# 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.
#
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)
# 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
# assert "%c3%a9%e3%81%82%e3%81%84%e3%81%86".from_percent_encoding == "éあいう"
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
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.
# 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 "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)
-
- 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
-
+ buf.capitalize(keep_upper=keep_upper, src=self)
return buf.to_s
end
end
# 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
# use only when the data has already been verified as valid UTF-8.
fun to_s_unsafe(length: nullable Int): String is abstract
- # Get a `String` from the raw `bytelen` bytes at `self` with `unilen` Unicode characters
+ # Get a `String` from the raw `byte_length` bytes at `self` with `unilen` Unicode characters
#
# The created `String` points to the data at `self`.
# This method should be used when `self` was allocated by the Nit GC,
# use only when the data has already been verified as valid UTF-8.
#
# SEE: `abstract_text::Text` for more info on the difference
- # between `Text::bytelen` and `Text::length`.
- fun to_s_full(bytelen, unilen: Int): String is abstract
+ # between `Text::byte_length` and `Text::length`.
+ fun to_s_full(byte_length, unilen: Int): String is abstract
# Copies the content of `src` to `self`
#
- # NOTE: `self` must be large enough to withold `self.bytelen` bytes
- fun fill_from(src: Text) do src.copy_to_native(self, src.bytelen, 0, 0)
+ # NOTE: `self` must be large enough to withold `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