import collection
intrude import collection::array
+in "C" `{
+ #include <stdio.h>
+ #include <stdlib.h>
+ #include <string.h>
+`}
+
# High-level abstraction for all text representations
abstract class Text
super Comparable
# 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
+
# Number of characters contained in self.
#
# 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
+
# 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"
#
# A `from` index < 0 will be replaced by 0.
# Unless a `count` value is > 0 at the same time.
var mypos = src_offset
var itspos = dest_offset
while n > 0 do
- dest[itspos] = self.chars[mypos]
+ dest[itspos] = self.bytes[mypos]
itspos += 1
mypos += 1
n -= 1
# Real items, used as cache for to_cstring is called
private var real_items: nullable NativeString = null
- # Returns a char* starting at position `index_from`
+ # Returns a char* starting at position `first_byte`
#
# WARNING: If you choose to use this service, be careful of the following.
#
redef var length = 0
+ redef var bytelen = 0
+
redef fun output
do
var i = 0
end
# Abstract class for the SequenceRead compatible
-# views on String and Buffer objects
+# views on the chars of any Text
private abstract class StringCharView
super SequenceRead[Char]
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.
# 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
+
+ # In Buffers, the internal sequence of bytes is mutable
+ # Thus, `bytes` can be used to modify the buffer.
+ redef fun bytes: Sequence[Byte] is abstract
end
-# View on Buffer objects, extends Sequence
+# View for chars on Buffer objects, extends Sequence
# for mutation operations
private abstract class BufferCharView
super StringCharView
end
+# View for bytes on Buffer objects, extends Sequence
+# for mutation operations
+private abstract class BufferByteView
+ super StringByteView
+ super Sequence[Byte]
+
+ redef type SELFTYPE: Buffer
+end
+
redef class Object
# User readable representation of `self`.
fun to_s: String do return inspect
redef class Byte
# C function to calculate the length of the `NativeString` to receive `self`
- private fun byte_to_s_len: Int is extern "native_byte_length_str"
+ 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) is extern "native_byte_to_s"
+ private fun native_byte_to_s(nstr: NativeString, strlen: Int) `{
+ snprintf(nstr, strlen, "0x%02x", self);
+ `}
# Displayable byte in its hexadecimal form (0x..)
#
redef fun to_s do
var nslen = byte_to_s_len
var ns = new NativeString(nslen + 1)
- ns[nslen] = '\0'
+ ns[nslen] = 0u8
native_byte_to_s(ns, nslen + 1)
return ns.to_s_with_length(nslen)
end
redef class Int
# Wrapper of strerror C function
- private fun strerror_ext: NativeString is extern "strerror"
+ private fun strerror_ext: NativeString `{ return strerror(self); `}
# Returns a string describing error number
fun strerror: String do return strerror_ext.to_s
end
# C function to calculate the length of the `NativeString` to receive `self`
- private fun int_to_s_len: Int is extern "native_int_length_str"
+ 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) is extern "native_int_to_s"
+ private fun native_int_to_s(nstr: NativeString, 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
end
redef class Char
+
+ # Length of `self` in a UTF-8 String
+ private fun u8char_len: Int do
+ var c = self.ascii
+ if c < 0x80 then return 1
+ if c <= 0x7FF then return 2
+ if c <= 0xFFFF then return 3
+ if c <= 0x10FFFF then return 4
+ # Bad character format
+ return 1
+ end
+
# assert 'x'.to_s == "x"
- redef fun to_s
- do
- var s = new Buffer.with_cap(1)
- s.chars[0] = self
- return s.to_s
+ redef fun to_s do
+ var ln = u8char_len
+ var ns = new NativeString(ln + 1)
+ u8char_tos(ns, ln)
+ return ns.to_s_with_length(ln)
end
+ private fun u8char_tos(r: NativeString, len: Int) `{
+ r[len] = '\0';
+ switch(len){
+ case 1:
+ r[0] = self;
+ break;
+ case 2:
+ r[0] = 0xC0 | ((self & 0x7C0) >> 6);
+ r[1] = 0x80 | (self & 0x3F);
+ break;
+ case 3:
+ r[0] = 0xE0 | ((self & 0xF000) >> 12);
+ r[1] = 0x80 | ((self & 0xFC0) >> 6);
+ r[2] = 0x80 | (self & 0x3F);
+ break;
+ case 4:
+ r[0] = 0xF0 | ((self & 0x1C0000) >> 18);
+ r[1] = 0x80 | ((self & 0x3F000) >> 12);
+ r[2] = 0x80 | ((self & 0xFC0) >> 6);
+ r[3] = 0x80 | (self & 0x3F);
+ break;
+ }
+ `}
+
# 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
+ #
+ # FIXME: Works on ASCII-range only
fun is_numeric: Bool
do
return self >= '0' and self <= '9'
# assert 'Z'.is_alpha
# assert not '0'.is_alpha
# assert not '?'.is_alpha
+ #
+ # FIXME: Works on ASCII-range only
fun is_alpha: Bool
do
return (self >= 'a' and self <= 'z') or (self >= 'A' and self <= 'Z')
# 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
return s.to_s
end
- # Concatenate and separate each elements with `sep`.
+ # Concatenate and separate each elements with `separator`.
+ #
+ # Only concatenate if `separator == null`.
#
- # assert [1, 2, 3].join(":") == "1:2:3"
- # assert [1..3].join(":") == "1:2:3"
- fun join(sep: Text): String
+ # 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
do
if is_empty then return ""
# Concat other items
i.next
while i.is_ok do
- s.append(sep)
+ if separator != null then s.append(separator)
e = i.item
if e != null then s.append(e.to_s)
i.next
end
redef class Map[K,V]
- # Concatenate couple of 'key value'.
- # key and value are separated by `couple_sep`.
- # each couple is separated each couple with `sep`.
+ # Concatenate couples of key value.
+ # Key and value are separated by `couple_sep`.
+ # Couples are separated by `sep`.
#
- # var m = new ArrayMap[Int, String]
+ # var m = new HashMap[Int, String]
# m[1] = "one"
# m[10] = "ten"
# assert m.join("; ", "=") == "1=one; 10=ten"
- fun join(sep: String, couple_sep: String): String is abstract
+ fun join(sep, couple_sep: String): String is abstract
end
redef class Sys
nitlanguage / nit.git / blobdiff