strutils

The system module defines several common functions for working with strings, such as:

• $ for converting other data-types to strings
• & for string concatenation
• add for adding a new character or a string to the existing one
• in (alias for contains) and notin for checking if a character is in a string

This module builds upon that, providing additional functionality in form of procedures, iterators and templates for strings.

import strutils

let
  numbers = @[867, 5309]
  multiLineString = "first line\nsecond line\nthird line"

let jenny = numbers.join("-")
assert jenny == "867-5309"

assert splitLines(multiLineString) ==
       @["first line", "second line", "third line"]
assert split(multiLineString) == @["first", "line", "second",
                                   "line", "third", "line"]
assert indent(multiLineString, 4) ==
       "    first line\n    second line\n    third line"
assert 'z'.repeat(5) == "zzzzz"

The chaining of functions is possible thanks to the method call syntax:

import strutils
from sequtils import map

let jenny = "867-5309"
assert jenny.split('-').map(parseInt) == @[867, 5309]

assert "Beetlejuice".indent(1).repeat(3).strip ==
       "Beetlejuice Beetlejuice Beetlejuice"

This module is available for the JavaScript target.

---

See also:

• strformat module for string interpolation and formatting
• unicode module for Unicode UTF-8 handling
• sequtils module for operations on container types (including strings)
• parsecsv module for a high-performance CSV parser
• parseutils module for lower-level parsing of tokens, numbers, identifiers, etc.
• parseopt module for command-line parsing
• pegs module for PEG (Parsing Expression Grammar) support
• strtabs module for efficient hash tables (dictionaries, in some programming languages) mapping from strings to strings
• ropes module for rope data type, which can represent very long strings efficiently
• re module for regular expression (regex) support
• strscans for scanf and scanp macros, which offer easier substring extraction than regular expressions

Imports

parseutils, math, algorithm, macros, unicode, since

Types

SkipTable = array[char, int]

Source Edit

FloatFormatMode = enum
  ffDefault,                ## use the shorter floating point notation
  ffDecimal,                ## use decimal floating point notation
  ffScientific               ## use scientific notation (using ``e`` character)

the different modes of floating point formatting Source Edit

BinaryPrefixMode = enum
  bpIEC, bpColloquial

the different names for binary prefixes Source Edit

Consts

Whitespace = {' ', '\t', '\v', '\c', '\n', '\f'}

All the characters that count as whitespace (space, tab, vertical tab, carriage return, new line, form feed) Source Edit

Letters = {'A'..'Z', 'a'..'z'}

the set of letters Source Edit

Digits = {'0'..'9'}

the set of digits Source Edit

HexDigits = {'0'..'9', 'A'..'F', 'a'..'f'}

the set of hexadecimal digits Source Edit

IdentChars = {'a'..'z', 'A'..'Z', '0'..'9', '_'}

the set of characters an identifier can consist of Source Edit

IdentStartChars = {'a'..'z', 'A'..'Z', '_'}

the set of characters an identifier can start with Source Edit

Newlines = {'\c', '\n'}

the set of characters a newline terminator can start with (carriage return, line feed) Source Edit

AllChars = {'\x00'..'\xFF'}

A set with all the possible characters.

Not very useful by its own, you can use it to create inverted sets to make the find proc find invalid characters in strings. Example:

let invalid = AllChars - Digits
doAssert "01234".find(invalid) == -1
doAssert "01A34".find(invalid) == 2

Source Edit

Procs

proc isAlphaAscii(c: char): bool {...}{.noSideEffect, gcsafe,
                                   extern: "nsuIsAlphaAsciiChar", raises: [],
                                   tags: [].}

Checks whether or not character c is alphabetical.

This checks a-z, A-Z ASCII characters only. Use Unicode module for UTF-8 support.

Example:

doAssert isAlphaAscii('e') == true
doAssert isAlphaAscii('E') == true
doAssert isAlphaAscii('8') == false

Source Edit

proc isAlphaNumeric(c: char): bool {...}{.noSideEffect, gcsafe,
                                     extern: "nsuIsAlphaNumericChar",
                                     raises: [], tags: [].}

Checks whether or not c is alphanumeric.

This checks a-z, A-Z, 0-9 ASCII characters only.

Example:

doAssert isAlphaNumeric('n') == true
doAssert isAlphaNumeric('8') == true
doAssert isAlphaNumeric(' ') == false

Source Edit

proc isDigit(c: char): bool {...}{.noSideEffect, gcsafe, extern: "nsuIsDigitChar",
                              raises: [], tags: [].}

Checks whether or not c is a number.

This checks 0-9 ASCII characters only.

Example:

doAssert isDigit('n') == false
doAssert isDigit('8') == true

Source Edit

proc isSpaceAscii(c: char): bool {...}{.noSideEffect, gcsafe,
                                   extern: "nsuIsSpaceAsciiChar", raises: [],
                                   tags: [].}

Checks whether or not c is a whitespace character.

Example:

doAssert isSpaceAscii('n') == false
doAssert isSpaceAscii(' ') == true
doAssert isSpaceAscii('\t') == true

Source Edit

proc isLowerAscii(c: char): bool {...}{.noSideEffect, gcsafe,
                                   extern: "nsuIsLowerAsciiChar", raises: [],
                                   tags: [].}

Checks whether or not c is a lower case character.

This checks ASCII characters only. Use Unicode module for UTF-8 support.

See also:

• toLowerAscii proc

Example:

doAssert isLowerAscii('e') == true
doAssert isLowerAscii('E') == false
doAssert isLowerAscii('7') == false

Source Edit

proc isUpperAscii(c: char): bool {...}{.noSideEffect, gcsafe,
                                   extern: "nsuIsUpperAsciiChar", raises: [],
                                   tags: [].}

Checks whether or not c is an upper case character.

This checks ASCII characters only. Use Unicode module for UTF-8 support.

See also:

• toUpperAscii proc

Example:

doAssert isUpperAscii('e') == false
doAssert isUpperAscii('E') == true
doAssert isUpperAscii('7') == false

Source Edit

proc toLowerAscii(c: char): char {...}{.noSideEffect, gcsafe,
                                   extern: "nsuToLowerAsciiChar", raises: [],
                                   tags: [].}

Returns the lower case version of character c.

This works only for the letters A-Z. See unicode.toLower for a version that works for any Unicode character.

See also:

• isLowerAscii proc
• toLowerAscii proc for converting a string

Example:

doAssert toLowerAscii('A') == 'a'
doAssert toLowerAscii('e') == 'e'

Source Edit

proc toLowerAscii(s: string): string {...}{.noSideEffect, gcsafe,
                                       extern: "nsuToLowerAsciiStr", raises: [],
                                       tags: [].}

Converts string s into lower case.

This works only for the letters A-Z. See unicode.toLower for a version that works for any Unicode character.

See also:

• normalize proc

Example:

doAssert toLowerAscii("FooBar!") == "foobar!"

Source Edit

proc toUpperAscii(c: char): char {...}{.noSideEffect, gcsafe,
                                   extern: "nsuToUpperAsciiChar", raises: [],
                                   tags: [].}

Converts character c into upper case.

This works only for the letters A-Z. See unicode.toUpper for a version that works for any Unicode character.

See also:

• isLowerAscii proc
• toUpperAscii proc for converting a string
• capitalizeAscii proc

Example:

doAssert toUpperAscii('a') == 'A'
doAssert toUpperAscii('E') == 'E'

Source Edit

proc toUpperAscii(s: string): string {...}{.noSideEffect, gcsafe,
                                       extern: "nsuToUpperAsciiStr", raises: [],
                                       tags: [].}

Converts string s into upper case.

This works only for the letters A-Z. See unicode.toUpper for a version that works for any Unicode character.

See also:

• capitalizeAscii proc

Example:

doAssert toUpperAscii("FooBar!") == "FOOBAR!"

Source Edit

proc capitalizeAscii(s: string): string {...}{.noSideEffect, gcsafe,
    extern: "nsuCapitalizeAscii", raises: [], tags: [].}

Converts the first character of string s into upper case.

This works only for the letters A-Z. Use Unicode module for UTF-8 support.

See also:

• toUpperAscii proc

Example:

doAssert capitalizeAscii("foo") == "Foo"
doAssert capitalizeAscii("-bar") == "-bar"

Source Edit

proc nimIdentNormalize(s: string): string {...}{.raises: [], tags: [].}

Normalizes the string s as a Nim identifier.

That means to convert to lower case and remove any '_' on all characters except first one.

Example:

doAssert nimIdentNormalize("Foo_bar") == "Foobar"

Source Edit

proc normalize(s: string): string {...}{.noSideEffect, gcsafe,
                                    extern: "nsuNormalize", raises: [], tags: [].}

Normalizes the string s.

That means to convert it to lower case and remove any '_'. This should NOT be used to normalize Nim identifier names.

See also:

• toLowerAscii proc

Example:

doAssert normalize("Foo_bar") == "foobar"
doAssert normalize("Foo Bar") == "foo bar"

Source Edit

proc cmpIgnoreCase(a, b: string): int {...}{.noSideEffect, gcsafe,
                                        extern: "nsuCmpIgnoreCase", raises: [],
                                        tags: [].}

Compares two strings in a case insensitive manner. Returns:

0 if a == b
< 0 if a < b
> 0 if a > b

Example:

doAssert cmpIgnoreCase("FooBar", "foobar") == 0
doAssert cmpIgnoreCase("bar", "Foo") < 0
doAssert cmpIgnoreCase("Foo5", "foo4") > 0

Source Edit

proc cmpIgnoreStyle(a, b: string): int {...}{.noSideEffect, gcsafe,
    extern: "nsuCmpIgnoreStyle", raises: [], tags: [].}

Semantically the same as cmp(normalize(a), normalize(b)). It is just optimized to not allocate temporary strings. This should NOT be used to compare Nim identifier names. Use macros.eqIdent for that.

Returns:

0 if a == b
< 0 if a < b
> 0 if a > b

Example:

doAssert cmpIgnoreStyle("foo_bar", "FooBar") == 0
doAssert cmpIgnoreStyle("foo_bar_5", "FooBar4") > 0

Source Edit

proc split(s: string; sep: char; maxsplit: int = -1): seq[string] {...}{.
    noSideEffect, gcsafe, extern: "nsuSplitChar", raises: [], tags: [].}

The same as the split iterator (see its documentation), but is a proc that returns a sequence of substrings.

See also:

• split iterator
• rsplit proc
• splitLines proc
• splitWhitespace proc

Example:

doAssert "a,b,c".split(',') == @["a", "b", "c"]
doAssert "".split(' ') == @[""]

Source Edit

proc split(s: string; seps: set[char] = Whitespace; maxsplit: int = -1): seq[
    string] {...}{.noSideEffect, gcsafe, extern: "nsuSplitCharSet", raises: [],
              tags: [].}

The same as the split iterator (see its documentation), but is a proc that returns a sequence of substrings.

See also:

• split iterator
• rsplit proc
• splitLines proc
• splitWhitespace proc

Example:

doAssert "a,b;c".split({',', ';'}) == @["a", "b", "c"]
doAssert "".split({' '}) == @[""]

Source Edit

proc split(s: string; sep: string; maxsplit: int = -1): seq[string] {...}{.
    noSideEffect, gcsafe, extern: "nsuSplitString", raises: [], tags: [].}

Splits the string s into substrings using a string separator.

Substrings are separated by the string sep. This is a wrapper around the split iterator.

See also:

• split iterator
• rsplit proc
• splitLines proc
• splitWhitespace proc

Example:

doAssert "a,b,c".split(",") == @["a", "b", "c"]
doAssert "a man a plan a canal panama".split("a ") == @["", "man ", "plan ", "canal panama"]
doAssert "".split("Elon Musk") == @[""]
doAssert "a  largely    spaced sentence".split(" ") == @["a", "", "largely",
    "", "", "", "spaced", "sentence"]
doAssert "a  largely    spaced sentence".split(" ", maxsplit = 1) == @["a", " largely    spaced sentence"]

Source Edit

proc rsplit(s: string; sep: char; maxsplit: int = -1): seq[string] {...}{.
    noSideEffect, gcsafe, extern: "nsuRSplitChar", raises: [], tags: [].}

The same as the rsplit iterator, but is a proc that returns a sequence of substrings.

A possible common use case for rsplit is path manipulation, particularly on systems that don't use a common delimiter.

For example, if a system had # as a delimiter, you could do the following to get the tail of the path:

var tailSplit = rsplit("Root#Object#Method#Index", '#', maxsplit=1)

Results in tailSplit containing:

@["Root#Object#Method", "Index"]

See also:

• rsplit iterator
• split proc
• splitLines proc
• splitWhitespace proc

Source Edit

proc rsplit(s: string; seps: set[char] = Whitespace; maxsplit: int = -1): seq[
    string] {...}{.noSideEffect, gcsafe, extern: "nsuRSplitCharSet", raises: [],
              tags: [].}

The same as the rsplit iterator, but is a proc that returns a sequence of substrings.

A possible common use case for rsplit is path manipulation, particularly on systems that don't use a common delimiter.

For example, if a system had # as a delimiter, you could do the following to get the tail of the path:

var tailSplit = rsplit("Root#Object#Method#Index", {'#'}, maxsplit=1)

Results in tailSplit containing:

@["Root#Object#Method", "Index"]

See also:

• rsplit iterator
• split proc
• splitLines proc
• splitWhitespace proc

Source Edit

proc rsplit(s: string; sep: string; maxsplit: int = -1): seq[string] {...}{.
    noSideEffect, gcsafe, extern: "nsuRSplitString", raises: [], tags: [].}

The same as the rsplit iterator, but is a proc that returns a sequence of substrings.

A possible common use case for rsplit is path manipulation, particularly on systems that don't use a common delimiter.

For example, if a system had # as a delimiter, you could do the following to get the tail of the path:

var tailSplit = rsplit("Root#Object#Method#Index", "#", maxsplit=1)

Results in tailSplit containing:

@["Root#Object#Method", "Index"]

See also:

• rsplit iterator
• split proc
• splitLines proc
• splitWhitespace proc

Example:

doAssert "a  largely    spaced sentence".rsplit(" ", maxsplit = 1) == @[
    "a  largely    spaced", "sentence"]
doAssert "a,b,c".rsplit(",") == @["a", "b", "c"]
doAssert "a man a plan a canal panama".rsplit("a ") == @["", "man ",
    "plan ", "canal panama"]
doAssert "".rsplit("Elon Musk") == @[""]
doAssert "a  largely    spaced sentence".rsplit(" ") == @["a", "",
    "largely", "", "", "", "spaced", "sentence"]

Source Edit

proc splitLines(s: string; keepEol = false): seq[string] {...}{.noSideEffect, gcsafe,
    extern: "nsuSplitLines", raises: [], tags: [].}

The same as the splitLines iterator (see its documentation), but is a proc that returns a sequence of substrings.

See also:

• splitLines iterator
• splitWhitespace proc
• countLines proc

Source Edit

proc splitWhitespace(s: string; maxsplit: int = -1): seq[string] {...}{.noSideEffect,
    gcsafe, extern: "nsuSplitWhitespace", raises: [], tags: [].}

The same as the splitWhitespace iterator (see its documentation), but is a proc that returns a sequence of substrings.

See also:

• splitWhitespace iterator
• splitLines proc

Source Edit

proc toBin(x: BiggestInt; len: Positive): string {...}{.noSideEffect, gcsafe,
    extern: "nsuToBin", raises: [], tags: [].}

Converts x into its binary representation.

The resulting string is always len characters long. No leading 0b prefix is generated.

Example:

let
  a = 29
  b = 257
doAssert a.toBin(8) == "00011101"
doAssert b.toBin(8) == "00000001"
doAssert b.toBin(9) == "100000001"

Source Edit

proc toOct(x: BiggestInt; len: Positive): string {...}{.noSideEffect, gcsafe,
    extern: "nsuToOct", raises: [], tags: [].}

Converts x into its octal representation.

The resulting string is always len characters long. No leading 0o prefix is generated.

Do not confuse it with toOctal proc.

Example:

let
  a = 62
  b = 513
doAssert a.toOct(3) == "076"
doAssert b.toOct(3) == "001"
doAssert b.toOct(5) == "01001"

Source Edit

proc toHex[T: SomeInteger](x: T; len: Positive): string {...}{.noSideEffect.}

Converts x to its hexadecimal representation.

The resulting string will be exactly len characters long. No prefix like 0x is generated. x is treated as an unsigned value.

Example:

let
  a = 62'u64
  b = 4097'u64
doAssert a.toHex(3) == "03E"
doAssert b.toHex(3) == "001"
doAssert b.toHex(4) == "1001"
doAssert toHex(62, 3) == "03E"
doAssert toHex(-8, 6) == "FFFFF8"

Source Edit

proc toHex[T: SomeInteger](x: T): string {...}{.noSideEffect.}

Shortcut for toHex(x, T.sizeof * 2)

Example:

doAssert toHex(1984'i64) == "00000000000007C0"
doAssert toHex(1984'i16) == "07C0"

Source Edit

proc toHex(s: string): string {...}{.noSideEffect, gcsafe, raises: [], tags: [].}

Converts a bytes string to its hexadecimal representation.

The output is twice the input long. No prefix like 0x is generated.

See also:

• parseHexStr proc for the reverse operation

Example:

let
  a = "1"
  b = "A"
  c = "\0\255"
doAssert a.toHex() == "31"
doAssert b.toHex() == "41"
doAssert c.toHex() == "00FF"

Source Edit

proc toOctal(c: char): string {...}{.noSideEffect, gcsafe, extern: "nsuToOctal",
                                raises: [], tags: [].}

Converts a character c to its octal representation.

The resulting string may not have a leading zero. Its length is always exactly 3.

Do not confuse it with toOct proc.

Example:

doAssert toOctal('1') == "061"
doAssert toOctal('A') == "101"
doAssert toOctal('a') == "141"
doAssert toOctal('!') == "041"

Source Edit

proc fromBin[T: SomeInteger](s: string): T

Parses a binary integer value from a string s.

If s is not a valid binary integer, ValueError is raised. s can have one of the following optional prefixes: 0b, 0B. Underscores within s are ignored.

Does not check for overflow. If the value represented by s is too big to fit into a return type, only the value of the rightmost binary digits of s is returned without producing an error.

Example:

let s = "0b_0100_1000_1000_1000_1110_1110_1001_1001"
doAssert fromBin[int](s) == 1216933529
doAssert fromBin[int8](s) == 0b1001_1001'i8
doAssert fromBin[int8](s) == -103'i8
doAssert fromBin[uint8](s) == 153
doAssert s.fromBin[:int16] == 0b1110_1110_1001_1001'i16
doAssert s.fromBin[:uint64] == 1216933529'u64

Source Edit

proc fromOct[T: SomeInteger](s: string): T

Parses an octal integer value from a string s.

If s is not a valid octal integer, ValueError is raised. s can have one of the following optional prefixes: 0o, 0O. Underscores within s are ignored.

Does not check for overflow. If the value represented by s is too big to fit into a return type, only the value of the rightmost octal digits of s is returned without producing an error.

Example:

let s = "0o_123_456_777"
doAssert fromOct[int](s) == 21913087
doAssert fromOct[int8](s) == 0o377'i8
doAssert fromOct[int8](s) == -1'i8
doAssert fromOct[uint8](s) == 255'u8
doAssert s.fromOct[:int16] == 24063'i16
doAssert s.fromOct[:uint64] == 21913087'u64

Source Edit

proc fromHex[T: SomeInteger](s: string): T

Parses a hex integer value from a string s.

If s is not a valid hex integer, ValueError is raised. s can have one of the following optional prefixes: 0x, 0X, #. Underscores within s are ignored.

Does not check for overflow. If the value represented by s is too big to fit into a return type, only the value of the rightmost hex digits of s is returned without producing an error.

Example:

let s = "0x_1235_8df6"
doAssert fromHex[int](s) == 305499638
doAssert fromHex[int8](s) == 0xf6'i8
doAssert fromHex[int8](s) == -10'i8
doAssert fromHex[uint8](s) == 246'u8
doAssert s.fromHex[:int16] == -29194'i16
doAssert s.fromHex[:uint64] == 305499638'u64

Source Edit

proc intToStr(x: int; minchars: Positive = 1): string {...}{.noSideEffect, gcsafe,
    extern: "nsuIntToStr", raises: [], tags: [].}

Converts x to its decimal representation.

The resulting string will be minimally minchars characters long. This is achieved by adding leading zeros.

Example:

doAssert intToStr(1984) == "1984"
doAssert intToStr(1984, 6) == "001984"

Source Edit

proc parseInt(s: string): int {...}{.noSideEffect, gcsafe, extern: "nsuParseInt",
                                raises: [ValueError], tags: [].}

Parses a decimal integer value contained in s.

If s is not a valid integer, ValueError is raised.

Example:

doAssert parseInt("-0042") == -42

Source Edit

proc parseBiggestInt(s: string): BiggestInt {...}{.noSideEffect, gcsafe,
    extern: "nsuParseBiggestInt", raises: [ValueError], tags: [].}

Parses a decimal integer value contained in s.

If s is not a valid integer, ValueError is raised.

Source Edit

proc parseUInt(s: string): uint {...}{.noSideEffect, gcsafe, extern: "nsuParseUInt",
                                  raises: [ValueError], tags: [].}

Parses a decimal unsigned integer value contained in s.

If s is not a valid integer, ValueError is raised.

Source Edit

proc parseBiggestUInt(s: string): BiggestUInt {...}{.noSideEffect, gcsafe,
    extern: "nsuParseBiggestUInt", raises: [ValueError], tags: [].}

Parses a decimal unsigned integer value contained in s.

If s is not a valid integer, ValueError is raised.

Source Edit

proc parseFloat(s: string): float {...}{.noSideEffect, gcsafe,
                                    extern: "nsuParseFloat",
                                    raises: [ValueError], tags: [].}

Parses a decimal floating point value contained in s.

If s is not a valid floating point number, ValueError is raised. NAN, INF, -INF are also supported (case insensitive comparison).

Example:

doAssert parseFloat("3.14") == 3.14
doAssert parseFloat("inf") == 1.0/0

Source Edit

proc parseBinInt(s: string): int {...}{.noSideEffect, gcsafe,
                                   extern: "nsuParseBinInt",
                                   raises: [ValueError], tags: [].}

Parses a binary integer value contained in s.

If s is not a valid binary integer, ValueError is raised. s can have one of the following optional prefixes: 0b, 0B. Underscores within s are ignored.

Example:

let
  a = "0b11_0101"
  b = "111"
doAssert a.parseBinInt() == 53
doAssert b.parseBinInt() == 7

Source Edit

proc parseOctInt(s: string): int {...}{.noSideEffect, gcsafe,
                                   extern: "nsuParseOctInt",
                                   raises: [ValueError], tags: [].}

Parses an octal integer value contained in s.

If s is not a valid oct integer, ValueError is raised. s can have one of the following optional prefixes: 0o, 0O. Underscores within s are ignored.

Source Edit

proc parseHexInt(s: string): int {...}{.noSideEffect, gcsafe,
                                   extern: "nsuParseHexInt",
                                   raises: [ValueError], tags: [].}

Parses a hexadecimal integer value contained in s.

If s is not a valid hex integer, ValueError is raised. s can have one of the following optional prefixes: 0x, 0X, #. Underscores within s are ignored.

Source Edit

proc parseHexStr(s: string): string {...}{.noSideEffect, gcsafe,
                                      extern: "nsuParseHexStr",
                                      raises: [ValueError], tags: [].}

Convert hex-encoded string to byte string, e.g.:

Raises ValueError for an invalid hex values. The comparison is case-insensitive.

See also:

• toHex proc for the reverse operation

Example:

let
  a = "41"
  b = "3161"
  c = "00ff"
doAssert parseHexStr(a) == "A"
doAssert parseHexStr(b) == "1a"
doAssert parseHexStr(c) == "\0\255"

Source Edit

proc parseBool(s: string): bool {...}{.raises: [ValueError], tags: [].}

Parses a value into a bool.

If s is one of the following values: y, yes, true, 1, on, then returns true. If s is one of the following values: n, no, false, 0, off, then returns false. If s is something else a ValueError exception is raised.

Example:

let a = "n"
doAssert parseBool(a) == false

Source Edit

proc parseEnum[T: enum](s: string): T

Parses an enum T. This errors at compile time, if the given enum type contains multiple fields with the same string value.

Raises ValueError for an invalid value in s. The comparison is done in a style insensitive way.

Example:

type
  MyEnum = enum
    first = "1st",
    second,
    third = "3rd"

doAssert parseEnum[MyEnum]("1_st") == first
doAssert parseEnum[MyEnum]("second") == second
doAssertRaises(ValueError):
  echo parseEnum[MyEnum]("third")

Source Edit

proc parseEnum[T: enum](s: string; default: T): T

Parses an enum T. This errors at compile time, if the given enum type contains multiple fields with the same string value.

Uses default for an invalid value in s. The comparison is done in a style insensitive way.

Example:

type
  MyEnum = enum
    first = "1st",
    second,
    third = "3rd"

doAssert parseEnum[MyEnum]("1_st") == first
doAssert parseEnum[MyEnum]("second") == second
doAssert parseEnum[MyEnum]("last", third) == third

Source Edit

proc repeat(c: char; count: Natural): string {...}{.noSideEffect, gcsafe,
    extern: "nsuRepeatChar", raises: [], tags: [].}

Returns a string of length count consisting only of the character c.

Example:

let a = 'z'
doAssert a.repeat(5) == "zzzzz"

Source Edit

proc repeat(s: string; n: Natural): string {...}{.noSideEffect, gcsafe,
    extern: "nsuRepeatStr", raises: [], tags: [].}

Returns string s concatenated n times.

Example:

doAssert "+ foo +".repeat(3) == "+ foo ++ foo ++ foo +"

Source Edit

proc spaces(n: Natural): string {...}{.inline, raises: [], tags: [].}

Returns a string with n space characters. You can use this proc to left align strings.

See also:

• align proc
• alignLeft proc
• indent proc
• center proc

Example:

let
  width = 15
  text1 = "Hello user!"
  text2 = "This is a very long string"
doAssert text1 & spaces(max(0, width - text1.len)) & "|" ==
         "Hello user!    |"
doAssert text2 & spaces(max(0, width - text2.len)) & "|" ==
         "This is a very long string|"

Source Edit

proc align(s: string; count: Natural; padding = ' '): string {...}{.noSideEffect,
    gcsafe, extern: "nsuAlignString", raises: [], tags: [].}

Aligns a string s with padding, so that it is of length count.

padding characters (by default spaces) are added before s resulting in right alignment. If s.len >= count, no spaces are added and s is returned unchanged. If you need to left align a string use the alignLeft proc.

See also:

• alignLeft proc
• spaces proc
• indent proc
• center proc

Example:

assert align("abc", 4) == " abc"
assert align("a", 0) == "a"
assert align("1232", 6) == "  1232"
assert align("1232", 6, '#') == "##1232"

Source Edit

proc alignLeft(s: string; count: Natural; padding = ' '): string {...}{.noSideEffect,
    raises: [], tags: [].}

Left-Aligns a string s with padding, so that it is of length count.

padding characters (by default spaces) are added after s resulting in left alignment. If s.len >= count, no spaces are added and s is returned unchanged. If you need to right align a string use the align proc.

See also:

• align proc
• spaces proc
• indent proc
• center proc

Example:

assert alignLeft("abc", 4) == "abc "
assert alignLeft("a", 0) == "a"
assert alignLeft("1232", 6) == "1232  "
assert alignLeft("1232", 6, '#') == "1232##"

Source Edit

proc center(s: string; width: int; fillChar: char = ' '): string {...}{.noSideEffect,
    gcsafe, extern: "nsuCenterString", raises: [], tags: [].}

Return the contents of s centered in a string width long using fillChar (default: space) as padding.

The original string is returned if width is less than or equal to s.len.

See also:

• align proc
• alignLeft proc
• spaces proc
• indent proc

Example:

let a = "foo"
doAssert a.center(2) == "foo"
doAssert a.center(5) == " foo "
doAssert a.center(6) == " foo  "

Source Edit

proc indent(s: string; count: Natural; padding: string = " "): string {...}{.
    noSideEffect, gcsafe, extern: "nsuIndent", raises: [], tags: [].}

Indents each line in s by count amount of padding.

Note: This does not preserve the new line characters used in s.

See also:

• align proc
• alignLeft proc
• spaces proc
• unindent proc
• dedent proc

Example:

doAssert indent("First line\c\l and second line.", 2) ==
         "  First line\l   and second line."

Source Edit

proc unindent(s: string; count: Natural = int.high; padding: string = " "): string {...}{.
    noSideEffect, gcsafe, extern: "nsuUnindent", raises: [], tags: [].}

Unindents each line in s by count amount of padding.

Note: This does not preserve the new line characters used in s.

See also:

• dedent proc
• align proc
• alignLeft proc
• spaces proc
• indent proc

Example:

let x = """
      Hello
        There
    """.unindent()

doAssert x == "Hello\nThere\n"

Source Edit

proc indentation(s: string): Natural {...}{.raises: [], tags: [].}

Returns the amount of indentation all lines of s have in common, ignoring lines that consist only of whitespace. Source Edit

proc dedent(s: string; count: Natural = indentation(s)): string {...}{.noSideEffect,
    gcsafe, extern: "nsuDedent", raises: [], tags: [].}

Unindents each line in s by count amount of padding. The only difference between this and the unindent proc is that this by default only cuts off the amount of indentation that all lines of s share as opposed to all indentation. It only supports spaces as padding.

Note: This does not preserve the new line characters used in s.

See also:

• unindent proc
• align proc
• alignLeft proc
• spaces proc
• indent proc

Example:

let x = """
      Hello
        There
    """.dedent()

doAssert x == "Hello\n  There\n"

Source Edit

proc delete(s: var string; first, last: int) {...}{.noSideEffect, gcsafe,
    extern: "nsuDelete", raises: [], tags: [].}

Deletes in s (must be declared as var) the characters at positions first ..last (both ends included).

This modifies s itself, it does not return a copy.

Example:

var a = "abracadabra"

a.delete(4, 5)
doAssert a == "abradabra"

a.delete(1, 6)
doAssert a == "ara"

a.delete(2, 999)
doAssert a == "ar"

Source Edit

proc startsWith(s: string; prefix: char): bool {...}{.noSideEffect, inline,
    raises: [], tags: [].}

Returns true if s starts with character prefix.

See also:

• endsWith proc
• continuesWith proc
• removePrefix proc

Example:

let a = "abracadabra"
doAssert a.startsWith('a') == true
doAssert a.startsWith('b') == false

Source Edit

proc startsWith(s, prefix: string): bool {...}{.noSideEffect, gcsafe,
    extern: "nsuStartsWith", raises: [], tags: [].}

Returns true if s starts with string prefix.

If prefix == "" true is returned.

See also:

• endsWith proc
• continuesWith proc
• removePrefix proc

Example:

let a = "abracadabra"
doAssert a.startsWith("abra") == true
doAssert a.startsWith("bra") == false

Source Edit

proc endsWith(s: string; suffix: char): bool {...}{.noSideEffect, inline, raises: [],
    tags: [].}

Returns true if s ends with suffix.

See also:

• startsWith proc
• continuesWith proc
• removeSuffix proc

Example:

let a = "abracadabra"
doAssert a.endsWith('a') == true
doAssert a.endsWith('b') == false

Source Edit

proc endsWith(s, suffix: string): bool {...}{.noSideEffect, gcsafe,
    extern: "nsuEndsWith", raises: [], tags: [].}

Returns true if s ends with suffix.

If suffix == "" true is returned.

See also:

• startsWith proc
• continuesWith proc
• removeSuffix proc

Example:

let a = "abracadabra"
doAssert a.endsWith("abra") == true
doAssert a.endsWith("dab") == false

Source Edit

proc continuesWith(s, substr: string; start: Natural): bool {...}{.noSideEffect,
    gcsafe, extern: "nsuContinuesWith", raises: [], tags: [].}

Returns true if s continues with substr at position start.

If substr == "" true is returned.

See also:

• startsWith proc
• endsWith proc

Example:

let a = "abracadabra"
doAssert a.continuesWith("ca", 4) == true
doAssert a.continuesWith("ca", 5) == false
doAssert a.continuesWith("dab", 6) == true

Source Edit

proc removePrefix(s: var string; chars: set[char] = Newlines) {...}{.gcsafe,
    extern: "nsuRemovePrefixCharSet", raises: [], tags: [].}

Removes all characters from chars from the start of the string s (in-place).

See also:

• removeSuffix proc

Example:

var userInput = "\r\n*~Hello World!"
userInput.removePrefix
doAssert userInput == "*~Hello World!"
userInput.removePrefix({'~', '*'})
doAssert userInput == "Hello World!"

var otherInput = "?!?Hello!?!"
otherInput.removePrefix({'!', '?'})
doAssert otherInput == "Hello!?!"

Source Edit

proc removePrefix(s: var string; c: char) {...}{.gcsafe,
    extern: "nsuRemovePrefixChar", raises: [], tags: [].}

Removes all occurrences of a single character (in-place) from the start of a string.

See also:

• removeSuffix proc
• startsWith proc

Example:

var ident = "pControl"
ident.removePrefix('p')
doAssert ident == "Control"

Source Edit

proc removePrefix(s: var string; prefix: string) {...}{.gcsafe,
    extern: "nsuRemovePrefixString", raises: [], tags: [].}

Remove the first matching prefix (in-place) from a string.

See also:

• removeSuffix proc
• startsWith proc

Example:

var answers = "yesyes"
answers.removePrefix("yes")
doAssert answers == "yes"

Source Edit

proc removeSuffix(s: var string; chars: set[char] = Newlines) {...}{.gcsafe,
    extern: "nsuRemoveSuffixCharSet", raises: [], tags: [].}

Removes all characters from chars from the end of the string s (in-place).

See also:

• removePrefix proc

Example:

var userInput = "Hello World!*~\r\n"
userInput.removeSuffix
doAssert userInput == "Hello World!*~"
userInput.removeSuffix({'~', '*'})
doAssert userInput == "Hello World!"

var otherInput = "Hello!?!"
otherInput.removeSuffix({'!', '?'})
doAssert otherInput == "Hello"

Source Edit

proc removeSuffix(s: var string; c: char) {...}{.gcsafe,
    extern: "nsuRemoveSuffixChar", raises: [], tags: [].}

Removes all occurrences of a single character (in-place) from the end of a string.

See also:

• removePrefix proc
• endsWith proc

Example:

var table = "users"
table.removeSuffix('s')
doAssert table == "user"

var dots = "Trailing dots......."
dots.removeSuffix('.')
doAssert dots == "Trailing dots"

Source Edit

proc removeSuffix(s: var string; suffix: string) {...}{.gcsafe,
    extern: "nsuRemoveSuffixString", raises: [], tags: [].}

Remove the first matching suffix (in-place) from a string.

See also:

• removePrefix proc
• endsWith proc

Example:

var answers = "yeses"
answers.removeSuffix("es")
doAssert answers == "yes"

Source Edit

proc addSep(dest: var string; sep = ", "; startLen: Natural = 0) {...}{.noSideEffect,
    inline, raises: [], tags: [].}

Adds a separator to dest only if its length is bigger than startLen.

A shorthand for:

if dest.len > startLen: add(dest, sep)

This is often useful for generating some code where the items need to be separated by sep. sep is only added if dest is longer than startLen. The following example creates a string describing an array of integers.

Example:

var arr = "["
for x in items([2, 3, 5, 7, 11]):
  addSep(arr, startLen = len("["))
  add(arr, $x)
add(arr, "]")
doAssert arr == "[2, 3, 5, 7, 11]"

Source Edit

proc allCharsInSet(s: string; theSet: set[char]): bool {...}{.raises: [], tags: [].}

Returns true if every character of s is in the set theSet.

Example:

doAssert allCharsInSet("aeea", {'a', 'e'}) == true
doAssert allCharsInSet("", {'a', 'e'}) == true

Source Edit

proc abbrev(s: string; possibilities: openArray[string]): int {...}{.raises: [],
    tags: [].}

Returns the index of the first item in possibilities which starts with s, if not ambiguous.

Returns -1 if no item has been found and -2 if multiple items match.

Example:

doAssert abbrev("fac", ["college", "faculty", "industry"]) == 1
doAssert abbrev("foo", ["college", "faculty", "industry"]) == -1 # Not found
doAssert abbrev("fac", ["college", "faculty", "faculties"]) == -2 # Ambiguous
doAssert abbrev("college", ["college", "colleges", "industry"]) == 0

Source Edit

proc join(a: openArray[string]; sep: string = ""): string {...}{.noSideEffect,
    gcsafe, extern: "nsuJoinSep", raises: [], tags: [].}

Concatenates all strings in the container a, separating them with sep.

Example:

doAssert join(["A", "B", "Conclusion"], " -> ") == "A -> B -> Conclusion"

Source Edit

proc join[T: not string](a: openArray[T]; sep: string = ""): string {...}{.
    noSideEffect, gcsafe.}

Converts all elements in the container a to strings using $, and concatenates them with sep.

Example:

doAssert join([1, 2, 3], " -> ") == "1 -> 2 -> 3"

Source Edit

proc initSkipTable(a: var SkipTable; sub: string) {...}{.noSideEffect, gcsafe,
    extern: "nsuInitSkipTable", raises: [], tags: [].}

Preprocess table a for sub. Source Edit

proc find(a: SkipTable; s, sub: string; start: Natural = 0; last = 0): int {...}{.
    noSideEffect, gcsafe, extern: "nsuFindStrA", raises: [], tags: [].}

Searches for sub in s inside range start..last using preprocessed table a. If last is unspecified, it defaults to s.high (the last element).

Searching is case-sensitive. If sub is not in s, -1 is returned.

Source Edit

proc find(s: string; sub: char; start: Natural = 0; last = 0): int {...}{.
    noSideEffect, gcsafe, extern: "nsuFindChar", raises: [], tags: [].}

Searches for sub in s inside range start..last (both ends included). If last is unspecified, it defaults to s.high (the last element).

Searching is case-sensitive. If sub is not in s, -1 is returned. Otherwise the index returned is relative to s[0], not start. Use s[start..last].rfind for a start-origin index.

See also:

• rfind proc
• replace proc

Source Edit

proc find(s: string; chars: set[char]; start: Natural = 0; last = 0): int {...}{.
    noSideEffect, gcsafe, extern: "nsuFindCharSet", raises: [], tags: [].}

Searches for chars in s inside range start..last (both ends included). If last is unspecified, it defaults to s.high (the last element).

If s contains none of the characters in chars, -1 is returned. Otherwise the index returned is relative to s[0], not start. Use s[start..last].find for a start-origin index.

See also:

• rfind proc
• multiReplace proc

Source Edit

proc find(s, sub: string; start: Natural = 0; last = 0): int {...}{.noSideEffect,
    gcsafe, extern: "nsuFindStr", raises: [], tags: [].}

Searches for sub in s inside range start..last (both ends included). If last is unspecified, it defaults to s.high (the last element).

Searching is case-sensitive. If sub is not in s, -1 is returned. Otherwise the index returned is relative to s[0], not start. Use s[start..last].find for a start-origin index.

See also:

• rfind proc
• replace proc

Source Edit

proc rfind(s: string; sub: char; start: Natural = 0; last = -1): int {...}{.
    noSideEffect, gcsafe, extern: "nsuRFindChar", raises: [], tags: [].}

Searches for sub in s inside range start..last (both ends included) in reverse -- starting at high indexes and moving lower to the first character or start. If last is unspecified, it defaults to s.high (the last element).

Searching is case-sensitive. If sub is not in s, -1 is returned. Otherwise the index returned is relative to s[0], not start. Use s[start..last].find for a start-origin index.

See also:

• find proc

Source Edit

proc rfind(s: string; chars: set[char]; start: Natural = 0; last = -1): int {...}{.
    noSideEffect, gcsafe, extern: "nsuRFindCharSet", raises: [], tags: [].}

Searches for chars in s inside range start..last (both ends included) in reverse -- starting at high indexes and moving lower to the first character or start. If last is unspecified, it defaults to s.high (the last element).

If s contains none of the characters in chars, -1 is returned. Otherwise the index returned is relative to s[0], not start. Use s[start..last].rfind for a start-origin index.

See also:

• find proc

Source Edit

proc rfind(s, sub: string; start: Natural = 0; last = -1): int {...}{.noSideEffect,
    gcsafe, extern: "nsuRFindStr", raises: [], tags: [].}

Searches for sub in s inside range start..last (both ends included) included) in reverse -- starting at high indexes and moving lower to the first character or start. If last is unspecified, it defaults to s.high (the last element).

Searching is case-sensitive. If sub is not in s, -1 is returned. Otherwise the index returned is relative to s[0], not start. Use s[start..last].rfind for a start-origin index.

See also:

• find proc

Source Edit

proc count(s: string; sub: char): int {...}{.noSideEffect, gcsafe,
                                        extern: "nsuCountChar", raises: [],
                                        tags: [].}

Count the occurrences of the character sub in the string s.

See also:

• countLines proc

Source Edit

proc count(s: string; subs: set[char]): int {...}{.noSideEffect, gcsafe,
    extern: "nsuCountCharSet", raises: [], tags: [].}

Count the occurrences of the group of character subs in the string s.

See also:

• countLines proc

Source Edit

proc count(s: string; sub: string; overlapping: bool = false): int {...}{.
    noSideEffect, gcsafe, extern: "nsuCountString", raises: [], tags: [].}

Count the occurrences of a substring sub in the string s. Overlapping occurrences of sub only count when overlapping is set to true (default: false).

See also:

• countLines proc

Source Edit

proc countLines(s: string): int {...}{.noSideEffect, gcsafe, extern: "nsuCountLines",
                                  raises: [], tags: [].}

Returns the number of lines in the string s.

This is the same as len(splitLines(s)), but much more efficient because it doesn't modify the string creating temporal objects. Every character literal newline combination (CR, LF, CR-LF) is supported.

In this context, a line is any string separated by a newline combination. A line can be an empty string.

See also:

• splitLines proc

Example:

doAssert countLines("First line\l and second line.") == 2

Source Edit

proc contains(s, sub: string): bool {...}{.noSideEffect, raises: [], tags: [].}

Same as find(s, sub) >= 0.

See also:

• find proc

Source Edit

proc contains(s: string; chars: set[char]): bool {...}{.noSideEffect, raises: [],
    tags: [].}

Same as find(s, chars) >= 0.

See also:

• find proc

Source Edit

proc replace(s, sub: string; by = ""): string {...}{.noSideEffect, gcsafe,
    extern: "nsuReplaceStr", raises: [], tags: [].}

Replaces sub in s by the string by.

See also:

• find proc
• replace proc for replacing single characters
• replaceWord proc
• multiReplace proc

Source Edit

proc replace(s: string; sub, by: char): string {...}{.noSideEffect, gcsafe,
    extern: "nsuReplaceChar", raises: [], tags: [].}

Replaces sub in s by the character by.

Optimized version of replace for characters.

See also:

• find proc
• replaceWord proc
• multiReplace proc

Source Edit

proc replaceWord(s, sub: string; by = ""): string {...}{.noSideEffect, gcsafe,
    extern: "nsuReplaceWord", raises: [], tags: [].}

Replaces sub in s by the string by.

Each occurrence of sub has to be surrounded by word boundaries (comparable to \b in regular expressions), otherwise it is not replaced.

Source Edit

proc multiReplace(s: string; replacements: varargs[(string, string)]): string {...}{.
    noSideEffect, raises: [], tags: [].}

Same as replace, but specialized for doing multiple replacements in a single pass through the input string.

multiReplace performs all replacements in a single pass, this means it can be used to swap the occurrences of "a" and "b", for instance.

If the resulting string is not longer than the original input string, only a single memory allocation is required.

The order of the replacements does matter. Earlier replacements are preferred over later replacements in the argument list.

Source Edit

proc insertSep(s: string; sep = '_'; digits = 3): string {...}{.noSideEffect, gcsafe,
    extern: "nsuInsertSep", raises: [], tags: [].}

Inserts the separator sep after digits characters (default: 3) from right to left.

Even though the algorithm works with any string s, it is only useful if s contains a number.

Example:

doAssert insertSep("1000000") == "1_000_000"

Source Edit

proc escape(s: string; prefix = "\""; suffix = "\""): string {...}{.noSideEffect,
    gcsafe, extern: "nsuEscape", raises: [], tags: [].}

Escapes a string s. See system.addEscapedChar for the escaping scheme.

The resulting string is prefixed with prefix and suffixed with suffix. Both may be empty strings.

See also:

• unescape proc for the opposite

operation

Source Edit

proc unescape(s: string; prefix = "\""; suffix = "\""): string {...}{.noSideEffect,
    gcsafe, extern: "nsuUnescape", raises: [ValueError], tags: [].}

Unescapes a string s.

This complements escape proc as it performs the opposite operations.

If s does not begin with prefix and end with suffix a ValueError exception will be raised.

Source Edit

proc validIdentifier(s: string): bool {...}{.noSideEffect, gcsafe,
                                        extern: "nsuValidIdentifier",
                                        raises: [], tags: [].}

Returns true if s is a valid identifier.

A valid identifier starts with a character of the set IdentStartChars and is followed by any number of characters of the set IdentChars.

Example:

doAssert "abc_def08".validIdentifier

Source Edit

proc formatBiggestFloat(f: BiggestFloat; format: FloatFormatMode = ffDefault;
                        precision: range[-1 .. 32] = 16; decimalSep = '.'): string {...}{.
    noSideEffect, gcsafe, extern: "nsu$1", raises: [], tags: [].}

Converts a floating point value f to a string.

If format == ffDecimal then precision is the number of digits to be printed after the decimal point. If format == ffScientific then precision is the maximum number of significant digits to be printed. precision's default value is the maximum number of meaningful digits after the decimal point for Nim's biggestFloat type.

If precision == -1, it tries to format it nicely.

Example:

let x = 123.456
doAssert x.formatBiggestFloat() == "123.4560000000000"
doAssert x.formatBiggestFloat(ffDecimal, 4) == "123.4560"
doAssert x.formatBiggestFloat(ffScientific, 2) == "1.23e+02"

Source Edit

proc formatFloat(f: float; format: FloatFormatMode = ffDefault;
                 precision: range[-1 .. 32] = 16; decimalSep = '.'): string {...}{.
    noSideEffect, gcsafe, extern: "nsu$1", raises: [], tags: [].}

Converts a floating point value f to a string.

If format == ffDecimal then precision is the number of digits to be printed after the decimal point. If format == ffScientific then precision is the maximum number of significant digits to be printed. precision's default value is the maximum number of meaningful digits after the decimal point for Nim's float type.

If precision == -1, it tries to format it nicely.

Example:

let x = 123.456
doAssert x.formatFloat() == "123.4560000000000"
doAssert x.formatFloat(ffDecimal, 4) == "123.4560"
doAssert x.formatFloat(ffScientific, 2) == "1.23e+02"

Source Edit

proc trimZeros(x: var string; decimalSep = '.') {...}{.noSideEffect, raises: [],
    tags: [].}

Trim trailing zeros from a formatted floating point value x (must be declared as var).

This modifies x itself, it does not return a copy.

Example:

var x = "123.456000000"
x.trimZeros()
doAssert x == "123.456"

Source Edit

proc formatSize(bytes: int64; decimalSep = '.'; prefix = bpIEC;
                includeSpace = false): string {...}{.noSideEffect, raises: [],
    tags: [].}

Rounds and formats bytes.

By default, uses the IEC/ISO standard binary prefixes, so 1024 will be formatted as 1KiB. Set prefix to bpColloquial to use the colloquial names from the SI standard (e.g. k for 1000 being reused as 1024).

includeSpace can be set to true to include the (SI preferred) space between the number and the unit (e.g. 1 KiB).

See also:

• strformat module for string interpolation and formatting

Example:

doAssert formatSize((1'i64 shl 31) + (300'i64 shl 20)) == "2.293GiB"
doAssert formatSize((2.234*1024*1024).int) == "2.234MiB"
doAssert formatSize(4096, includeSpace = true) == "4 KiB"
doAssert formatSize(4096, prefix = bpColloquial, includeSpace = true) == "4 kB"
doAssert formatSize(4096) == "4KiB"
doAssert formatSize(5_378_934, prefix = bpColloquial, decimalSep = ',') == "5,13MB"

Source Edit

proc formatEng(f: BiggestFloat; precision: range[0 .. 32] = 10;
               trim: bool = true; siPrefix: bool = false; unit: string = "";
               decimalSep = '.'; useUnitSpace = false): string {...}{.noSideEffect,
    raises: [], tags: [].}

Converts a floating point value f to a string using engineering notation.

Numbers in of the range -1000.0<f<1000.0 will be formatted without an exponent. Numbers outside of this range will be formatted as a significand in the range -1000.0<f<1000.0 and an exponent that will always be an integer multiple of 3, corresponding with the SI prefix scale k, M, G, T etc for numbers with an absolute value greater than 1 and m, μ, n, p etc for numbers with an absolute value less than 1.

The default configuration (trim=true and precision=10) shows the shortest form that precisely (up to a maximum of 10 decimal places) displays the value. For example, 4.100000 will be displayed as 4.1 (which is mathematically identical) whereas 4.1000003 will be displayed as 4.1000003.

If trim is set to true, trailing zeros will be removed; if false, the number of digits specified by precision will always be shown.

precision can be used to set the number of digits to be shown after the decimal point or (if trim is true) the maximum number of digits to be shown.

formatEng(0, 2, trim=false) == "0.00"
formatEng(0, 2) == "0"
formatEng(0.053, 0) == "53e-3"
formatEng(52731234, 2) == "52.73e6"
formatEng(-52731234, 2) == "-52.73e6"

If siPrefix is set to true, the number will be displayed with the SI prefix corresponding to the exponent. For example 4100 will be displayed as "4.1 k" instead of "4.1e3". Note that u is used for micro- in place of the greek letter mu (μ) as per ISO 2955. Numbers with an absolute value outside of the range 1e-18<f<1000e18 (1a<f<1000E) will be displayed with an exponent rather than an SI prefix, regardless of whether siPrefix is true.

If useUnitSpace is true, the provided unit will be appended to the string (with a space as required by the SI standard). This behaviour is slightly different to appending the unit to the result as the location of the space is altered depending on whether there is an exponent.

formatEng(4100, siPrefix=true, unit="V") == "4.1 kV"
formatEng(4.1, siPrefix=true, unit="V") == "4.1 V"
formatEng(4.1, siPrefix=true) == "4.1" # Note lack of space
formatEng(4100, siPrefix=true) == "4.1 k"
formatEng(4.1, siPrefix=true, unit="") == "4.1 " # Space with unit=""
formatEng(4100, siPrefix=true, unit="") == "4.1 k"
formatEng(4100) == "4.1e3"
formatEng(4100, unit="V") == "4.1e3 V"
formatEng(4100, unit="", useUnitSpace=true) == "4.1e3 " # Space with useUnitSpace=true

decimalSep is used as the decimal separator.

See also:

• strformat module for string interpolation and formatting

Source Edit

proc addf(s: var string; formatstr: string; a: varargs[string, `$`]) {...}{.
    noSideEffect, gcsafe, extern: "nsuAddf", raises: [ValueError], tags: [].}

The same as add(s, formatstr % a), but more efficient. Source Edit

proc `%`(formatstr: string; a: openArray[string]): string {...}{.noSideEffect,
    gcsafe, extern: "nsuFormatOpenArray", raises: [ValueError], tags: [].}

Interpolates a format string with the values from a.

The substitution operator performs string substitutions in formatstr and returns a modified formatstr. This is often called string interpolation.

This is best explained by an example:

"$1 eats $2." % ["The cat", "fish"]

Results in:

"The cat eats fish."

The substitution variables (the thing after the $) are enumerated from 1 to a.len. To produce a verbatim $, use $$. The notation $# can be used to refer to the next substitution variable:

"$# eats $#." % ["The cat", "fish"]

Substitution variables can also be words (that is [A-Za-z_]+[A-Za-z0-9_]*) in which case the arguments in a with even indices are keys and with odd indices are the corresponding values. An example:

"$animal eats $food." % ["animal", "The cat", "food", "fish"]

Results in:

"The cat eats fish."

The variables are compared with cmpIgnoreStyle. ValueError is raised if an ill-formed format string has been passed to the % operator.

See also:

• strformat module for string interpolation and formatting

Source Edit

proc `%`(formatstr, a: string): string {...}{.noSideEffect, gcsafe,
    extern: "nsuFormatSingleElem", raises: [ValueError], tags: [].}

This is the same as formatstr % [a] (see % proc). Source Edit

proc format(formatstr: string; a: varargs[string, `$`]): string {...}{.noSideEffect,
    gcsafe, extern: "nsuFormatVarargs", raises: [ValueError], tags: [].}

This is the same as formatstr % a (see % proc) except that it supports auto stringification.

See also:

• strformat module for string interpolation and formatting

Source Edit

proc strip(s: string; leading = true; trailing = true;
           chars: set[char] = Whitespace): string {...}{.noSideEffect, gcsafe,
    extern: "nsuStrip", raises: [], tags: [].}

Strips leading or trailing chars (default: whitespace characters) from s and returns the resulting string.

If leading is true (default), leading chars are stripped. If trailing is true (default), trailing chars are stripped. If both are false, the string is returned unchanged.

See also:

• stripLineEnd proc

Example:

let a = "  vhellov   "
let b = strip(a)
doAssert b == "vhellov"

doAssert a.strip(leading = false) == "  vhellov"
doAssert a.strip(trailing = false) == "vhellov   "

doAssert b.strip(chars = {'v'}) == "hello"
doAssert b.strip(leading = false, chars = {'v'}) == "vhello"

let c = "blaXbla"
doAssert c.strip(chars = {'b', 'a'}) == "laXbl"
doAssert c.strip(chars = {'b', 'a', 'l'}) == "X"

Source Edit

proc stripLineEnd(s: var string) {...}{.raises: [], tags: [].}

Returns s stripped from one of these suffixes: \r, \n, \r\n, \f, \v (at most once instance). For example, can be useful in conjunction with osproc.execCmdEx. aka: chomp

Example:

var s = "foo\n\n"
s.stripLineEnd
doAssert s == "foo\n"
s = "foo\r\n"
s.stripLineEnd
doAssert s == "foo"

Source Edit

proc isEmptyOrWhitespace(s: string): bool {...}{.noSideEffect, gcsafe,
    extern: "nsuIsEmptyOrWhitespace", raises: [], tags: [].}

Checks if s is empty or consists entirely of whitespace characters. Source Edit

Iterators

iterator split(s: string; sep: char; maxsplit: int = -1): string {...}{.raises: [],
    tags: [].}

Splits the string s into substrings using a single separator.

Substrings are separated by the character sep. The code:

for word in split(";;this;is;an;;example;;;", ';'):
  writeLine(stdout, word)

Results in:

""
""
"this"
"is"
"an"
""
"example"
""
""
""

See also:

• rsplit iterator
• splitLines iterator
• splitWhitespace iterator
• split proc

Source Edit

iterator split(s: string; seps: set[char] = Whitespace; maxsplit: int = -1): string {...}{.
    raises: [], tags: [].}

Splits the string s into substrings using a group of separators.

Substrings are separated by a substring containing only seps.

for word in split("this\lis an\texample"):
  writeLine(stdout, word)

...generates this output:

"this"
"is"
"an"
"example"

And the following code:

for word in split("this:is;an$example", {';', ':', '$'}):
  writeLine(stdout, word)

...produces the same output as the first example. The code:

let date = "2012-11-20T22:08:08.398990"
let separators = {' ', '-', ':', 'T'}
for number in split(date, separators):
  writeLine(stdout, number)

...results in:

"2012"
"11"
"20"
"22"
"08"
"08.398990"

See also:

• rsplit iterator
• splitLines iterator
• splitWhitespace iterator
• split proc

Source Edit

iterator split(s: string; sep: string; maxsplit: int = -1): string {...}{.raises: [],
    tags: [].}

Splits the string s into substrings using a string separator.

Substrings are separated by the string sep. The code:

for word in split("thisDATAisDATAcorrupted", "DATA"):
  writeLine(stdout, word)

Results in:

"this"
"is"
"corrupted"

See also:

• rsplit iterator
• splitLines iterator
• splitWhitespace iterator
• split proc

Source Edit

iterator rsplit(s: string; sep: char; maxsplit: int = -1): string {...}{.raises: [],
    tags: [].}

Splits the string s into substrings from the right using a string separator. Works exactly the same as split iterator except in reverse order.

for piece in "foo:bar".rsplit(':'):
  echo piece

Results in:

"bar"
"foo"

Substrings are separated from the right by the char sep.

See also:

• split iterator
• splitLines iterator
• splitWhitespace iterator
• rsplit proc

Source Edit

iterator rsplit(s: string; seps: set[char] = Whitespace; maxsplit: int = -1): string {...}{.
    raises: [], tags: [].}

Splits the string s into substrings from the right using a string separator. Works exactly the same as split iterator except in reverse order.

for piece in "foo bar".rsplit(WhiteSpace):
  echo piece

Results in:

"bar"
"foo"

Substrings are separated from the right by the set of chars seps

See also:

• split iterator
• splitLines iterator
• splitWhitespace iterator
• rsplit proc

Source Edit

iterator rsplit(s: string; sep: string; maxsplit: int = -1;
                keepSeparators: bool = false): string {...}{.raises: [], tags: [].}

Splits the string s into substrings from the right using a string separator. Works exactly the same as split iterator except in reverse order.

for piece in "foothebar".rsplit("the"):
  echo piece

Results in:

"bar"
"foo"

Substrings are separated from the right by the string sep

See also:

• split iterator
• splitLines iterator
• splitWhitespace iterator
• rsplit proc

Source Edit

iterator splitLines(s: string; keepEol = false): string {...}{.raises: [], tags: [].}

Splits the string s into its containing lines.

Every character literal newline combination (CR, LF, CR-LF) is supported. The result strings contain no trailing end of line characters unless parameter keepEol is set to true.

Example:

for line in splitLines("\nthis\nis\nan\n\nexample\n"):
  writeLine(stdout, line)

Results in:

""
"this"
"is"
"an"
""
"example"
""

See also:

• splitWhitespace iterator
• splitLines proc

Source Edit

iterator splitWhitespace(s: string; maxsplit: int = -1): string {...}{.raises: [],
    tags: [].}

Splits the string s at whitespace stripping leading and trailing whitespace if necessary. If maxsplit is specified and is positive, no more than maxsplit splits is made.

The following code:

let s = "  foo \t bar  baz  "
for ms in [-1, 1, 2, 3]:
  echo "------ maxsplit = ", ms, ":"
  for item in s.splitWhitespace(maxsplit=ms):
    echo '"', item, '"'

...results in:

------ maxsplit = -1:
"foo"
"bar"
"baz"
------ maxsplit = 1:
"foo"
"bar  baz  "
------ maxsplit = 2:
"foo"
"bar"
"baz  "
------ maxsplit = 3:
"foo"
"bar"
"baz"

See also:

• splitLines iterator
• splitWhitespace proc

Source Edit

iterator tokenize(s: string; seps: set[char] = Whitespace): tuple[token: string,
    isSep: bool] {...}{.raises: [], tags: [].}

Tokenizes the string s into substrings.

Substrings are separated by a substring containing only seps. Example:

for word in tokenize("  this is an  example  "):
  writeLine(stdout, word)

Results in:

("  ", true)
("this", false)
(" ", true)
("is", false)
(" ", true)
("an", false)
("  ", true)
("example", false)
("  ", true)

Source Edit

Exports

toLower, toLower, toUpper, toUpper