C++ boost::format
Description: As in the c language printf("%3d",&a) is used for the format in which we want to print for this only the alternative is boost::format which is used here for the higher order numbers . boost::format is a class in which there is a string containing special characters to control formatting which is passed to the constructor of boost::function class.The data that replaces these special characters in the output is linked via the operator%.
//writing two customised namespace for easy readability
using namespace std;
namespace MyNS_ForOutput {
using std::cout;
using std::cerr;
using std::string;
using std::endl;
using std::flush;
using boost::format;
using boost::io::group;
using boost::io::str;
}
namespace MyNS_Manips {
using std::setfill;
using std::setw;
using std::hex ;
using std::dec ;
}
int main()
{
using namespace MyNS_ForOutput;// these are two customised namespace names
using namespace MyNS_Manips;
cout<<"these simple use of boost::format:"<<endl;
/*
as I explained in the derivatives of format string %spec so some of the examples are explained below
as %$1 %$2 it means that it should take first and second argument respectively .
after that it has + sign which will print the number as it is and 5 width as we notice output it will show
that it has that 5 width space for -23 and same for 35 . and though it is not decimal point but still the
conversion specifier is d because it covers almost all types so when we are not sure we gave d
and if when it comes to alignment then by default it right side alignment and others we can see in output
*/
cout<<"normal use \n";
cout <<boost::format("(x,y) = (%1$+5d,%2$-5d ,%2$=5d) \n") % -23 % 35; // positional syntax
/* Note: conversion specifier does not strictly imposed on the arguments but it is used for ios flag
as u can see for string it is s but if u change it d also it does not effect our output
as the clear difference is there in next line where 18 is shown as x ,o , s
*/
cout<<"\n use of different conversion specifier\n";
cout << boost::format("writing %s, x=%d : %d-th step \n") % "toto" % 40.23 % 50;
cout <<boost::format("_%1$4d_ is : _%1$#4x_ in hexadecimal, _%1$#4o_ in octal, and _%1 in string \n") % 18;
// it is the example of %N% where spec format is same
cout<<"\n use of %N% \n";
cout << boost::format("%1%.%2%.%3%") % 12 % 5 % 2014 << '\n';
// it is the example of %|spec| and here also spec has same format as first one
cout<<"\n use of %|spec| \n";
cout << boost::format("%|1$1| %|2$20|") % "Hello" % 3 << std::endl;
cout<<"here below it is shown that same output can be there with different format writing"<<endl;
cout <<boost::format("(x,y) = (%+5d,%+5d) \n") % -23 % 35; //non positional syntax
cout << boost::format("(x,y) = (%|+5|,%|+5|) \n") % -23 % 35;
cout << boost::format("(x,y) = (%|1$+5|,%|2$+5|) \n") % -23 % 35;
/* know here it is shown that boost::format can be used with ios flag manipulators
as the internal format stream can be changed by manipulators only using the group function
*/
cout<<"\n use of manipulators \n";
cout <<boost::format("%2% %1% %2%\n") % 1 % group(setfill('X'), hex, setw(4), 16+3) ;
/* know below we can see that assertion is used as we know assertion is a sanity check that you
can turn on and off when you have finished testing
so by putting the format under one string and checking it
*/
cout<<"\n use of assertion \n";
std::string s;
s= str( format(" %d %d ") % 11 % 22 );
assert( s == " 11 22 ");
cout << format("%%##%#x ") % 20 << endl;
cout<<"added features in printf"<<endl<<endl;
cout<<"same arguments can be used number of times \n";
cout << format("%1% %2% %3% %2% %1% \n") % "o" % "oo" % "O";
// Tabulations : "%|Nt|" => tabulation of N spaces.
// "%|NTf|" => tabulation of N times the character <f>.
/* we want to print multiple values at a time so we can use the simple vector concpet where
parameterized constructor is used as container of size 1 is created
*/
cout<<"\n use of vector\n";
vector<string> names(1, "sunita"),
surname(1,"konada"),
tel(1, "6745218890");
// apart from the container size we can manually add the values using push_back function
names.push_back("chaitanya");
surname.push_back("gunda sai");
tel.push_back("6777290123");
for( int i=0; i<names.size(); ++i) // as we can see boost::format is used with vector
cout << format("%1%, %2%, %|20t|%3%\n") % names[i] % surname[i] % tel[i]; // and with vector it will use 3rd syntax thats why additionally
cout << "\n"; //width is used
// truncation s of strings as s takes the string and trunc it to two characters and for c
// it will take the single character from the string user
cout<<"\n truncation of string \n";
cout << format("%|.2s| %|8c|.\n") % "root" % "user";
// here we can see the nesting of the format string and the manipulators
cout<<"\n nested output \n";
cout << format("%2$014x [%1%] %2$05s\n") % (format("%05s / %s") % -18 % 7)
% group(showbase, -100);
}
The format object is constructed from a format-string and is then given arguments through repeated calls to operator%.Each of these arguments are then converted to strings ,who are in turn combined into one string ,according to format-string
Syntax: General syntax of format string is:
boost::format(format string)%arg1,%arg2,....%argN
Directives in format string:
1.printf format string : In this %spec is used as the general syntax .the printf format supported boost.format follows the Unix98 open group printf precise syntax rather than the standard c language
The specification spec has the form :
[ N$ ] [ flags ] [ width ] [.precision] [argument-type] conversion-specifier. (any of the field any be optional)
> [ N$ ] : it is known as positional format specification .if it is not present then the argument will be taken one by one
> [flag] : flag is sequence of any of these:
> [width]: it specifies the minimal width of the string resulting from the conversion.
>[.precision]: it sets the stream precision .when output is the float type number,it sets the maximum number of digits after decimal point in fixed or scientific mood and when in default mood(eg: %g)
>[argument-type]: in printf family it is used to process the argument but in boost::format it is used with % operator which allows the template to carry the argument type.Argument type hh,h,l,ll,j,z and L are recognized as the Microsoft extensions w,I,i32,i64.
>conversion-specifier:
2.%|spec| : where spec is same as printf format specification as i explained in first point.
the pip delimited syntax is introduced to improve the readability of the format string but primarily to make the type conversion optional in spec
3.%N%: this is the simple positional notation request the formatting of the nth argument without any formatting option
Manipulators : Here the arguments can be modified using ios flags through group function in the group function(group()) the last argument in the function is the actual numbers for ios manipulators go to the link https://doc.bccnsoft.com/docs/cppreference_en/io_flags.html
Advantages:
1.safe : type-safe,and throws an exception ,for example the specification of too-many or too few arguments
2.extensible : work for any type that can be streamed
3.convenient: standard posix and similar format string
program:
//writing two customised namespace for easy readability
using namespace std;
namespace MyNS_ForOutput {
using std::cout;
using std::cerr;
using std::string;
using std::endl;
using std::flush;
using boost::format;
using boost::io::group;
using boost::io::str;
}
namespace MyNS_Manips {
using std::setfill;
using std::setw;
using std::hex ;
using std::dec ;
}
int main()
{
using namespace MyNS_ForOutput;// these are two customised namespace names
using namespace MyNS_Manips;
cout<<"these simple use of boost::format:"<<endl;
/*
as I explained in the derivatives of format string %spec so some of the examples are explained below
as %$1 %$2 it means that it should take first and second argument respectively .
after that it has + sign which will print the number as it is and 5 width as we notice output it will show
that it has that 5 width space for -23 and same for 35 . and though it is not decimal point but still the
conversion specifier is d because it covers almost all types so when we are not sure we gave d
and if when it comes to alignment then by default it right side alignment and others we can see in output
*/
cout<<"normal use \n";
cout <<boost::format("(x,y) = (%1$+5d,%2$-5d ,%2$=5d) \n") % -23 % 35; // positional syntax
/* Note: conversion specifier does not strictly imposed on the arguments but it is used for ios flag
as u can see for string it is s but if u change it d also it does not effect our output
as the clear difference is there in next line where 18 is shown as x ,o , s
*/
cout<<"\n use of different conversion specifier\n";
cout << boost::format("writing %s, x=%d : %d-th step \n") % "toto" % 40.23 % 50;
cout <<boost::format("_%1$4d_ is : _%1$#4x_ in hexadecimal, _%1$#4o_ in octal, and _%1 in string \n") % 18;
// it is the example of %N% where spec format is same
cout<<"\n use of %N% \n";
cout << boost::format("%1%.%2%.%3%") % 12 % 5 % 2014 << '\n';
// it is the example of %|spec| and here also spec has same format as first one
cout<<"\n use of %|spec| \n";
cout << boost::format("%|1$1| %|2$20|") % "Hello" % 3 << std::endl;
cout<<"here below it is shown that same output can be there with different format writing"<<endl;
cout <<boost::format("(x,y) = (%+5d,%+5d) \n") % -23 % 35; //non positional syntax
cout << boost::format("(x,y) = (%|+5|,%|+5|) \n") % -23 % 35;
cout << boost::format("(x,y) = (%|1$+5|,%|2$+5|) \n") % -23 % 35;
/* know here it is shown that boost::format can be used with ios flag manipulators
as the internal format stream can be changed by manipulators only using the group function
*/
cout<<"\n use of manipulators \n";
cout <<boost::format("%2% %1% %2%\n") % 1 % group(setfill('X'), hex, setw(4), 16+3) ;
/* know below we can see that assertion is used as we know assertion is a sanity check that you
can turn on and off when you have finished testing
so by putting the format under one string and checking it
*/
cout<<"\n use of assertion \n";
std::string s;
s= str( format(" %d %d ") % 11 % 22 );
assert( s == " 11 22 ");
cout << format("%%##%#x ") % 20 << endl;
cout<<"added features in printf"<<endl<<endl;
cout<<"same arguments can be used number of times \n";
cout << format("%1% %2% %3% %2% %1% \n") % "o" % "oo" % "O";
// Tabulations : "%|Nt|" => tabulation of N spaces.
// "%|NTf|" => tabulation of N times the character <f>.
/* we want to print multiple values at a time so we can use the simple vector concpet where
parameterized constructor is used as container of size 1 is created
*/
cout<<"\n use of vector\n";
vector<string> names(1, "sunita"),
surname(1,"konada"),
tel(1, "6745218890");
// apart from the container size we can manually add the values using push_back function
names.push_back("chaitanya");
surname.push_back("gunda sai");
tel.push_back("6777290123");
for( int i=0; i<names.size(); ++i) // as we can see boost::format is used with vector
cout << format("%1%, %2%, %|20t|%3%\n") % names[i] % surname[i] % tel[i]; // and with vector it will use 3rd syntax thats why additionally
cout << "\n"; //width is used
// truncation s of strings as s takes the string and trunc it to two characters and for c
// it will take the single character from the string user
cout<<"\n truncation of string \n";
cout << format("%|.2s| %|8c|.\n") % "root" % "user";
// here we can see the nesting of the format string and the manipulators
cout<<"\n nested output \n";
cout << format("%2$014x [%1%] %2$05s\n") % (format("%05s / %s") % -18 % 7)
% group(showbase, -100);
}
output:
these simple use of boost::format:
normal use
(x,y) = ( -23,35 , 35 )
use of different conversion specifier
writing toto, x=40.23 : 50-th step
_ 18_ is : _0x12_ in hexadecimal, _ 022_ in octal, and _18_ in string
use of %N%
12.5.2014
use of %|spec|
Hello 3
here below it is shown that same output can be there with different format writing
(x,y) = ( -23, +35)
(x,y) = ( -23, +35)
(x,y) = ( -23, +35)
use of manipulators
XX13 1 XX13
use of assertion
%##0x14
added features in printf
same arguments can be used number of times
o oo O oo o
use of vector
sunita, konada, 6745218890
chaitanya, gunda sai, 6777290123
truncation of string
ro u.nested output
0x0000ffffff9c [-0018 / 7] -0100
Note: As we seen there non-positional format(%d) and positional format (%$4) and they should not be used together // unsupported printf directives %n and asterisk-fields are purely ignored.
// do *NOT* provide an argument for them, it is an error.
cout << format("|%5d| %n") % 7 << endl;
// prints "| 7| "
cout << format("|%*.*d|") % 7 << endl;
// prints "|7|"
Comments