1 # This file is part of NIT ( http://www.nitlanguage.org ).
3 # This file is free software, which comes along with NIT. This software is
4 # distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY;
5 # without even the implied warranty of MERCHANTABILITY or FITNESS FOR A
6 # PARTICULAR PURPOSE. You can modify it is you want, provided this header
7 # is kept unaltered, and a notification of the changes is added.
8 # You are allowed to redistribute it and sell it, alone or is a part of
11 # Abstract class for manipulation of sequences of characters
17 intrude import collection
::array
25 # High-level abstraction for all text representations
30 redef type OTHER: Text
32 # Type of self (used for factorization of several methods, ex : substring_from, empty...)
35 # Gets a view on the chars of the Text object
37 # assert "hello".chars.to_a == ['h', 'e', 'l', 'l', 'o']
38 fun chars
: SequenceRead[Char] is abstract
40 # Gets a view on the bytes of the Text object
42 # assert "hello".bytes.to_a == [104u8, 101u8, 108u8, 108u8, 111u8]
43 fun bytes
: SequenceRead[Byte] is abstract
45 # Number of characters contained in self.
47 # assert "12345".length == 5
48 # assert "".length == 0
49 # assert "あいうえお".length == 5
50 fun length
: Int is abstract
52 # Number of bytes in `self`
54 # assert "12345".byte_length == 5
55 # assert "あいうえお".byte_length == 15
56 fun byte_length
: Int is abstract
60 # assert "abcd".substring(1, 2) == "bc"
61 # assert "abcd".substring(-1, 2) == "a"
62 # assert "abcd".substring(1, 0) == ""
63 # assert "abcd".substring(2, 5) == "cd"
64 # assert "あいうえお".substring(1,3) == "いうえ"
66 # A `from` index < 0 will be replaced by 0.
67 # Unless a `count` value is > 0 at the same time.
68 # In this case, `from += count` and `count -= from`.
69 fun substring
(from
: Int, count
: Int): SELFTYPE is abstract
71 # Iterates on the substrings of self if any
72 private fun substrings
: Iterator[FlatText] is abstract
74 # Is the current Text empty (== "")
77 # assert not "foo".is_empty
78 fun is_empty
: Bool do return self.length
== 0
80 # Returns an empty Text of the right type
82 # This method is used internally to get the right
83 # implementation of an empty string.
84 protected fun empty
: SELFTYPE is abstract
86 # Gets the first char of the Text
88 # DEPRECATED : Use self.chars.first instead
89 fun first
: Char do return self.chars
[0]
91 # Access a character at `index` in the string.
93 # assert "abcd"[2] == 'c'
95 # DEPRECATED : Use self.chars.[] instead
96 fun [](index
: Int): Char do return self.chars
[index
]
98 # Gets the index of the first occurence of 'c'
100 # Returns -1 if not found
102 # DEPRECATED : Use self.chars.index_of instead
103 fun index_of
(c
: Char): Int
105 return index_of_from
(c
, 0)
108 # Gets the last char of self
110 # DEPRECATED : Use self.chars.last instead
111 fun last
: Char do return self.chars
[length-1
]
113 # Gets the index of the first occurence of ´c´ starting from ´pos´
115 # Returns -1 if not found
117 # DEPRECATED : Use self.chars.index_of_from instead
118 fun index_of_from
(c
: Char, pos
: Int): Int
120 var iter
= self.chars
.iterator_from
(pos
)
122 if iter
.item
== c
then return iter
.index
128 # Gets the last index of char ´c´
130 # Returns -1 if not found
132 # DEPRECATED : Use self.chars.last_index_of instead
133 fun last_index_of
(c
: Char): Int
135 return last_index_of_from
(c
, length
- 1)
138 # Return a null terminated char *
139 fun to_cstring
: NativeString is abstract
141 # The index of the last occurrence of an element starting from pos (in reverse order).
143 # var s = "/etc/bin/test/test.nit"
144 # assert s.last_index_of_from('/', s.length-1) == 13
145 # assert s.last_index_of_from('/', 12) == 8
147 # Returns -1 if not found
149 # DEPRECATED : Use self.chars.last_index_of_from instead
150 fun last_index_of_from
(item
: Char, pos
: Int): Int do return chars
.last_index_of_from
(item
, pos
)
152 # Gets an iterator on the chars of self
154 # DEPRECATED : Use self.chars.iterator instead
155 fun iterator
: Iterator[Char]
157 return self.chars
.iterator
161 # Gets an Array containing the chars of self
163 # DEPRECATED : Use self.chars.to_a instead
164 fun to_a
: Array[Char] do return chars
.to_a
166 # Create a substring from `self` beginning at the `from` position
168 # assert "abcd".substring_from(1) == "bcd"
169 # assert "abcd".substring_from(-1) == "abcd"
170 # assert "abcd".substring_from(2) == "cd"
172 # As with substring, a `from` index < 0 will be replaced by 0
173 fun substring_from
(from
: Int): SELFTYPE
175 if from
>= self.length
then return empty
176 if from
< 0 then from
= 0
177 return substring
(from
, length
- from
)
180 # Does self have a substring `str` starting from position `pos`?
182 # assert "abcd".has_substring("bc",1) == true
183 # assert "abcd".has_substring("bc",2) == false
185 # Returns true iff all characters of `str` are presents
186 # at the expected index in `self.`
187 # The first character of `str` being at `pos`, the second
188 # character being at `pos+1` and so on...
190 # This means that all characters of `str` need to be inside `self`.
192 # assert "abcd".has_substring("xab", -1) == false
193 # assert "abcd".has_substring("cdx", 2) == false
195 # And that the empty string is always a valid substring.
197 # assert "abcd".has_substring("", 2) == true
198 # assert "abcd".has_substring("", 200) == true
199 fun has_substring
(str
: String, pos
: Int): Bool
201 if str
.is_empty
then return true
202 if pos
< 0 or pos
+ str
.length
> length
then return false
203 var myiter
= self.chars
.iterator_from
(pos
)
204 var itsiter
= str
.chars
.iterator
205 while myiter
.is_ok
and itsiter
.is_ok
do
206 if myiter
.item
!= itsiter
.item
then return false
210 if itsiter
.is_ok
then return false
214 # Is this string prefixed by `prefix`?
216 # assert "abcd".has_prefix("ab") == true
217 # assert "abcbc".has_prefix("bc") == false
218 # assert "ab".has_prefix("abcd") == false
219 fun has_prefix
(prefix
: String): Bool do return has_substring
(prefix
,0)
221 # Is this string suffixed by `suffix`?
223 # assert "abcd".has_suffix("abc") == false
224 # assert "abcd".has_suffix("bcd") == true
225 fun has_suffix
(suffix
: String): Bool do return has_substring
(suffix
, length
- suffix
.length
)
227 # Returns `self` as the corresponding integer
229 # assert "123".to_i == 123
230 # assert "-1".to_i == -1
231 # assert "0x64".to_i == 100
232 # assert "0b1100_0011".to_i== 195
233 # assert "--12".to_i == 12
235 # REQUIRE: `self`.`is_int`
236 fun to_i
: Int is abstract
238 # If `self` contains a float, return the corresponding float
240 # assert "123".to_f == 123.0
241 # assert "-1".to_f == -1.0
242 # assert "-1.2e-3".to_f == -0.0012
246 return to_s
.to_cstring
.atof
249 # If `self` contains only digits and alpha <= 'f', return the corresponding integer.
251 # assert "ff".to_hex == 255
252 fun to_hex
(pos
, ln
: nullable Int): Int do
254 if pos
== null then pos
= 0
255 if ln
== null then ln
= length
- pos
257 for i
in [pos
.. max
[ do
259 res
+= self[i
].from_hex
264 # If `self` contains only digits <= '7', return the corresponding integer.
266 # assert "714".to_oct == 460
267 fun to_oct
: Int do return a_to
(8)
269 # If `self` contains only '0' et '1', return the corresponding integer.
271 # assert "101101".to_bin == 45
272 fun to_bin
: Int do return a_to
(2)
274 # If `self` contains only digits '0' .. '9', return the corresponding integer.
276 # assert "108".to_dec == 108
277 fun to_dec
: Int do return a_to
(10)
279 # If `self` contains only digits and letters, return the corresponding integer in a given base
281 # assert "120".a_to(3) == 15
282 fun a_to
(base
: Int) : Int
287 for j
in [0..length
[ do
309 # Is this string in a valid numeric format compatible with `to_f`?
311 # assert "123".is_numeric == true
312 # assert "1.2".is_numeric == true
313 # assert "-1.2".is_numeric == true
314 # assert "-1.23e-2".is_numeric == true
315 # assert "1..2".is_numeric == false
316 # assert "".is_numeric == false
319 var has_point
= false
321 for i
in [0..length
[ do
323 if not c
.is_numeric
then
324 if c
== '.' and not has_point
then
326 else if c
== 'e' and e_index
== -1 and i
> 0 and i
< length
- 1 and chars
[i-1
] != '-' then
328 else if c
== '-' and i
== e_index
+ 1 and i
< length
- 1 then
337 # Returns `true` if the string contains only Hex chars
339 # assert "048bf".is_hex == true
340 # assert "ABCDEF".is_hex == true
341 # assert "0G".is_hex == false
344 for i
in [0..length
[ do
346 if not (c
>= 'a' and c
<= 'f') and
347 not (c
>= 'A' and c
<= 'F') and
348 not (c
>= '0' and c
<= '9') then return false
353 # Returns `true` if the string contains only Binary digits
355 # assert "1101100".is_bin == true
356 # assert "1101020".is_bin == false
358 for i
in chars
do if i
!= '0' and i
!= '1' then return false
362 # Returns `true` if the string contains only Octal digits
364 # assert "213453".is_oct == true
365 # assert "781".is_oct == false
367 for i
in chars
do if i
< '0' or i
> '7' then return false
371 # Returns `true` if the string contains only Decimal digits
373 # assert "10839".is_dec == true
374 # assert "164F".is_dec == false
376 for i
in chars
do if i
< '0' or i
> '9' then return false
380 # Are all letters in `self` upper-case ?
382 # assert "HELLO WORLD".is_upper == true
383 # assert "%$&%!".is_upper == true
384 # assert "hello world".is_upper == false
385 # assert "Hello World".is_upper == false
388 for i
in [0..length
[ do
390 if char
.is_lower
then return false
395 # Are all letters in `self` lower-case ?
397 # assert "hello world".is_lower == true
398 # assert "%$&%!".is_lower == true
399 # assert "Hello World".is_lower == false
402 for i
in [0..length
[ do
404 if char
.is_upper
then return false
409 # Removes the whitespaces at the beginning of self
411 # assert " \n\thello \n\t".l_trim == "hello \n\t"
413 # `Char::is_whitespace` determines what is a whitespace.
416 var iter
= self.chars
.iterator
418 if not iter
.item
.is_whitespace
then break
421 if iter
.index
== length
then return self.empty
422 return self.substring_from
(iter
.index
)
425 # Removes the whitespaces at the end of self
427 # assert " \n\thello \n\t".r_trim == " \n\thello"
429 # `Char::is_whitespace` determines what is a whitespace.
432 var iter
= self.chars
.reverse_iterator
434 if not iter
.item
.is_whitespace
then break
437 if iter
.index
< 0 then return self.empty
438 return self.substring
(0, iter
.index
+ 1)
441 # Trims trailing and preceding white spaces
443 # assert " Hello World ! ".trim == "Hello World !"
444 # assert "\na\nb\tc\t".trim == "a\nb\tc"
446 # `Char::is_whitespace` determines what is a whitespace.
447 fun trim
: SELFTYPE do return (self.l_trim
).r_trim
449 # Is the string non-empty but only made of whitespaces?
451 # assert " \n\t ".is_whitespace == true
452 # assert " hello ".is_whitespace == false
453 # assert "".is_whitespace == false
455 # `Char::is_whitespace` determines what is a whitespace.
456 fun is_whitespace
: Bool
458 if is_empty
then return false
459 for c
in self.chars
do
460 if not c
.is_whitespace
then return false
465 # Returns `self` removed from its last line terminator (if any).
467 # assert "Hello\n".chomp == "Hello"
468 # assert "Hello".chomp == "Hello"
470 # assert "\n".chomp == ""
471 # assert "".chomp == ""
473 # Line terminators are `"\n"`, `"\r\n"` and `"\r"`.
474 # A single line terminator, the last one, is removed.
476 # assert "\r\n".chomp == ""
477 # assert "\r\n\n".chomp == "\r\n"
478 # assert "\r\n\r\n".chomp == "\r\n"
479 # assert "\r\n\r".chomp == "\r\n"
481 # Note: unlike with most IO methods like `Reader::read_line`,
482 # a single `\r` is considered here to be a line terminator and will be removed.
486 if len
== 0 then return self
487 var l
= self.chars
.last
489 return substring
(0, len-1
)
490 else if l
!= '\n' then
492 else if len
> 1 and self.chars
[len-2
] == '\r' then
493 return substring
(0, len-2
)
495 return substring
(0, len-1
)
499 # Justify `self` in a space of `length`
501 # `left` is the space ratio on the left side.
502 # * 0.0 for left-justified (no space at the left)
503 # * 1.0 for right-justified (all spaces at the left)
504 # * 0.5 for centered (half the spaces at the left)
506 # `char`, or `' '` by default, is repeated to pad the empty space.
510 # assert "hello".justify(10, 0.0) == "hello "
511 # assert "hello".justify(10, 1.0) == " hello"
512 # assert "hello".justify(10, 0.5) == " hello "
513 # assert "hello".justify(10, 0.5, '.') == "..hello..."
515 # If `length` is not enough, `self` is returned as is.
517 # assert "hello".justify(2, 0.0) == "hello"
519 # REQUIRE: `left >= 0.0 and left <= 1.0`
520 # ENSURE: `self.length <= length implies result.length == length`
521 # ENSURE: `self.length >= length implies result == self`
522 fun justify
(length
: Int, left
: Float, char
: nullable Char): String
524 var pad
= (char
or else ' ').to_s
525 var diff
= length
- self.length
526 if diff
<= 0 then return to_s
527 assert left
>= 0.0 and left
<= 1.0
528 var before
= (diff
.to_f
* left
).to_i
529 return pad
* before
+ self + pad
* (diff-before
)
532 # Mangle a string to be a unique string only made of alphanumeric characters and underscores.
534 # This method is injective (two different inputs never produce the same
535 # output) and the returned string always respect the following rules:
537 # * Contains only US-ASCII letters, digits and underscores.
538 # * Never starts with a digit.
539 # * Never ends with an underscore.
540 # * Never contains two contiguous underscores.
542 # assert "42_is/The answer!".to_cmangle == "_52d2_is_47dThe_32danswer_33d"
543 # assert "__".to_cmangle == "_95d_95d"
544 # assert "__d".to_cmangle == "_95d_d"
545 # assert "_d_".to_cmangle == "_d_95d"
546 # assert "_42".to_cmangle == "_95d42"
547 # assert "foo".to_cmangle == "foo"
548 # assert "".to_cmangle == ""
549 fun to_cmangle
: String
551 if is_empty
then return ""
553 var underscore
= false
557 if c
>= '0' and c
<= '9' then
559 res
.append
(c
.code_point
.to_s
)
563 for i
in [start
..length
[ do
565 if (c
>= 'a' and c
<= 'z') or (c
>='A' and c
<= 'Z') then
571 res
.append
('_'.code_point
.to_s
)
574 if c
>= '0' and c
<= '9' then
577 else if c
== '_' then
582 res
.append
(c
.code_point
.to_s
)
588 res
.append
('_'.code_point
.to_s
)
594 # Escape `"` `\` `'`, trigraphs and non printable characters using the rules of literal C strings and characters
596 # assert "abAB12<>&".escape_to_c == "abAB12<>&"
597 # assert "\n\"'\\".escape_to_c == "\\n\\\"\\'\\\\"
598 # assert "allo???!".escape_to_c == "allo??\\?!"
599 # assert "??=??/??'??(??)".escape_to_c == "?\\?=?\\?/??\\'?\\?(?\\?)"
600 # assert "??!??<??>??-".escape_to_c == "?\\?!?\\?<?\\?>?\\?-"
602 # Most non-printable characters (bellow ASCII 32) are escaped to an octal form `\nnn`.
603 # Three digits are always used to avoid following digits to be interpreted as an element
604 # of the octal sequence.
606 # assert "{0.code_point}{1.code_point}{8.code_point}{31.code_point}{32.code_point}".escape_to_c == "\\000\\001\\010\\037 "
608 # The exceptions are the common `\t` and `\n`.
609 fun escape_to_c
: String
612 for i
in [0..length
[ do
616 else if c
== '\t' then
618 else if c
== '"' then
620 else if c == '\'' then
622 else if c == '\\
' then
624 else if c == '?' then
625 # Escape if it is the last question mark of a ANSI C trigraph.
629 # We ignore `??'` because it will be escaped as `??\
'`.
642 else if c.code_point < 32 then
644 var oct = c.code_point.to_base(8)
645 # Force 3 octal digits since it is the
646 # maximum allowed in the C specification
647 if oct.length == 1 then
650 else if oct.length == 2 then
661 # Escape additionnal characters
662 # The result might no be legal in C but be used in other languages
664 # assert "ab|\{\}".escape_more_to_c("|\{\}") == "ab\\|\\\{\\\}"
665 # assert "allo???!".escape_more_to_c("") == "allo??\\?!"
666 fun escape_more_to_c(chars: String): String
669 for c in escape_to_c.chars do
670 if chars.chars.has(c) then
678 # Escape to C plus braces
680 # assert "\n\"'\\\
{\}".escape_to_nit == "\\n\\\
"\\'\\\\\\\{\\\}"
681 fun escape_to_nit
: String do return escape_more_to_c
("\{\}")
683 # Escape to POSIX Shell (sh).
685 # Abort if the text contains a null byte.
687 # assert "\n\"'\\\{\}0".escape_to_sh == "'\n\"'\\''\\\{\}0'"
688 fun escape_to_sh
: String do
691 for i in [0..length[ do
696 assert without_null_byte
: c
!= '\0'
704 # Escape to include in a Makefile
706 # Unfortunately, some characters are not escapable in Makefile.
707 # These characters are `;`, `|`, `\`, and the non-printable ones.
708 # They will be rendered as `"?{hex}"`.
709 fun escape_to_mk: String do
711 for i in [0..length[ do
715 else if c == ':' or c == ' ' or c == '#' then
718 else if c
.code_point
< 32 or c
== ';' or c
== '|' or c
== '\\' or c
== '=' then
719 b
.append
("?{c.code_point.to_base(16)}")
727 # Return a string where Nit escape sequences are transformed.
730 # assert s.length == 2
731 # var u = s.unescape_nit
732 # assert u.length == 1
733 # assert u.chars[0].code_point == 10 # (the ASCII value of the "new line" character)
734 fun unescape_nit
: String
736 var res
= new Buffer.with_cap
(self.length
)
737 var was_slash
= false
738 for i
in [0..length
[ do
740 if not was_slash
then
751 else if c
== 'r' then
753 else if c
== 't' then
755 else if c
== '0' then
764 # Returns `self` with all characters escaped with their UTF-16 representation
766 # assert "Aèあ𐏓".escape_to_utf16 == "\\u0041\\u00e8\\u3042\\ud800\\udfd3"
767 fun escape_to_utf16
: String do
769 for i
in chars
do buf
.append i
.escape_to_utf16
773 # Returns the Unicode char escaped by `self`
775 # assert "\\u0041".from_utf16_escape == 'A'
776 # assert "\\ud800\\udfd3".from_utf16_escape == '𐏓'
777 # assert "\\u00e8".from_utf16_escape == 'è'
778 # assert "\\u3042".from_utf16_escape == 'あ'
779 fun from_utf16_escape
(pos
, ln
: nullable Int): Char do
780 if pos
== null then pos
= 0
781 if ln
== null then ln
= length
- pos
782 if ln
< 6 then return 0xFFFD.code_point
783 var cp
= from_utf16_digit
(pos
+ 2)
784 if cp
< 0xD800 then return cp
.code_point
785 if cp
> 0xDFFF then return cp
.code_point
786 if cp
> 0xDBFF then return 0xFFFD.code_point
787 if ln
== 6 then return 0xFFFD.code_point
788 if ln
< 12 then return 0xFFFD.code_point
790 cp
+= from_utf16_digit
(pos
+ 8)
791 var cplo
= cp
& 0xFFFF
792 if cplo
< 0xDC00 then return 0xFFFD.code_point
793 if cplo
> 0xDFFF then return 0xFFFD.code_point
794 return cp
.from_utf16_surr
.code_point
797 # Returns a UTF-16 escape value
799 # var s = "\\ud800\\udfd3"
800 # assert s.from_utf16_digit(2) == 0xD800
801 # assert s.from_utf16_digit(8) == 0xDFD3
802 fun from_utf16_digit
(pos
: nullable Int): Int do
803 if pos
== null then pos
= 0
804 return to_hex
(pos
, 4)
807 # Encode `self` to percent (or URL) encoding
809 # assert "aBc09-._~".to_percent_encoding == "aBc09-._~"
810 # assert "%()< >".to_percent_encoding == "%25%28%29%3c%20%3e"
811 # assert ".com/post?e=asdf&f=123".to_percent_encoding == ".com%2fpost%3fe%3dasdf%26f%3d123"
812 # assert "éあいう".to_percent_encoding == "%c3%a9%e3%81%82%e3%81%84%e3%81%86"
813 fun to_percent_encoding
: String
817 for i
in [0..length
[ do
819 if (c
>= '0' and c
<= '9') or
820 (c
>= 'a' and c
<= 'z') or
821 (c
>= 'A' and c
<= 'Z') or
822 c
== '-' or c
== '.' or
827 var bytes
= c
.to_s
.bytes
828 for b
in bytes
do buf
.append
"%{b.to_i.to_hex}"
835 # Decode `self` from percent (or URL) encoding to a clear string
837 # Replace invalid use of '%' with '?'.
839 # assert "aBc09-._~".from_percent_encoding == "aBc09-._~"
840 # assert "%25%28%29%3c%20%3e".from_percent_encoding == "%()< >"
841 # assert ".com%2fpost%3fe%3dasdf%26f%3d123".from_percent_encoding == ".com/post?e=asdf&f=123"
842 # assert "%25%28%29%3C%20%3E".from_percent_encoding == "%()< >"
843 # assert "incomplete %".from_percent_encoding == "incomplete ?"
844 # assert "invalid % usage".from_percent_encoding == "invalid ? usage"
845 # assert "%c3%a9%e3%81%82%e3%81%84%e3%81%86".from_percent_encoding == "éあいう"
846 fun from_percent_encoding
: String
848 var len
= byte_length
849 var has_percent
= false
857 # If no transformation is needed, return self as a string
858 if not has_percent
then return to_s
860 var buf
= new NativeString(len
)
866 if i
+ 2 >= length
then
867 # What follows % has been cut off
871 var hex_s
= substring
(i
, 2)
873 var hex_i
= hex_s
.to_hex
877 # What follows a % is not Hex
882 else buf
[l
] = c
.ascii
888 return buf
.to_s_unsafe
(l
)
891 # Escape the characters `<`, `>`, `&`, `"`, `'` and `/` as HTML/XML entity references.
893 # assert "a&b-<>\"x\"/'".html_escape == "a&b-<>"x"/'"
895 # SEE: <https://www.owasp.org/index.php/XSS_%28Cross_Site_Scripting%29_Prevention_Cheat_Sheet#RULE_.231_-_HTML_Escape_Before_Inserting_Untrusted_Data_into_HTML_Element_Content>
896 fun html_escape
: String
900 for i
in [0..length
[ do
904 else if c
== '<' then
906 else if c
== '>' then
908 else if c
== '"' then
910 else if c
== '\'' then
912 else if c == '/' then
921 # Two pieces of text are equals if thez have the same characters in the same order.
923 # assert "hello" == "hello"
924 # assert "hello" != "HELLO"
925 # assert "hello" == "hel"+"lo"
927 # Things that are not Text are not equal.
930 # assert "9" != ['9']
933 # assert "9".chars.first == '9' # equality of Char
934 # assert "9".chars == ['9'] # equality of Sequence
935 # assert "9".to_i == 9 # equality of Int
938 if o == null then return false
939 if not o isa Text then return false
940 if self.is_same_instance(o) then return true
941 if self.length != o.length then return false
942 return self.chars == o.chars
945 # Lexicographical comparaison
947 # assert "abc" < "xy"
948 # assert "ABC" < "abc"
951 var self_chars = self.chars.iterator
952 var other_chars = other.chars.iterator
954 while self_chars.is_ok and other_chars.is_ok do
955 if self_chars.item < other_chars.item then return true
956 if self_chars.item > other_chars.item then return false
961 if self_chars.is_ok then
968 # Escape string used in labels for graphviz
970 # assert ">><<".escape_to_dot == "\\>\\>\\<\\<"
971 fun escape_to_dot: String
973 return escape_more_to_c("|\{\}<>")
976 private var hash_cache: nullable Int = null
980 if hash_cache == null then
981 # djb2 hash algorithm
984 for i in [0..length[ do
986 h = (h << 5) + h + char.code_point
991 return hash_cache.as(not null)
994 # Format `self` by replacing each `%n` with the `n`th item of `args`
996 # The character `%` followed by something other than a number are left as is.
997 # To represent a `%` followed by a number, double the `%`, as in `%%7`.
999 # assert "This %0 is a %1.".format("String", "formatted String") == "This String is a formatted String."
1000 # assert "Do not escape % nor %%1".format("unused") == "Do not escape % nor %1"
1001 fun format(args: Object...): String do
1002 var s = new Array[Text]
1006 if self[i] == '%' then
1010 while i < length and self[i].is_numeric do
1014 var ciph_len = i - ciph_st
1015 if ciph_len == 0 then
1016 # What follows '%' is not a number.
1017 s.push substring(curr_st, i - curr_st)
1018 if i < length and self[i] == '%' then
1026 var arg_index = substring(ciph_st, ciph_len).to_i
1027 if arg_index >= args.length then continue
1029 s.push substring(curr_st, fmt_st - curr_st)
1030 s.push args[arg_index].to_s
1037 s.push substring(curr_st, length - curr_st)
1041 # Return the Levenshtein distance between two strings
1044 # assert "abcd".levenshtein_distance("abcd") == 0
1045 # assert "".levenshtein_distance("abcd") == 4
1046 # assert "abcd".levenshtein_distance("") == 4
1047 # assert "abcd".levenshtein_distance("xyz") == 4
1048 # assert "abcd".levenshtein_distance("xbdy") == 3
1050 fun levenshtein_distance(other: String): Int
1052 var slen = self.length
1053 var olen = other.length
1056 if slen == 0 then return olen
1057 if olen == 0 then return slen
1058 if self == other then return 0
1060 # previous row of distances
1061 var v0 = new Array[Int].with_capacity(olen+1)
1063 # current row of distances
1064 var v1 = new Array[Int].with_capacity(olen+1)
1066 for j in [0..olen] do
1067 # prefix insert cost
1071 for i in [0..slen[ do
1073 # prefix delete cost
1076 for j in [0..olen[ do
1078 var cost1 = v1[j] + 1
1080 var cost2 = v0[j + 1] + 1
1081 # same char cost (+0)
1084 if self[i] != other[j] then cost3 += 1
1086 v1[j+1] = cost1.min(cost2).min(cost3)
1090 # * v1 become v0 in the next iteration
1091 # * old v0 is reused as the new v1
1100 # Copies `n` bytes from `self` at `src_offset` into `dest` starting at `dest_offset`
1102 # Basically a high-level synonym of NativeString::copy_to
1104 # REQUIRE: `n` must be large enough to contain `len` bytes
1106 # var ns = new NativeString(8)
1107 # "Text is String".copy_to_native(ns, 8, 2, 0)
1108 # assert ns.to_s_unsafe(8) == "xt is St"
1110 fun copy_to_native(dest: NativeString, n, src_offset, dest_offset: Int) do
1111 var mypos = src_offset
1112 var itspos = dest_offset
1114 dest[itspos] = self.bytes[mypos]
1121 # Packs the content of a string in packs of `ln` chars.
1122 # This variant ensures that only the last element might be smaller than `ln`
1125 # var s = "abcdefghijklmnopqrstuvwxyz"
1126 # assert s.pack_l(4) == ["abcd","efgh","ijkl","mnop","qrst","uvwx","yz"]
1128 fun pack_l(ln: Int): Array[Text] do
1130 var retarr = new Array[Text].with_capacity(length / ln + length % ln)
1131 while st < length do
1132 retarr.add(substring(st, ln))
1138 # Packs the content of a string in packs of `ln` chars.
1139 # This variant ensures that only the first element might be smaller than `ln`
1142 # var s = "abcdefghijklmnopqrstuvwxyz"
1143 # assert s.pack_r(4) == ["ab","cdef","ghij","klmn","opqr","stuv","wxyz"]
1145 fun pack_r(ln: Int): Array[Text] do
1147 var retarr = new Array[Text].with_capacity(length / ln + length % ln)
1149 retarr.add(substring(st - ln, ln))
1152 return retarr.reversed
1156 # All kinds of array-based text representations.
1157 abstract class FlatText
1160 # Underlying C-String (`char*`)
1162 # Warning : Might be void in some subclasses, be sure to check
1163 # if set before using it.
1164 var items: NativeString is noinit
1166 # Returns a char* starting at position `first_byte`
1168 # WARNING: If you choose to use this service, be careful of the following.
1170 # Strings and NativeString are *ideally* always allocated through a Garbage Collector.
1171 # Since the GC tracks the use of the pointer for the beginning of the char*, it may be
1172 # deallocated at any moment, rendering the pointer returned by this function invalid.
1173 # Any access to freed memory may very likely cause undefined behaviour or a crash.
1174 # (Failure to do so will most certainly result in long and painful debugging hours)
1176 # The only safe use of this pointer is if it is ephemeral (e.g. read in a C function
1177 # then immediately return).
1179 # As always, do not modify the content of the String in C code, if this is what you want
1180 # copy locally the char* as Nit Strings are immutable.
1181 fun fast_cstring: NativeString is abstract
1183 redef var length = 0
1185 redef var byte_length = 0
1196 redef fun copy_to_native(dest, n, src_offset, dest_offset) do
1197 items.copy_to(dest, n, src_offset, dest_offset)
1201 # Abstract class for the SequenceRead compatible
1202 # views on the chars of any Text
1203 private abstract class StringCharView
1204 super SequenceRead[Char]
1208 var target: SELFTYPE
1210 redef fun is_empty do return target.is_empty
1212 redef fun length do return target.length
1214 redef fun iterator: IndexedIterator[Char] do return self.iterator_from(0)
1216 redef fun reverse_iterator do return self.reverse_iterator_from(self.length - 1)
1219 # Abstract class for the SequenceRead compatible
1220 # views on the bytes of any Text
1221 private abstract class StringByteView
1222 super SequenceRead[Byte]
1226 var target: SELFTYPE
1228 redef fun is_empty do return target.is_empty
1230 redef fun length do return target.byte_length
1232 redef fun iterator do return self.iterator_from(0)
1234 redef fun reverse_iterator do return self.reverse_iterator_from(target.byte_length - 1)
1237 # Immutable sequence of characters.
1239 # String objects may be created using literals.
1241 # assert "Hello World!" isa String
1242 abstract class String
1245 redef type SELFTYPE: String is fixed
1247 redef fun to_s do return self
1249 redef fun clone do return self
1250 # Concatenates `o` to `self`
1252 # assert "hello" + "world" == "helloworld"
1253 # assert "" + "hello" + "" == "hello"
1254 fun +(o: Text): SELFTYPE is abstract
1256 # Concatenates self `i` times
1258 # assert "abc" * 4 == "abcabcabcabc"
1259 # assert "abc" * 1 == "abc"
1260 # assert "abc" * 0 == ""
1261 fun *(i: Int): SELFTYPE is abstract
1263 # Insert `s` at `pos`.
1265 # assert "helloworld".insert_at(" ", 5) == "hello world"
1266 fun insert_at(s: String, pos: Int): SELFTYPE is abstract
1268 redef fun substrings is abstract
1270 # Returns a reversed version of self
1272 # assert "hello".reversed == "olleh"
1273 # assert "bob".reversed == "bob"
1274 # assert "".reversed == ""
1275 fun reversed: SELFTYPE is abstract
1277 # A upper case version of `self`
1279 # assert "Hello World!".to_upper == "HELLO WORLD!"
1280 fun to_upper: SELFTYPE is abstract
1282 # A lower case version of `self`
1284 # assert "Hello World!".to_lower == "hello world!"
1285 fun to_lower : SELFTYPE is abstract
1287 # Takes a camel case `self` and converts it to snake case
1289 # assert "randomMethodId".to_snake_case == "random_method_id"
1291 # The rules are the following:
1293 # An uppercase is always converted to a lowercase
1295 # assert "HELLO_WORLD".to_snake_case == "hello_world"
1297 # An uppercase that follows a lowercase is prefixed with an underscore
1299 # assert "HelloTheWORLD".to_snake_case == "hello_the_world"
1301 # An uppercase that follows an uppercase and is followed by a lowercase, is prefixed with an underscore
1303 # assert "HelloTHEWorld".to_snake_case == "hello_the_world"
1305 # All other characters are kept as is; `self` does not need to be a proper CamelCased string.
1307 # assert "=-_H3ll0Th3W0rld_-=".to_snake_case == "=-_h3ll0th3w0rld_-="
1308 fun to_snake_case: SELFTYPE
1310 if self.is_lower then return self
1312 var new_str = new Buffer.with_cap(self.length)
1313 var prev_is_lower = false
1314 var prev_is_upper = false
1316 for i in [0..length[ do
1318 if char.is_lower then
1320 prev_is_lower = true
1321 prev_is_upper = false
1322 else if char.is_upper then
1323 if prev_is_lower then
1325 else if prev_is_upper and i+1 < length and chars[i+1].is_lower then
1328 new_str.add(char.to_lower)
1329 prev_is_lower = false
1330 prev_is_upper = true
1333 prev_is_lower = false
1334 prev_is_upper = false
1341 # Takes a snake case `self` and converts it to camel case
1343 # assert "random_method_id".to_camel_case == "randomMethodId"
1345 # If the identifier is prefixed by an underscore, the underscore is ignored
1347 # assert "_private_field".to_camel_case == "_privateField"
1349 # If `self` is upper, it is returned unchanged
1351 # assert "RANDOM_ID".to_camel_case == "RANDOM_ID"
1353 # If there are several consecutive underscores, they are considered as a single one
1355 # assert "random__method_id".to_camel_case == "randomMethodId"
1356 fun to_camel_case: SELFTYPE
1358 if self.is_upper then return self
1360 var new_str = new Buffer
1361 var is_first_char = true
1362 var follows_us = false
1364 for i in [0..length[ do
1366 if is_first_char then
1368 is_first_char = false
1369 else if char == '_
' then
1371 else if follows_us then
1372 new_str.add(char.to_upper)
1382 # Returns a capitalized `self`
1384 # Letters that follow a letter are lowercased
1385 # Letters that follow a non-letter are upcased.
1387 # If `keep_upper = true`, already uppercase letters are not lowercased.
1389 # SEE : `Char::is_letter` for the definition of letter.
1391 # assert "jAVASCRIPT".capitalized == "Javascript"
1392 # assert "i am root".capitalized == "I Am Root"
1393 # assert "ab_c -ab0c ab\nc".capitalized == "Ab_C -Ab0C Ab\nC"
1394 # assert "preserve my ACRONYMS".capitalized(keep_upper=true) == "Preserve My ACRONYMS"
1395 fun capitalized(keep_upper: nullable Bool): SELFTYPE do
1396 if length == 0 then return self
1398 var buf = new Buffer.with_cap(length)
1399 buf.capitalize(keep_upper=keep_upper, src=self)
1404 # A mutable sequence of characters.
1405 abstract class Buffer
1408 # Returns an arbitrary subclass of `Buffer` with default parameters
1411 # Returns an instance of a subclass of `Buffer` with `i` base capacity
1412 new with_cap(i: Int) is abstract
1414 redef type SELFTYPE: Buffer is fixed
1416 # Copy-On-Write flag
1418 # If the `Buffer` was to_s'd
, the next in-place altering
1419 # operation will cause the current `Buffer` to be re-allocated.
1421 # The flag will then be set at `false`.
1422 protected var written
= false
1424 # Modifies the char contained at pos `index`
1426 # DEPRECATED : Use self.chars.[]= instead
1427 fun []=(index
: Int, item
: Char) is abstract
1430 # var b = new Buffer
1431 # b.append("Buffer!")
1436 var cln
= new Buffer.with_cap
(byte_length
)
1441 # Adds a char `c` at the end of self
1443 # DEPRECATED : Use self.chars.add instead
1444 fun add
(c
: Char) is abstract
1448 # var b = new Buffer
1450 # assert not b.is_empty
1453 fun clear
is abstract
1455 # Enlarges the subsequent array containing the chars of self
1456 fun enlarge
(cap
: Int) is abstract
1458 # Adds the content of text `s` at the end of self
1460 # var b = new Buffer
1463 # assert b == "helloworld"
1464 fun append
(s
: Text) is abstract
1466 # `self` is appended in such a way that `self` is repeated `r` times
1468 # var b = new Buffer
1471 # assert b == "hellohellohello"
1472 fun times
(r
: Int) is abstract
1474 # Reverses itself in-place
1476 # var b = new Buffer
1479 # assert b == "olleh"
1480 fun reverse
is abstract
1482 # Changes each lower-case char in `self` by its upper-case variant
1484 # var b = new Buffer
1485 # b.append("Hello World!")
1487 # assert b == "HELLO WORLD!"
1488 fun upper
is abstract
1490 # Changes each upper-case char in `self` by its lower-case variant
1492 # var b = new Buffer
1493 # b.append("Hello World!")
1495 # assert b == "hello world!"
1496 fun lower
is abstract
1498 # Capitalizes each word in `self`
1500 # Letters that follow a letter are lowercased
1501 # Letters that follow a non-letter are upcased.
1503 # If `keep_upper = true`, uppercase letters are not lowercased.
1505 # When `src` is specified, this method reads from `src` instead of `self`
1506 # but it still writes the result to the beginning of `self`.
1507 # This requires `self` to have the capacity to receive all of the
1508 # capitalized content of `src`.
1510 # SEE: `Char::is_letter` for the definition of a letter.
1512 # var b = new FlatBuffer.from("jAVAsCriPt")
1514 # assert b == "Javascript"
1515 # b = new FlatBuffer.from("i am root")
1517 # assert b == "I Am Root"
1518 # b = new FlatBuffer.from("ab_c -ab0c ab\nc")
1520 # assert b == "Ab_C -Ab0C Ab\nC"
1522 # b = new FlatBuffer.from("12345")
1523 # b.capitalize(src="foo")
1524 # assert b == "Foo45"
1526 # b = new FlatBuffer.from("preserve my ACRONYMS")
1527 # b.capitalize(keep_upper=true)
1528 # assert b == "Preserve My ACRONYMS"
1529 fun capitalize
(keep_upper
: nullable Bool, src
: nullable Text) do
1530 src
= src
or else self
1531 var length
= src
.length
1532 if length
== 0 then return
1533 keep_upper
= keep_upper
or else false
1535 var c
= src
[0].to_upper
1538 for i
in [1 .. length
[ do
1541 if prev
.is_letter
then
1545 self[i
] = c
.to_lower
1548 self[i
] = c
.to_upper
1553 # In Buffers, the internal sequence of character is mutable
1554 # Thus, `chars` can be used to modify the buffer.
1555 redef fun chars
: Sequence[Char] is abstract
1557 # Appends `length` chars from `s` starting at index `from`
1560 # var b = new Buffer
1561 # b.append_substring("abcde", 1, 2)
1563 # b.append_substring("vwxyz", 2, 3)
1564 # assert b == "bcxyz"
1565 # b.append_substring("ABCDE", 4, 300)
1566 # assert b == "bcxyzE"
1567 # b.append_substring("VWXYZ", 400, 1)
1568 # assert b == "bcxyzE"
1570 fun append_substring
(s
: Text, from
, length
: Int) do
1576 if (length
+ from
) > ln
then length
= ln
- from
1577 if length
<= 0 then return
1578 append_substring_impl
(s
, from
, length
)
1581 # Unsafe version of `append_substring` for performance
1583 # NOTE: Use only if sure about `from` and `length`, no checks
1584 # or bound recalculation is done
1585 fun append_substring_impl
(s
: Text, from
, length
: Int) do
1587 for i
in [0 .. length
[ do
1594 # View for chars on Buffer objects, extends Sequence
1595 # for mutation operations
1596 private abstract class BufferCharView
1597 super StringCharView
1598 super Sequence[Char]
1600 redef type SELFTYPE: Buffer
1604 # View for bytes on Buffer objects, extends Sequence
1605 # for mutation operations
1606 private abstract class BufferByteView
1607 super StringByteView
1609 redef type SELFTYPE: Buffer
1613 # User readable representation of `self`.
1614 fun to_s
: String do return inspect
1616 # The class name of the object in NativeString format.
1617 private fun native_class_name
: NativeString is intern
1619 # The class name of the object.
1621 # assert 5.class_name == "Int"
1622 fun class_name
: String do return native_class_name
.to_s
1624 # Developer readable representation of `self`.
1625 # Usually, it uses the form "<CLASSNAME:#OBJECTID bla bla bla>"
1628 return "<{inspect_head}>"
1631 # Return "CLASSNAME:#OBJECTID".
1632 # This function is mainly used with the redefinition of the inspect method
1633 protected fun inspect_head
: String
1635 return "{class_name}:#{object_id.to_hex}"
1640 # assert true.to_s == "true"
1641 # assert false.to_s == "false"
1653 # C function to calculate the length of the `NativeString` to receive `self`
1654 private fun byte_to_s_len
: Int `{
1655 return snprintf(NULL, 0, "0x%02x", self);
1658 # C function to convert an nit Int to a NativeString (char*)
1659 private fun native_byte_to_s
(nstr
: NativeString, strlen
: Int) `{
1660 snprintf(nstr, strlen, "0x%02x", self);
1663 # Displayable byte in its hexadecimal form (0x..)
1665 # assert 1.to_b.to_s == "0x01"
1666 # assert (-123).to_b.to_s == "0x85"
1668 var nslen
= byte_to_s_len
1669 var ns
= new NativeString(nslen
+ 1)
1671 native_byte_to_s
(ns
, nslen
+ 1)
1672 return ns
.to_s_unsafe
(nslen
)
1678 # Wrapper of strerror C function
1679 private fun strerror_ext
: NativeString `{ return strerror((int)self); `}
1681 # Returns a string describing error number
1682 fun strerror: String do return strerror_ext.to_s
1684 # Fill `s
` with the digits in base `base
` of `self` (and with the '-' sign if negative).
1685 # assume < to_c max const of char
1686 private fun fill_buffer(s: Buffer, base: Int)
1693 else if self == 0 then
1700 var pos = digit_count(base) - 1
1701 while pos >= 0 and n > 0 do
1702 s.chars[pos] = (n % base).to_c
1708 # C function to calculate the length of the `NativeString` to receive `self`
1709 private fun int_to_s_len: Int `{
1710 return snprintf
(NULL, 0, "%ld", self);
1713 # C function to convert an nit Int to a NativeString (char*)
1714 private fun native_int_to_s(nstr: NativeString, strlen: Int) `{
1715 snprintf
(nstr
, strlen
, "%ld", self);
1718 # String representation of `self` in the given `base
`
1721 # assert 15.to_base(10) == "15"
1722 # assert 15.to_base(16) == "f"
1723 # assert 15.to_base(2) == "1111"
1724 # assert (-10).to_base(3) == "-101"
1726 fun to_base(base: Int): String
1728 var l = digit_count(base)
1731 for x in [0..l[ do s.add(' ')
1732 fill_buffer(s, base)
1737 # return displayable int in hexadecimal
1739 # assert 1.to_hex == "1"
1740 # assert (-255).to_hex == "-ff"
1741 fun to_hex: String do return to_base(16)
1745 # Pretty representation of `self`, with decimals as needed from 1 to a maximum of 3
1747 # assert 12.34.to_s == "12.34"
1748 # assert (-0120.030).to_s == "-120.03"
1750 # see `to_precision
` for a custom precision.
1752 var str = to_precision( 3 )
1753 if is_inf != 0 or is_nan then return str
1754 var len = str.length
1755 for i in [0..len-1] do
1757 var c = str.chars[j]
1760 else if c == '.' then
1761 return str.substring( 0, j+2 )
1763 return str.substring( 0, j+1 )
1769 # `String` representation of `self` with the given number of `decimals
`
1771 # assert 12.345.to_precision(0) == "12"
1772 # assert 12.345.to_precision(3) == "12.345"
1773 # assert (-12.345).to_precision(3) == "-12.345"
1774 # assert (-0.123).to_precision(3) == "-0.123"
1775 # assert 0.999.to_precision(2) == "1.00"
1776 # assert 0.999.to_precision(4) == "0.9990"
1777 fun to_precision(decimals: Int): String
1779 if is_nan then return "nan"
1781 var isinf = self.is_inf
1784 else if isinf == -1 then
1788 if decimals == 0 then return self.to_i.to_s
1790 for i in [0..decimals[ do f = f * 10.0
1797 if i == 0 then return "0." + "0"*decimals
1799 # Prepare both parts of the float, before and after the "."
1804 if sl > decimals then
1805 # Has something before the "."
1806 p1 = s.substring(0, sl-decimals)
1807 p2 = s.substring(sl-decimals, decimals)
1810 p2 = "0"*(decimals-sl) + s
1813 if i < 0 then p1 = "-" + p1
1815 return p1 + "." + p2
1821 # Returns a sequence with the UTF-8 bytes of `self`
1823 # assert 'a'.bytes == [0x61u8]
1824 # assert 'ま'.bytes == [0xE3u8, 0x81u8, 0xBEu8]
1825 fun bytes: SequenceRead[Byte] do return to_s.bytes
1827 # Is `self` an UTF-16 surrogate pair ?
1828 fun is_surrogate: Bool do
1830 return cp >= 0xD800 and cp <= 0xDFFF
1833 # Is `self` a UTF-16 high surrogate ?
1834 fun is_hi_surrogate: Bool do
1836 return cp >= 0xD800 and cp <= 0xDBFF
1839 # Is `self` a UTF-16 low surrogate ?
1840 fun is_lo_surrogate: Bool do
1842 return cp >= 0xDC00 and cp <= 0xDFFF
1845 # Length of `self` in a UTF-8 String
1846 fun u8char_len: Int do
1847 var c = self.code_point
1848 if c < 0x80 then return 1
1849 if c <= 0x7FF then return 2
1850 if c <= 0xFFFF then return 3
1851 if c <= 0x10FFFF then return 4
1852 # Bad character format
1856 # assert 'x'.to_s == "x"
1859 var ns = new NativeString(ln + 1)
1861 return ns.to_s_unsafe(ln)
1864 # Returns `self` escaped to UTF-16
1866 # i.e. Represents `self`.`code_point
` using UTF-16 codets escaped
1869 # assert 'A'.escape_to_utf16 == "\\u0041"
1870 # assert 'è'.escape_to_utf16 == "\\u00e8"
1871 # assert 'あ'.escape_to_utf16 == "\\u3042"
1872 # assert '𐏓'.escape_to_utf16 == "\\ud800\\udfd3"
1873 fun escape_to_utf16: String do
1876 if cp < 0xD800 or (cp >= 0xE000 and cp <= 0xFFFF) then
1877 buf = new Buffer.with_cap(6)
1878 buf.append("\\u0000")
1881 for i in hx.chars.reverse_iterator do
1886 buf = new Buffer.with_cap(12)
1887 buf.append("\\u0000\\u0000")
1888 var lo = (((cp - 0x10000) & 0x3FF) + 0xDC00).to_hex
1889 var hi = ((((cp - 0x10000) & 0xFFC00) >> 10) + 0xD800).to_hex
1904 private fun u8char_tos(r: NativeString, len: Int) `{
1911 r
[0] = 0xC0 | ((self & 0x7C0) >> 6);
1912 r
[1] = 0x80 | (self & 0x3F);
1915 r
[0] = 0xE0 | ((self & 0xF000) >> 12);
1916 r
[1] = 0x80 | ((self & 0xFC0) >> 6);
1917 r
[2] = 0x80 | (self & 0x3F);
1920 r
[0] = 0xF0 | ((self & 0x1C0000) >> 18);
1921 r
[1] = 0x80 | ((self & 0x3F000) >> 12);
1922 r
[2] = 0x80 | ((self & 0xFC0) >> 6);
1923 r
[3] = 0x80 | (self & 0x3F);
1928 # Returns true if the char is a numerical digit
1930 # assert '0'.is_numeric
1931 # assert '9'.is_numeric
1932 # assert not 'a'.is_numeric
1933 # assert not '?'.is_numeric
1935 # FIXME: Works on ASCII-range only
1936 fun is_numeric: Bool
1938 return self >= '0' and self <= '9'
1941 # Returns true if the char is an alpha digit
1943 # assert 'a'.is_alpha
1944 # assert 'Z'.is_alpha
1945 # assert not '0'.is_alpha
1946 # assert not '?'.is_alpha
1948 # FIXME: Works on ASCII-range only
1951 return (self >= 'a' and self <= 'z') or (self >= 'A' and self <= 'Z')
1954 # Is `self` an hexadecimal digit ?
1956 # assert 'A'.is_hexdigit
1957 # assert not 'G'.is_hexdigit
1958 # assert 'a'.is_hexdigit
1959 # assert not 'g'.is_hexdigit
1960 # assert '5'.is_hexdigit
1961 fun is_hexdigit: Bool do return (self >= '0' and self <= '9') or (self >= 'A' and self <= 'F') or
1962 (self >= 'a' and self <= 'f')
1964 # Returns true if the char is an alpha or a numeric digit
1966 # assert 'a'.is_alphanumeric
1967 # assert 'Z'.is_alphanumeric
1968 # assert '0'.is_alphanumeric
1969 # assert '9'.is_alphanumeric
1970 # assert not '?'.is_alphanumeric
1972 # FIXME: Works on ASCII-range only
1973 fun is_alphanumeric: Bool
1975 return self.is_numeric or self.is_alpha
1978 # Returns `self` to its int value
1980 # REQUIRE: `is_hexdigit
`
1981 fun from_hex: Int do
1982 if self >= '0' and self <= '9' then return code_point - 0x30
1983 if self >= 'A' and self <= 'F' then return code_point - 0x37
1984 if self >= 'a' and self <= 'f' then return code_point - 0x57
1985 # Happens if self is not a hexdigit
1986 assert self.is_hexdigit
1987 # To make flow analysis happy
1992 redef class Collection[E]
1993 # String representation of the content of the collection.
1995 # The standard representation is the list of elements separated with commas.
1998 # assert [1,2,3].to_s == "[1,2,3]"
1999 # assert [1..3].to_s == "[1,2,3]"
2000 # assert (new Array[Int]).to_s == "[]" # empty collection
2003 # Subclasses may return a more specific string representation.
2006 return "[" + join(",") + "]"
2009 # Concatenate elements without separators
2012 # assert [1,2,3].plain_to_s == "123"
2013 # assert [11..13].plain_to_s == "111213"
2014 # assert (new Array[Int]).plain_to_s == "" # empty collection
2016 fun plain_to_s: String
2019 for e in self do if e != null then s.append(e.to_s)
2023 # Concatenate and separate each elements with `separator
`.
2025 # Only concatenate if `separator
== null`.
2027 # assert [1, 2, 3].join(":") == "1:2:3"
2028 # assert [1..3].join(":") == "1:2:3"
2029 # assert [1..3].join == "123"
2031 # if `last_separator
` is given, then it is used to separate the last element.
2033 # assert [1, 2, 3, 4].join(", ", " and ") == "1, 2, 3 and 4"
2034 fun join(separator: nullable Text, last_separator: nullable Text): String
2036 if is_empty then return ""
2038 var s = new Buffer # Result
2043 if e != null then s.append(e.to_s)
2045 if last_separator == null then last_separator = separator
2047 # Concat other items
2053 if separator != null then s.append(separator)
2055 if last_separator != null then s.append(last_separator)
2057 if e != null then s.append(e.to_s)
2063 redef class Map[K,V]
2064 # Concatenate couples of key value.
2065 # Key and value are separated by `couple_sep
`.
2066 # Couples are separated by `sep
`.
2068 # var m = new HashMap[Int, String]
2071 # assert m.join("; ", "=") == "1=one; 10=ten"
2072 fun join(sep, couple_sep: String): String is abstract
2076 private var args_cache: nullable Sequence[String] = null
2078 # The arguments of the program as given by the OS
2079 fun program_args: Sequence[String]
2081 if _args_cache == null then init_args
2082 return _args_cache.as(not null)
2085 # The name of the program as given by the OS
2086 fun program_name: String
2088 return native_argv(0).to_s
2091 # Initialize `program_args
` with the contents of `native_argc
` and `native_argv
`.
2092 private fun init_args
2094 var argc = native_argc
2095 var args = new Array[String].with_capacity(0)
2098 args[i-1] = native_argv(i).to_s
2104 # First argument of the main C function.
2105 private fun native_argc: Int is intern
2107 # Second argument of the main C function.
2108 private fun native_argv(i: Int): NativeString is intern
2111 # Comparator that efficienlty use `to_s
` to compare things
2113 # The comparaison call `to_s
` on object and use the result to order things.
2115 # var a = [1, 2, 3, 10, 20]
2116 # (new CachedAlphaComparator).sort(a)
2117 # assert a == [1, 10, 2, 20, 3]
2119 # Internally the result of `to_s
` is cached in a HashMap to counter
2120 # uneficient implementation of `to_s
`.
2122 # Note: it caching is not usefull, see `alpha_comparator
`
2123 class CachedAlphaComparator
2125 redef type COMPARED: Object
2127 private var cache = new HashMap[Object, String]
2129 private fun do_to_s(a: Object): String do
2130 if cache.has_key(a) then return cache[a]
2136 redef fun compare(a, b) do
2137 return do_to_s(a) <=> do_to_s(b)
2141 # see `alpha_comparator
`
2142 private class AlphaComparator
2144 redef fun compare(a, b) do
2145 if a == b then return 0
2146 if a == null then return -1
2147 if b == null then return 1
2148 return a.to_s <=> b.to_s
2152 # Stateless comparator that naively use `to_s
` to compare things.
2154 # Note: the result of `to_s
` is not cached, thus can be invoked a lot
2155 # on a single instace. See `CachedAlphaComparator` as an alternative.
2157 # var a = [1, 2, 3, 10, 20]
2158 # alpha_comparator.sort(a)
2159 # assert a == [1, 10, 2, 20, 3]
2160 fun alpha_comparator: Comparator do return once new AlphaComparator
2162 # The arguments of the program as given by the OS
2163 fun args: Sequence[String]
2165 return sys.program_args
2168 redef class NativeString
2169 # Get a `String` from the data at `self` copied into Nit memory
2171 # Require: `self` is a null-terminated string.
2172 fun to_s_with_copy: String is abstract
2174 # Get a `String` from `length
` bytes at `self`
2176 # The result may point to the data at `self` or
2177 # it may make a copy in Nit controlled memory.
2178 # This method should only be used when `self` was allocated by the Nit GC,
2179 # or when manually controlling the deallocation of `self`.
2180 fun to_s_with_length(length: Int): String is abstract
2182 # Get a `String` from the raw `length
` bytes at `self`
2184 # The default value of `length
` is the number of bytes before
2185 # the first null character.
2187 # The created `String` points to the data at `self`.
2188 # This method should be used when `self` was allocated by the Nit GC,
2189 # or when manually controlling the deallocation of `self`.
2191 # /!\: This service does not clean the items for compliance with UTF-8,
2192 # use only when the data has already been verified as valid UTF-8.
2193 fun to_s_unsafe(length: nullable Int): String is abstract
2195 # Get a `String` from the raw `byte_length
` bytes at `self` with `unilen
` Unicode characters
2197 # The created `String` points to the data at `self`.
2198 # This method should be used when `self` was allocated by the Nit GC,
2199 # or when manually controlling the deallocation of `self`.
2201 # /!\: This service does not clean the items for compliance with UTF-8,
2202 # use only when the data has already been verified as valid UTF-8.
2204 # SEE: `abstract_text
::Text` for more info on the difference
2205 # between `Text::byte_length
` and `Text::length
`.
2206 fun to_s_full(byte_length, unilen: Int): String is abstract
2208 # Copies the content of `src
` to `self`
2210 # NOTE: `self` must be large enough to withold `self.byte_length
` bytes
2211 fun fill_from(src: Text) do src.copy_to_native(self, src.byte_length, 0, 0)
2214 redef class NativeArray[E]
2215 # Join all the elements using `to_s
`
2217 # REQUIRE: `self isa NativeArray[String]`
2218 # REQUIRE: all elements are initialized
2219 fun native_to_s: String is abstract