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 (== "")
#
# Returns -1 if not found
#
# 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
+ fun last_index_of_from(item: Char, pos: Int): Int do return chars.last_index_of_from(item, pos)
# Gets an iterator on the chars of self
#
return res.to_s
end
+ # Returns `self` with all characters escaped with their UTF-16 representation
+ #
+ # assert "Aรจใ๐".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 == '๐'
+ # assert "\\u00e8".from_utf16_escape == 'รจ'
+ # assert "\\u3042".from_utf16_escape == 'ใ'
+ fun from_utf16_escape: Char do
+ var ln = length
+ if ln != 6 and ln != 12 then return 0xFFFD.code_point
+ var cphi = substring(2, 4).to_hex
+ if cphi < 0xD800 then return cphi.code_point
+ if cphi > 0xDFFF then return cphi.code_point
+ if cphi > 0xDBFF then return 0xFFFD.code_point
+ var cp = 0
+ cp += (cphi - 0xD800) << 10
+ var cplo = substring(8, 4).to_hex
+ if cplo < 0xDC00 then return 0xFFFD.code_point
+ if cplo > 0xDFFF then return 0xFFFD.code_point
+ cp += cplo - 0xDC00
+ cp += 0x10000
+ return cp.code_point
+ end
+
# Encode `self` to percent (or URL) encoding
#
# assert "aBc09-._~".to_percent_encoding == "aBc09-._~"
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
#
# 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: NativeString is noinit
# Returns a char* starting at position `first_byte`
#
#
# 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: NativeString is abstract
redef var length = 0
return ns.to_s_with_length(ln)
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 'รจ'.escape_to_utf16 == "\\u00e8"
+ # assert 'ใ'.escape_to_utf16 == "\\u3042"
+ # assert '๐'.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) `{
r[len] = '\0';
switch(len){
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