I am having a map like this
typedef vector< pair<int, string> > Categories;
map<string,Categories> cats;
but when I am trying to read elements of the map like
for(map<string,Categories>::const_iterator it = cats.begin(); it != cats.end(); ++it)
{
std::cout << it->first << " " << it->second.first<<"\n";
}
I do get errors
error: 'const class std::vector<std::pair<int, std::basic_string<char>
' has no member named 'first' std::cout << it->first << " " << it-> second.first<<"\n";
error: 'const class std::vector ' has no member named 'first'
std::cout << it->first << " " << it->second.first<<"\n";
Its clear as Crystal, that you might have many elements in your values of your map, which is a std::vector< std::pair<int, std::string>>
Then how would you print elements of a vector? The options are:
random access (i.e, vec[index])
iterator (i.e, std::vector< std::pair<int, std::string>>::const_iterator itr;)
or by a range based for loop (i.e, for(const auto& it: vec) )
In your case, if you wanna have something simple and easy code is using a range based loop:
for(const auto& it: cats)
{
std::cout << it.first << " = "; // keys
for(const auto& ve: it.second) // values vec
std::cout << ve.first << "-" << ve.second << "\t";
std::cout << std::endl;
}
If you still wanna have long iterator loops, here is it.
see live action here: https://www.ideone.com/3bS1kR
#include <string>
#include <iostream>
#include <vector>
#include <map>
#include <iterator>
typedef std::vector< std::pair<int, std::string>> Categories;
int main()
{
std::map<std::string, Categories> cats;
cats["key1"] = {std::make_pair(1, "pair1"), std::make_pair(1, "pair2"), std::make_pair(1, "par3")};
cats["key2"] = {std::make_pair(2, "pair1"), std::make_pair(2, "pair2")};
cats["key3"] = {std::make_pair(3, "pair1")};
std::cout << "Range based loop \n";
for(const auto& it: cats)
{
std::cout << it.first << " = "; // keys
for(const auto& ve: it.second) // values vec
std::cout << ve.first << "-" << ve.second << "\t";
std::cout << std::endl;
}
std::cout << "\nIterator loop \n";
std::map<std::string, Categories>::const_iterator it;
std::vector< std::pair<int, std::string>>::const_iterator curr_val_it;
for(it = cats.cbegin(); it != cats.cend(); ++it)
{
std::cout << it->first << " = "; // keys
for(curr_val_it = it->second.cbegin(); curr_val_it != it->second.cend(); ++curr_val_it )
std::cout << curr_val_it->first << "-" << curr_val_it->second << "\t"; // values vec
std::cout << std::endl;
}
return 0;
}
you need to access a element of the vector first, then the pair within.
... it->second[0].first<< ...
better impl of loop:
for(const auto& cat : cats)
{
string mapidx = cat.first;
vector<pair<int, std::string>> catvect = cat.second;
}
then you can have a seperate loop to read the contents of the vector:
for(const auto& cat : cats)
{
string mapidx = cat.first;
vector<pair<int, std::string>> catvect = cat.second;
for (const auto& entry : catvect)
{
int number = entry.first;
string whatever = entry.second;
}
}
the temp variables are just for readability, no need for all the copies ;)
Error is exacly what compiler told you:
const class std::vector ' has no member named 'first'
so, you have do decide how to print your map by overloading ostream opeartor, below example how it can be achived:
#include <iostream>
#include <map>
#include <string>
#include <utility>
#include <vector>
typedef std::vector<std::pair<int, std::string>> Categories;
std::map<std::string,Categories> cats;
std::ostream& operator << (std::ostream& os, const std::vector<std::pair<int, std::string>>& v)
{
os << "[";
for (auto& el : v) {
os << " " << el.first << " : " << el.second;
}
os << "]";
return os;
}
int main() {
cats.emplace("cat1", std::vector<std::pair<int, std::string>>(1, std::make_pair(1, "category1")));
for(auto& cat : cats) {
std::cout << cat.first << " " << cat.second << "\n";
}
}
Since we are storing Vector Categories inside a Map we will have to iterate Vector too:
for(map<string,Categories>::const_iterator it = cats.begin(); it != cats.end(); ++it)
{
//iterator for vector (it->second)
for(Categories::const_iterator it2 = it->second.begin(); it2 != it->second.end(); it2++ )
{
std::cout << it->first << " " << it2->first<<" "<<it2->second <<"\n";
}
}
Related
While putting sorted vector elements into map, it inserts them in ascending order after it has been sorted in descending order with the compare function. I know that map is in ascending order by default based on its key, but I am not sure why it sorts them again in ascending order based on its value. Is this common in std::map<std::string, int>?
I simply want to return a map in descending order by its value, I need to display the directory strings with a number of items found by decreasing order.
#include <iostream>
#include <vector>
#include <map>
#include <string>
#include <algorithm>
bool compare_dir(std::pair<std::string, int>&a,
std::pair<std::string, int>&b)
{
return a.second > b.second;
}
void sort_func(std::map<std::string, int>& M)
{
std::map<std::string, int>::iterator itr;
std::vector<std::pair<std::string, int>> temp;
std::map<std::string, int> result;
for(auto &it : M)
temp.push_back(it);
sort(temp.begin(), temp.end(), compare_dir);
for(auto &el : temp)
std::cout << el.first << " " << el.second << std::endl;
std::cout << std::endl;
for(auto & el : temp)
result.insert(el);
for(itr = result.begin(); itr != result.end(); itr++)
std::cout << itr->first << " " << itr->second << std::endl;
}
int main()
{
std::map<std::string, int> my_map{{"dir1", 1}, {"dir2", 2}, {"dir3", 3}};
for (const auto& [key, value] : my_map) {
std::cout << '[' << key << "] = " << value << "; " << std::endl;
}
std::cout << std::endl;
sort_func(my_map);
}
First one being:
map <int,int> m;
//... some elements inserted
auto i= m.begin();
cout<<(*i).first<<(*i).second;
Here we are required to use the dereference operator *
Second:
map <int,int> m;
//... some elements inserted
for(auto i: m)
cout<<i.first<<i.second;
Why am I not required to use the * operator this time?
One more doubt:
for(auto &i: m)
what difference does '&' make here?
auto i=m.begin() will give you iterator .. which is accessed more like a pointer (syntactically) when you want to access the value...
for(auto i:m) will copy current element of m (a pair) into i , i is a copy of element, not the element itself...
for (auto &i: m) will work on a reference, the original map is affected
As explained in the below code snippet, the first i is of type iterator where as the i in the for loops are of pair type.
#include <iostream>
#include <map>
int main()
{
std::map <int,int> m;
m[1] = 5;
m[10] = 60;
m[100] = 800;
// Below i is map iterator (std::map<int, int>::iterator)
auto i = m.begin();
std::cout << typeid(i).name() << '\n';
std::cout << (*i).first << " : " << (*i).second << '\n';
std::cout << i->first << " : " << i->second << '\n';
for(auto i: m) {
// Below i is pair<int, int>
std::cout << typeid(i).name() << '\n';
std::cout << i.first << " : " << i.second << '\n';
}
for(auto& i: m) {
// Below i is reference of pair<int, int>)
// modifying this would result in updated values in the map.
std::cout << typeid(i).name() << '\n';
std::cout << i.first << " : " << i.second << '\n';
}
return 0;
}
I want to iterate through each element in the map<string, int> without knowing any of its string-int values or keys.
What I have so far:
void output(map<string, int> table)
{
map<string, int>::iterator it;
for (it = table.begin(); it != table.end(); it++)
{
//How do I access each element?
}
}
You can achieve this like following :
map<string, int>::iterator it;
for (it = symbolTable.begin(); it != symbolTable.end(); it++)
{
std::cout << it->first // string (key)
<< ':'
<< it->second // string's value
<< std::endl;
}
With C++11 ( and onwards ),
for (auto const& x : symbolTable)
{
std::cout << x.first // string (key)
<< ':'
<< x.second // string's value
<< std::endl;
}
With C++17 ( and onwards ),
for (auto const& [key, val] : symbolTable)
{
std::cout << key // string (key)
<< ':'
<< val // string's value
<< std::endl;
}
Try the following
for ( const auto &p : table )
{
std::cout << p.first << '\t' << p.second << std::endl;
}
The same can be written using an ordinary for loop
for ( auto it = table.begin(); it != table.end(); ++it )
{
std::cout << it->first << '\t' << it->second << std::endl;
}
Take into account that value_type for std::map is defined the following way
typedef pair<const Key, T> value_type
Thus in my example p is a const reference to the value_type where Key is std::string and T is int
Also it would be better if the function would be declared as
void output( const map<string, int> &table );
The value_type of a map is a pair containing the key and value as it's first and second member, respectively.
map<string, int>::iterator it;
for (it = symbolTable.begin(); it != symbolTable.end(); it++)
{
std::cout << it->first << ' ' << it->second << '\n';
}
Or with C++11, using range-based for:
for (auto const& p : symbolTable)
{
std::cout << p.first << ' ' << p.second << '\n';
}
As #Vlad from Moscow says,
Take into account that value_type for std::map is defined the following way:
typedef pair<const Key, T> value_type
This then means that if you wish to replace the keyword auto with a more explicit type specifier, then you could this;
for ( const pair<const string, int> &p : table ) {
std::cout << p.first << '\t' << p.second << std::endl;
}
Just for understanding what auto will translate to in this case.
As P0W has provided complete syntax for each C++ version, I would like to add couple of more points by looking at your code
Always take const & as argument as to avoid extra copies of the same object.
use unordered_map as its always faster to use. See this discussion
here is a sample code:
#include <iostream>
#include <unordered_map>
using namespace std;
void output(const auto& table)
{
for (auto const & [k, v] : table)
{
std::cout << "Key: " << k << " Value: " << v << std::endl;
}
}
int main() {
std::unordered_map<string, int> mydata = {
{"one", 1},
{"two", 2},
{"three", 3}
};
output(mydata);
return 0;
}
if you just want to iterate over the content without changing values
do:
for(const auto & variable_name : container_name(//here it is map name)){
cout << variable_name.first << " : " << variable_name.second << endl;
}
If you want to modify the contents of the map, remove the const and keep & (if you want to modify directly the contents inside container). If you want to work with a copy of the container values, remove the & sign too; after that, you can access them by using .first and .second on "variable_name".
it can even be done with a classic for loop.
advancing the iterator manually.
typedef std::map<int, int> Map;
Map mymap;
mymap['a']=50;
mymap['b']=100;
mymap['c']=150;
mymap['d']=200;
bool itexist = false;
int sizeMap = static_cast<int>(mymap.size());
auto it = mymap.begin();
for(int i = 0; i < sizeMap; i++){
std::cout << "Key: " << it->first << " Value: " << it->second << std::endl;
it++;
}
Other way :
map <int, string> myMap = {
{ 1,"Hello" },
{ 2,"stackOverflow" }
};
for (auto iter = cbegin(myMap); iter != cend(myMap); ++iter) {
cout << iter->second << endl;
}
How can I print the contents of a nested map? I am counting the number of times a word appears in a file, reporting it by line number and number of times per line. The words, lines, and occurrences per line are being stored in the following container:
map<string, map<int, int>> tokens;
However, I'm not sure on the syntax. I am printing the outer map that lists all of the words using the following code, but can't figure out how to print the inner values (the line number and number of times the word appears on each line) as well. I assume I can just include it inline in the for loop, but I can't figure out how:
for (map <string, map<int, int>>::iterator it = tokens.begin(); it != tokens.end(); ++it){
cout << it->first << " : " << /* assume I can include another statement here to print the values? */ endl;
}
I am trying to get an output similar to this:
(word : line:occurrences, line:occurrences, ...)
about : 16:1, 29:1, 166:1, 190:1, 191:1
above : 137:1
accompanied : 6:1
across : 26:1
admit : 20:1
advancing : 170:1
.
.
.
It is actually pretty simple.
You just get the internal map with it->second, and you iterate through that the same way.
Thereby, you would write something like this:
for (map <string, map<int, int>>::iterator it = tokens.begin(); it != tokens.end(); ++it){
cout << it->first << " : ";
map<int, int> &internal_map = it->second;
for (map<int, int>::iterator it2 = internal_map.begin(); it2 != internal_map.end(); ++it2){
if (it2 != internal_map.begin())
cout << ",";
cout << it2->first << ":" << it2->second;
}
cout << endl;
}
You could write something like this if you have C++11 support:
for (auto it : tokens) {
cout << it->first << " : ";
map<int, int> &internal_map = it->second;
for (auto it2: internal_map) {
if (it2 != internal_map.begin())
cout << ",";
cout << it2->first << ":" << it2->second;
}
cout << endl;
}
C++17
Since C++17 you can use range-based for loops together with structured bindings for iterating over maps.
This way, the readability of lpapp's C++11 solution can be further improved as follows:
for (auto const &[k1, v1] : tokens) {
std::cout << k1 << " : ";
for (auto const &[k2, v2] : v1) {
if (&k2 != &v1.begin()->first)
std::cout << ", ";
std::cout << k2 << ":" << v2;
}
std::cout << std::endl;
}
Note: I admit, the check for printing a comma is a bit messy. If you are looking for a nicer solution, you
might want to take a look at this Q&A.
Code on Coliru
Since we are only printing the elements, not modifying it, I would make more use of const, (a) const reference and (b) const iterators.
Also, for pre-C++11, it helps to define typedef's for complex types.
#include <string>
#include <map>
#include <iostream>
using namespace std;
typedef map<int, int> InnerMap;
typedef map<string, InnerMap> OuterMap;
void printClassic( OuterMap const & tokens ) {
for( OuterMap::const_iterator cit = tokens.begin();
cit != tokens.end(); ++cit ) {
cout << cit->first << " : ";
InnerMap const & imap = cit->second;
for( InnerMap::const_iterator cit2 = imap.begin();
cit2 != imap.end(); ++cit2 ) {
cout << cit2->first << ":" << cit2->second << ",";
}
cout << endl;
}
}
void printCpp11( OuterMap const & tokens ) {
for( auto const & cit : tokens ) {
cout << cit.first << " : ";
auto const & imap = cit.second;
for( auto const & cit2 : imap ) {
cout << cit2.first << ":" << cit2.second << ",";
}
cout << endl;
}
}
I have a map declared as follows:
map < string , list < string > > mapex ; list< string > li;
How can I display the items stored in the above map on the console?
Update (Back to the future): with C++11 range-based for loops –
std::map<Key, Value> m { ... /* initialize it */ ... };
for (const auto &p : m) {
std::cout << "m[" << p.first << "] = " << p.second << '\n';
}
Well it depends on how you want to display them, but you can always iterate them easily:
typedef map<string, list<string>>::const_iterator MapIterator;
for (MapIterator iter = mapex.begin(); iter != mapex.end(); iter++)
{
cout << "Key: " << iter->first << endl << "Values:" << endl;
typedef list<string>::const_iterator ListIterator;
for (ListIterator list_iter = iter->second.begin(); list_iter != iter->second.end(); list_iter++)
cout << " " << *list_iter << endl;
}
I'd try the following
void dump_list(const std::list<string>& l) {
for ( std::list<string>::const_iterator it = l.begin(); l != l.end(); l++ ) {
cout << *l << endl;
}
}
void dump_map(const std::map<string, std::list<string>>& map) {
for ( std::map<string,std::list<string>>::const_iterator it = map.begin(); it != map.end(); it++) {
cout << "Key: " << it->first << endl;
cout << "Values" << endl;
dump_list(it->second);
}
I'm a little off topic here...
I guess you want to dump the map content for debugging. I like to mention that the next gdb release (version 7.0) will have a built in python interpreter which will be used by the gcc libstdc++ to provide stl pretty printers. Here is an example for your case
#include <map>
#include <map>
#include <list>
#include <string>
using namespace std;
int main()
{
typedef map<string, list<string> > map_type;
map_type mymap;
list<string> mylist;
mylist.push_back("item 1");
mylist.push_back("item 2");
mymap["foo"] = mylist;
mymap["bar"] = mylist;
return 0; // stopped here
}
which results in
(gdb) print mymap
$1 = std::map with 2 elements = {
["bar"] = std::list = {
[0] = "item 1",
[1] = "item 2"
},
["foo"] = std::list = {
[0] = "item 1",
[1] = "item 2"
}
}
Yay!
Another form, using <algorithm>:
void printPair(const pair<string, list<string> > &p)
{
cout << "Key: " << p.first << endl;
copy(p.second.begin(), p.second.end(), ostream_iterator<string>(cout, "\n"));
}
for_each(mapex.begin(), mapex.end(), printPair);
Test program:
#include <iostream>
#include <map>
#include <list>
#include <iterator>
#include <algorithm>
using namespace std;
void printPair(const pair<string, list<string> > &p)
{
cout << "Key: " << p.first << endl;
copy(p.second.begin(), p.second.end(), ostream_iterator<string>(cout, "\n"));
}
int main()
{
map<string, list<string> > mapex;
list<string> mylist1;
mylist1.push_back("item 1");
mylist1.push_back("item 2");
mapex["foo"] = mylist1;
list<string> mylist2;
mylist2.push_back("item 3");
mylist2.push_back("item 4");
mylist2.push_back("item 5");
mapex["bar"] = mylist2;
for_each(mapex.begin(), mapex.end(), printPair);
}
You can write a quite generic overloaded function, which is good for two purposes:
It works with any map.
It allows for using <<.
The function is
template<class key_t, class value_t>
ostream& operator<<(ostream& os, const map<key_t, value_t>& m) {
for (typename map<key_t, value_t>::const_iterator it = m.begin();
it != m.end(); it++) {
os << "Key: " << it->first << ", Value: " << it->second;
}
return os;
}
cout << will work with any map for which << is defined for typenames key_t and value_t. In your case, this is not defined for value_t (= list<string>), so you also have to define it.
In a similar spirit, you can use
template<class T>
ostream& operator<<(ostream& os, const list<T>& l) {
for (typename list<T>::const_iterator it = l.begin(); it != l.end(); it++) {
os << "\"" << *it << "\", ";
}
return os;
}
So, you may:
Add these two functions.
Add the prototypes where needed.
Use using namespace std; (or add std:: as needed).
Use, e.g.,
cout << mapex << endl;
cout << li << endl;
Remember that if there is any other viable candidate for the <<s just defined (which I take there is not, otherwise you would likely not ask this question), it may take precedence over the present ones.
If you can use C++11 features, then I think range-based for loops as proposed in The Paramagnetic Croissant's answer provide the most readable option. However, if C++17 is available to you, then you can combine those loops with structured bindings to further increase readability, because you no longer need to use the first and second members. For your specific use case, my solution would look as follows:
std::map<std::string, std::list<std::string>> mapex;
mapex["a"] = { "1", "2", "3", "4" };
mapex["b"] = { "5", "6", "7" };
for (const auto &[k, v] : mapex) {
std::cout << "m[" << k.c_str() << "] =";
for (const auto &s : v)
std::cout << " " << s.c_str();
std::cout << std::endl;
}
Output:
m[a] = 1 2 3 4
m[b] = 5 6 7
Code on Coliru