# High-level abstraction for all text representations
abstract class Text
super Comparable
+ super Cloneable
redef type OTHER: Text
# 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.
#
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 (== "")
#
# 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
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).
#
# DEPRECATED : Use self.chars.last_index_of_from instead
fun last_index_of_from(item: Char, pos: Int): Int do return chars.last_index_of_from(item, pos)
+ # Concatenates `o` to `self`
+ #
+ # 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
# 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)
+ 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.
#
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 == 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
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, 0.5, '.') == "..hello..."
#
# If `length` is not enough, `self` is returned as is.
#
# 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.
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('_')
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
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 "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
b.append("\\\'")
else if c == '\\' then
b.append("\\\\")
+ 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.code_point.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
# The result might no be legal in C but be used in other languages
#
# 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
b.add('\\')
b.add(c)
else if c.code_point < 32 or c == ';' or c == '|' or c == '\\' or c == '=' then
- b.append("?{c.code_point.to_base(16, false)}")
+ b.append("?{c.code_point.to_base(16)}")
else
b.add(c)
end
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-._~"
# 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 "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 len = bytelen
+ var len = byte_length
var has_percent = false
for c in chars do
if c == '%' then
# If no transformation is needed, return self as a string
if not has_percent then return to_s
- var buf = new NativeString(len)
+ var buf = new CString(len)
var i = 0
var l = 0
while i < length do
if c == '%' then
if i + 2 >= length then
# What follows % has been cut off
- buf[l] = '?'.ascii
+ buf[l] = '%'.ascii
else
i += 1
var hex_s = substring(i, 2)
i += 1
else
# What follows a % is not Hex
- buf[l] = '?'.ascii
+ buf[l] = '%'.ascii
i -= 1
end
end
l += 1
end
- return buf.to_s_with_length(l)
+ return buf.to_s_unsafe(l, copy=false)
end
# Escape the characters `<`, `>`, `&`, `"`, `'` and `/` as HTML/XML entity references.
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`
+ #
+ # 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 %1 is a %2.".format("String", "formatted String") == "This String is a formatted String."
- # assert "\\%1 This string".format("String") == "\\%1 This string"
+ # 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)
+ # 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: NativeString, n, src_offset, dest_offset: Int) do
+ 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
+ # Packs the content of a string in packs of `ln` chars.
+ # This variant ensures that only the last element might be smaller than `ln`
+ #
+ # ~~~nit
+ # 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
+
+ # Packs the content of a string in packs of `ln` chars.
+ # This variant ensures that only the first element might be smaller than `ln`
+ #
+ # ~~~nit
+ # 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
+
+ # Concatenates self `i` times
+ #
+ #~~~nit
+ # 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
+
+ # Insert `s` at `pos`.
+ #
+ #~~~nit
+ # 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
+
+ # 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 lower case version of `self`
+ #
+ # 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"
+ #
+ # The rules are the following:
+ #
+ # An uppercase is always converted to a lowercase
+ #
+ # 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"
+ #
+ # 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"
+ #
+ # 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 is abstract
+
+ # Takes a snake case `self` and converts it to camel case
+ #
+ # 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"
+ #
+ # If `self` is upper, it is returned unchanged
+ #
+ # 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 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"
+ # 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
# All kinds of array-based text representations.
abstract class FlatText
super Text
- # Underlying C-String (`char*`)
+ # Underlying CString (`char*`)
#
- # Warning : Might be void in some subclasses, be sure to check
+ # 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 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 NativeString are *ideally* always allocated through a Garbage Collector.
+ # 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.
#
# 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
+ fun fast_cstring: CString is abstract
redef var length = 0
- redef var bytelen = 0
+ redef var byte_length = 0
redef fun output
do
redef fun is_empty do return target.is_empty
- redef fun length do return target.bytelen
+ 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.bytelen - 1)
+ redef fun reverse_iterator do return self.reverse_iterator_from(target.byte_length - 1)
end
# Immutable sequence of characters.
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 lower case version of `self`
- #
- # 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"
- #
- # The rules are the following:
- #
- # An uppercase is always converted to a lowercase
- #
- # 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"
- #
- # 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"
- #
- # 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
+ redef fun clone do return self
- 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
+ redef fun to_buffer do return new Buffer.from_text(self)
- # Takes a snake case `self` and converts it to camel case
- #
- # 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"
- #
- # If `self` is upper, it is returned unchanged
- #
- # 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
+ redef fun to_camel_case 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
-
+ var new_str = new Buffer.with_cap(length)
+ new_str.append self
+ new_str.camel_case
return new_str.to_s
end
- # Returns a capitalized `self`
- #
- # Letters that follow a letter are lowercased
- # Letters that follow a non-letter are upcased.
- #
- # 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
- if length == 0 then return self
-
- var buf = new Buffer.with_cap(length)
-
- var curr = chars[0].to_upper
- var prev = curr
- buf[0] = curr
-
- 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
- end
+ redef fun to_snake_case do
+ if self.is_lower then return self
- return buf.to_s
+ 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
#
# DEPRECATED : Use self.chars.[]= instead
fun []=(index: Int, item: Char) is abstract
+ redef fun to_buffer do return clone
+
+ #~~~nit
+ # 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
# 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 = new FlatBuffer.from("ab_c -ab0c ab\nc")
# b.capitalize
# assert b == "Ab_C -Ab0C Ab\nC"
- fun capitalize do
+ #
+ # 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`
+ #
+ # ~~~nit
+ # 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`
+ #
+ #~~~nit
+ # 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`
+ #
+ #~~~nit
+ # 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
+ #
+ #~~~nit
+ # 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.
#
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);
`}
# assert (-123).to_b.to_s == "0x85"
redef fun to_s do
var nslen = byte_to_s_len
- var ns = new NativeString(nslen + 1)
+ var ns = new CString(nslen + 1)
ns[nslen] = 0u8
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)
+ fun to_hex: String do return to_base(16)
end
redef class Float
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
+ 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
- if i < 0 then p1 = "-" + p1
+ # 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);
+ `}
- return p1 + "." + p2
- end
+ # Fill `cstr` with `self` and `nb` decimals
+ private fun to_precision_fill(nb, size: Int, cstr: CString) `{
+ snprintf(cstr, size, "%.*f", (int)nb, self);
+ `}
end
redef class Char
# assert 'ま'.bytes == [0xE3u8, 0x81u8, 0xBEu8]
fun bytes: SequenceRead[Byte] 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
+ fun u8char_len: Int do
var c = self.code_point
if c < 0x80 then return 1
if c <= 0x7FF then return 2
# 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:
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
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]
# 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
+ #
+ # 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
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
# 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.
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
+
+ # 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
- # Returns `self` as a String of `length`.
- fun to_s_with_length(length: Int): 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]