# High-level abstraction for all text representations
abstract class Text
super Comparable
+ super Cloneable
redef type OTHER: Text
# Gets a view on the chars of the Text object
#
- # assert "hello".chars.to_a == ['h', 'e', 'l', 'l', 'o']
+ # ~~~
+ # assert "hello".chars.to_a == ['h', 'e', 'l', 'l', 'o']
+ # ~~~
fun chars: SequenceRead[Char] is abstract
# Gets a view on the bytes of the Text object
#
- # assert "hello".bytes.to_a == [104u8, 101u8, 108u8, 108u8, 111u8]
- fun bytes: SequenceRead[Byte] is abstract
+ # ~~~
+ # assert "hello".bytes.to_a == [104, 101, 108, 108, 111]
+ # ~~~
+ fun bytes: SequenceRead[Int] is abstract
# Number of characters contained in self.
#
- # assert "12345".length == 5
- # assert "".length == 0
- # assert "あいうえお".length == 5
+ # ~~~
+ # assert "12345".length == 5
+ # assert "".length == 0
+ # assert "あいうえお".length == 5
+ # ~~~
fun length: Int is abstract
# Number of bytes in `self`
#
- # assert "12345".bytelen == 5
- # assert "あいうえお".bytelen == 15
- fun bytelen: Int is abstract
+ # ~~~
+ # assert "12345".byte_length == 5
+ # assert "あいうえお".byte_length == 15
+ # ~~~
+ fun byte_length: Int is abstract
# Create a substring.
#
- # assert "abcd".substring(1, 2) == "bc"
- # assert "abcd".substring(-1, 2) == "a"
- # assert "abcd".substring(1, 0) == ""
- # assert "abcd".substring(2, 5) == "cd"
- # assert "あいうえお".substring(1,3) == "いうえ"
+ # ~~~
+ # assert "abcd".substring(1, 2) == "bc"
+ # assert "abcd".substring(-1, 2) == "a"
+ # assert "abcd".substring(1, 0) == ""
+ # assert "abcd".substring(2, 5) == "cd"
+ # assert "あいうえお".substring(1,3) == "いうえ"
+ # ~~~
#
# A `from` index < 0 will be replaced by 0.
# Unless a `count` value is > 0 at the same time.
fun substring(from: Int, count: Int): SELFTYPE is abstract
# Iterates on the substrings of self if any
- fun substrings: Iterator[FlatText] is abstract
+ private fun substrings: Iterator[FlatText] is abstract
# Is the current Text empty (== "")
#
- # assert "".is_empty
- # assert not "foo".is_empty
+ # ~~~
+ # assert "".is_empty
+ # assert not "foo".is_empty
+ # ~~~
fun is_empty: Bool do return self.length == 0
# Returns an empty Text of the right type
# implementation of an empty string.
protected fun empty: SELFTYPE is abstract
+ # Returns a copy of `self` as a Buffer
+ fun to_buffer: Buffer is abstract
+
# Gets the first char of the Text
- #
- # DEPRECATED : Use self.chars.first instead
fun first: Char do return self.chars[0]
# Access a character at `index` in the string.
#
- # assert "abcd"[2] == 'c'
- #
- # DEPRECATED : Use self.chars.[] instead
+ # ~~~
+ # assert "abcd"[2] == 'c'
+ # ~~~
fun [](index: Int): Char do return self.chars[index]
# Gets the index of the first occurence of 'c'
#
# Returns -1 if not found
- #
- # DEPRECATED : Use self.chars.index_of instead
fun index_of(c: Char): Int
do
return index_of_from(c, 0)
end
# Gets the last char of self
- #
- # DEPRECATED : Use self.chars.last instead
fun last: Char do return self.chars[length-1]
# Gets the index of the first occurence of ´c´ starting from ´pos´
#
# Returns -1 if not found
- #
- # DEPRECATED : Use self.chars.index_of_from instead
fun index_of_from(c: Char, pos: Int): Int
do
var iter = self.chars.iterator_from(pos)
# Gets the last index of char ´c´
#
# Returns -1 if not found
- #
- # DEPRECATED : Use self.chars.last_index_of instead
fun last_index_of(c: Char): Int
do
return last_index_of_from(c, length - 1)
end
# Return a null terminated char *
- fun to_cstring: NativeString is abstract
+ fun to_cstring: CString is abstract
# The index of the last occurrence of an element starting from pos (in reverse order).
#
- # var s = "/etc/bin/test/test.nit"
- # assert s.last_index_of_from('/', s.length-1) == 13
- # assert s.last_index_of_from('/', 12) == 8
+ # ~~~
+ # var s = "/etc/bin/test/test.nit"
+ # assert s.last_index_of_from('/', s.length-1) == 13
+ # assert s.last_index_of_from('/', 12) == 8
+ # ~~~
#
# Returns -1 if not found
+ fun last_index_of_from(item: Char, pos: Int): Int do return chars.last_index_of_from(item, pos)
+
+ # Concatenates `o` to `self`
#
- # DEPRECATED : Use self.chars.last_index_of_from instead
- fun last_index_of_from(item: Char, pos: Int): Int
- do
- var iter = self.chars.reverse_iterator_from(pos)
- while iter.is_ok do
- if iter.item == item then return iter.index
- iter.next
- end
- return -1
- end
+ # ~~~
+ # assert "hello" + "world" == "helloworld"
+ # assert "" + "hello" + "" == "hello"
+ # ~~~
+ fun +(o: Text): SELFTYPE is abstract
# Gets an iterator on the chars of self
- #
- # DEPRECATED : Use self.chars.iterator instead
fun iterator: Iterator[Char]
do
return self.chars.iterator
# Gets an Array containing the chars of self
- #
- # DEPRECATED : Use self.chars.to_a instead
fun to_a: Array[Char] do return chars.to_a
# Create a substring from `self` beginning at the `from` position
#
- # assert "abcd".substring_from(1) == "bcd"
- # assert "abcd".substring_from(-1) == "abcd"
- # assert "abcd".substring_from(2) == "cd"
+ # ~~~
+ # assert "abcd".substring_from(1) == "bcd"
+ # assert "abcd".substring_from(-1) == "abcd"
+ # assert "abcd".substring_from(2) == "cd"
+ # ~~~
#
# As with substring, a `from` index < 0 will be replaced by 0
fun substring_from(from: Int): SELFTYPE
# Does self have a substring `str` starting from position `pos`?
#
- # assert "abcd".has_substring("bc",1) == true
- # assert "abcd".has_substring("bc",2) == false
+ # ~~~
+ # assert "abcd".has_substring("bc",1) == true
+ # assert "abcd".has_substring("bc",2) == false
+ # ~~~
#
# Returns true iff all characters of `str` are presents
# at the expected index in `self.`
#
# This means that all characters of `str` need to be inside `self`.
#
- # assert "abcd".has_substring("xab", -1) == false
- # assert "abcd".has_substring("cdx", 2) == false
+ # ~~~
+ # assert "abcd".has_substring("xab", -1) == false
+ # assert "abcd".has_substring("cdx", 2) == false
+ # ~~~
#
# And that the empty string is always a valid substring.
#
- # assert "abcd".has_substring("", 2) == true
- # assert "abcd".has_substring("", 200) == true
+ # ~~~
+ # assert "abcd".has_substring("", 2) == true
+ # assert "abcd".has_substring("", 200) == true
+ # ~~~
fun has_substring(str: String, pos: Int): Bool
do
if str.is_empty then return true
# Is this string prefixed by `prefix`?
#
- # assert "abcd".has_prefix("ab") == true
- # assert "abcbc".has_prefix("bc") == false
- # assert "ab".has_prefix("abcd") == false
+ # ~~~
+ # assert "abcd".has_prefix("ab") == true
+ # assert "abcbc".has_prefix("bc") == false
+ # assert "ab".has_prefix("abcd") == false
+ # ~~~
fun has_prefix(prefix: String): Bool do return has_substring(prefix,0)
# Is this string suffixed by `suffix`?
#
- # assert "abcd".has_suffix("abc") == false
- # assert "abcd".has_suffix("bcd") == true
+ # ~~~
+ # assert "abcd".has_suffix("abc") == false
+ # assert "abcd".has_suffix("bcd") == true
+ # ~~~
fun has_suffix(suffix: String): Bool do return has_substring(suffix, length - suffix.length)
# Returns `self` as the corresponding integer
#
- # assert "123".to_i == 123
- # assert "-1".to_i == -1
- # assert "0x64".to_i == 100
- # assert "0b1100_0011".to_i== 195
- # assert "--12".to_i == 12
+ # ~~~
+ # assert "123".to_i == 123
+ # assert "-1".to_i == -1
+ # assert "0x64".to_i == 100
+ # assert "0b1100_0011".to_i== 195
+ # assert "--12".to_i == 12
+ # assert "+45".to_i == 45
+ # ~~~
#
# REQUIRE: `self`.`is_int`
fun to_i: Int is abstract
# If `self` contains a float, return the corresponding float
#
- # assert "123".to_f == 123.0
- # assert "-1".to_f == -1.0
- # assert "-1.2e-3".to_f == -0.0012
+ # ~~~
+ # assert "123".to_f == 123.0
+ # assert "-1".to_f == -1.0
+ # assert "-1.2e-3".to_f == -0.0012
+ # ~~~
fun to_f: Float
do
# Shortcut
# If `self` contains only digits and alpha <= 'f', return the corresponding integer.
#
- # assert "ff".to_hex == 255
- fun to_hex: Int do return a_to(16)
+ # ~~~
+ # assert "ff".to_hex == 255
+ # ~~~
+ fun to_hex(pos, ln: nullable Int): Int do
+ var res = 0
+ if pos == null then pos = 0
+ if ln == null then ln = length - pos
+ var max = pos + ln
+ for i in [pos .. max[ do
+ res <<= 4
+ res += self[i].from_hex
+ end
+ return res
+ end
# If `self` contains only digits <= '7', return the corresponding integer.
#
- # assert "714".to_oct == 460
+ # ~~~
+ # assert "714".to_oct == 460
+ # ~~~
fun to_oct: Int do return a_to(8)
# If `self` contains only '0' et '1', return the corresponding integer.
#
- # assert "101101".to_bin == 45
+ # ~~~
+ # assert "101101".to_bin == 45
+ # ~~~
fun to_bin: Int do return a_to(2)
# If `self` contains only digits '0' .. '9', return the corresponding integer.
#
- # assert "108".to_dec == 108
+ # ~~~
+ # assert "108".to_dec == 108
+ # ~~~
fun to_dec: Int do return a_to(10)
# If `self` contains only digits and letters, return the corresponding integer in a given base
#
- # assert "120".a_to(3) == 15
+ # ~~~
+ # assert "120".a_to(3) == 15
+ # ~~~
fun a_to(base: Int) : Int
do
var i = 0
end
end
- # Returns `true` if the string contains only Numeric values (and one "," or one "." character)
+ # Is this string in a valid numeric format compatible with `to_f`?
#
- # assert "123".is_numeric == true
- # assert "1.2".is_numeric == true
- # assert "1,2".is_numeric == true
- # assert "1..2".is_numeric == false
+ # ~~~
+ # assert "123".is_numeric == true
+ # assert "1.2".is_numeric == true
+ # assert "-1.2".is_numeric == true
+ # assert "-1.23e-2".is_numeric == true
+ # assert "1..2".is_numeric == false
+ # assert "".is_numeric == false
+ # ~~~
fun is_numeric: Bool
do
- var has_point_or_comma = false
+ var has_point = false
+ var e_index = -1
for i in [0..length[ do
var c = chars[i]
if not c.is_numeric then
- if (c == '.' or c == ',') and not has_point_or_comma then
- has_point_or_comma = true
+ if c == '.' and not has_point then
+ has_point = true
+ else if c == 'e' and e_index == -1 and i > 0 and i < length - 1 and chars[i-1] != '-' then
+ e_index = i
+ else if c == '-' and i == e_index + 1 and i < length - 1 then
else
return false
end
end
end
- return true
+ return not is_empty
end
# Returns `true` if the string contains only Hex chars
#
- # assert "048bf".is_hex == true
- # assert "ABCDEF".is_hex == true
- # assert "0G".is_hex == false
+ # ~~~
+ # assert "048bf".is_hex == true
+ # assert "ABCDEF".is_hex == true
+ # assert "0G".is_hex == false
+ # ~~~
fun is_hex: Bool
do
for i in [0..length[ do
# Returns `true` if the string contains only Binary digits
#
- # assert "1101100".is_bin == true
- # assert "1101020".is_bin == false
+ # ~~~
+ # assert "1101100".is_bin == true
+ # assert "1101020".is_bin == false
+ # ~~~
fun is_bin: Bool do
for i in chars do if i != '0' and i != '1' then return false
return true
# Returns `true` if the string contains only Octal digits
#
- # assert "213453".is_oct == true
- # assert "781".is_oct == false
+ # ~~~
+ # assert "213453".is_oct == true
+ # assert "781".is_oct == false
+ # ~~~
fun is_oct: Bool do
for i in chars do if i < '0' or i > '7' then return false
return true
# Returns `true` if the string contains only Decimal digits
#
- # assert "10839".is_dec == true
- # assert "164F".is_dec == false
+ # ~~~
+ # assert "10839".is_dec == true
+ # assert "164F".is_dec == false
+ # ~~~
fun is_dec: Bool do
for i in chars do if i < '0' or i > '9' then return false
return true
# Are all letters in `self` upper-case ?
#
- # assert "HELLO WORLD".is_upper == true
- # assert "%$&%!".is_upper == true
- # assert "hello world".is_upper == false
- # assert "Hello World".is_upper == false
+ # ~~~
+ # assert "HELLO WORLD".is_upper == true
+ # assert "%$&%!".is_upper == true
+ # assert "hello world".is_upper == false
+ # assert "Hello World".is_upper == false
+ # ~~~
fun is_upper: Bool
do
for i in [0..length[ do
# Are all letters in `self` lower-case ?
#
- # assert "hello world".is_lower == true
- # assert "%$&%!".is_lower == true
- # assert "Hello World".is_lower == false
+ # ~~~
+ # assert "hello world".is_lower == true
+ # assert "%$&%!".is_lower == true
+ # assert "Hello World".is_lower == false
+ # ~~~
fun is_lower: Bool
do
for i in [0..length[ do
# Removes the whitespaces at the beginning of self
#
- # assert " \n\thello \n\t".l_trim == "hello \n\t"
+ # ~~~
+ # assert " \n\thello \n\t".l_trim == "hello \n\t"
+ # ~~~
#
# `Char::is_whitespace` determines what is a whitespace.
fun l_trim: SELFTYPE
# Removes the whitespaces at the end of self
#
- # assert " \n\thello \n\t".r_trim == " \n\thello"
+ # ~~~
+ # assert " \n\thello \n\t".r_trim == " \n\thello"
+ # ~~~
#
# `Char::is_whitespace` determines what is a whitespace.
fun r_trim: SELFTYPE
# Trims trailing and preceding white spaces
#
- # assert " Hello World ! ".trim == "Hello World !"
- # assert "\na\nb\tc\t".trim == "a\nb\tc"
+ # ~~~
+ # assert " Hello World ! ".trim == "Hello World !"
+ # assert "\na\nb\tc\t".trim == "a\nb\tc"
+ # ~~~
#
# `Char::is_whitespace` determines what is a whitespace.
fun trim: SELFTYPE do return (self.l_trim).r_trim
# Is the string non-empty but only made of whitespaces?
#
- # assert " \n\t ".is_whitespace == true
- # assert " hello ".is_whitespace == false
- # assert "".is_whitespace == false
+ # ~~~
+ # assert " \n\t ".is_whitespace == true
+ # assert " hello ".is_whitespace == false
+ # assert "".is_whitespace == false
+ # ~~~
#
# `Char::is_whitespace` determines what is a whitespace.
fun is_whitespace: Bool
# Returns `self` removed from its last line terminator (if any).
#
- # assert "Hello\n".chomp == "Hello"
- # assert "Hello".chomp == "Hello"
+ # ~~~
+ # assert "Hello\n".chomp == "Hello"
+ # assert "Hello".chomp == "Hello"
#
- # assert "\n".chomp == ""
- # assert "".chomp == ""
+ # assert "\n".chomp == ""
+ # assert "".chomp == ""
+ # ~~~
#
# Line terminators are `"\n"`, `"\r\n"` and `"\r"`.
# A single line terminator, the last one, is removed.
#
- # assert "\r\n".chomp == ""
- # assert "\r\n\n".chomp == "\r\n"
- # assert "\r\n\r\n".chomp == "\r\n"
- # assert "\r\n\r".chomp == "\r\n"
+ # ~~~
+ # assert "\r\n".chomp == ""
+ # assert "\r\n\n".chomp == "\r\n"
+ # assert "\r\n\r\n".chomp == "\r\n"
+ # assert "\r\n\r".chomp == "\r\n"
+ # ~~~
#
# Note: unlike with most IO methods like `Reader::read_line`,
# a single `\r` is considered here to be a line terminator and will be removed.
end
end
- # Justify a self in a space of `length`
+ # Justify `self` in a space of `length`
#
# `left` is the space ratio on the left side.
# * 0.0 for left-justified (no space at the left)
# * 1.0 for right-justified (all spaces at the left)
# * 0.5 for centered (half the spaces at the left)
#
+ # `char`, or `' '` by default, is repeated to pad the empty space.
+ #
# Examples
#
- # assert "hello".justify(10, 0.0) == "hello "
- # assert "hello".justify(10, 1.0) == " hello"
- # assert "hello".justify(10, 0.5) == " hello "
+ # ~~~
+ # assert "hello".justify(10, 0.0) == "hello "
+ # assert "hello".justify(10, 1.0) == " hello"
+ # assert "hello".justify(10, 0.5) == " hello "
+ # assert "hello".justify(10, 0.5, '.') == "..hello..."
+ # ~~~
#
# If `length` is not enough, `self` is returned as is.
#
- # assert "hello".justify(2, 0.0) == "hello"
+ # ~~~
+ # assert "hello".justify(2, 0.0) == "hello"
+ # ~~~
#
# REQUIRE: `left >= 0.0 and left <= 1.0`
# ENSURE: `self.length <= length implies result.length == length`
# ENSURE: `self.length >= length implies result == self`
- fun justify(length: Int, left: Float): String
+ fun justify(length: Int, left: Float, char: nullable Char): String
do
+ var pad = (char or else ' ').to_s
var diff = length - self.length
if diff <= 0 then return to_s
assert left >= 0.0 and left <= 1.0
var before = (diff.to_f * left).to_i
- return " " * before + self + " " * (diff-before)
+ return pad * before + self + pad * (diff-before)
end
# Mangle a string to be a unique string only made of alphanumeric characters and underscores.
# * Never ends with an underscore.
# * Never contains two contiguous underscores.
#
- # assert "42_is/The answer!".to_cmangle == "_52d2_is_47dThe_32danswer_33d"
- # assert "__".to_cmangle == "_95d_95d"
- # assert "__d".to_cmangle == "_95d_d"
- # assert "_d_".to_cmangle == "_d_95d"
- # assert "_42".to_cmangle == "_95d42"
- # assert "foo".to_cmangle == "foo"
- # assert "".to_cmangle == ""
+ # Examples:
+ #
+ # ~~~
+ # assert "42_is/The answer!".to_cmangle == "_52d2_is_47dThe_32danswer_33d"
+ # assert "__".to_cmangle == "_95d_95d"
+ # assert "__d".to_cmangle == "_95d_d"
+ # assert "_d_".to_cmangle == "_d_95d"
+ # assert "_42".to_cmangle == "_95d42"
+ # assert "foo".to_cmangle == "foo"
+ # assert "".to_cmangle == ""
+ # ~~~
fun to_cmangle: String
do
if is_empty then return ""
var res = new Buffer
var underscore = false
var start = 0
- var c = chars[0]
+ var c = self[0]
if c >= '0' and c <= '9' then
res.add('_')
- res.append(c.ascii.to_s)
+ res.append(c.code_point.to_s)
res.add('d')
start = 1
end
for i in [start..length[ do
- c = chars[i]
+ c = self[i]
if (c >= 'a' and c <= 'z') or (c >='A' and c <= 'Z') then
res.add(c)
underscore = false
continue
end
if underscore then
- res.append('_'.ascii.to_s)
+ res.append('_'.code_point.to_s)
res.add('d')
end
if c >= '0' and c <= '9' then
underscore = true
else
res.add('_')
- res.append(c.ascii.to_s)
+ res.append(c.code_point.to_s)
res.add('d')
underscore = false
end
end
if underscore then
- res.append('_'.ascii.to_s)
+ res.append('_'.code_point.to_s)
res.add('d')
end
return res.to_s
end
- # Escape " \ ' and non printable characters using the rules of literal C strings and characters
+ # Escape `"` `\` `'`, trigraphs and non printable characters using the rules of literal C strings and characters
#
- # assert "abAB12<>&".escape_to_c == "abAB12<>&"
- # assert "\n\"'\\".escape_to_c == "\\n\\\"\\'\\\\"
+ # ~~~
+ # assert "abAB12<>&".escape_to_c == "abAB12<>&"
+ # assert "\n\"'\\".escape_to_c == "\\n\\\"\\'\\\\"
+ # assert "allo???!".escape_to_c == "allo??\\?!"
+ # assert "??=??/??'??(??)".escape_to_c == "?\\?=?\\?/??\\'?\\?(?\\?)"
+ # assert "??!??<??>??-".escape_to_c == "?\\?!?\\?<?\\?>?\\?-"
+ # ~~~
#
# Most non-printable characters (bellow ASCII 32) are escaped to an octal form `\nnn`.
# Three digits are always used to avoid following digits to be interpreted as an element
# of the octal sequence.
#
- # assert "{0.ascii}{1.ascii}{8.ascii}{31.ascii}{32.ascii}".escape_to_c == "\\000\\001\\010\\037 "
+ # ~~~
+ # assert "{0.code_point}{1.code_point}{8.code_point}{31.code_point}{32.code_point}".escape_to_c == "\\000\\001\\010\\037 "
+ # ~~~
#
# The exceptions are the common `\t` and `\n`.
fun escape_to_c: String
b.append("\\\'")
else if c == '\\' then
b.append("\\\\")
- else if c.ascii < 32 then
+ else if c == '?' then
+ # Escape if it is the last question mark of a ANSI C trigraph.
+ var j = i + 1
+ if j < length then
+ var next = chars[j]
+ # We ignore `??'` because it will be escaped as `??\'`.
+ if
+ next == '!' or
+ next == '(' or
+ next == ')' or
+ next == '-' or
+ next == '/' or
+ next == '<' or
+ next == '=' or
+ next == '>'
+ then b.add('\\')
+ end
+ b.add('?')
+ else if c.code_point < 32 then
b.add('\\')
- var oct = c.ascii.to_base(8, false)
+ var oct = c.code_point.to_base(8)
# Force 3 octal digits since it is the
# maximum allowed in the C specification
if oct.length == 1 then
# Escape additionnal characters
# The result might no be legal in C but be used in other languages
#
- # assert "ab|\{\}".escape_more_to_c("|\{\}") == "ab\\|\\\{\\\}"
+ # ~~~
+ # assert "ab|\{\}".escape_more_to_c("|\{\}") == "ab\\|\\\{\\\}"
+ # assert "allo???!".escape_more_to_c("") == "allo??\\?!"
+ # ~~~
fun escape_more_to_c(chars: String): String
do
var b = new Buffer
# Escape to C plus braces
#
- # assert "\n\"'\\\{\}".escape_to_nit == "\\n\\\"\\'\\\\\\\{\\\}"
+ # ~~~
+ # assert "\n\"'\\\{\}".escape_to_nit == "\\n\\\"\\'\\\\\\\{\\\}"
+ # ~~~
fun escape_to_nit: String do return escape_more_to_c("\{\}")
# Escape to POSIX Shell (sh).
#
# Abort if the text contains a null byte.
#
- # assert "\n\"'\\\{\}0".escape_to_sh == "'\n\"'\\''\\\{\}0'"
+ # ~~~
+ # assert "\n\"'\\\{\}0".escape_to_sh == "'\n\"'\\''\\\{\}0'"
+ # ~~~
fun escape_to_sh: String do
var b = new Buffer
b.chars.add '\''
else if c == ':' or c == ' ' or c == '#' then
b.add('\\')
b.add(c)
- else if c.ascii < 32 or c == ';' or c == '|' or c == '\\' or c == '=' then
- b.append("?{c.ascii.to_base(16, false)}")
+ else if c.code_point < 32 or c == ';' or c == '|' or c == '\\' then
+ b.append("?{c.code_point.to_base(16)}")
else
b.add(c)
end
# Return a string where Nit escape sequences are transformed.
#
- # var s = "\\n"
- # assert s.length == 2
- # var u = s.unescape_nit
- # assert u.length == 1
- # assert u.chars[0].ascii == 10 # (the ASCII value of the "new line" character)
+ # ~~~
+ # var s = "\\n"
+ # assert s.length == 2
+ # var u = s.unescape_nit
+ # assert u.length == 1
+ # assert u.chars[0].code_point == 10 # (the ASCII value of the "new line" character)
+ # ~~~
fun unescape_nit: String
do
var res = new Buffer.with_cap(self.length)
return res.to_s
end
+ # Returns `self` with all characters escaped with their UTF-16 representation
+ #
+ # ~~~
+ # assert "Aèあ𐏓".escape_to_utf16 == "\\u0041\\u00e8\\u3042\\ud800\\udfd3"
+ # ~~~
+ fun escape_to_utf16: String do
+ var buf = new Buffer
+ for i in chars do buf.append i.escape_to_utf16
+ return buf.to_s
+ end
+
+ # Returns the Unicode char escaped by `self`
+ #
+ # ~~~
+ # assert "\\u0041".from_utf16_escape == 'A'
+ # assert "\\ud800\\udfd3".from_utf16_escape == '𐏓'
+ # assert "\\u00e8".from_utf16_escape == 'è'
+ # assert "\\u3042".from_utf16_escape == 'あ'
+ # ~~~
+ fun from_utf16_escape(pos, ln: nullable Int): Char do
+ if pos == null then pos = 0
+ if ln == null then ln = length - pos
+ if ln < 6 then return 0xFFFD.code_point
+ var cp = from_utf16_digit(pos + 2).to_u32
+ if cp < 0xD800u32 then return cp.code_point
+ if cp > 0xDFFFu32 then return cp.code_point
+ if cp > 0xDBFFu32 then return 0xFFFD.code_point
+ if ln == 6 then return 0xFFFD.code_point
+ if ln < 12 then return 0xFFFD.code_point
+ cp <<= 16
+ cp += from_utf16_digit(pos + 8).to_u32
+ var cplo = cp & 0xFFFFu32
+ if cplo < 0xDC00u32 then return 0xFFFD.code_point
+ if cplo > 0xDFFFu32 then return 0xFFFD.code_point
+ return cp.from_utf16_surr.code_point
+ end
+
+ # Returns a UTF-16 escape value
+ #
+ # ~~~
+ # var s = "\\ud800\\udfd3"
+ # assert s.from_utf16_digit(2) == 0xD800
+ # assert s.from_utf16_digit(8) == 0xDFD3
+ # ~~~
+ fun from_utf16_digit(pos: nullable Int): Int do
+ if pos == null then pos = 0
+ return to_hex(pos, 4)
+ end
+
# Encode `self` to percent (or URL) encoding
#
- # assert "aBc09-._~".to_percent_encoding == "aBc09-._~"
- # assert "%()< >".to_percent_encoding == "%25%28%29%3c%20%3e"
- # assert ".com/post?e=asdf&f=123".to_percent_encoding == ".com%2fpost%3fe%3dasdf%26f%3d123"
+ # ~~~
+ # assert "aBc09-._~".to_percent_encoding == "aBc09-._~"
+ # assert "%()< >".to_percent_encoding == "%25%28%29%3c%20%3e"
+ # assert ".com/post?e=asdf&f=123".to_percent_encoding == ".com%2fpost%3fe%3dasdf%26f%3d123"
+ # assert "éあいう".to_percent_encoding == "%c3%a9%e3%81%82%e3%81%84%e3%81%86"
+ # ~~~
fun to_percent_encoding: String
do
var buf = new Buffer
c == '_' or c == '~'
then
buf.add c
- else buf.append "%{c.ascii.to_hex}"
+ else
+ var bytes = c.to_s.bytes
+ for b in bytes do buf.append "%{b.to_i.to_hex}"
+ end
end
return buf.to_s
# Decode `self` from percent (or URL) encoding to a clear string
#
- # Replace invalid use of '%' with '?'.
+ # Invalid '%' are not decoded.
#
- # assert "aBc09-._~".from_percent_encoding == "aBc09-._~"
- # assert "%25%28%29%3c%20%3e".from_percent_encoding == "%()< >"
- # assert ".com%2fpost%3fe%3dasdf%26f%3d123".from_percent_encoding == ".com/post?e=asdf&f=123"
- # assert "%25%28%29%3C%20%3E".from_percent_encoding == "%()< >"
- # assert "incomplete %".from_percent_encoding == "incomplete ?"
- # assert "invalid % usage".from_percent_encoding == "invalid ? usage"
+ # ~~~
+ # assert "aBc09-._~".from_percent_encoding == "aBc09-._~"
+ # assert "%25%28%29%3c%20%3e".from_percent_encoding == "%()< >"
+ # assert ".com%2fpost%3fe%3dasdf%26f%3d123".from_percent_encoding == ".com/post?e=asdf&f=123"
+ # assert "%25%28%29%3C%20%3E".from_percent_encoding == "%()< >"
+ # assert "incomplete %".from_percent_encoding == "incomplete %"
+ # assert "invalid % usage".from_percent_encoding == "invalid % usage"
+ # assert "%c3%a9%e3%81%82%e3%81%84%e3%81%86".from_percent_encoding == "éあいう"
+ # assert "%1 %A %C3%A9A9".from_percent_encoding == "%1 %A éA9"
+ # ~~~
fun from_percent_encoding: String
do
- var buf = new Buffer
+ var len = byte_length
+ var has_percent = false
+ for c in chars do
+ if c == '%' then
+ len -= 2
+ has_percent = true
+ end
+ end
+
+ # If no transformation is needed, return self as a string
+ if not has_percent then return to_s
+ var buf = new CString(len)
var i = 0
+ var l = 0
while i < length do
var c = chars[i]
if c == '%' then
if i + 2 >= length then
# What follows % has been cut off
- buf.add '?'
+ buf[l] = u'%'
else
i += 1
var hex_s = substring(i, 2)
if hex_s.is_hex then
var hex_i = hex_s.to_hex
- buf.add hex_i.ascii
+ buf[l] = hex_i
i += 1
else
# What follows a % is not Hex
- buf.add '?'
+ buf[l] = u'%'
i -= 1
end
end
- else buf.add c
+ else buf[l] = c.code_point
i += 1
+ l += 1
end
- return buf.to_s
+ return buf.to_s_unsafe(l, copy=false)
end
# Escape the characters `<`, `>`, `&`, `"`, `'` and `/` as HTML/XML entity references.
#
- # assert "a&b-<>\"x\"/'".html_escape == "a&b-<>"x"/'"
+ # ~~~
+ # assert "a&b-<>\"x\"/'".html_escape == "a&b-<>"x"/'"
+ # ~~~
#
# SEE: <https://www.owasp.org/index.php/XSS_%28Cross_Site_Scripting%29_Prevention_Cheat_Sheet#RULE_.231_-_HTML_Escape_Before_Inserting_Untrusted_Data_into_HTML_Element_Content>
fun html_escape: String
do
- var buf = new Buffer
+ var buf: nullable Buffer = null
for i in [0..length[ do
var c = chars[i]
+ var sub = null
if c == '&' then
- buf.append "&"
+ sub = "&"
else if c == '<' then
- buf.append "<"
+ sub = "<"
else if c == '>' then
- buf.append ">"
+ sub = ">"
else if c == '"' then
- buf.append """
+ sub = """
else if c == '\'' then
- buf.append "'"
+ sub = "'"
else if c == '/' then
- buf.append "/"
- else buf.add c
+ sub = "/"
+ else
+ if buf != null then buf.add c
+ continue
+ end
+ if buf == null then
+ buf = new Buffer
+ for j in [0..i[ do buf.add chars[j]
+ end
+ buf.append sub
end
+ if buf == null then return self.to_s
return buf.to_s
end
# Equality of text
# Two pieces of text are equals if thez have the same characters in the same order.
#
- # assert "hello" == "hello"
- # assert "hello" != "HELLO"
- # assert "hello" == "hel"+"lo"
+ # ~~~
+ # assert "hello" == "hello"
+ # assert "hello" != "HELLO"
+ # assert "hello" == "hel"+"lo"
+ # ~~~
#
# Things that are not Text are not equal.
#
- # assert "9" != '9'
- # assert "9" != ['9']
- # assert "9" != 9
+ # ~~~
+ # assert "9" != '9'
+ # assert "9" != ['9']
+ # assert "9" != 9
#
- # assert "9".chars.first == '9' # equality of Char
- # assert "9".chars == ['9'] # equality of Sequence
- # assert "9".to_i == 9 # equality of Int
+ # assert "9".chars.first == '9' # equality of Char
+ # assert "9".chars == ['9'] # equality of Sequence
+ # assert "9".to_i == 9 # equality of Int
+ # ~~~
redef fun ==(o)
do
if o == null then return false
# Lexicographical comparaison
#
- # assert "abc" < "xy"
- # assert "ABC" < "abc"
+ # ~~~
+ # assert "abc" < "xy"
+ # assert "ABC" < "abc"
+ # ~~~
redef fun <(other)
do
var self_chars = self.chars.iterator
# Escape string used in labels for graphviz
#
- # assert ">><<".escape_to_dot == "\\>\\>\\<\\<"
+ # ~~~
+ # assert ">><<".escape_to_dot == "\\>\\>\\<\\<"
+ # ~~~
fun escape_to_dot: String
do
return escape_more_to_c("|\{\}<>")
for i in [0..length[ do
var char = chars[i]
- h = (h << 5) + h + char.ascii
+ h = (h << 5) + h + char.code_point
end
hash_cache = h
return hash_cache.as(not null)
end
- # Gives the formatted string back as a Nit string with `args` in place
+ # Format `self` by replacing each `%n` with the `n`th item of `args`
#
- # assert "This %1 is a %2.".format("String", "formatted String") == "This String is a formatted String."
- # assert "\\%1 This string".format("String") == "\\%1 This string"
+ # The character `%` followed by something other than a number are left as is.
+ # To represent a `%` followed by a number, double the `%`, as in `%%7`.
+ #
+ # ~~~
+ # assert "This %0 is a %1.".format("String", "formatted String") == "This String is a formatted String."
+ # assert "Do not escape % nor %%1".format("unused") == "Do not escape % nor %1"
+ # ~~~
fun format(args: Object...): String do
var s = new Array[Text]
var curr_st = 0
var i = 0
while i < length do
- # Skip escaped characters
- if self[i] == '\\' then
- i += 1
- # In case of format
- else if self[i] == '%' then
+ if self[i] == '%' then
var fmt_st = i
i += 1
var ciph_st = i
while i < length and self[i].is_numeric do
i += 1
end
- i -= 1
- var fmt_end = i
- var ciph_len = fmt_end - ciph_st + 1
- var arg_index = substring(ciph_st, ciph_len).to_i - 1
+ var ciph_len = i - ciph_st
+ if ciph_len == 0 then
+ # What follows '%' is not a number.
+ s.push substring(curr_st, i - curr_st)
+ if i < length and self[i] == '%' then
+ # Skip the next `%`
+ i += 1
+ end
+ curr_st = i
+ continue
+ end
+
+ var arg_index = substring(ciph_st, ciph_len).to_i
if arg_index >= args.length then continue
s.push substring(curr_st, fmt_st - curr_st)
s.push args[arg_index].to_s
- curr_st = i + 1
+
+ curr_st = i
+ i -= 1
end
i += 1
end
return s.plain_to_s
end
+ # Return the Levenshtein distance between two strings
+ #
+ # ~~~
+ # assert "abcd".levenshtein_distance("abcd") == 0
+ # assert "".levenshtein_distance("abcd") == 4
+ # assert "abcd".levenshtein_distance("") == 4
+ # assert "abcd".levenshtein_distance("xyz") == 4
+ # assert "abcd".levenshtein_distance("xbdy") == 3
+ # ~~~
+ fun levenshtein_distance(other: String): Int
+ do
+ var slen = self.length
+ var olen = other.length
+
+ # fast cases
+ if slen == 0 then return olen
+ if olen == 0 then return slen
+ if self == other then return 0
+
+ # previous row of distances
+ var v0 = new Array[Int].with_capacity(olen+1)
+
+ # current row of distances
+ var v1 = new Array[Int].with_capacity(olen+1)
+
+ for j in [0..olen] do
+ # prefix insert cost
+ v0[j] = j
+ end
+
+ for i in [0..slen[ do
+
+ # prefix delete cost
+ v1[0] = i + 1
+
+ for j in [0..olen[ do
+ # delete cost
+ var cost1 = v1[j] + 1
+ # insert cost
+ var cost2 = v0[j + 1] + 1
+ # same char cost (+0)
+ var cost3 = v0[j]
+ # change cost
+ if self[i] != other[j] then cost3 += 1
+ # keep the min
+ v1[j+1] = cost1.min(cost2).min(cost3)
+ end
+
+ # Switch columns:
+ # * v1 become v0 in the next iteration
+ # * old v0 is reused as the new v1
+ var tmp = v1
+ v1 = v0
+ v0 = tmp
+ end
+
+ return v0[olen]
+ end
+
# Copies `n` bytes from `self` at `src_offset` into `dest` starting at `dest_offset`
#
- # Basically a high-level synonym of NativeString::copy_to
+ # Basically a high-level synonym of CString::copy_to
#
# REQUIRE: `n` must be large enough to contain `len` bytes
#
- # var ns = new NativeString(8)
- # "Text is String".copy_to_native(ns, 8, 2, 0)
- # assert ns.to_s_with_length(8) == "xt is St"
- #
- fun copy_to_native(dest: NativeString, n, src_offset, dest_offset: Int) do
+ # ~~~
+ # var ns = new CString(8)
+ # "Text is String".copy_to_native(ns, 8, 2, 0)
+ # assert ns.to_s_with_length(8) == "xt is St"
+ # ~~~
+ fun copy_to_native(dest: CString, n, src_offset, dest_offset: Int) do
var mypos = src_offset
var itspos = dest_offset
while n > 0 do
end
end
-end
-
-# All kinds of array-based text representations.
-abstract class FlatText
- super Text
-
- # Underlying C-String (`char*`)
+ # Packs the content of a string in packs of `ln` chars.
+ # This variant ensures that only the last element might be smaller than `ln`
#
- # Warning : Might be void in some subclasses, be sure to check
- # if set before using it.
- private var items: NativeString is noinit
-
- # Real items, used as cache for to_cstring is called
- private var real_items: nullable NativeString = null
+ # ~~~
+ # var s = "abcdefghijklmnopqrstuvwxyz"
+ # assert s.pack_l(4) == ["abcd","efgh","ijkl","mnop","qrst","uvwx","yz"]
+ # ~~~
+ fun pack_l(ln: Int): Array[Text] do
+ var st = 0
+ var retarr = new Array[Text].with_capacity(length / ln + length % ln)
+ while st < length do
+ retarr.add(substring(st, ln))
+ st += ln
+ end
+ return retarr
+ end
- # Returns a char* starting at position `first_byte`
- #
- # WARNING: If you choose to use this service, be careful of the following.
- #
- # Strings and NativeString are *ideally* always allocated through a Garbage Collector.
- # Since the GC tracks the use of the pointer for the beginning of the char*, it may be
- # deallocated at any moment, rendering the pointer returned by this function invalid.
- # Any access to freed memory may very likely cause undefined behaviour or a crash.
- # (Failure to do so will most certainly result in long and painful debugging hours)
+ # Packs the content of a string in packs of `ln` chars.
+ # This variant ensures that only the first element might be smaller than `ln`
#
- # The only safe use of this pointer is if it is ephemeral (e.g. read in a C function
- # then immediately return).
- #
- # As always, do not modify the content of the String in C code, if this is what you want
- # copy locally the char* as Nit Strings are immutable.
- private fun fast_cstring: NativeString is abstract
-
- redef var length = 0
-
- redef var bytelen = 0
-
- redef fun output
- do
- var i = 0
- while i < length do
- items[i].output
- i += 1
+ # ~~~
+ # var s = "abcdefghijklmnopqrstuvwxyz"
+ # assert s.pack_r(4) == ["ab","cdef","ghij","klmn","opqr","stuv","wxyz"]
+ # ~~~
+ fun pack_r(ln: Int): Array[Text] do
+ var st = length
+ var retarr = new Array[Text].with_capacity(length / ln + length % ln)
+ while st >= 0 do
+ retarr.add(substring(st - ln, ln))
+ st -= ln
end
+ return retarr.reversed
end
- redef fun copy_to_native(dest, n, src_offset, dest_offset) do
- items.copy_to(dest, n, src_offset, dest_offset)
- end
-end
+ # Concatenates self `i` times
+ #
+ # ~~~
+ # assert "abc" * 4 == "abcabcabcabc"
+ # assert "abc" * 1 == "abc"
+ # assert "abc" * 0 == ""
+ # var b = new Buffer
+ # b.append("天地")
+ # b = b * 4
+ # assert b == "天地天地天地天地"
+ # ~~~
+ fun *(i: Int): SELFTYPE is abstract
-# Abstract class for the SequenceRead compatible
-# views on the chars of any Text
-private abstract class StringCharView
- super SequenceRead[Char]
+ # Insert `s` at `pos`.
+ #
+ # ~~~
+ # assert "helloworld".insert_at(" ", 5) == "hello world"
+ # var b = new Buffer
+ # b.append("Hello世界")
+ # b = b.insert_at(" beautiful ", 5)
+ # assert b == "Hello beautiful 世界"
+ # ~~~
+ fun insert_at(s: String, pos: Int): SELFTYPE is abstract
- type SELFTYPE: Text
+ # Returns a reversed version of self
+ #
+ # ~~~
+ # assert "hello".reversed == "olleh"
+ # assert "bob".reversed == "bob"
+ # assert "".reversed == ""
+ # ~~~
+ fun reversed: SELFTYPE is abstract
- var target: SELFTYPE
-
- redef fun is_empty do return target.is_empty
-
- redef fun length do return target.length
-
- redef fun iterator: IndexedIterator[Char] do return self.iterator_from(0)
-
- redef fun reverse_iterator do return self.reverse_iterator_from(self.length - 1)
-end
-
-# Abstract class for the SequenceRead compatible
-# views on the bytes of any Text
-private abstract class StringByteView
- super SequenceRead[Byte]
-
- type SELFTYPE: Text
-
- var target: SELFTYPE
-
- redef fun is_empty do return target.is_empty
-
- redef fun length do return target.length
-
- redef fun iterator do return self.iterator_from(0)
-
- redef fun reverse_iterator do return self.reverse_iterator_from(target.bytelen - 1)
-end
-
-# Immutable sequence of characters.
-#
-# String objects may be created using literals.
-#
-# assert "Hello World!" isa String
-abstract class String
- super Text
-
- redef type SELFTYPE: String is fixed
-
- redef fun to_s do return self
-
- # Concatenates `o` to `self`
- #
- # assert "hello" + "world" == "helloworld"
- # assert "" + "hello" + "" == "hello"
- fun +(o: Text): SELFTYPE is abstract
-
- # Concatenates self `i` times
- #
- # assert "abc" * 4 == "abcabcabcabc"
- # assert "abc" * 1 == "abc"
- # assert "abc" * 0 == ""
- fun *(i: Int): SELFTYPE is abstract
-
- # Insert `s` at `pos`.
- #
- # assert "helloworld".insert_at(" ", 5) == "hello world"
- fun insert_at(s: String, pos: Int): SELFTYPE is abstract
-
- redef fun substrings is abstract
-
- # Returns a reversed version of self
- #
- # assert "hello".reversed == "olleh"
- # assert "bob".reversed == "bob"
- # assert "".reversed == ""
- fun reversed: SELFTYPE is abstract
-
- # A upper case version of `self`
- #
- # assert "Hello World!".to_upper == "HELLO WORLD!"
- fun to_upper: SELFTYPE is abstract
+ # A upper case version of `self`
+ #
+ # ~~~
+ # assert "Hello World!".to_upper == "HELLO WORLD!"
+ # ~~~
+ fun to_upper: SELFTYPE is abstract
# A lower case version of `self`
#
- # assert "Hello World!".to_lower == "hello world!"
+ # ~~~
+ # assert "Hello World!".to_lower == "hello world!"
+ # ~~~
fun to_lower : SELFTYPE is abstract
# Takes a camel case `self` and converts it to snake case
#
- # assert "randomMethodId".to_snake_case == "random_method_id"
+ # ~~~
+ # assert "randomMethodId".to_snake_case == "random_method_id"
+ # ~~~
#
# The rules are the following:
#
# An uppercase is always converted to a lowercase
#
- # assert "HELLO_WORLD".to_snake_case == "hello_world"
+ # ~~~
+ # assert "HELLO_WORLD".to_snake_case == "hello_world"
+ # ~~~
#
# An uppercase that follows a lowercase is prefixed with an underscore
#
- # assert "HelloTheWORLD".to_snake_case == "hello_the_world"
+ # ~~~
+ # assert "HelloTheWORLD".to_snake_case == "hello_the_world"
+ # ~~~
#
# An uppercase that follows an uppercase and is followed by a lowercase, is prefixed with an underscore
#
- # assert "HelloTHEWorld".to_snake_case == "hello_the_world"
+ # ~~~
+ # assert "HelloTHEWorld".to_snake_case == "hello_the_world"
+ # ~~~
#
# All other characters are kept as is; `self` does not need to be a proper CamelCased string.
#
- # assert "=-_H3ll0Th3W0rld_-=".to_snake_case == "=-_h3ll0th3w0rld_-="
- fun to_snake_case: SELFTYPE
- do
- if self.is_lower then return self
-
- var new_str = new Buffer.with_cap(self.length)
- var prev_is_lower = false
- var prev_is_upper = false
-
- for i in [0..length[ do
- var char = chars[i]
- if char.is_lower then
- new_str.add(char)
- prev_is_lower = true
- prev_is_upper = false
- else if char.is_upper then
- if prev_is_lower then
- new_str.add('_')
- else if prev_is_upper and i+1 < length and chars[i+1].is_lower then
- new_str.add('_')
- end
- new_str.add(char.to_lower)
- prev_is_lower = false
- prev_is_upper = true
- else
- new_str.add(char)
- prev_is_lower = false
- prev_is_upper = false
- end
- end
-
- return new_str.to_s
- end
+ # ~~~
+ # assert "=-_H3ll0Th3W0rld_-=".to_snake_case == "=-_h3ll0th3w0rld_-="
+ # ~~~
+ fun to_snake_case: SELFTYPE is abstract
# Takes a snake case `self` and converts it to camel case
#
- # assert "random_method_id".to_camel_case == "randomMethodId"
+ # ~~~
+ # assert "random_method_id".to_camel_case == "randomMethodId"
+ # ~~~
#
# If the identifier is prefixed by an underscore, the underscore is ignored
#
- # assert "_private_field".to_camel_case == "_privateField"
+ # ~~~
+ # assert "_private_field".to_camel_case == "_privateField"
+ # ~~~
#
# If `self` is upper, it is returned unchanged
#
- # assert "RANDOM_ID".to_camel_case == "RANDOM_ID"
+ # ~~~
+ # assert "RANDOM_ID".to_camel_case == "RANDOM_ID"
+ # ~~~
#
# If there are several consecutive underscores, they are considered as a single one
#
- # assert "random__method_id".to_camel_case == "randomMethodId"
- fun to_camel_case: SELFTYPE
- do
- if self.is_upper then return self
-
- var new_str = new Buffer
- var is_first_char = true
- var follows_us = false
-
- for i in [0..length[ do
- var char = chars[i]
- if is_first_char then
- new_str.add(char)
- is_first_char = false
- else if char == '_' then
- follows_us = true
- else if follows_us then
- new_str.add(char.to_upper)
- follows_us = false
- else
- new_str.add(char)
- end
- end
-
- return new_str.to_s
- end
+ # ~~~
+ # assert "random__method_id".to_camel_case == "randomMethodId"
+ # ~~~
+ fun to_camel_case: SELFTYPE is abstract
# Returns a capitalized `self`
#
# Letters that follow a letter are lowercased
# Letters that follow a non-letter are upcased.
#
+ # If `keep_upper = true`, already uppercase letters are not lowercased.
+ #
# SEE : `Char::is_letter` for the definition of letter.
#
- # assert "jAVASCRIPT".capitalized == "Javascript"
- # assert "i am root".capitalized == "I Am Root"
- # assert "ab_c -ab0c ab\nc".capitalized == "Ab_C -Ab0C Ab\nC"
- fun capitalized: SELFTYPE do
+ # ~~~
+ # assert "jAVASCRIPT".capitalized == "Javascript"
+ # assert "i am root".capitalized == "I Am Root"
+ # assert "ab_c -ab0c ab\nc".capitalized == "Ab_C -Ab0C Ab\nC"
+ # assert "preserve my ACRONYMS".capitalized(keep_upper=true) == "Preserve My ACRONYMS"
+ # ~~~
+ fun capitalized(keep_upper: nullable Bool): SELFTYPE do
if length == 0 then return self
var buf = new Buffer.with_cap(length)
+ buf.capitalize(keep_upper=keep_upper, src=self)
+ return buf.to_s
+ end
+end
- var curr = chars[0].to_upper
- var prev = curr
- buf[0] = curr
+# All kinds of array-based text representations.
+abstract class FlatText
+ super Text
- for i in [1 .. length[ do
- prev = curr
- curr = self[i]
- if prev.is_letter then
- buf[i] = curr.to_lower
- else
- buf[i] = curr.to_upper
- end
+ # Underlying CString (`char*`)
+ #
+ # Warning: Might be void in some subclasses, be sure to check
+ # if set before using it.
+ var items: CString is noinit
+
+ # Returns a char* starting at position `first_byte`
+ #
+ # WARNING: If you choose to use this service, be careful of the following.
+ #
+ # Strings and CString are *ideally* always allocated through a Garbage Collector.
+ # Since the GC tracks the use of the pointer for the beginning of the char*, it may be
+ # deallocated at any moment, rendering the pointer returned by this function invalid.
+ # Any access to freed memory may very likely cause undefined behaviour or a crash.
+ # (Failure to do so will most certainly result in long and painful debugging hours)
+ #
+ # The only safe use of this pointer is if it is ephemeral (e.g. read in a C function
+ # then immediately return).
+ #
+ # As always, do not modify the content of the String in C code, if this is what you want
+ # copy locally the char* as Nit Strings are immutable.
+ fun fast_cstring: CString is abstract
+
+ redef var length = 0
+
+ redef var byte_length = 0
+
+ redef fun output
+ do
+ var i = 0
+ while i < length do
+ items[i].output
+ i += 1
end
+ end
- return buf.to_s
+ redef fun copy_to_native(dest, n, src_offset, dest_offset) do
+ items.copy_to(dest, n, src_offset, dest_offset)
+ end
+end
+
+# Abstract class for the SequenceRead compatible
+# views on the chars of any Text
+private abstract class StringCharView
+ super SequenceRead[Char]
+
+ type SELFTYPE: Text
+
+ var target: SELFTYPE
+
+ redef fun is_empty do return target.is_empty
+
+ redef fun length do return target.length
+
+ redef fun iterator: IndexedIterator[Char] do return self.iterator_from(0)
+
+ redef fun reverse_iterator do return self.reverse_iterator_from(self.length - 1)
+end
+
+# Abstract class for the SequenceRead compatible
+# views on the bytes of any Text
+private abstract class StringByteView
+ super SequenceRead[Int]
+
+ type SELFTYPE: Text
+
+ var target: SELFTYPE
+
+ redef fun is_empty do return target.is_empty
+
+ redef fun length do return target.byte_length
+
+ redef fun iterator do return self.iterator_from(0)
+
+ redef fun reverse_iterator do return self.reverse_iterator_from(target.byte_length - 1)
+end
+
+# Immutable sequence of characters.
+#
+# String objects may be created using literals.
+#
+# ~~~
+# assert "Hello World!" isa String
+# ~~~
+abstract class String
+ super Text
+
+ redef type SELFTYPE: String is fixed
+
+ redef fun to_s do return self
+
+ redef fun clone do return self
+
+ redef fun to_buffer do return new Buffer.from_text(self)
+
+ redef fun to_camel_case do
+ if self.is_upper then return self
+
+ var new_str = new Buffer.with_cap(length)
+ new_str.append self
+ new_str.camel_case
+ return new_str.to_s
+ end
+
+ redef fun to_snake_case do
+ if self.is_lower then return self
+
+ var new_str = new Buffer.with_cap(self.length)
+ new_str.append self
+ new_str.snake_case
+ return new_str.to_s
end
end
# Returns an instance of a subclass of `Buffer` with `i` base capacity
new with_cap(i: Int) is abstract
- redef type SELFTYPE: Buffer is fixed
+ # Returns an instance of a subclass of `Buffer` with `t` as content
+ new from_text(t: Text) do
+ var ret = new Buffer.with_cap(t.byte_length)
+ ret.append t
+ return ret
+ end
- # Specific implementations MUST set this to `true` in order to invalidate caches
- protected var is_dirty = true
+ redef type SELFTYPE: Buffer is fixed
# Copy-On-Write flag
#
protected var written = false
# Modifies the char contained at pos `index`
- #
- # DEPRECATED : Use self.chars.[]= instead
fun []=(index: Int, item: Char) is abstract
+ redef fun to_buffer do return clone
+
+ # ~~~
+ # var b = new Buffer
+ # b.append("Buffer!")
+ # var c = b.clone
+ # assert b == c
+ # ~~~
+ redef fun clone do
+ var cln = new Buffer.with_cap(byte_length)
+ cln.append self
+ return cln
+ end
+
# Adds a char `c` at the end of self
- #
- # DEPRECATED : Use self.chars.add instead
fun add(c: Char) is abstract
# Clears the buffer
#
- # var b = new Buffer
- # b.append "hello"
- # assert not b.is_empty
- # b.clear
- # assert b.is_empty
+ # ~~~
+ # var b = new Buffer
+ # b.append "hello"
+ # assert not b.is_empty
+ # b.clear
+ # assert b.is_empty
+ # ~~~
fun clear is abstract
# Enlarges the subsequent array containing the chars of self
# Adds the content of text `s` at the end of self
#
- # var b = new Buffer
- # b.append "hello"
- # b.append "world"
- # assert b == "helloworld"
+ # ~~~
+ # var b = new Buffer
+ # b.append "hello"
+ # b.append "world"
+ # assert b == "helloworld"
+ # ~~~
fun append(s: Text) is abstract
# `self` is appended in such a way that `self` is repeated `r` times
#
- # var b = new Buffer
- # b.append "hello"
- # b.times 3
- # assert b == "hellohellohello"
+ # ~~~
+ # var b = new Buffer
+ # b.append "hello"
+ # b.times 3
+ # assert b == "hellohellohello"
+ # ~~~
fun times(r: Int) is abstract
# Reverses itself in-place
#
- # var b = new Buffer
- # b.append("hello")
- # b.reverse
- # assert b == "olleh"
+ # ~~~
+ # var b = new Buffer
+ # b.append("hello")
+ # b.reverse
+ # assert b == "olleh"
+ # ~~~
fun reverse is abstract
# Changes each lower-case char in `self` by its upper-case variant
#
- # var b = new Buffer
- # b.append("Hello World!")
- # b.upper
- # assert b == "HELLO WORLD!"
+ # ~~~
+ # var b = new Buffer
+ # b.append("Hello World!")
+ # b.upper
+ # assert b == "HELLO WORLD!"
+ # ~~~
fun upper is abstract
# Changes each upper-case char in `self` by its lower-case variant
#
- # var b = new Buffer
- # b.append("Hello World!")
- # b.lower
- # assert b == "hello world!"
+ # ~~~
+ # var b = new Buffer
+ # b.append("Hello World!")
+ # b.lower
+ # assert b == "hello world!"
+ # ~~~
fun lower is abstract
# Capitalizes each word in `self`
# Letters that follow a letter are lowercased
# Letters that follow a non-letter are upcased.
#
+ # If `keep_upper = true`, uppercase letters are not lowercased.
+ #
+ # When `src` is specified, this method reads from `src` instead of `self`
+ # but it still writes the result to the beginning of `self`.
+ # This requires `self` to have the capacity to receive all of the
+ # capitalized content of `src`.
+ #
# SEE: `Char::is_letter` for the definition of a letter.
#
- # var b = new FlatBuffer.from("jAVAsCriPt")
- # b.capitalize
- # assert b == "Javascript"
- # b = new FlatBuffer.from("i am root")
- # b.capitalize
- # assert b == "I Am Root"
- # b = new FlatBuffer.from("ab_c -ab0c ab\nc")
- # b.capitalize
- # assert b == "Ab_C -Ab0C Ab\nC"
- fun capitalize do
+ # ~~~
+ # var b = new FlatBuffer.from("jAVAsCriPt")
+ # b.capitalize
+ # assert b == "Javascript"
+ # b = new FlatBuffer.from("i am root")
+ # b.capitalize
+ # assert b == "I Am Root"
+ # b = new FlatBuffer.from("ab_c -ab0c ab\nc")
+ # b.capitalize
+ # assert b == "Ab_C -Ab0C Ab\nC"
+ #
+ # b = new FlatBuffer.from("12345")
+ # b.capitalize(src="foo")
+ # assert b == "Foo45"
+ #
+ # b = new FlatBuffer.from("preserve my ACRONYMS")
+ # b.capitalize(keep_upper=true)
+ # assert b == "Preserve My ACRONYMS"
+ # ~~~
+ fun capitalize(keep_upper: nullable Bool, src: nullable Text) do
+ src = src or else self
+ var length = src.length
if length == 0 then return
- var c = self[0].to_upper
+ keep_upper = keep_upper or else false
+
+ var c = src[0].to_upper
self[0] = c
var prev = c
for i in [1 .. length[ do
prev = c
- c = self[i]
+ c = src[i]
if prev.is_letter then
- self[i] = c.to_lower
+ if keep_upper then
+ self[i] = c
+ else
+ self[i] = c.to_lower
+ end
else
self[i] = c.to_upper
end
end
end
- redef fun hash
- do
- if is_dirty then hash_cache = null
- return super
- end
-
# In Buffers, the internal sequence of character is mutable
# Thus, `chars` can be used to modify the buffer.
redef fun chars: Sequence[Char] is abstract
+
+ # Appends `length` chars from `s` starting at index `from`
+ #
+ # ~~~
+ # var b = new Buffer
+ # b.append_substring("abcde", 1, 2)
+ # assert b == "bc"
+ # b.append_substring("vwxyz", 2, 3)
+ # assert b == "bcxyz"
+ # b.append_substring("ABCDE", 4, 300)
+ # assert b == "bcxyzE"
+ # b.append_substring("VWXYZ", 400, 1)
+ # assert b == "bcxyzE"
+ # ~~~
+ fun append_substring(s: Text, from, length: Int) do
+ if from < 0 then
+ length += from
+ from = 0
+ end
+ var ln = s.length
+ if (length + from) > ln then length = ln - from
+ if length <= 0 then return
+ append_substring_impl(s, from, length)
+ end
+
+ # Unsafe version of `append_substring` for performance
+ #
+ # NOTE: Use only if sure about `from` and `length`, no checks
+ # or bound recalculation is done
+ fun append_substring_impl(s: Text, from, length: Int) do
+ var max = from + length
+ for i in [from .. max[ do add s[i]
+ end
+
+ redef fun *(i) do
+ var ret = new Buffer.with_cap(byte_length * i)
+ for its in [0 .. i[ do ret.append self
+ return ret
+ end
+
+ redef fun insert_at(s, pos) do
+ var obuf = new Buffer.with_cap(byte_length + s.byte_length)
+ obuf.append_substring(self, 0, pos)
+ obuf.append s
+ obuf.append_substring(self, pos, length - pos)
+ return obuf
+ end
+
+ # Inserts `s` at position `pos`
+ #
+ # ~~~
+ # var b = new Buffer
+ # b.append "美しい世界"
+ # b.insert(" nit ", 3)
+ # assert b == "美しい nit 世界"
+ # ~~~
+ fun insert(s: Text, pos: Int) is abstract
+
+ # Inserts `c` at position `pos`
+ #
+ # ~~~
+ # var b = new Buffer
+ # b.append "美しい世界"
+ # b.insert_char(' ', 3)
+ # assert b == "美しい 世界"
+ # ~~~
+ fun insert_char(c: Char, pos: Int) is abstract
+
+ # Removes a substring from `self` at position `pos`
+ #
+ # NOTE: `length` defaults to 1, expressed in chars
+ #
+ # ~~~
+ # var b = new Buffer
+ # b.append("美しい 世界")
+ # b.remove_at(3)
+ # assert b == "美しい世界"
+ # b.remove_at(1, 2)
+ # assert b == "美世界"
+ # ~~~
+ fun remove_at(pos: Int, length: nullable Int) is abstract
+
+ redef fun reversed do
+ var ret = clone
+ ret.reverse
+ return ret
+ end
+
+ redef fun to_upper do
+ var ret = clone
+ ret.upper
+ return ret
+ end
+
+ redef fun to_lower do
+ var ret = clone
+ ret.lower
+ return ret
+ end
+
+ redef fun to_snake_case do
+ var ret = clone
+ ret.snake_case
+ return ret
+ end
+
+ # Takes a camel case `self` and converts it to snake case
+ #
+ # SEE: `to_snake_case`
+ fun snake_case do
+ if self.is_lower then return
+ var prev_is_lower = false
+ var prev_is_upper = false
+
+ var i = 0
+ while i < length do
+ var char = chars[i]
+ if char.is_lower then
+ prev_is_lower = true
+ prev_is_upper = false
+ else if char.is_upper then
+ if prev_is_lower then
+ insert_char('_', i)
+ i += 1
+ else if prev_is_upper and i + 1 < length and self[i + 1].is_lower then
+ insert_char('_', i)
+ i += 1
+ end
+ self[i] = char.to_lower
+ prev_is_lower = false
+ prev_is_upper = true
+ else
+ prev_is_lower = false
+ prev_is_upper = false
+ end
+ i += 1
+ end
+ end
+
+ redef fun to_camel_case
+ do
+ var new_str = clone
+ new_str.camel_case
+ return new_str
+ end
+
+ # Takes a snake case `self` and converts it to camel case
+ #
+ # SEE: `to_camel_case`
+ fun camel_case do
+ if is_upper then return
+
+ var underscore_count = 0
+
+ var pos = 1
+ while pos < length do
+ var char = self[pos]
+ if char == '_' then
+ underscore_count += 1
+ else if underscore_count > 0 then
+ pos -= underscore_count
+ remove_at(pos, underscore_count)
+ self[pos] = char.to_upper
+ underscore_count = 0
+ end
+ pos += 1
+ end
+ if underscore_count > 0 then remove_at(pos - underscore_count - 1, underscore_count)
+ end
+
+ redef fun capitalized(keep_upper) do
+ if length == 0 then return self
+
+ var buf = new Buffer.with_cap(byte_length)
+ buf.capitalize(keep_upper=keep_upper, src=self)
+ return buf
+ end
end
# View for chars on Buffer objects, extends Sequence
# User readable representation of `self`.
fun to_s: String do return inspect
- # The class name of the object in NativeString format.
- private fun native_class_name: NativeString is intern
+ # The class name of the object in CString format.
+ private fun native_class_name: CString is intern
# The class name of the object.
#
- # assert 5.class_name == "Int"
+ # ~~~
+ # assert 5.class_name == "Int"
+ # ~~~
fun class_name: String do return native_class_name.to_s
# Developer readable representation of `self`.
end
redef class Bool
- # assert true.to_s == "true"
- # assert false.to_s == "false"
+ # ~~~
+ # assert true.to_s == "true"
+ # assert false.to_s == "false"
+ # ~~~
redef fun to_s
do
if self then
end
redef class Byte
- # C function to calculate the length of the `NativeString` to receive `self`
+ # C function to calculate the length of the `CString` to receive `self`
private fun byte_to_s_len: Int `{
return snprintf(NULL, 0, "0x%02x", self);
`}
- # C function to convert an nit Int to a NativeString (char*)
- private fun native_byte_to_s(nstr: NativeString, strlen: Int) `{
+ # C function to convert an nit Int to a CString (char*)
+ private fun native_byte_to_s(nstr: CString, strlen: Int) `{
snprintf(nstr, strlen, "0x%02x", self);
`}
# Displayable byte in its hexadecimal form (0x..)
#
- # assert 1.to_b.to_s == "0x01"
- # assert (-123).to_b.to_s == "0x85"
+ # ~~~
+ # assert 1.to_b.to_s == "0x01"
+ # assert (-123).to_b.to_s == "0x85"
+ # ~~~
redef fun to_s do
var nslen = byte_to_s_len
- var ns = new NativeString(nslen + 1)
- ns[nslen] = 0u8
+ var ns = new CString(nslen + 1)
+ ns[nslen] = 0
native_byte_to_s(ns, nslen + 1)
- return ns.to_s_with_length(nslen)
+ return ns.to_s_unsafe(nslen, copy=false, clean=false)
end
end
redef class Int
# Wrapper of strerror C function
- private fun strerror_ext: NativeString `{ return strerror(self); `}
+ private fun strerror_ext: CString `{ return strerror((int)self); `}
# Returns a string describing error number
fun strerror: String do return strerror_ext.to_s
- # Fill `s` with the digits in base `base` of `self` (and with the '-' sign if 'signed' and negative).
+ # Fill `s` with the digits in base `base` of `self` (and with the '-' sign if negative).
# assume < to_c max const of char
- private fun fill_buffer(s: Buffer, base: Int, signed: Bool)
+ private fun fill_buffer(s: Buffer, base: Int)
do
var n: Int
# Sign
end
end
- # C function to calculate the length of the `NativeString` to receive `self`
+ # C function to calculate the length of the `CString` to receive `self`
private fun int_to_s_len: Int `{
return snprintf(NULL, 0, "%ld", self);
`}
- # C function to convert an nit Int to a NativeString (char*)
- private fun native_int_to_s(nstr: NativeString, strlen: Int) `{
+ # C function to convert an nit Int to a CString (char*)
+ private fun native_int_to_s(nstr: CString, strlen: Int) `{
snprintf(nstr, strlen, "%ld", self);
`}
- # return displayable int in base base and signed
- fun to_base(base: Int, signed: Bool): String is abstract
+ # String representation of `self` in the given `base`
+ #
+ # ~~~
+ # assert 15.to_base(10) == "15"
+ # assert 15.to_base(16) == "f"
+ # assert 15.to_base(2) == "1111"
+ # assert (-10).to_base(3) == "-101"
+ # ~~~
+ fun to_base(base: Int): String
+ do
+ var l = digit_count(base)
+ var s = new Buffer
+ s.enlarge(l)
+ for x in [0..l[ do s.add(' ')
+ fill_buffer(s, base)
+ return s.to_s
+ end
+
# return displayable int in hexadecimal
#
- # assert 1.to_hex == "1"
- # assert (-255).to_hex == "-ff"
- fun to_hex: String do return to_base(16,false)
+ # ~~~
+ # assert 1.to_hex == "1"
+ # assert (-255).to_hex == "-ff"
+ # ~~~
+ fun to_hex: String do return to_base(16)
end
redef class Float
# Pretty representation of `self`, with decimals as needed from 1 to a maximum of 3
#
- # assert 12.34.to_s == "12.34"
- # assert (-0120.030).to_s == "-120.03"
+ # ~~~
+ # assert 12.34.to_s == "12.34"
+ # assert (-0120.030).to_s == "-120.03"
+ # ~~~
#
# see `to_precision` for a custom precision.
redef fun to_s do
- var str = to_precision( 3 )
+ var str = to_precision(3)
if is_inf != 0 or is_nan then return str
var len = str.length
for i in [0..len-1] do
# `String` representation of `self` with the given number of `decimals`
#
- # assert 12.345.to_precision(0) == "12"
- # assert 12.345.to_precision(3) == "12.345"
- # assert (-12.345).to_precision(3) == "-12.345"
- # assert (-0.123).to_precision(3) == "-0.123"
- # assert 0.999.to_precision(2) == "1.00"
- # assert 0.999.to_precision(4) == "0.9990"
+ # ~~~
+ # assert 12.345.to_precision(0) == "12"
+ # assert 12.345.to_precision(3) == "12.345"
+ # assert (-12.345).to_precision(3) == "-12.345"
+ # assert (-0.123).to_precision(3) == "-0.123"
+ # assert 0.999.to_precision(2) == "1.00"
+ # assert 0.999.to_precision(4) == "0.9990"
+ # ~~~
fun to_precision(decimals: Int): String
do
if is_nan then return "nan"
return "-inf"
end
- if decimals == 0 then return self.to_i.to_s
- var f = self
- for i in [0..decimals[ do f = f * 10.0
- if self > 0.0 then
- f = f + 0.5
- else
- f = f - 0.5
- end
- var i = f.to_i
- if i == 0 then return "0." + "0"*decimals
-
- # Prepare both parts of the float, before and after the "."
- var s = i.abs.to_s
- var sl = s.length
- var p1
- var p2
- if sl > decimals then
- # Has something before the "."
- p1 = s.substring(0, sl-decimals)
- p2 = s.substring(sl-decimals, decimals)
- else
- p1 = "0"
- p2 = "0"*(decimals-sl) + s
- end
-
- if i < 0 then p1 = "-" + p1
+ var size = to_precision_size(decimals)
+ var cstr = new CString(size + 1)
+ to_precision_fill(decimals, size + 1, cstr)
+ return cstr.to_s_unsafe(byte_length = size, copy = false)
+ end
- return p1 + "." + p2
+ # Returns the hexadecimal (`String`) representation of `self` in exponential notation
+ #
+ # ~~~
+ # assert 12.345.to_hexa_exponential_notation == "0x1.8b0a3d70a3d71p+3"
+ # assert 12.345.to_hexa_exponential_notation.to_f == 12.345
+ # ~~~
+ fun to_hexa_exponential_notation: String
+ do
+ var size = to_precision_size_hexa
+ var cstr = new CString(size + 1)
+ to_precision_fill_hexa(size + 1, cstr)
+ return cstr.to_s_unsafe(byte_length = size, copy = false)
end
+
+ # Required string length to hold `self` with `nb` decimals
+ #
+ # The length does not include the terminating null byte.
+ private fun to_precision_size(nb: Int): Int `{
+ return snprintf(NULL, 0, "%.*f", (int)nb, self);
+ `}
+
+ # Fill `cstr` with `self` and `nb` decimals
+ private fun to_precision_fill(nb, size: Int, cstr: CString) `{
+ snprintf(cstr, size, "%.*f", (int)nb, self);
+ `}
+
+ # The lenght of `self` in exponential hexadecimal notation
+ private fun to_precision_size_hexa: Int`{
+ return snprintf(NULL, 0, "%a", self);
+ `}
+
+ # Fill `cstr` with `self` in exponential hexadecimal notation
+ private fun to_precision_fill_hexa(size: Int, cstr: CString) `{
+ snprintf(cstr, size, "%a", self);
+ `}
end
redef class Char
+ # Returns a sequence with the UTF-8 bytes of `self`
+ #
+ # ~~~
+ # assert 'a'.bytes == [0x61]
+ # assert 'ま'.bytes == [0xE3, 0x81, 0xBE]
+ # ~~~
+ fun bytes: SequenceRead[Int] do return to_s.bytes
+
+ # Is `self` an UTF-16 surrogate pair ?
+ fun is_surrogate: Bool do
+ var cp = code_point
+ return cp >= 0xD800 and cp <= 0xDFFF
+ end
+
+ # Is `self` a UTF-16 high surrogate ?
+ fun is_hi_surrogate: Bool do
+ var cp = code_point
+ return cp >= 0xD800 and cp <= 0xDBFF
+ end
+
+ # Is `self` a UTF-16 low surrogate ?
+ fun is_lo_surrogate: Bool do
+ var cp = code_point
+ return cp >= 0xDC00 and cp <= 0xDFFF
+ end
+
# Length of `self` in a UTF-8 String
- private fun u8char_len: Int do
- var c = self.ascii
+ fun u8char_len: Int do
+ var c = self.code_point
if c < 0x80 then return 1
if c <= 0x7FF then return 2
if c <= 0xFFFF then return 3
return 1
end
- # assert 'x'.to_s == "x"
+ # ~~~
+ # assert 'x'.to_s == "x"
+ # ~~~
redef fun to_s do
var ln = u8char_len
- var ns = new NativeString(ln + 1)
+ var ns = new CString(ln + 1)
u8char_tos(ns, ln)
- return ns.to_s_with_length(ln)
+ return ns.to_s_unsafe(ln, copy=false, clean=false)
+ end
+
+ # Returns `self` escaped to UTF-16
+ #
+ # i.e. Represents `self`.`code_point` using UTF-16 codets escaped
+ # with a `\u`
+ #
+ # ~~~
+ # assert 'A'.escape_to_utf16 == "\\u0041"
+ # assert 'è'.escape_to_utf16 == "\\u00e8"
+ # assert 'あ'.escape_to_utf16 == "\\u3042"
+ # assert '𐏓'.escape_to_utf16 == "\\ud800\\udfd3"
+ # ~~~
+ fun escape_to_utf16: String do
+ var cp = code_point
+ var buf: Buffer
+ if cp < 0xD800 or (cp >= 0xE000 and cp <= 0xFFFF) then
+ buf = new Buffer.with_cap(6)
+ buf.append("\\u0000")
+ var hx = cp.to_hex
+ var outid = 5
+ for i in hx.chars.reverse_iterator do
+ buf[outid] = i
+ outid -= 1
+ end
+ else
+ buf = new Buffer.with_cap(12)
+ buf.append("\\u0000\\u0000")
+ var lo = (((cp - 0x10000) & 0x3FF) + 0xDC00).to_hex
+ var hi = ((((cp - 0x10000) & 0xFFC00) >> 10) + 0xD800).to_hex
+ var out = 2
+ for i in hi do
+ buf[out] = i
+ out += 1
+ end
+ out = 8
+ for i in lo do
+ buf[out] = i
+ out += 1
+ end
+ end
+ return buf.to_s
end
- private fun u8char_tos(r: NativeString, len: Int) `{
+ private fun u8char_tos(r: CString, len: Int) `{
r[len] = '\0';
switch(len){
case 1:
# Returns true if the char is a numerical digit
#
- # assert '0'.is_numeric
- # assert '9'.is_numeric
- # assert not 'a'.is_numeric
- # assert not '?'.is_numeric
+ # ~~~
+ # assert '0'.is_numeric
+ # assert '9'.is_numeric
+ # assert not 'a'.is_numeric
+ # assert not '?'.is_numeric
+ # ~~~
#
# FIXME: Works on ASCII-range only
fun is_numeric: Bool
# Returns true if the char is an alpha digit
#
- # assert 'a'.is_alpha
- # assert 'Z'.is_alpha
- # assert not '0'.is_alpha
- # assert not '?'.is_alpha
+ # ~~~
+ # assert 'a'.is_alpha
+ # assert 'Z'.is_alpha
+ # assert not '0'.is_alpha
+ # assert not '?'.is_alpha
+ # ~~~
#
# FIXME: Works on ASCII-range only
fun is_alpha: Bool
return (self >= 'a' and self <= 'z') or (self >= 'A' and self <= 'Z')
end
+ # Is `self` an hexadecimal digit ?
+ #
+ # ~~~
+ # assert 'A'.is_hexdigit
+ # assert not 'G'.is_hexdigit
+ # assert 'a'.is_hexdigit
+ # assert not 'g'.is_hexdigit
+ # assert '5'.is_hexdigit
+ # ~~~
+ fun is_hexdigit: Bool do return (self >= '0' and self <= '9') or (self >= 'A' and self <= 'F') or
+ (self >= 'a' and self <= 'f')
+
# Returns true if the char is an alpha or a numeric digit
#
- # assert 'a'.is_alphanumeric
- # assert 'Z'.is_alphanumeric
- # assert '0'.is_alphanumeric
- # assert '9'.is_alphanumeric
- # assert not '?'.is_alphanumeric
+ # ~~~
+ # assert 'a'.is_alphanumeric
+ # assert 'Z'.is_alphanumeric
+ # assert '0'.is_alphanumeric
+ # assert '9'.is_alphanumeric
+ # assert not '?'.is_alphanumeric
+ # ~~~
#
# FIXME: Works on ASCII-range only
fun is_alphanumeric: Bool
do
return self.is_numeric or self.is_alpha
end
+
+ # Returns `self` to its int value
+ #
+ # REQUIRE: `is_hexdigit`
+ fun from_hex: Int do
+ if self >= '0' and self <= '9' then return code_point - 0x30
+ if self >= 'A' and self <= 'F' then return code_point - 0x37
+ if self >= 'a' and self <= 'f' then return code_point - 0x57
+ # Happens if self is not a hexdigit
+ assert self.is_hexdigit
+ # To make flow analysis happy
+ abort
+ end
end
redef class Collection[E]
#
# Only concatenate if `separator == null`.
#
- # assert [1, 2, 3].join(":") == "1:2:3"
- # assert [1..3].join(":") == "1:2:3"
- # assert [1..3].join == "123"
- fun join(separator: nullable Text): String
+ # ~~~
+ # assert [1, 2, 3].join(":") == "1:2:3"
+ # assert [1..3].join(":") == "1:2:3"
+ # assert [1..3].join == "123"
+ # ~~~
+ #
+ # if `last_separator` is given, then it is used to separate the last element.
+ #
+ # ~~~
+ # assert [1, 2, 3, 4].join(", ", " and ") == "1, 2, 3 and 4"
+ # ~~~
+ fun join(separator: nullable Text, last_separator: nullable Text): String
do
if is_empty then return ""
var e = i.item
if e != null then s.append(e.to_s)
+ if last_separator == null then last_separator = separator
+
# Concat other items
i.next
while i.is_ok do
- if separator != null then s.append(separator)
e = i.item
- if e != null then s.append(e.to_s)
i.next
+ if i.is_ok then
+ if separator != null then s.append(separator)
+ else
+ if last_separator != null then s.append(last_separator)
+ end
+ if e != null then s.append(e.to_s)
end
return s.to_s
end
# Key and value are separated by `couple_sep`.
# Couples are separated by `sep`.
#
- # var m = new HashMap[Int, String]
- # m[1] = "one"
- # m[10] = "ten"
- # assert m.join("; ", "=") == "1=one; 10=ten"
+ # ~~~
+ # var m = new HashMap[Int, String]
+ # m[1] = "one"
+ # m[10] = "ten"
+ # assert m.join("; ", "=") == "1=one; 10=ten"
+ # ~~~
fun join(sep, couple_sep: String): String is abstract
end
private fun native_argc: Int is intern
# Second argument of the main C function.
- private fun native_argv(i: Int): NativeString is intern
+ private fun native_argv(i: Int): CString is intern
end
# Comparator that efficienlty use `to_s` to compare things
#
# The comparaison call `to_s` on object and use the result to order things.
#
-# var a = [1, 2, 3, 10, 20]
-# (new CachedAlphaComparator).sort(a)
-# assert a == [1, 10, 2, 20, 3]
+# ~~~
+# var a = [1, 2, 3, 10, 20]
+# (new CachedAlphaComparator).sort(a)
+# assert a == [1, 10, 2, 20, 3]
+# ~~~
#
# Internally the result of `to_s` is cached in a HashMap to counter
# uneficient implementation of `to_s`.
# see `alpha_comparator`
private class AlphaComparator
super Comparator
- redef fun compare(a, b) do return a.to_s <=> b.to_s
+ redef fun compare(a, b) do
+ if a == b then return 0
+ if a == null then return -1
+ if b == null then return 1
+ return a.to_s <=> b.to_s
+ end
end
# Stateless comparator that naively use `to_s` to compare things.
# Note: the result of `to_s` is not cached, thus can be invoked a lot
# on a single instace. See `CachedAlphaComparator` as an alternative.
#
-# var a = [1, 2, 3, 10, 20]
-# alpha_comparator.sort(a)
-# assert a == [1, 10, 2, 20, 3]
+# ~~~
+# var a = [1, 2, 3, 10, 20]
+# alpha_comparator.sort(a)
+# assert a == [1, 10, 2, 20, 3]
+# ~~~
fun alpha_comparator: Comparator do return once new AlphaComparator
# The arguments of the program as given by the OS
return sys.program_args
end
-redef class NativeString
- # Returns `self` as a new String.
- fun to_s_with_copy: String is abstract
+redef class CString
- # Returns `self` as a String of `length`.
- fun to_s_with_length(length: Int): String is abstract
+ # Get a `String` from the data at `self` (with unsafe options)
+ #
+ # The default behavior is the safest and equivalent to `to_s`.
+ #
+ # Options:
+ #
+ # * Set `byte_length` to the number of bytes to use as data.
+ # Otherwise, this method searches for a terminating null byte.
+ #
+ # * Set `char_length` to the number of Unicode character in the string.
+ # Otherwise, the data is read to count the characters.
+ # Ignored if `clean == true`.
+ #
+ # * If `copy == true`, the default, copies the data at `self` in the
+ # Nit GC allocated memory. Otherwise, the return may still point to
+ # the data at `self`.
+ #
+ # * If `clean == true`, the default, the string is cleaned of invalid UTF-8
+ # characters. If cleaning is necessary, the data is copied into Nit GC
+ # managed memory, whether or not `copy == true`.
+ # Don't clean only when the data has already been verified as valid UTF-8,
+ # other library services rely on UTF-8 compliant characters.
+ fun to_s_unsafe(byte_length, char_length: nullable Int, copy, clean: nullable Bool): String is abstract
+
+ # Retro-compatibility service use by execution engines
+ #
+ # TODO remove this method at the next c_src regen.
+ private fun to_s_full(byte_length, char_length: Int): String do return to_s_unsafe(byte_length, char_length, false, false)
- # Returns `self` as a String with `bytelen` and `length` set
+ # Copies the content of `src` to `self`
#
- # SEE: `abstract_text::Text` for more infos on the difference
- # between `Text::bytelen` and `Text::length`
- fun to_s_full(bytelen, unilen: Int): String is abstract
+ # NOTE: `self` must be large enough to contain `self.byte_length` bytes
+ fun fill_from(src: Text) do src.copy_to_native(self, src.byte_length, 0, 0)
end
redef class NativeArray[E]
nitlanguage / nit.git / blobdiff