# 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.
#
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)
# 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.
redef type SELFTYPE: Buffer is fixed
- # Specific implementations MUST set this to `true` in order to invalidate caches
- protected var is_dirty = true
-
# Copy-On-Write flag
#
# If the `Buffer` was to_s'd, the next in-place altering
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
# 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.
# 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.byte_length` bytes
+ fun fill_from(src: Text) do src.copy_to_native(self, src.byte_length, 0, 0)
end
redef class NativeArray[E]