It is a helper, lightweight JavaScript-like C++ type used for routine tasks.
The key word for the intention to use this type 'var' is 'sometimes', you should
not plug all the 'holes' with this type
( this is especially true for beginners,experienced developers know the price of wrappers and will not abuse it ),
but 'sometimes' some problems can be easily and conveniently solved by this type.
In my opinion, the correct scope is:
-
while initialization process: ( for further use )
- string parsing
- creation of complex arrays, strings, maps ( as storage objects )
- config files
- ...
-
sometimes as complex/variable return result and/or complex/variable parameter(s) for functions
-
two sided network communication using a special map string (map_str)
-
complex arrays, maps to maintain:
- errors
- logs
- exceptions
- states
- ...
If it seems to you that something is clearly not working as it should,
please write me.
( as short example/description/other way )
- include "var.h" before using
#include "your_path/var.h"- use:
var S = "Hello, World!";
...- example:
// HW.cpp
#include "var.h"
int main()
{
var S = "Hello, World!";
S.prnt(); // "Hello, World!"
return 0;
}- Compiling a program ( all files in the same directory )
win-case:
c++ HW.cpp -o HW.exe -L. -lvar_win
./HW.exe
linux-case:
c++ HW.cpp -o HW -L. -lvar_lnx
./HW
bsd-case:
c++ HW.cpp -o HW -L. -lvar_bsd
./HW
macOS-case:
c++ HW.cpp -o HW -L. -lvar_mac -std=c++17
./HW
Since var:arr( array ) is a wrapper of the std::vector, ( almost ) all
std::vector interfaces are implemented and can be used. ( for more details see var.h )
var s = "STR";
var a = { { "arr" }, { 2 } }, udf, map, pmap;
map.map_set(""
" { math: { const: pi: 3.141593; "
" e: 2.718282; "
" }"
" }");
pmap = map.map_get("math:const:e");
// int(longlong)
// | string iterator: I_str another arrays
// | | | | / / |
var A = { 1, 0.555, "..text..", 'c', s.begin(), udf, a, { { 3 }, { a.begin() } },
// | | | |
// maps double char undefined iterator: I_arr
// |
map, true, var( { }), pmap };
// | | |
// bool empty array map-pointer: var:P_map
// note: all var:X are acceptable
//
A._str().prnt(); // {1,0.555000,"..text..",'c',| [var:I_str] val: 'S' |,
// |udf|,{ {"arr"},{2}},{ {3},{| [var:I_arr] val: {"arr"} |}},
// {math:{const:e:2.718282;pi:3.141593;} },true,{},
// | [var:P_map] val: 2.718282 |}
// or
A.prnt(); // [ 12 ]{ 1,
// 0.555000,
// "..text..", ("STR" -> first character)
// 'c', ^
// | [var:I_str] val: S |,
// |udf|,
// [ 2 ]{ [ 1 ]{ "arr" },
// [ 1 ]{ 2 } },
// [ 2 ]{ [ 1 ]{ 3 },
// [ 1 ]{ | [var:I_arr] val: {"arr"} | } },
// { math: { const: e: 2.718282;
// pi: 3.141593;
// }
// },
// true,
// { },
// | [var:P_map] val: 2.718282 | }
...deep access
var A = { '>', { '>', { '>', { " ! ! " }, '<' }, '<' }, '<' };
//
A._str().prnt(); // {'>',{'>',{'>',{" ! ! "},'<'},'<'},'<'}
//
// All var-interfaces (supported by this type) will work like this one.
// ^ note: Some IDE (not the compiler)
// | marking this way of access as error.
// | (need 'dancing with a tambourine' to solve this :) )
// |
A[1][1][1][0].insert(3, "deep insert");
A._str().prnt(); // {'>',{'>',{'>',{" ! deep insert ! "},'<'},'<'},'<'}
A[1][1].at(1).at(0).push_back(":) ");
A._str().prnt(); // {'>',{'>',{'>',{" ! deep insert ! :) "},'<'},'<'},'<'}
//
var x = A.join();
x.prnt(); // >>> ! deep insert ! :) <<<
...Since var:str( string ) is a wrapper of the std::string, ( almost ) all
std::string interfaces are implemented and can be used. ( for more details see var.h )
// Some features from high-level languages: ( new features will be added with time )
var S = " s t r i n g ";
S.trim().prnt(); // "s t r i n g"
S.ssplit(' ');
S._str().prnt(); // {'s','t','r','i','n','g'}
S.push_back('!');
S._str().prnt(); // {'s','t','r','i','n','g','!'}
S.sjoin();
S.prnt(); // "string!"
var i = 12345678, s;
s = i._str();
s.prnt(); // "12345678"
int ii = s.substr(3, 3)._int();
std::cout << ii << std::endl; // 456
...Maps are convenient tool with access by field-name.
In my opinion var:map could be used:
- for config files
- for transfer through network (as special 'map string')
- maintenance of the: errors,logs,exceptions,...
- while runtime as storage objects
- ... etc
// MAP_SET: - Takes c++ 'like' string (null-terminated) and parse it to the var:map.
var x;
// I. Common map_set syntax:
x.map_set(""
" { cars: { germany: { MercedesBenz: B_Class: 2; "
" C_Class: 3; "
" } "
" Opel: 4; "
" BMW: 5; "
" } "
" { american: Dodge: 5; "
" Lincoln: 9; "
" Ford: 7; "
" } "
" other: 6; " // ................... (1)
" } " // | |
" { movies: { action: _1: \" No Time to Die \"; "
" _2: \" Mission: Impossible - Fallout \"; "
" _3: \"...\"; "
" } "
" { fiction: _1: \" Edge Of Tomorrow \"; "
" _2: \" The Martian \"; "
" _3: \"...\"; "
" } "
" undefined:; "
" } "
" { math: pi: 3.141593; "
" e: 2.718282; "
" }");
// ^
x.prnt(); // the output will have the same format, but without extra slashes
// !!!!!!!!!!!!! Note: data are always stored in sorted ascii-order !!!!!!!!!!!!!
// II. Simple map_set syntax: // ...
// { germany: BMW: 5;
x.map_set("cars:germany:Volkswagen:3;"); // { MercedesBenz: B_Class: 2;
// C_Class: 3;
// }
// Opel: 4;
// Volkswagen: 3; <-- reassigned
// }
// ...
// III. Complex-mask map_set syntax: // { cars: { Italy: Fiat: 2; <-- new
// }
x.map_set("{ cars: { france: Renault: 7;" // { american: Dodge: 0; <-- reassigned
" Peugeot: 4;" // Ford: 0; <-- reassigned
" }" // Lincoln: 0; <-- reassigned
" { american: Dodge: 0;" // }
" Lincoln: 0;" // { france: Peugeot: 7; <-- new
" Ford: 0;" // Renault: 4; <-- new
" }" // }
" other: 10;" // { germany: BMW: 4;
" Italy: Fiat: 2;" // { MercedesBenz: B_Class: 2;
"}"); // C_Class: 3;
// }
// Opel: 4;
// Volkswagen: 3;
// }
// other: 10; <-- reassigned
// }
// ...
// IV. Direct map_set case:
if (var w = x.map_get( "cars:germany:MercedesBenz:C_Class")) { // ...
// { MercedesBenz: B_Class: 2;
*w = 100; // C_Class: 100;
// ^
} else var("fail").prnt(); // } reassigned
// ...
// V. List map_set case:
var lm; // s1: "s1";
// s2: "s2";
lm = var().set_map(""); // s3: "...";
m = var().map_set("v1:1;v2:2;v3:3;s1:\"s1\";s2:\"s2\";s3:\"...\";"); // v1: 1;
// v2: 2;
// v3: 3;
/*
* Notes about slashes '\':
*
* You have to use slashes for const char* literals (1) and use
* map_fread()-rules ( see map_fread() [R] ) for hand made strings
* like: char*(null-terminated),std::string.
*
* const char* literals:
* 3+" 3+\
* the rules are: \\\" for " and \\\\ for \
*/
// ...
x.map_set("dir: \" D:\\\\music\\\\Accept \";"); // dir: " D:\music\Accept "; // (win)
// ...
x.map_set("txt: \" Saying \\\"Stop the car\\\" was a mistake \";");
// ...
// txt: " Saying "Stop the car" was a mistake ";
// ...
/* 3+3+\\ 3+\+3+"
* Also possible: \\\\\\\\ -> \\ or \\\\\\\" -> \" and so on.
*/
...// MAP_GET: - Getting values from the var:map.
var x;
x.map_set(""
" { cars: { germany: { MercedesBenz: B_Class: 2; "
" C_Class: 3; "
" } "
" } "
" } "
" { movies: { action: _1: \" No Time to Die \"; "
" } "
" } ");
// Good to use: (with checks)
var w;
// ...
if ((w = x.map_get("cars:germany"))) { // { MercedesBenz: B_Class: 2;
(*w).prnt(); // print --> // C_Class: 3;
} else var("fail").prnt(); // }
// ...
if ((w = x.map_get("cars:germany:MercedesBenz:C_Class"))) {
*w = 100; // direct map_set
} else var("fail").prnt();
if ((w = x.map_get("movies:action:_1"))) {
(*w).push_back(":) :) :)"); // just changes value
} else var("fail").prnt();
// { cars: { germany: { MercedesBenz: B_Class: 2;
// C_Class: 100; <-- reassigned
// }
// }
// }
// { movies: { action: _1: " No Time to Die :) :) :)"; <-- changed
// }
// }
// excp: // excp: [ invalid_map_pointer ] [ var:P_map ]
//w = x.map_get("movies: fiction:_1"); <- excp // at: var& operator*()
//(*w).prnt(); // <--- excp // dscr: field not exist or became invalid
...// MAP_REM: - Remove value(s) from the var:map.
var x;
x.map_set(""
" { cars: { germany: { MercedesBenz: B_Class: 2; "
" C_Class: 3; "
" } "
" Opel: 4; "
" BMW: 5; "
" } "
" { american: Dodge: 5; "
" Lincoln: 9; "
" Ford: 7; "
" } "
" other: 6; "
" } ");
// I. Removes a field:
x.map_rem("cars:germany:MercedesBenz:C_Class");
// { cars: { american: Dodge: 5;
// Ford: 7;
// Lincoln: 9;
// }
// { germany: BMW: 5;
// { MercedesBenz: B_Class: 2;
// }
// Opel: 4;
// }
// other: 6;
// }
// II. Removes a group of fields:
x.map_rem("cars:american");
// { cars: { germany: BMW: 5;
// { MercedesBenz: B_Class: 2;
// }
// Opel: 4;
// }
// other: 6;
// }
// III. Removes a complex group of fields:
x.map_rem("cars");
// { } <-- Or you can use x.map_free() to get the same result.
...// MAP_FOREACH: - Iterations on the var:map fields. (main purpose is collection of information)
var x;
//depth: 1 2 3 4
x.map_set(""
" { cars: { germany: { MercedesBenz: B_Class: 2; "
" C_Class: 3; "
" } "
" Opel: 4; "
" BMW: 5; "
" } "
" } "
//depth: 1 2 3
" { movies: { action: _1: \" No Time to Die \"; "
" _2: \" Mission: Impossible - Fallout \"; "
" _3: \"...\"; "
" } "
" }"
" val: 123; ");
struct CTX {
int tot = 0;
// ...
} _ctx;
// ----- iterations order:
x.map_foreach([](auto field, auto val, auto ctx) { // cars |
// // germany |
CTX *_ctx_ = (CTX*) ctx; // explicit cast // BMW |
// // MercedesBenz |
++_ctx_->tot; // doing the work // B_Class |
var(field).prnt(); // C_Class |
// // Opel |
return var::map::nxt; // continue // movies |
// // action |
},&_ctx); // _1 |
// ^ pass the context pointer // _2 |
// // _3 |
// // val V
// -----
x.map_foreach([](auto field, auto val, auto ctx) { // cars |
// germany |
var(field).prnt(); // movies |
return var::map::nxt; // action |
// val V
},0, 2);
// ^ set the depth
// ----- ex* If you really want to change value while iterating you need
// to drop 'const' i.e. const_cast<var&>(val).do_something().
// ( it's safe, but (!) )
x.map_foreach([](auto field, auto val, auto ctx) {
if (val.is_map()) {
const_cast<var&>(val) = //
var().map_set("{new_field: pi: 3.141593; e: 2.718282;}");
// (!)
// When you set one var:map to another var:map(current) you can get
// 'infinite recursion' and catch [ map_max_depth ]
// (because 'map_foreach' will go into newly created var:map etc .. ).
// So you need to use var::map::skp to skip the newly created var:map
// or take control of it by yourself
//
// You can also skip (by var::map::skp) the current var:map if
// it doesn't make sense to go into it.
return var::map::skp;
}
// other cases:
const_cast<var&>(val) = 777;
return var::map::nxt;
},&_ctx);
// reassigned
// ^
// |
// { cars: { new_field: e: 2.718282;
// | pi: 3.141593;
// } |
// } |
// { movies: { new_field: e: 2.718282;
// pi: 3.141593;
// }
// }
// val: 777; <-- reassigned
...// MAP_FREAD: - Getting data from the file with name 'fname' and parse it to the var:map.
file: "map" (example of the file)
{ cars: { germany: { MercedesBenz: B_Class: 2;
C_Class: 3;
}
Opel: 4;
BMW: 5;
}
{ american: Dodge: 5;
Lincoln: 9;
Ford: 7;
}
other: 6; .................. (1)
} | |
{ movies: { action: _1: " No Time to Die ";
_2: " Mission: Impossible - Fallout ";
_3: "...";
}
{ fiction: _1: " Edge Of Tomorrow ";
_2: " The Martian ";
_3: "...";
}
undefined:;
}
{ math: pi: 3.141593;
e: 2.718282;
}
temperature: -20;
// Good to use: (with checks)
var x, w;
if ((w = x.map_fread("bad_map_name"))[0]) {
x.prnt();
} else w[1].prnt(); // err: Unable to open file
if ((w = x.map_fread("map"))[0]) {
x.prnt(); // ok -> now it can be used like var:map
} else w[1].prnt();
/*
* Notes about slashes '\':
*
* - no need to use slashes for string values (1)
* 1+" 1+\
* - [R] the rules are: \" for " and \\ for \
* 1+\+1+\ 1+\+1+"
* - also possible: \\\\ -> \\ or \\\" -> \" and so on
*
*/
// MAP_FWRITE: - Writes data to the file with name 'fname'.
// NOTE: - All var:I_str,var:Ir_str,var:I_arr,var:Ir_arr,var:P_map
// will be defined as var::udf(undefined). (because of nothing to store)
// - var:arr (array) also will be defined as var::udf. (at the moment)
// ! but you can freely use them at runtime !
// Some changes:
srand((*x.map_get("cars:germany:MercedesBenz:C_Class")). //
_str().sample()._int());
*(x.map_get("temperature")) = rand();
*(x.map_get("cars:germany:MercedesBenz:C_Class")) = rand();
// write to file:
if (var w = x.map_fwrite("map")[0]) {
var("ok").prnt();
} else w[1].prnt(); // err-string
// NOTE: map_fread/map_fwrite are not thread safe (at the moment)
...// MAP_STR: - Makes special 'map-string' that can be easily parse back to the var:map.
// It can be used: for network transfer, in files defined by user,
// in databases, other ways ..
var x, s;
x.map_set(""
" { cars: { germany: { MercedesBenz: B_Class: 2; "
" C_Class: 3; "
" } "
" Opel: 4; "
" BMW: 5; "
" } "
" { american: Dodge: 5; "
" Lincoln: 9; "
" Ford: 7; "
" } "
" str: \" Saying \\\"Stop the car\\\" \"; "
" udf:; "
" } ");
// {cars:{american:Dodge:5;Ford:7;Lincoln:9;}{germany:BMW:5;
s = x.map_str(); // {MercedesBenz:B_Class:2;C_Class:3;}Opel:4;}str:" Saying \"Stop
// the car\" ";udf:;}
s = x.map_str(true); // formatted output (with spaces and newlines)
...// INVALID POINTER OF THE MAP.
// The are four main causes of invalid map pointers:
// I/II. Set/Remove cases:
var map;
var p1; // <-------------is become invalidate-----------------.
var p2; // <-------------is become invalidate---------------. |
// | |
map.map_set("" // | |
" { cars: { germany: MercedesBenz: 6; "// | |
" Opel: 4; "// | |
" BMW: 5; "// | |
" }"// | |
" }");// | |
// | |
p1 = map.map_get("cars:germany:Opel"); // | |
p2 = map.map_get("cars:germany:BMW"); // | |
// | |
map.map_rem("cars:germany:Opel"); //-----case(rem)---------|-'
p1.prnt(); // | [var:P_map] is invalid | |
map.map_set("cars:germany:0;"); //-----case(reasign)-----'
p2.prnt(); // | [var:P_map] is invalid |
// III. Source destroy case:
var ext_p1; // <-----------------------------------------------.
// are become invalidate |
var ext_p2; // <-----------------------------------------------.
// |
{ // |
var __map; // |
__map.map_set("" // |
" { cars: { germany: MercedesBenz: 6; "// |
" Opel: 4; "// |
" BMW: 5; "// |
" }"// |
" }");// |
// |
ext_p1 = __map.map_get("cars:germany:Opel"); // |
ext_p1.prnt(); // | [var:P_map] val: 4 | |
ext_p2 = __map.map_get("cars:germany:BMW"); // |
ext_p2.prnt(); // | [var:P_map] val: 5 | |
// |
}// |
// cause: |
// end of the block scope -> source (__map) is destroyed --->'
ext_p1.prnt(); // | [var:P_map] is invalid |
ext_p2.prnt(); // | [var:P_map] is invalid |
// IV. Field not exist:
var p = map.map_get("cars:xxx");
p.prnt(); // | [var:P_map] is invalid |
// use
// excp: [ invalid_map_pointer ] [ var:P_map ]
//(*p); // at: var& operator*()
// dscr: field not exist or became invalid
...// MAP_EQUAL/MAP_MISMATCH:
var m1, m2, m3, m4, x;
m1.map_set(""
" { cars: { germany: MercedesBenz: 6; "
" Opel: 4; "
" BMW: 5; "
" }"
" }");
m3.map_set(""
" { cars: { germany: MercedesBenz: 6; " // different
" Opel: 4; " // sizes
" }"
" }");
m4.map_set(""
" { cars: { germany: MercedesBenz: 6; " // different
" Opel: 4; " // fields
" BMW_x: 5; "
" }"
" }");
m2 = m1; // copy: m1 -> m2
x = m1.map_mismatch(m2); // false
x = m1.map_equal(m2), ; // true
m2.map_set("cars:germany:Opel:777;"); // change value
x = m1.map_mismatch(m2); // ------> [ 3 ]{ "cars:germany:Opel",
x = m1.map_equal(m2); // false | [var:P_map] val: 4 |,
// | [var:P_map] val: 777 | }
x = m1.map_mismatch(m3); // ------> [ 3 ]{ "cars:germany",
x = m1.map_equal(m3); // false | [var:P_map] val: BMW:5;MercedesBenz:6;Opel:4; |,
// | [var:P_map] val: MercedesBenz:6;Opel:4; | }
x = m1.map_mismatch(m4); // ------> [ 3 ]{ "cars:germany:BMW",
x = m1.map_equal(m4); // false | [var:P_map] val: 5 |,
// | [var:P_map] is invalid | }
// ^
// sign of fields mismatch
// ( not exist )
...// SORT/S(elf)SORT:
var s = "str", sb = s.begin(), srb = s.rbegin(), udf;
var a = { 1 }, ab = a.begin(), arb = a.rbegin();
var map, pmap1, pmap2;
map.map_set(""
" { cars: { germany: MercedesBenz: 6; "
" Opel: 4; "
" BMW: 5; "
" }"
" }");
pmap1 = map.map_get("cars:germany:BMW");
pmap2 = map.map_get("cars:germany:Opel");
var A = { a, sb, "AB", ab, 2.5, "ABC", false, var( { }), srb, { 1, 2 }, //
pmap2, 'A', udf, 1, map, var(), arb, 'B', true, pmap1 };
var ra = A.sort(); // ra = "[ 20 ]{ false, booleans
// true, |
// 1, numbers
// 2.500000, |
// 'A', chars
// 'B', |
// "AB", strings
// "ABC", |
// { }, arrays
// [ 1 ]{ 1 }, |
// [ 2 ]{ 1, |
// 2 }, |
// { cars: { germany: BMW: 5; maps
// MercedesBenz: 6; |
// Opel: 4; |
// } |
// }, |
// | [var:I_str] val: s |, iterators
// | [var:Ir_str] val: r |, |
// | [var:I_arr] val: 1 |, |
// | [var:Ir_arr] val: 1 |, |
// | [var:P_map] val: 4 |, map-pointers
// | [var:P_map] val: 5 |, |
// |udf|, undefined
// |udf| } |
var S = "cB3bC2A1a";
var rs = S.sort(); // rs = "123ABCabc"
// or
a = { 6, 3, 7, 2, 9, 4, 1, 8, 0, 5 };
a.ssort_cmp([](auto a, auto b) {
return b < a;
},true);
a._str().prnt(); // {9,8,7,6,5,4,3,2,1,0}
a.ssort_cmp([](auto a, auto b) {
return b < a;
},false);
a._str().prnt(); // {0,1,2,3,4,5,6,7,8,9}
a = { 3, 7, "Abc", 2, 9, "ABC", 1, 0 };
a.ssort_cmp([](auto a, auto b) {
if (a.is_str()) {
if (b.is_str()) return b < a;
} else if (a.is_int() || a.is_double()) {
if (b.is_int() || b.is_double()) return b < a;
else if (b.is_str()) return true;
}
return false;
});
a._str().prnt(); // {"ABC","Abc",0,1,2,3,7,9}
// {9,7,3,2,1,0,"Abc","ABC"} (reverse=true)
...// SAMPLE/SAMPLE_UNQ:
var a = { 1, 2, 3, 4, 5, 6, 7, 8, 9 }, x;
x = a.sample(5); // x = "{7,8,2,7,6}"
x = a.sample(); // x = "{8,6,5,7,5,1,7,4,6}"
var s = "ABCDEFGHIJKLMNOPQ";
x = s.sample(10); // x = "LMIEOLPAMN"
x = s.sample(); // x = "CCDQOMNLKHECMFJLN"
// unique sample
x = a.sample_unq(5); // x = "{5,2,6,3,9}"
x = a.sample_unq(); // x = "{2,6,3,4,1,5,7,9,8}"
x = s.sample_unq(10); // x = "NGCDBKJLHM"
x = s.sample_unq(); // x = "JMCQBAKNGILEOFHPD"
...// EQUAL/MISMATCH:
// arr:
// <--------------------
// -------------------->
var a = { 'r', 'a', 'd', 'a', 'r' }, udf, x;
res = a.equal(a.rbegin(), a.rend(), 0); // true: "radar" is a palindrome
var a1 = { 1, 2, 3, 4, 5, 6, 7, 8, 9 }, a2 = { 3, 4, 5, 6, 7 };
// ''''''''' '''''''''
x = a1.equal(a2.begin() += 1, a2.end() -= 1, 3, 3); // true
// different types example:
var m = var().map_set("x:1;y:2;z:3;");
var a3 = { 1, 2.2, udf, true, { { { 1 } } }, "str", m, 'c', a.begin() };
var a4 = { 1, 2.2, udf, true, { { { 1 } } }, "str", m, 'c', a.begin() };
x = a3.equal(a4.begin(), a4.end()); // true
// str: (same rules as for arrays)
var s = "level";
x = s.equal(s.rbegin(), s.rend()); // true: "level" is a palindrome
// ----
a1 = { 1, 2, 3, 4 };
a2 = { 1, 2, 3, 4 };
x = a1.mismatch(a2.begin(), a2.end()); // false
a1 = { 1, 2, 3, 4 }; // x = [ 4 ]{ | [var:I_arr] val: 4 |,
a2 = { 1, 2, 3, 5 }; // | [var:I_arr] val: 5 |,
// |udf|,
x = a1.mismatch(a2.begin(), a2.end()); // |udf| }
a1 = { 1, 2, { 1, 2, 3 }, 3 }; // x = [ 4 ]{ | [var:I_arr] val: {1,2,3} |,
a2 = { 1, 2, { 1, 2, 4 }, 3 }; // | [var:I_arr] val: {1,2,4} |,
// | [var:I_arr] val: 3 |,
x = a1.mismatch(a2.begin(), a2.end()); // | [var:I_arr] val: 4 | }
a1 = { 1, 2, { 1, 2, 3 }, 3 }; // x = [ 4 ]{ | [var:I_arr] val: {1,2,3} |,
a2 = { 1, 2, { 1, 2 }, 3 }; // | [var:I_arr] val: {1,2} |,
// | [var:I_arr] val: 3 |,
x = a1.mismatch(a2.begin(), a2.end()); // | [var:I_arr] at the end() | }
a1 = { 1, { 1, { 1, { 2, 3, { { { { 5 } } } } } } }, 2 };
a2 = { 1, { 1, { 1, { 2, 3, { { { { 4 } } } } } } }, 2 };
x = a1.mismatch(a2.begin(), a2.end());
// x = [ 4 ]{ | [var:I_arr] val: {1,{1,{2,3,{ { { {5}}}}}}} |,
// | [var:I_arr] val: {1,{1,{2,3,{ { { {4}}}}}}} |,
// | [var:I_arr] val: 5 |,
// | [var:I_arr] val: 4 | }
a1 = { 1, 2, 3, { 5, var().map_set("x:1;y:2;z:3;"), 6 }, 4 };
a2 = { 1, 2, 3, { 5, var().map_set("x:1;y:777;z:3;"), 6 }, 4 };
x = a1.mismatch(a2.begin(), a2.end());
// x = [ 4 ]{ | [var:I_arr] val: {5,x:1;y:2;z:3;,6} |,
// | [var:I_arr] val: {5,x:1;y:777;z:3;,6} |,
// | [var:I_arr] val: x:1;y:2;z:3; |,
// | [var:I_arr] val: x:1;y:777;z:3; | }
...// REVERS/S(elf)REVERS:
var x;
var a = { 1, 2, 3, 4, 5, 6, 7, 8, 9, 0 };
x = a.reverse(); // x = "{0,9,8,7,6,5,4,3,2,1}"
a.sreverse(); // a = "{0,9,8,7,6,5,4,3,2,1}"
var s = "ABCDEFGHIJKLMNOPQ";
x = s.reverse(); // x = "QPONMLKJIHGFEDCBA
s.sreverse(); // s = "QPONMLKJIHGFEDCBA
...// FILL:
var a = { 1, 1, 1, 1, 1, 1, };
a.fill(2, 1, 2); // "{1,2,2,1,1,1}"
a.fill(3, 6, 2); // "{1,2,2,1,1,1,3,3}"
var s = "......";
s.fill('!', 1, 2); // ".!!..."
s.fill('!', 6, 2); // ".!!...??"
...var base;
base.initArr();
// strict types of input parameters
// |
// ...................................
auto add = [&](cch *item, double price, int amount) {
if (price >= 0 && amount >= 0) {
//
// simple addition of the complex arrays
// |
base.push_back( { item, price, amount });
// .-------------------> this is copy of the
// status | item base[base.length() - 1]
//return var( { "ok", base[base.length() - 1] });
// | .-------------------> this is iterator (var:I_arr)
// or | | pointing to the
// | status | item base[base.length() - 1] -> directly
return var( { "ok", --base.end() });
// |
// verbose return
// |
} // | status reason
return var( { "fail", "why: bad 'price' or 'amount'" });
};
add("book", 1.55, 10)._str().prnt(); // {"ok",| [var:I_arr] val: {"book",1.550000,10} |}
add("apple", 0.3, 100)._str().prnt(); // {"ok",| [var:I_arr] val: {"apple",0.300000,100} |}
add("cup", -5, 4)._str().prnt(); // {"fail","why: bad 'price' or 'amount'"}
// JS approach
// dynamic resolution of input parameters
// (additional checks required)
// |
// ...............................
auto _add = [&](var item, var price, var amount) {
if (price >= 0 && amount >= 0) {
base.push_back( { item, price, amount });
return var( { "ok", base[base.length() - 1] });
}
return var( { "fail", "why: bad 'price' or 'amount'" });
};
_add("book", 1.55, 10)._str().prnt(); // {"ok",{"book",1.550000,10}}
//_add(1.55, "apple", 10); // exception thrown '[ unsupported_operation ]'
_add(0.3, 100, 777)._str().prnt(); // {"ok",{0.300000,100,777}}
// // (slippage of a bad argument '0.3')
// P.S.
// JS approach is bad in this case, given the capabilities of C++
...var S = "string", b = S.begin(), e = S.end();
while (b != e) {
(*b).prnt(); // s t r i n g
++b;
}
b = S.rbegin(), e = S.rend();
while (b != e) {
(*b).prnt(); // g n i r t s
++b;
}
var A = { 1, 2.2, 'A', "BCD", true };
b = A.begin(), e = A.end();
while (b != e) {
(*b).prnt(); // 1 2.2 A BCD true
++b;
}
b = A.rbegin(), e = A.rend();
while (b != e) {
(*b).prnt(); // true BCD A 2.2 1
++b;
}
...shorter versions:
b = S.begin();
do {
(*b).prnt(); // s t r i n g
} while (b.next());
b = A.rbegin();
do {
(*b).prnt(); // true BCD A 2.2 1
} while (b.next());
// or
for (var i = 0; i < S.length(); ++i) {
S[i].prnt(); // s t r i n g
}
// or
for (auto i : A) {
i.prnt(); // 1 2.2 A BCD true
}
// or
b = S.begin();
while (b) {
(*b).prnt(); // s t r i n g
++b;
}
...note: Use prefix version of operator++,--() e.g ++i,--i , because an additional temporary object is being created for i++,i-- .
Invalidated iterators: ( some cases )
var ab_ext;
var sb_ext;
{
var s = "STR"; //
var sb = s.begin(), se = s.end(); // <----------are become invalidate----------.
// |
var a = { 1, 2 }; // |
var ab = a.begin(), ae = a.end(); // <----------are become invalidate----------|--.
// | |
var A = { 3, sb, 2, se, 1, { ab, ae } }; // | |
var Ab = A.begin(), Ae = A.end(); // <----------are become invalidate-------. | |
// | | |
// push own-iterators | | |
A.push_back( { Ab, Ae }); // -------------------cause---------------------->| | |
// | | |
A.prnt(); // [ 7 ]{ 3, | | |
// | [var:I_str] val: S |, | | |
// 2, | | |
// | [var:I_str] at the end() |, | | |
// 1, | | |
// [ 2 ]{ | [var:I_arr] val: 1 |, | | |
// | [var:I_arr] at the end() | }, | | |
// [ 2 ]{ | [var:I_arr] is invalid |, |--' | |
// | [var:I_arr] is invalid | } } | | |
// | |
a.push_back("!"); // ---------------------------cause--------------------------|->|
// | |
s.resize(20); // -------------------------------cause------------------------->| |
// | |
A.prnt(); // [ 7 ]{ 3, | |
// | [var:I_str] is invalid |, |----------------| |
// 2, | |
// | [var:I_str] is invalid |, |----------------' |
// 1, |
// [ 2 ]{ | [var:I_arr] is invalid |, |-----------'
// | [var:I_arr] is invalid | }, |
// [ 2 ]{ | [var:I_arr] is invalid |,
// | [var:I_arr] is invalid | } }
// any 'direct' operations on invalid iterators will throw exception:
//*sb; // excp: [ invalid_iterator ]
//se--; // excp: [ invalid_iterator ]
//ab + 1; // excp: [ invalid_iterator ]
//ae -= 1; // excp: [ invalid_iterator ]
// ...
// It possible to check of the valid, but i don't recommend to do it.
// This exceptions serve to show that the program has bad design (regarding iterators)
// and can (have to) be fixed.
(bool) sb; // false (if(sb) ...)
!sb; // true (if(!sb)...)
// NOTE: **************************************************************************
var q = ab; // copy-constructor var(var&)/(const var&) (1) | no
var w; // | exceptions are
w = ab; // assignment operator operator=(var&)/(const var&) (2) | thrown
//*q; // excp: [ invalid_iterator ] | 'direct'
//*w; // excp: [ invalid_iterator ] | operations
// As you can see from the above, only 'direct' operations on the iterators can throw
// exceptions, and (1)/(2) look like propagators of invalid iterators ,
// but this is a compromise with memory leaks sometimes occurring while exceptions
// in constructors (1) and assignment operator (2) --> so keep it in mind.
// **********************************************************************************
// right now they are valid: (another case)
ab_ext = s.begin(); // <--------------------------.
// are become invalidate |
sb_ext = a.begin(); // <--------------------------|
// |
} // |
// cause: |
// end of the block scope -> source is destroyed -->'
//*ab_ext; // excp: [ invalid_iterator ]
//*sb_ext; // excp: [ invalid_iterator ]
...one more case
var s = "string";
// "trin" cause:
var itr = s.substr(1, 4).find("i"); // temporary substring (will be destroyed)---.
// |
//*itr; // excp: [ invalid_iterator ] <-------is become invalidate---------------'
// one of the solutions ...
var ss;
// "trin"
itr = (ss = s.substr(1, 4)).find("i");
*itr; // 'i'
itr.pos(); // 2
...var i = 3, ii = 4;
i /= 2, i.prnt(); // int -> double : 1.5
i *= 2, i.prnt(); // double -> int : 3
i += "str", i.prnt(); // int -> string : "3str"
ii += 'c', ii.prnt(); // int -> string : "4c"
//
var c = 'a', cc = 'b', ccc = 'c';
c += 123, c.prnt(); // char -> string : a123
cc += ccc, cc.prnt(); // char -> string : bc
ccc += "!!!", ccc.prnt(); // char -> string : c!!!
//
var b = true, bb = false, bbb = true;
b -= 100, b.prnt(); // bool -> int : -99
bb += 'c', bb.prnt(); // bool -> string : 0c
bbb += "XYZ", bbb.prnt(); // bool -> string : 1XYZ
... Good idea: ( to improve performance )
- as function parameters:
void some_fun(var &a, var *b)void some_fun(const var &a, const var *b)- ...
- maybe loops:
var A = { 1, 2, 3 }, *p;
int i = 0;
while (i < (int) A.length()) {
//
p = &A[i]; // take once
//
*p += 1; // |
p->prnt(); // | use as you want
// ... |
++i;
}
A._str().prnt(); // {2,3,4}
... more details:
- arrays
// > & <
var rA = { 1, 'c', "str" };
// .----^ ^-----. ^-------------.
// | | |
var &ridx0 = rA[0], &ridx1 = rA[1], &ridx2 = rA[2]; // direct references (1)
ridx0.prnt(), ridx1.prnt(), ridx2.prnt(); // 1 c str
// bad use:
rA.push_back("push"); // -> memory relocation <- (or reassign e.g. rA="some val")
// V
//ridx0.prnt(); // undefined behavior
// > * < (same problem as with &)
var pA = { 1, 'c', "str" };
// .---^ ^-----. ^---------------.
// | | |
var *pidx0 = &pA[0], *pidx1 = &pA[1], *pidx2 = &pA[2]; // direct pointers (2)
pidx0->prnt(), pidx1->prnt(), pidx2->prnt(); // 1 c str
// bad use:
pA.push_back("push"); // -> memory relocation <- (or reassign e.g. pA="some val")
// V
//pidx0->prnt(); // undefined behavior
...- strings
// There is nuance: ! keep it in mind !
// Strings in the current implementation cannot address each character access
// as an array does for each element (1)/(2), so all accesses through
// opreator[..] and at(..) (3) functions calls go through a single internal
// buffer [var:char] ----. , which means all references and pointers will contain
// the value (and address)| the last use of the functions (3).
// |
// > & < |
// ............... |
// ||| | V
var S = "STR"; // [var:char] -> single internal buffer
// ^
// |
// direct reference ( or pointer) last use
// .------------>^<-------------. |
// | | | |
var &ridx0 = S[0], &ridx1 = S[1], &ridx2 = S[2];
// 'S'->'T'->'R' 'T'->'R' 'R'
//
ridx0.prnt(), ridx1.prnt(), ridx2.prnt(); // !!! all three refer to 'R' !!!
// bad use:
S.resize(20); // -> memory relocation <- (or reassign e.g. S="some val")
// V
//ridx0 = '?'; // undefined behavior (write to unknown area)
// > * < (same problem as with &) ...
...Conclusion: If use of &,* is conscious -> then why not :)
var S = "v1 : i = 1 ; v2:d = 3.141 ; v3:s = string ; v4:c = A ;"; // incoming string
(var("in : \"") += S + "\"").prnt();
var in = S.split(';'), beg = in.begin(), l, r, A;
A.initArr();
do {
l = (*beg).split(':'), r = l[1].split('='), l[1].trim();
if (l[1] == 'i') A.push_back( { l[0].trim(), r[1].trim()._int() });
else if (l[1] == 'd') A.push_back( { l[0].trim(), r[1].trim()._double() });
else if (l[1] == 's') A.push_back( { l[0].trim(), r[1].trim() });
else if (l[1] == 'c') A.push_back( { l[0].trim(), r[1].trim()[0] });
} while (beg.next());
(var("out: ") += A._str()).prnt();
// in : "v1 : i = 1 ; v2:d = 3.141 ; v3:s = string ; v4:c = A ;"
// out: { {"v1",1},{"v2",3.141000},{"v3","string"},{"v4",'A'}}
// and back again:
beg = A.begin();
S = "";
do {
l = (*beg)[0], r = (*beg)[1];
if (r.is_int()) S += l += var(":i=") += r;
else if (r.is_double()) {
S += l += var(":d=");
var a = r._str().split('.');
S += a[0];
var res = a[1].find_last_not_of("0"); // <---.
if (res) { // |
S += '.'; // |
S += a[1].substr(0, res.pos() + 1); // <--discard zeros
}
} else if (r.is_str()) S += l += var(":s=") += r;
else if (r.is_char()) S += l += var(":c=") += r;
S += ';';
} while (beg.next());
(var("back: \"") += S + "\"").prnt();
// back: "v1:i=1;v2:d=3.141;v3:s=string;v4:c=A;"
...local typedef:
const char -> cch;
std::string -> str;
long long -> llong
unsigned long long -> ullong
local define:
std::initializer_list -> __iniL
- common interfaces:
int size() const noexcept;
cch* gtid() const noexcept;
bool is_def() const noexcept;
bool is_bool() const noexcept;
bool is_int() const noexcept;
bool is_double() const noexcept;
bool is_char() const noexcept;
bool is_str() const noexcept;
bool is_arr() const noexcept;
bool is_map() const noexcept;
bool is_P_map() const noexcept;
bool is_itr() const noexcept;
bool is_I_str() const noexcept;
bool is_Ir_str() const noexcept;
bool is_I_arr() const noexcept;
bool is_Ir_arr() const noexcept;
void prnt() const noexcept;- string/array shared interfaces:
var& at(size_t index);
const var& at(size_t index) const;
var& front();
const var& front() const;
var& back();
const var& back() const;
var begin();
const var begin() const;
var end();
const var end() const;
var rbegin();
const var rbegin() const;
var rend();
const var rend() const;
bool next() const;
llong pos() const;
bool empty() const;
size_t length() const;
size_t max_size() const;
void reserve(size_t new_cap);
size_t capacity() const;
void shrink_to_fit();
void clear();
void swap(var &with);
..
void resize(size_t count);
void resize(size_t count, const var &X);
void resize(size_t count, int X);
void resize(size_t count, float X);
void resize(size_t count, double X);
void resize(size_t count, long X);
void resize(size_t count, llong X);
void resize(size_t count, ullong X);
void resize(size_t count, char X);
void resize(size_t count, cch *X);
void resize(size_t count, const str &X);
void resize(size_t count, __iniL<var> X);
..
var insert(size_t index, const var &src_fist, const var &src_last);
var insert(size_t index, const var &X);
var insert(size_t index, int X);
var insert(size_t index, float X);
var insert(size_t index, double X);
var insert(size_t index, long X);
var insert(size_t index, llong X);
var insert(size_t index, ullong X);
var insert(size_t index, char X);
var insert(size_t index, cch *X);
var insert(size_t index, const str &X);
var insert(size_t index, __iniL<var> X);
var insert(size_t index, size_t count, const var &X);
var insert(size_t index, size_t count, int X);
var insert(size_t index, size_t count, float X);
var insert(size_t index, size_t count, double X);
var insert(size_t index, size_t count, long X);
var insert(size_t index, size_t count, llong X);
var insert(size_t index, size_t count, ullong X);
var insert(size_t index, size_t count, char X);
var insert(size_t index, size_t count, cch *X);
var insert(size_t index, size_t count, const str &X);
var insert(size_t index, size_t count, __iniL<var> X);
..
var erase(size_t index);
var erase(size_t first, size_t last);
..
var push_back(const var &X);
var push_back(int X);
var push_back(float X);
var push_back(double X);
var push_back(long X);
var push_back(llong X);
var push_back(ullong X);
var push_back(char X);
var push_back(cch *X);
var push_back(const str &X);
var push_back( __iniL<var> X);
..
var push_front(const var &X);
var push_front(int X);
var push_front(float X);
var push_front(double X);
var push_front(long X);
var push_front(llong X);
var push_front(ullong X);
var push_front(char X);
var push_front(cch *X);
var push_front(const str &X);
var push_front( __iniL<var> X);
...
var pop_back();
var pop_front();
..
var sort_cmp(bool (*cmp)(var &a, var &b), bool reverse = false) const;
var& ssort_cmp(bool (*cmp)(var &a, var &b),bool reverse = false);
var sort(bool reverse = false) const;
var& ssort(bool reverse = false);
var sample(size_t num = 0) const;
var sample_unq(size_t num = 0) const;
var reverse();
var& sreverse();
bool equal(const var &src_first, const var &src_last, //
size_t first = 0, size_t count = 0) const;
var mismatch(const var &src_first, const var &src_last, //
size_t index = 0, size_t count = 0) const;
..
var& fill(const var &X, size_t index, size_t count);
var& fill(int X, size_t index, size_t count);
var& fill(float X, size_t index, size_t count);
var& fill(double X, size_t index, size_t count);
var& fill(long X, size_t index, size_t count);
var& fill(llong X, size_t index, size_t count);
var& fill(ullong X, size_t index, size_t count);
var& fill(char X, size_t index, size_t count);
var& fill(cch *X, size_t index, size_t count);
var& fill(const str &X, size_t index, size_t count);
var& fill(__iniL<var> X, size_t index, size_t count);- array interfaces:
void initArr() noexcept;
var join(char sep = 0) const;
var& sjoin(char sep = 0);- map interfaces:
var& map_init() noexcept;
var& map_set(cch *map);
var map_get(cch *map) const;
var map_rem(cch *map);
var& map_free();
void map_foreach(var::map (*fun)(char *field,const var &val, void *ctx), //
void *ctx = 0, int depth = 0) const
var map_fread(cch *fname);
var map_fwrite(cch *fname) const;
var map_str(bool format = 0) const;
bool map_equal(const var& map) const;
var map_mismatch(const var& map) const;- string interfaces:
cch* data() const;
var split(char sep) const;
var& ssplit(char sep);
var& ltrim();
var& rtrim();
var& trim();
var insert(size_t index, cch *src, size_t src_index, size_t src_count);
..
var& append(char ch, size_t count = 1);
var& append(const var &src, size_t src_index = 0, size_t src_count = 0);
var& append(cch *src, size_t src_index = 0, size_t src_count = 0);
var& append(str &src, size_t src_index = 0, size_t src_count = 0);
var& append(const var &src_first, const var &src_last);
..
var& replace(size_t index, size_t count, const var &src, size_t src_index = 0,
size_t src_count = 0);
var& replace(size_t index, size_t count, cch *src, size_t src_index = 0,
size_t src_count = 0);
var& replace(size_t index, size_t count, str &src, size_t src_index = 0,
size_t src_count = 0);
var& replace(size_t index, size_t count, const var &src_first,
const var &src_last);
..
int compare(const var &with) const;
int compare(cch *with) const;
int compare(str &with) const;
int compare(size_t index, size_t count, const var &with, size_t with_index = 0,
size_t with_count = 0) const;
int compare(size_t index, size_t count, cch *with, size_t with_index = 0,
size_t with_count = 0) const;
int compare(size_t index, size_t count, str &with, size_t with_index = 0,
size_t with_count = 0) const;
..
bool starts_with(const var &prefix) const;
bool starts_with(char prefix) const;
bool starts_with(cch *prefix) const;
..
bool ends_with(const var &suffix) const;
bool ends_with(char suffix) const;
bool ends_with(cch *suffix) const;
..
var substr(size_t index = 0, size_t count = 0);
size_t copy(char *dest, size_t index = 0, size_t count = 0,
bool zero = false) const;
..
var find(const var &src, size_t index = 0) const;
var find(char src, size_t index = 0) const;
var find(cch *src, size_t index = 0) const;
var find(str &src, size_t index = 0) const;
..
var rfind(const var &src, size_t index = 0) const;
var rfind(char src, size_t index = 0) const;
var rfind(cch *src, size_t index = 0) const;
var rfind(str &src, size_t index = 0) const;
..
var find_first_of(const var &src, size_t index = 0) const;
var find_first_of(char src, size_t index = 0) const;
var find_first_of(cch *src, size_t index = 0) const;
var find_first_of(str &src, size_t index = 0) const;
..
var find_first_not_of(const var &src, size_t index = 0) const;
var find_first_not_of(char src, size_t index = 0) const;
var find_first_not_of(cch *src, size_t index = 0) const;
..
var find_last_of(const var &src, size_t index = 0) const;
var find_last_of(char src, size_t index = 0) const;
var find_last_of(cch *src, size_t index = 0) const;
var find_last_of(str &src, size_t index = 0) const;
..
var find_last_not_of(const var &src, size_t index = 0) const;
var find_last_not_of(char src, size_t index = 0) const;
var find_last_not_of(cch *src, size_t index = 0) const;
var find_last_not_of(str &src, size_t index = 0) const;- explicit conversions:
int _int() const;
llong _llong() const;
ullong _ullong() const;
float _float() const;
double _double() const;
var _str() const;If this project receives feedback ( if it will be of interest to at least
someone at all ), or financially supported (see var.h) => this motivates me to devote
more time to the project, up to full time.
( current bugs will be fixed immediately, regardless of the above -^ )
to start:
-
extension of interfaces:
- ...
- popular algorithms from the C++ Algorithm Library:
- ...
- your suggestions ... ( write me )
-
extension of var:internal types:
- var:map:
- extension of interfaces <- [now in process]
begin()/end()iterations <- [now in process]- ...
- var:storage: ( access to data through var:map )
- var:storage <-> hdd/ssd
- var:storage:client <-> var:storage:server (maybe)
- ...
- your suggestions ... ( write me )
- var:map:
long-term: ( to improve overall performance )
- replace
std::stringwith my own implementation - replace
std::vectorwith my own implementation - ...
P.S. Sorry about my English :)