comma
[[fallthrough]];
default: //if it is a normal character
if (encountered && isspace(temp)) curr_parse.push_back(temp); //if we have found a single quote put the whitespace, we don't care
else if (isspace(temp)) break; //if we haven't found a single quote, trash the whitespace and go to the next character
else if (temp == '\'') break; //if the current character is a single quote, trash it and go to the next character.
else curr_parse.push_back(temp); //if all of the above failed, put the character into the current string
break; //go to the next character
}
}
for (int i = 0; i < result.size(); ++i) {
if (result[i] == "") result.erase(result.begin() + i);
//check that there are no empty strings in the vector
//if there are, delete them
}
return result;
}
This parses your string as you want it to, and returns a vector. Then, you can use it in your program:
#include <iostream>
int main() {
string s = "1, 10, 'abc', 'test, 1'";
vector<string> v = parse_string(s);
for (int i = 0; i < v.size(); ++i) {
cout << v[i] << endl;
}
}
and it properly prints out:
1
10
abc
test, 1
A proper solution would require a parser implementation. If you need a quick hack, just write a cell reading function (demo). The c++14's std::quoted manipulator is of great help here. The only problem is the manipulator requires a stream. This is easily solved with istringstream - see the second function. Note that the format of your string is CELL COMMA CELL COMMA... CELL.
istream& get_cell(istream& is, string& s)
{
char c;
is >> c; // skips ws
is.unget(); // puts back in the stream the last read character
if (c == '\'')
return is >> quoted(s, '\'', '\\'); // the first character of the cell is ' - read quoted
else
return getline(is, s, ','), is.unget(); // read unqoted, but put back comma - we need it later, in get function
}
vector<string> get(const string& s)
{
istringstream iss{ s };
string cell;
vector<string> r;
while (get_cell(iss, cell))
{
r.push_back( cell );
char comma;
iss >> comma; // expect a cell separator
if (comma != ',')
break; // cell separator not found; we are at the end of stream/string - break the loop
}
if (char c; iss >> c) // we reached the end of what we understand - probe the end of stream
throw "ill formed";
return r;
}
And this is how you use it:
int main()
{
string s = "1, 10, 'abc', 'test, 1'";
try
{
auto v = get(s);
}
catch (const char* e)
{
cout << e;
}
}
I was reading the following question Parsing a comma-delimited std::string on how to split a string by a comma (Someone gave me the link from my previous question) and one of the answers was:
stringstream ss( "1,1,1,1, or something else ,1,1,1,0" );
vector<string> result;
while( ss.good() )
{
string substr;
getline( ss, substr, ',' );
result.push_back( substr );
}
But what if my string was like the following, and I wanted to separate values only by the bold commas and ignoring what appears inside <>?
<a,b>,<c,d>,,<d,l>,
I want to get:
<a,b>
<c,d>
"" //Empty string
<d,l>
""
Given:<a,b>,,<c,d> It should return: <a,b> and "" and <c,d>
Given:<a,b>,<c,d> It should return:<a,b> and <c,d>
Given:<a,b>, It should return:<a,b> and ""
Given:<a,b>,,,<c,d> It should return:<a,b> and "" and "" and <c,d>
In other words, my program should behave just like the given solution above separated by , (Supposing there is no other , except the bold ones)
Here are some suggested solution and their problems:
Delete all bold commas: This will result in treating the following 2 inputs the same way while they shouldn't
<a,b>,<c,d>
<a,b>,,<c,d>
Replace all bold commas with some char and use the above algorithm: I can't select some char to replace the commas with since any value could appear in the rest of my string
Adding to #Carlos' answer, apart from regex (take a look at my comment); you can implement the substitution like the following (Here, I actually build a new string):
#include <algorithm>
#include <iostream>
#include <string>
int main() {
std::string str;
getline(std::cin,str);
std::string str_builder;
for (auto it = str.begin(); it != str.end(); it++) {
static bool flag = false;
if (*it == '<') {
flag = true;
}
else if (*it == '>') {
flag = false;
str_builder += *it;
}
if (flag) {
str_builder += *it;
}
}
}
Why not replace one set of commas with some known-to-not-clash character, then split it by the other commas, then reverse the replacement?
So replace the commas that are inside the <> with something, do the string split, replace again.
I think what you want is something like this:
vector<string> result;
string s = "<a,b>,,<c,d>"
int in_string = 0;
int latest_comma = 0;
for (int i = 0; i < s.size(); i++) {
if(s[i] == '<'){
result.push_back(s[i]);
in_string = 1;
latest_comma = 0;
}
else if(s[i] == '>'){
result.push_back(s[i]);
in_string = 0;
}
else if(!in_string && s[i] == ','){
if(latest_comma == 1)
result.push_back('\n');
else
latest_comma = 1;
}
else
result.push_back(s[i]);
}
Here is a possible code that scans a string one char at a time and splits it on commas (',') unless they are masked between brackets ('<' and '>').
Algo:
assume starting outside brackets
loop for each character:
if not a comma, or if inside brackets
store the character in the current item
if a < bracket: note that we are inside brackets
if a > bracket: note that we are outside brackets
else (an unmasked comma)
store the current item as a string into the resulting vector
clear the current item
store the last item into the resulting vector
Only 10 lines and my rubber duck agreed that it should work...
C++ implementation: I will use a vector to handle the current item because it is easier to build it one character at a time
std::vector<std::string> parse(const std::string& str) {
std::vector<std::string> result;
bool masked = false;
std::vector<char> current; // stores chars of the current item
for (const char c : str) {
if (masked || (c != ',')) {
current.push_back(c);
switch (c) {
case '<': masked = true; break;
case '>': masked = false;
}
}
else { // unmasked comma: store item and prepare next
current.push_back('\0'); // a terminating null for the vector data
result.push_back(std::string(¤t[0]));
current.clear();
}
}
// do not forget the last item...
current.push_back('\0');
result.push_back(std::string(¤t[0]));
return result;
}
I tested it with all your example strings and it gives the expected results.
Seems quite straight forward to me.
vector<string> customSplit(string s)
{
vector<string> results;
int level = 0;
std::stringstream ss;
for (char c : s)
{
switch (c)
{
case ',':
if (level == 0)
{
results.push_back(ss.str());
stringstream temp;
ss.swap(temp); // Clear ss for the new string.
}
else
{
ss << c;
}
break;
case '<':
level += 2;
case '>':
level -= 1;
default:
ss << c;
}
}
results.push_back(ss.str());
return results;
}
I would like to make a program that asks for text (a paragraph with several words) that would be separated by commas.
To transform the text and add a tag between the two, like to format the text to html text
Example:
word1, word2, word3
to
<a> word1 </a>, <a> word2 </a>, <a> word3 </a>
So I started doing this code but I do not know how to continue. How can I test the text to find the front of the word? I imagine with ASCII tests?
Maybe with a table that will test every case ?
I do not necessarily ask the complete answer but maybe a direction to follow could help.
#include <iostream>
#include <iomanip>
#include <string> //For getline()
using namespace std;
// Creating class
class GetText
{
public:
string text;
string line; //Using this as a buffer
void userText()
{
cout << "Please type a message: ";
do
{
getline(cin, line);
text += line;
}
while(line != "");
}
void to_string()
{
cout << "\n" << "User's Text: " << "\n" << text << endl;
}
};
int main() {
GetText test;
test.userText();
test.to_string();
system("pause");
return 0;
}
The next thing you would need to do is to split your input by a deltimeter (in your case ',') into a vector and later combine everything with pre and posfixes. C++ does not support splitting by default, you would have to be creative or search for a solution like here.
If you want to keep it really simple, you can detect word boundaries by checking two characters at a time. Here's a working example.
using namespace std;
#include <iostream>
#include <string>
#include <cctype>
typedef enum boundary_type_e {
E_BOUNDARY_TYPE_ERROR = -1,
E_BOUNDARY_TYPE_NONE,
E_BOUNDARY_TYPE_LEFT,
E_BOUNDARY_TYPE_RIGHT,
} boundary_type_t;
typedef struct boundary_s {
boundary_type_t type;
int pos;
} boundary_t;
bool is_word_char(int c) {
return ' ' <= c && c <= '~' && !isspace(c) && c != ',';
}
boundary_t maybe_word_boundary(string str, int pos) {
int len = str.length();
if (pos < 0 || pos >= len) {
return (boundary_t){.type = E_BOUNDARY_TYPE_ERROR};
} else {
if (pos == 0 && is_word_char(str[pos])) {
// if the first character is word-y, we have a left boundary at the beginning
return (boundary_t){.type = E_BOUNDARY_TYPE_LEFT, .pos = pos};
} else if (pos == len - 1 && is_word_char(str[pos])) {
// if the last character is word-y, we have a right boundary left of the null terminator
return (boundary_t){.type = E_BOUNDARY_TYPE_RIGHT, .pos = pos + 1};
} else if (!is_word_char(str[pos]) && is_word_char(str[pos + 1])) {
// if we have a delimiter followed by a word char, we have a left boundary left of the word char
return (boundary_t){.type = E_BOUNDARY_TYPE_LEFT, .pos = pos + 1};
} else if (is_word_char(str[pos]) && !is_word_char(str[pos + 1])) {
// if we have a word char followed by a delimiter, we have a right boundary right of the word char
return (boundary_t){.type = E_BOUNDARY_TYPE_RIGHT, .pos = pos + 1};
}
return (boundary_t){.type = E_BOUNDARY_TYPE_NONE};
}
}
int main() {
string str;
string ins_left("<tag>");
string ins_right("</tag>");
getline(cin, str);
// can't use length for the loop condition without recalculating it all the time
for (int i = 0; str[i] != '\0'; i++) {
boundary_t boundary = maybe_word_boundary(str, i);
if (boundary.type == E_BOUNDARY_TYPE_LEFT) {
str.insert(boundary.pos, ins_left);
i += ins_left.length();
} else if (boundary.type == E_BOUNDARY_TYPE_RIGHT) {
str.insert(boundary.pos, ins_right);
i += ins_right.length();
}
}
}
It would be better to use enum class but I forgot the notation. You can also copy to a buffer instead of generating the new string in-place, I was just trying to keep it simple. Feel free to expand it to a class based C++ style. To get your exact desired output, strip the spaces first and add spaces to ins_left and ins_right.
I am writing a program that should receive 3 parameters by User: file_upload "local_path" "remote_path"
code example:
std::vector split(std::string str, char delimiter) {
std::vector<string> v;
std::stringstream src(str);
std::string buf;
while(getline(src, buf, delimiter)) {
v.push_back(buf);
}
return v;
}
void function() {
std::string input
getline(std::cin, input);
// user input like this: file_upload /home/Space Dir/file c:\dir\file
std::vector<std::string> v_input = split(input, ' ');
// the code will do something like this
if(v_input[0].compare("file_upload") == 0) {
FILE *file;
file = fopen(v_input[1].c_str(), "rb");
send_upload_dir(v_input[2].c_str());
// bla bla bla
}
}
My question is: the second and third parameter are directories, then they can contain spaces in name. How can i make the split function does not change the spaces of the second and third parameter?
I thought to put quotes in directories and make a function to recognize, but not work 100% because the program has other functions that take only 2 parameters not three. can anyone help?
EDIT: /home/user/Space Dir/file.out <-- path with space name.
If this happens the vector size is greater than expected, and the path to the directory will be broken.. this can not happen..
the vector will contain something like this:
vector[1] = /home/user/Space
vector[2] = Dir/file.out
and what I want is this:
vector[1] = /home/user/Space Dir/file.out
Since you need to accept three values from a single string input, this is a problem of encoding.
Encoding is sometimes done by imposing fixed-width requirements on some or all fields, but that's clearly not appropriate here, since we need to support variable-width file system paths, and the first value (which appears to be some kind of mode specifier) may be variable-width as well. So that's out.
This leaves 4 possible solutions for variable-width encoding:
1: Unambiguous delimiter.
If you can select a separator character that is guaranteed never to show up in the delimited values, then you can split on that. For example, if NUL is guaranteed never to be part of the mode value or the path values, then we can do this:
std::vector<std::string> v_input = split(input,'\0');
Or maybe the pipe character:
std::vector<std::string> v_input = split(input,'|');
Hence the input would have to be given like this (for the pipe character):
file_upload|/home/user/Space Dir/file.out|/home/user/Other Dir/blah
2: Escaping.
You can write the code to iterate through the input line and properly split it on unescaped instances of the separator character. Escaped instances will not be considered separators. You can parameterize the escape character. For example:
std::vector<std::string> escapedSplit(std::string str, char delimiter, char escaper ) {
std::vector<std::string> res;
std::string cur;
for (size_t i = 0; i < str.size(); ++i) {
if (str[i] == delimiter) {
res.push_back(cur);
cur.clear();
} else if (str[i] == escaper) {
++i;
if (i == str.size()) break;
cur.push_back(str[i]);
} else {
cur.push_back(str[i]);
} // end if
} // end for
if (!cur.empty()) res.push_back(cur);
return res;
} // end escapedSplit()
std::vector<std::string> v_input = escapedSplit(input,' ','\\');
With input as:
file_upload /home/user/Space\ Dir/file.out /home/user/Other\ Dir/blah
3: Quoting.
You can write the code to iterate through the input line and properly split it on unquoted instances of the separator character. Quoted instances will not be considered separators. You can parameterize the quote character.
A complication of this approach is that it is not possible to include the quote character itself inside a quoted extent unless you introduce an escaping mechanism, similar to solution #2. A common strategy is to allow repetition of the quote character to escape it. For example:
std::vector<std::string> quotedSplit(std::string str, char delimiter, char quoter ) {
std::vector<std::string> res;
std::string cur;
for (size_t i = 0; i < str.size(); ++i) {
if (str[i] == delimiter) {
res.push_back(cur);
cur.clear();
} else if (str[i] == quoter) {
++i;
for (; i < str.size(); ++i) {
if (str[i] == quoter) {
if (i+1 == str.size() || str[i+1] != quoter) break;
++i;
cur.push_back(quoter);
} else {
cur.push_back(str[i]);
} // end if
} // end for
} else {
cur.push_back(str[i]);
} // end if
} // end for
if (!cur.empty()) res.push_back(cur);
return res;
} // end quotedSplit()
std::vector<std::string> v_input = quotedSplit(input,' ','"');
With input as:
file_upload "/home/user/Space Dir/file.out" "/home/user/Other Dir/blah"
Or even just:
file_upload /home/user/Space" "Dir/file.out /home/user/Other" "Dir/blah
4: Length-value.
Finally, you can write the code to take a length before each value, and only grab that many characters. We could require a fixed-width length specifier, or skip a delimiting character following the length specifier. For example (note: light on error checking):
std::vector<std::string> lengthedSplit(std::string str) {
std::vector<std::string> res;
size_t i = 0;
while (i < str.size()) {
size_t len = std::atoi(str.c_str());
if (len == 0) break;
i += (size_t)std::log10(len)+2; // +1 to get base-10 digit count, +1 to skip delim
res.push_back(str.substr(i,len));
i += len;
} // end while
return res;
} // end lengthedSplit()
std::vector<std::string> v_input = lengthedSplit(input);
With input as:
11:file_upload29:/home/user/Space Dir/file.out25:/home/user/Other Dir/blah
I had similar problem few days ago and solve it like this:
First I've created a copy, Then replace the quoted strings in the copy with some padding to avoid white spaces, finally I split the original string according to the white space indexes from the copy.
Here is my full solution:
you may want to also remove the double quotes, trim the original string and so on:
#include <sstream>
#include<iostream>
#include<vector>
#include<string>
using namespace std;
string padString(size_t len, char pad)
{
ostringstream ostr;
ostr.fill(pad);
ostr.width(len);
ostr<<"";
return ostr.str();
}
void splitArgs(const string& s, vector<string>& result)
{
size_t pos1=0,pos2=0,len;
string res = s;
pos1 = res.find_first_of("\"");
while(pos1 != string::npos && pos2 != string::npos){
pos2 = res.find_first_of("\"",pos1+1);
if(pos2 != string::npos ){
len = pos2-pos1+1;
res.replace(pos1,len,padString(len,'X'));
pos1 = res.find_first_of("\"");
}
}
pos1=res.find_first_not_of(" \t\r\n",0);
while(pos1 < s.length() && pos2 < s.length()){
pos2 = res.find_first_of(" \t\r\n",pos1+1);
if(pos2 == string::npos ){
pos2 = res.length();
}
len = pos2-pos1;
result.push_back(s.substr(pos1,len));
pos1 = res.find_first_not_of(" \t\r\n",pos2+1);
}
}
int main()
{
string s = "234 \"5678 91\" 8989";
vector<string> args;
splitArgs(s,args);
cout<<"original string:"<<s<<endl;
for(size_t i=0;i<args.size();i++)
cout<<"arg "<<i<<": "<<args[i]<<endl;
return 0;
}
and this is the output:
original string:234 "5678 91" 8989
arg 0: 234
arg 1: "5678 91"
arg 2: 8989
I am programming some automated test equipment (ATE) and I'm trying to extract the following values out of an example response from the ATE:
DCRE? 1,
DCRE P, 10.3, (pin1)
DCRE F, 200.1, (pin2)
DCRE P, 20.4, (pin3)
From each line, I only care about the pin and the measured result value. So for the case above, I want to store the following pieces of information in a map<std::string, double> results;
results["pin1"] = 50.3;
results["pin2"] = 30.8;
results["pin3"] = 70.3;
I made the following code to parse the response:
void parseResultData(map<Pin*, double> &pinnametoresult, string &datatoparse) {
char *p = strtok((char*) datatoparse.c_str(), " \n");
string lastread;
string current;
while (p) {
current = p;
if(current.find('(') != string::npos) {
string substring = lastread.substr(1);
const char* last = substring.c_str();
double value = strtod(last, NULL);
unsigned short number = atoi(current.substr(4, current.size()-2).c_str());
pinnametoresult[&pinlookupmap[number]] = value;
}
lastread = p;
p = strtok(NULL, " \n");
}
}
It works, but it's not very efficient. Is there a way to make the function more efficient for this specific case? I don't care about the DCRE or P/F value on each line. I thought about using Boost regex library, but not sure if that would be more efficient.
In order to make this a bit more efficient, try to avoid copying. In particular, calls to substring, assignments etc can cause havoc on the performance. If you look at your code, you will see that the content of datatoparse are repeatedly assigned to lastread and current, each time with one line less at the beginning. So, on average you copy half of the original string times the number of lines, making just that part an O(n^2) algorithm. This isn't relevant if you have three or four line (not even on 100 lines!) but if you have a few more, performance degrades rapidly.
Try this approach instead:
string::size_type p0 = 0;
string::size_type p1 = input.find('\n', p0);
while (p1 != string::npos) {
// extract the line
string line = input.substr(p0, p1 - p0);
// move to the next line
p0 = p1 + 1;
p1 = input.find('\n', p0);
}
Notes:
Note that the algorithm still copies all input once, but each line only once, making it O(n).
Since you have a copy of the line, you can insert '\0' as artificial separator in order to give a substring to e.g. atoi() or strtod().
I'm not 100% sure of the order of parameters for string::find() and too lazy to look it up, but the idea is to start searching at a certain position. Look at the various overloads of find-like functions.
When handling a line, search the indices of the parts you need and then extract and parse them.
If you have line fragments (i.e. a partial line without a newline) at the end, you will have to modify the loop slightly. Create tests!
This is what I did:
#include <cstdlib>
#include <string>
#include <vector>
#include <unordered_map>
#include <sstream>
#include <iostream>
using namespace std;
struct Pin {
string something;
Pin() {}
};
vector<Pin*> pins = { new Pin(), new Pin(), new Pin() };
typedef unordered_map<Pin*, double> CONT_T;
inline bool OfInterest(const string& line) {
return line.find("(") != string::npos;
}
void parseResultData(CONT_T& pinnametoresult, const string& datatoparse)
{
istringstream is(datatoparse);
string line;
while (getline(is, line)) {
if (OfInterest(line)) {
double d = 0.0;
unsigned int pinid;
size_t firstComma = line.find(",")+2; // skip space
size_t secondComma = line.find(",", firstComma);
istringstream is2(line.substr(firstComma, secondComma-firstComma));
is2 >> d;
size_t paren = line.find("(")+4; // skip pin
istringstream is3(line.substr(paren, (line.length()-paren)-1));
is3 >> pinid;
--pinid;
Pin* pin = pins[pinid];
pinnametoresult[pin] = d;
}
}
}
/*
*
*/
int main(int argc, char** argv) {
string datatoparse = "DCRE? 1, \n"
"DCRE P, 10.3, (pin1)\n"
"DCRE F, 200.1, (pin2)\n"
"DCRE P, 20.4, (pin3)\n";
CONT_T results;
parseResultData(results, datatoparse);
return 0;
}
Here's my final result. Does not involve any copying, but it will destroy the string.
void parseResultData3(map<std::string, double> &pinnametoresult, std::string &datatoparse) {
char* str = (char*) datatoparse.c_str();
int length = datatoparse.size();
double lastdouble = 0.0;
char* startmarker = NULL; //beginning of next pin to parse
for(int pos = 0; pos < length; pos++, str++) {
if(str[0] == '(') {
startmarker = str + 1;
//get previous value
bool triggered = false;
for(char* lookback = str - 1; ; lookback--) {
if(!triggered && (isdigit(lookback[0]) || lookback[0] == '.')) {
triggered = true;
*(lookback + 1) = '\0';
}
else if(triggered && (!isdigit(lookback[0]) && lookback[0] != '.')) {
lastdouble = strtod(lookback, NULL);
break;
}
}
}
else if(startmarker != NULL) {
if(str[0] == ')') {
str[0] = '\0';
pinnametoresult[startmarker] = lastdouble;
startmarker = NULL;
}
if(str[0] == ',') {
str[0] = '\0';
pinnametoresult[startmarker] = lastdouble;
startmarker = str + 1;
}
}
}
}