ref k
kei

2019-09-20 14:00:39 2core 3gb avx2 © shakti l test
$k [-p 1234] [-v] [f.k]                      .z.i(pid) .z.x(arg) .z.e(env)

Verb                       Adverb                Noun              Type
:  assign                  '  each      n bar    char  " ab"       `c
+  add         flip        /  over      n div    name  ``a`b       `n
-  subtract    negate      \  scan      n mod    int   Ø 0 2       `i
*  multiply    first       ': eachprior peach    float ø 2.3 π ∞   `f
%  divideby    inverse     /: eachright sv|join  date 2019-06-28   `D .z.D
&  min|and     where       \: eachleft  vs|split time 12:34:56.789 `t .z.t
|  max|or      reverse
<  less        up          I/O (0:i close)       list (2;3.4;`c)   `.
>  more        down        0: read/write line    dict ``b!(2;`c)   `a
=  equal       group       1: read/write byte    tabl +{b:2 3 4}   `A
~  match       not         2: read/write data    expr :32+9*f%5    `0
!  key         enum        3: set/conn (.Z.s)    func {(+/x)%#x}   `1
,  catenate    enlist      4: get/http (.Z.gGP)
^  except      asc
#  take(f#)    count       #[t;c;b[;a]] select   \l f.k  load       \h help
_  drop(f_)    floor       _[t;c;b[;a]] update   \t[:n]x milli/time \l log
?  find        unique      ?[x;i;f[;y]] splice   \u[:n]x micro/trace
@  index       type        @[x;i;f[;y]] amend    \v [d] vars        \f [d] fns
.  apply       value       .[x;i;f[;y]] dmend    \lf file \lc char  \ll line
$  cast|pad    string      $[c;t;f]     cond     \cd [d] get[set]   \gr x file

util: in within bin like find freq prm cmb
math: sqrt sin cos abs [n]log [n]exp [n]rand
aggr: count first last min max sum avg var dev med
tabl: [f]asc [f]dsc [f]key [select|update]A by B from T where C

datetime:YMDHRSTUV+duration:ymdhrstuv; .z.D+.z.t day:7\ mth:12\ bar:5'
`year`month`date`hour`minute`second`millisecond`microsecond`nanosecond

Ktab: (+`a!2 3)!+`b!3 4
Xtab: `a`b!+`c`d!(2 3;4 5)

1way: `p`m`crc`sha`rip`bad   e.g. hash: `sha@"crypto"
2way:``j`k`csv`b64`hex`aes   e.g. json: `j?`j@{a:2;b:`c}

if[c;..];while[c;..]

2+3   /comment
2+ \3 /display

\\ exit
trap: [e]f/:x
time: [n]f\:x
attr: ^x ?x
syns: count first last sum min max asc

Verbs

Force the monadic case by applying a : suffix.

Force the dyadic case by putting the verb in parentheses:

@1
`i
(@)1
1@

Monadic and dyadic versions of a character are different verbs - it's not just a case of how many arguments they are called with.

For example, with a bracket call with one argument, the monadic verb will work fine, while the dyadic verb will project:

@:["abc"]
`c
@["abc"]
"abc"@

For dyadic verbs, x is on the left and y is on the right:

1,2
1 2
,[1;2]
1 2

Project over the first argument by omitting it entirely:

@[;1] "abc"
"b"

Triadic versions of a verb are the same as the dyadic version (except for cond):

(?) . ("abcb"; "b")
1
(?) . ("abcb"; 1; "d")
"adbcb"

Assign :

Where an assignment is used in a function, consider rewriting code like this (source):

f: {a: reusedtempdata; a*a}

As this:

f: {a*a:reusedtempdata}

Or this:

f: { {x*x} reusedtempdata }

You can assign a value to a list index, or a dictionary key:

a:!3
a[1]:4
a
0 4 2

d:`a`b!1 2
d[`a`c]:3 4
d
a:3
b:2
c:4

You can combine assignment with other verbs:

a:1
a+:1
a
2

l:()
l,:1 2 3
l
1 2 3

l[0],:0
l
1 0
2
3

Assign doesn't appear to be callable with square brackets; in fact the parse tree suggests it has an unconventional evaluation order:

`p "I[0]:4"
::
(`I;0)
4

Under normal bracket precedence, you'd lose info about where to assign to before the assign verb is invoked, because (`I;0) would resolve to the first element of I.

Add +

{a:1}+{a:2}
{a:3}
{a:1}+{b:2}
a|1
b|2

"a"+`c$25
"z"

Flip +:

 ("a";"bc")
"a"
"bc"

+("a";"bc")
ab
ac

+(1 2 3;4 5)
1 4
2 5
3 Ø

*++("##";"###")
"## "

First *:

First of an empty list is null of the type of the list:

*0#,1
Ø

Divide by %

1 % 2
0.5

Min &

2 & 1
1
2 & 1 3
1 2
1 3 & 2
1 2
1 3 & 2 0
1 0

Where &:

Generate an ascending list of numbers, where the number of times number i appears is equal to the number at position i of the input list.

&1 0 2 1
^0 2 2 3

Up <:

< "abracadabra"
?0 3 5 7 10 1 8 4 6 2 9
s@<s: "abracadabra"
"aaaaabbcdrr"

Equal =

1=1f
1

Group =:

With a list, get back the indices at which each unique element appears:

="abracadabra"
a|0 3 5 7 10
b|1 8
c|,4
d|,6
r|2 9

Match ~

Test whether data is identical. Can handle nested structures.

(1;"c";(;{a:2}))~(1;"c";(;{a:2}))
1

With one value different in the nested structure:

(1;"c";(;{a:2}))~(1;"c";(;{a:1}))
0

Unlike equal =, match compares types:

1~1f
0

Not ~:

~3
0
~~3
1

Enumerate !:

!3
!3
,/!3
0 1 2
!2 3
^0 0 0 1 1 1
0 1 2 0 1 2

Given a dict, get its keys:

! `a`b!1 2
`a`b

Catentate ,

Joins two atoms. Prepends/appends an atom to a list. Merges lists.

1,2
1 2
1 2,3
1 2 3
1,2 3
1 2 3
1 2,3 4
1 2 3 4

If you want to make a list nested in the output, enlist it first:

1,,2 3
1
2 3

Catenate merges dicts (where keys collide, right overrides left):

(`a`b!1 2),(`a`c!3 4)
a:3
b:2
c:4

Tables are lists of dicts, so:

t:+`a!,1
u:+`a!,2
t,u
a
-
1
2

Key !

Generate /a dictionary from lists of keys and values:

`k @ `a`b!1 2
"{a:1;b:2}"

Except ^

(!4)^1
0 2 3
(!4)^1 3
0 2
"abracadabra" ^ "bc"
"araadara"

Sort ascending ^:

^1 3 4 2
^1 2 3 4

Take #

3#"abracadabra"
"abr"

Given a filter function (returns 1 or 0 given x):

(2\)#!9
1 3 5 7
{~2\x}#!9      
0 2 4 6 8

Taking zero elements of a list generates an empty list of the type of the first element of the list (or if the list is empty, the type of the list):

0#(1;`a)
!0
0#!0
!0

Drop _

4_"abracadabra"
"cadabra"

I_ is cut:

1 2 6_"abracadabra"
b
raca
dabra

Floor _:

_ 1.23 -1.23 0
1 -2 0

Find ?

2 3 1 4 0 ? 3
1
22 3 1 4 0 ? 3 1 0
1 2 4

i? is draw (with replacement), -i? is deal (without replacement):

^ 22 9 ? 9
^1 3 5 5 6 6 7 7 8
2^ -9 ? 9
^?0 1 2 3 4 5 6 7 8

10 ? 9
1 0 1 7 3 6 2 2 5 0
-10 ? 9
-10 ? 9
   ^
length error

You can also draw or deal from a list:

3 ? "abcd"
"bdb"
-3 ? "abcd"
"cab"

Unique ?:

? "abracadabra"
?"abrcd"
? ("hi"; 1 2 3; `a`b; 0; "hi"; `a`b; 1 2; 0)
"hi"
1 2 3
`a`b
0
1 2

Index @

With an atom, applies that atom to the left function:

(1+) @ 2
3

With a list, applies each element of that list to the function ('shallow' application, ie doesn't go beyond x of xyz):

(1+) @ 2 4 7
3 5 8

Apply .

{z,x,y} . `a`b`c
`c`a`b

Apply can also be used to 'drill' into a data structure in a similar way to bracket indexing. For example:

:: m: 3 3 # !9
0 1 2
3 4 5
6 7 8

m[1]
3 4 5

m . ,1
3 4 5

m[1 2]
3 4 5
6 7 8

m . ,1 2
3 4 5
6 7 8

 / Atoms at x and y positions
m[1; 2]
5
m . 1 2
5

Value .:

. "1+2"
3

d:1
. "d+:1"
d
2

.`a`b!1 2
1 2

. {:x+y}
{:x+y}

. {x+y+a:1}
{x+y+a:1}

Cast $ (with a non-string)

"YMD"$0
2024Y
2024-01M
2024-01-01

(`Y`M`D$0)~"YMD"$0
1

`c $ 65+!3
"ABC"

However, casting chars to ints/floats/etc is more like a parse than a cast:

`i$"123"
123
`i$"ABC"
Ø

If you want ASCII numbers of characters:

0+"ABC"
65 66 67
0f+"ABC"
65 66 67f

Pad $ (with string)

Pads with spaces on right-hand side, or truncates from right:

4$"abc"
"abc "
2$"abc"
"ab"

String $:

$1
,"1"

$`name
"name"

$`a`b!(1 2;3)
a:(,"1";,"2")
b:,"3"

:: s:$+`a`b!(1 2;3 4)
a b
- -
1 3
2 4

s[0][`a]
,"1"

Adverbs

There should be no spaces between an adverb and the expression on its left-hand side (source).

(This is how the / adverb is distinguished from /, ie the start of a comment.)

Adverbs can be called in some similar ways to verbs:

+/ 1 2 3
6
(/ (+)) 1 2 3
6

But expect differences when trying to use adverbs as verbs, due to their many overloads.

Each '

Make a function apply to each element of its input lists, rather than the list itself.

"~" , "abc"
"~abc"
"~" ,' "abc"
~a
~b
~c

'Each' will apply to pairings of its input lists:

"ab" ,' "cd"
ac
bd

{x,y,z}'["ab";"cd";"ef"]
ace
bdf

Over /

n f/x

3 {x+1}/10
13

p f/x

{4>#x} {x,"k"}/"o"
"okkk"

Scan \

n f\x

3 {x+1}\10
10 11 12 13

p f\x

{4>#x} {x,"k"}\"o"
"o"
"ok"
"okk"
"okkk"

Each prior ':

The null chosen for the first pairing is of the type of the first member of the list.

{y,x}': !3
Ø 0
0 1
1 2
{y,x}': 1.0 2.1 4.7
ø 1
1 2.1
2.1 4.7
{y,x}': ("a";1;`ok)
" a"
("a";1)
(1;`ok)

Specify a number to seed the scan:

+':!5
0 1 3 5 7
8+':!5
8 1 3 5 7

': can also provide a 'sliding window' of size x over y:

3': !5
0 1 2
1 2 3
2 3 4

Each right /:

Compare to each:

"ab" ,' "cd"
ac
bd
"ab" ,/: "cd"
abc
abd

Each left \:

"ab" ,\: "cd"
acd
bcd

Join /:

The following work with lists of strings.

Prepend the separator to /:, without a space in between.

"-"/: ("la";"li";"lu";"le";"lo")
"la-li-lu-le-lo"

With empty symbol as 'separator', appends a carriage return and newline to each string and joins. (This is also known as 'sS' or 'string from String', with capitalisation following the k7 convention of lower for atom, upper for list of atoms of same type).

`/: ("ab";"cd")
"ab\r\ncd\r\n"

With three characters instead of just a separator, prepends/appends the first and last:

"(;)"/: (,:' "abcd")
"(a;b;c;d)"

That means 'join' can't be used for multi-character separators, but you can always do this (source):

{y,x,z}[", "]/$`ab`cd`ef
"ab, cd, ef"

Split \:

Prepend the separator to \:, without a space in between.

"-"\: "la-li-lu-le-lo"
la
li
lu
le
lo

With empty symbol as 'separator', splits on newlines, including (if present) at the end of the last word. (This is also known as 'Ss' or 'String from string'.)

`\: "ab\ncd\n"
ab
cd
`\: "ab\ncd"
ab
cd

Scalar from vector (sv) /:

Mnemonic tip: read 'sv' in k evaluation order, ie right to left.

Convert a vector of numbers from a specified base into base 10:

2/: 1 0 1 0
10

Zero is represented as an empty integer list !0, not 0:

10/: !0
0
10/: 0
{ {z+y*x} / [0;x;y] }
     ^
class error

Vector from scalar (vs) \:

Mnemonic tip: read 'vs' in k evaluation order, ie right to left.

Convert a number from base 10 into a specified base:

10\: 1000
1 0 0 0
2\: 10
1 0 1 0

Conversion of zero into a base may be surprising (empty list), but is consistent:

10\: 0
!0

Nouns

Names

Names start with a backtick `. Name literals can be:

Names are also known as symbols.

Floats

All of 1.0, .5, 1., 1f are valid float literals.

Ø=ø
1
Ø~ø
0

Lists

Although two lists may be zero length, and have the same type, they may not be the same:

(@()) = @0 2 # 0
1
(#()) = #0 2 # 0
1
() ~ 0 2 # 0
0

The 'extra dimensions' can matter for matrix multiplication:

() $ (0 2 # 0)
0#,ø ø
(!0) $ (0 2 # 0)
0 0f
(0 2 # 0) $ !0
(0 2 # 0) $ !0
          ^
length error

Empty list () is the same as 0 0#"", but not 0#"" or 0 0 0#"":

0#""
""
0 0#""
()
()~0 0#""
1
()~0 0 0#""
0

(2;3.4;`c) (or any list of atoms, functions etc) can also be written 2,3.4,`c:

(2;3.4;`c)~2,3.4,`c
1

But it breaks down when you include nested lists such as strings:

2,"hi",`c
2
"h"
"i"
`c

Dictionaries

Dicts are ordered:

|`a`b!1 2
b:2
a:1

Recover the keys and values using !: and .::

!`a`b!1 2
`a`b
.`a`b!1 2
1 2

Functions

Functions can call themselves by using their own name in their definition. Naive example:

factorial: {$[x<2;1;x*factorial[x-1]]}
factorial 4
24

Expressions

Exprs can be executed on tables, eg:

t: +`a`b!(1 2 3;4 5 6)
t :a>1
0 1 1
t@&t :a>1
a b
- -
2 5
3 6

Some expr functionality is NYI. For more info, see this forum post.

Utilities

in

`c`d in `a`b`c
1 0

"abcz" in "abracadabra"
1 1 1 0

2 5 in !5
1 0
2 5 in (1 2; 2 5; 2 4)
1
2 5 in (1 2; 2 4)
0

within

`p within `A`z
1
`p within `A`Z
0

Includes lower bound, excludes upper bound:

1 within 1 2
1
2 within 1 2
0
1 2 3 within 1 3
1 1 0

like

"abc" like "*b?"
1
"abcd" like "*b?"
0

(,"1") like "[0-3]"
1
(,"4") like "[0-3]"
0

Frequency histogram freq

Counts of each item in the list.

freq "alibaba"
a|3
b|2
i|1
l|1

find

"abracadabra" find "bra"
1 3
8 3

Sort ascending [f]asc

asc "abracadabra"
^"aaaaabbcdrr"

Sort descending [f]dsc

dsc "abracadabra"
"rrdcbbaaaaa"

Key [f]key

Turn a table into a keyed table:

`a`c key (`a`b`c!1 2 3;`a`b`c!4 5 6)
a c|b
- -|-
1 3|2
4 6|5

Math

Absolute value abs

abs 1.23
1.23
abs -1.23
1.23

Permutations prm

Generates permutation indices.

prm 3
?0 1 2
?1 0 2
?1 2 0
?0 2 1
?2 0 1
?2 1 0

Combinations

2 cmb 3
0 1
0 2
1 2

Natural logarithm (log:) and logarithm (log)

Monadic: natural logarithm, ie the power you'd need to raise e by to get x.

log 2
0.6931472
log (exp 1)
1f
(exp 1) exp (log 2)
2f

Dyadic: logarithm, ie the number you'd need to raise the left number to to get the right number.

2 log 2
1f
2 log 4
2f

Exponential (exp:) and power (exp)

Monadic: e to the power of x.

exp 1
2.718282
exp 2
7.389056
(exp 1) exp 2
7.389056

Dyadic: left to the power of right.

2 exp 4
16f
2 exp -1
0.5

Random [n]rand

Uniform distribution rand i or i rand i:

rand 3
0.5 0.9078156 0.269656
3 rand 10
5 5 1

Normal distribution rand -i:

`m @ rand -5
1.157194e-13
-0.5821508
1.846709
-1.075539
0.890081

Bar i'

Round down to the nearest i.

5'9
5
5'10
10

Integer division i/

5/9
1
5/10
2

Mod i\

2\!10
0 1 0 1 0 1 0 1 0 1

Aggregations

Median med

The value in the middle if the data was sorted.

If the count of the data is even, return the value on the right of the middle.

med !3
1
med !4
2
med 2 3 1
2

Tables and KSQL

Tables

A table is a list of dicts where each dict has the same keys in the same order.

A table can also be considered as a flipped dict of lists, where each list is of equal length.

t:(`a`b!1 2;`a`b!3 4)
u:+`a`b!(1 3;2 4)
t~u
1

You can access rows or columns of the table by indexing using the row number or column key:

t:(`a`b!1 2;`a`b!3 4)
t[1]~`a`b!3 4
1
t[`b]~2 4
1
t[1;`a]~3
1

Key tables are dictionaries where the rows of one table map to the rows of another table.

You can also use key to set the key columns after creation.

k:(`a`c!1 3;`a`c!4 6)
v:(`b!2;`b!5)
kt: k!v
kt~(+`a`c!(1 4;3 6))!+`b!2 5
1
@kt
`a

t:(`a`b`c!1 2 3;`a`b`c!4 5 6)
tk:`a`c key t
tk~kt
1

Access rows of the value table by indexing into the keytable:

kt[`a`c!4 6]  / {b:5}

KSQL {ksql}

t:+`a`b!(1 3;2 4)
update b:a*2 from t
a b
- -
1 2
3 6

t / t not updated in-place
a b
- -
1 2
3 4

Use by to group rows or aggregations:

t:+`a`b`c!(1 2 1;2 3 4;3 4 5);t
a b c
- - -
1 2 3
2 3 4
1 4 5

select by a from t
a|
-|--------------
1|+{b:2 4;c:3 5}
2|+{b:,3;c:,4}

select sum b by a from t
a|b
-|-
1|6
2|3

Commands

Access any shell command by putting a \ in front of it:

\seq 3
1
2
3

List files \lf

\lf
afile.txt
yet another file

List character counts \lc

\lc
afile.txt       |29
yet another file|50

You can't assign the result of \lc directly (ie a: \lc doesn't work). But you can capture its output and see that it is in fact a dictionary:

{(x;@x)} @ . "\\lc"
(,"k";"test.csv";"test.k";"test.txt";"testfile")!211392 12 4 12 24j
`a

List line counts \ll

\ll
afile.txt       |1
yet another file|3

Help \h

The official help included in the k binary. It's the navigation to this site!

Changelog \l

Find the latest changelog on GitHub.

IO and IPC

Read/write line 0:

Given the following in test.csv:

1,ABC,1.23
3,DEF,4567.89

We can read it in as lists of type inf respectively and separator ,:

("inf";",")0:"test.csv"
1 3
`ABC`DEF
1.23 4567.89

The filename can be given as `test.csv instead of "test.csv" (one char shorter!).

We can also write lists of strings to a file (verify output using a text editor):

"test.txt" 0: ("hello";"world")

And that includes saving tables to CSV, if we first convert the table to a list of strings:

"test.csv" 0: `csv @ (+`a`b!(1 2;3 4))

You can also use 0: to deserialise in-memory lists of strings with a common separator. Arthur's example:

 ("ii";"|")0:("2|3";"3|4";"4|5")
2 3 4
3 4 5

Keep in mind it's reading the data into columns, not rows. The first row of the console output is the first item in a list of columns.

Read/write bytes 1:

"testfile" 1: 0x0123456789abcdef
1: "testfile"
0x0123456789abcdef

You can verify 1: works with raw bytes using an external tool:

$ hexdump -C testfile
00000000  01 23 45 67 89 ab cd ef
00000008

And to break down what's happening in the 'prompt' example from the official tutorial, ie:

name: 1: ("" 1: "What is your name? ")
What is your name? Me
name
"Me"

On the last point: if just writing to stdout, make sure to put a semicolon at the end to suppress outputting the characters on the left of 1: to the REPL.

Read/write data 2:

"testfile" 2: (1 2 3 4)
2: "testfile"
1 2 3 4

You can see what testfile looks like in bytes with 1::

1: "testfile"
0x000000070400000001000000020000000300000004000000

Inter-process communication 3: and 4:

Start a k process running with port 1234 (that's k -p 1234).

Then in another k process, set up a connection to the first k process, send it commands, and get responses:

2019-04-18 15:45:55 2core 1gb avx2 © shakti l2.0 test
conn: 3: 1234
conn 4: "life: 42"
conn 4: "life"
42

3+ arguments

Select #[t;c;b[;a]]

Get all rows of t where c is true.

t: +`a`b!(1 2 3;4 5 6)
#[:a>1;t]
a b
- -
2 5
3 6

 / But since there are just two arguments,
 / we can use # as an infix verb:

(:a>1)#t
a b
- -
2 5
3 6

(:(a>1)&(b<6)) # t
+{a:,2;b:,5}
(:(a>1)&(b<5)) # t
+{a:!0;b:!0}

Update _[t;c;b[;a]]

Via Arthur:

t:+`b!2 3
_[t;();`b! :b+1]
b
-
3
4

Splice ?[x;i;f[;y]]

Insert y into x at index i.

If an element was at that index before, move it right.

?[!3;2;`abc]
0
1
`abc
2
?[!3;2;"abc"]
0
1
"a"
"b"
"c"
2

Amend @[x;i;f[;y]]

Apply function f to the elements at each i of x (with argument y, if f is dyadic).

@[!3; 0 2; 1+]
1 1 3
@[!3; 0 2; +; 2 7]
2 1 9

To do the update in-place, use the data's name symbol instead of the name directly (`name vs name).

v:!3
@[v;1;7]
0 7 2
v / original assignment unchanged
!3
@[`v;1;7]
`v
v
0 7 2

Dmend .[x;i;f[;y]]

Think of dmend as 'deep' or 'drill' amend - it's similar to amend in the same way that dyadic . and @ are similar.

i is a list that governs which slices of x to apply f to.

Apply f to slices 0 and 2 on the first 'depth' level down:

.[!3; ,0 2; 7]
7 1 7

Apply to rows or specific elements of a matrix:

:: m: 3 3 # !9
0 1 2
3 4 5
6 7 8

 / Row 1:

.[m; ,1; 1+]
0 1 2
4 5 6
6 7 8

 / Row 1, element 2:

.[m; 1 2; 1+]
0 1 2
3 4 6
6 7 8

 / Row 0, elements 1 and 2:

.[m; (0; 1 2); 1+]
0 2 3
3 4 5
6 7 8

 / In each of rows 1 and 2, elements 1 and 2:

.[m; (1 2; 1 2); 1+]
0 1 2
3 5 6
6 8 9

 / In all rows, elements 1 and 2:

.[m; (; 1 2); 1+]
0 2 3
3 5 6
6 8 9

Cond $[c;t;f]

If the true expression is returned, the false expression never executes (and vice versa):

a:1;b:1
$[1;a+:1;b+:1]; (a;b)
2 1
$[0;a+:1;b+:1]; (a;b)
2 2

Unlike other triadics, triadic $ is not the same as dyadic $:

($) . (`n; "a")
`a
($) . (1; "a"; "b")
($) . (1; "a"; "b")
    ^
nyi error

Simulate a vector cond:

{$[x;y;z]}'[1 0 1; "abc"; "def"]
"aec"

Or (modified version of Arthur's - no $ needed!):

{(+(z;y))@'x}[1 0 1;"abc";"def"]
"aec"

Datetimes

A datetime looks like this:

2019-05-04T13:13:12.313 /.z.T
2019-05-04T13:13:12.313

Datetime literals are designed to match ISO 8601 format.

Dates start from 2024-01-01:

`D $ 0
2024-01-01

'0' is Monday. Get the day of the week with 7 mod.

d:`Mon`Tue`Wed`Thu`Fri`Sat`Sun
d @ 7\2024-01-01
`Mon

You can also use duration literals (requires short-form code), and do date arithmetic:

2024-01-01 + 2m
2024-03-01

Datetime and duration names:

Long-form duration Short-form duration Datetime
`year `y `Y
`month `m `M
`date `d `D
`hour `h `H
`minute `r `R
`second `s `S
`millisecond `t `T
`microsecond `u `U
`nanosecond `v `V

Convert to/from/between datetimes and durations using $. It takes a name or string (short version only):

`year $ .z.t / .z.t doesn't contain year
0y
`Y    $ .z.D
2019Y
`Y    $ .z.D+.z.t
2019Y
`y    $ 2019
2019y
"y"   $ 2019
2019y

Convert between dates and year, month date digits using `/: and `\:.

Current date .z.d

.z.d / 2019-04-02

Current time .z.t

Greenwich Mean Time.

.z.t / 11:18:14.857

Converters

One-way conversions

Convert to with `x@data or `x data or `x[data], where `x is the relevant name.

Parse `p

`p @ "5*!3"
*
5
(!:;3)

A parse tree can be executed with .:

. `p @ "5*!3"
0 5 10

Although parse trees can be converted into strings with `p ?, it is not a lossless conversion.

The parser enlists names (and lists of names) to indicate they should not be evaluated.

Examples:

e: "1;.1;`n;\"c\";1 2;1. 2.;`n`o;\"chars\";"
e,:":;v;1+2;+/1 2;{x+2*y};a:1;:a+2*b; /comment;\\h"

{`string`parsetree`type!(x;`p x;@`p x)}' ";"\: e
string    parsetree   type
--------- ----------- ----
1         1           `i
.1        0.1         `f
`n        ,`n         `n
"c"       "c"         `c
1 2       1 2         `i
1. 2.     1 2f        `f
`n`o      ,`n`o       `.
"chars"   "chars"     `c
:         :           `2
v         `v          `n
1+2       (+;1;2)     `.
+/1 2     ((/;+);1 2) `.
{x+2*y}   {x+2*y}     `2
a:1       (::;`a;1)   `.
:a+2*b    :a+2*b      `0
/comment             `
\h        (;`h)       `.

Matrix display `m

d: 1 2 3
d
1 2 3
`m d
1
2
3

Two-way conversions

Convert to with `x@data or `x data, where `x is the relevant name.

Convert from with `x?serialdata.

Binary `

` "a"
0x0161
` `a
0x0f6100
` 10
0x070a000000

JSON `j

`j(+`a`b!(1,"x";2,`z))
"[{\"a\":1,\"b\":2},{\"a\":\"x\",\"b\":\"z\"}]"

KSON `k

`k 2*!3
"0 2 4"
`k(+`a`b!(1,"x";2,`z))
"+{a:(1;\"x\");b:(2;`z)}"

`csv

`csv ? ("a,b";"1,2";"3,4")
a b
- -
1 2
3 4

FAQ

How can I test if something is an atom, or a list?

isatom:{x~*x}

How do I see the output of an assignment expression?

Use the identity function, :::

a:1+2  / no output
:: a:1+2
3
@(::)
`1

How do I make an empty typed list?

For characters, it's just "":

#""
0
@""
`c

For integers and floats, use !:

@!0
`i
#!0
0
@!.0
`f
#!.0
0

For other types, take 0 elements of an atom of that type (may be a better way?). For example, for names:

@0#`
`n
#0#`
0

How can I make k's output syntax more 'raw'?

Start k with argument -v:

(base) chris@chris-VirtualBox:~$ k
2019-06-08 13:39:07 2core 4gb avx2 © shakti l2.0 test
 `a`b!1 2
a:1
b:2

(base) chris@chris-VirtualBox:~$ k -v
2019-06-08 13:39:07 2core 4gb avx2 © shakti l2.0 test
 `a`b!1 2
{a:1;b:2}

.z namespace

.z.x {.z.x}

In the REPL, contains an empty list by default:

(base) chris@chris-VirtualBox:~$ k
2019-04-28 15:03:42 2core 1gb avx2 © shakti l2.0 test
 .z.x
()

But in a file, it lists the filename and any args after it. If testarg.k comprises the line .z.x:

(base) chris@chris-VirtualBox:~$ k testarg.k not.a.real.file
2019-04-28 15:03:42 2core 1gb avx2 © shakti l2.0 test
testarg.k
not.a.real.file

Therefore, .z.x can be used to pass arguments to a k script.

Also, note those arguments won't appear if we load testarg.k in another k process:

(base) chris@chris-VirtualBox:~$ k
2019-04-28 15:03:42 2core 1gb avx2 © shakti l2.0 test
 \l testarg.k
()

So you can also use .z.x to identify whether a script was run in its own right, or merely \l-ed into another k process.

Secrets!

In addition to `b64, there's `b58, used in the `bad implementation:

`b58 "helloworld"
"6sBRWyteSSzHrs"

Are all characters in the string in the ASCII set?

`ascii @ "123"
1
`ascii @ "∞"
0

There is also a corresponding `utf8 verb.