1 # This file is part of NIT ( http://www.nitlanguage.org ).
3 # Copyright 2004-2008 Jean Privat <jean@pryen.org>
4 # Copyright 2006-2008 Floréal Morandat <morandat@lirmm.fr>
6 # This file is free software, which comes along with NIT. This software is
7 # distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY;
8 # without even the implied warranty of MERCHANTABILITY or FITNESS FOR A
9 # PARTICULAR PURPOSE. You can modify it is you want, provided this header
10 # is kept unaltered, and a notification of the changes is added.
11 # You are allowed to redistribute it and sell it, alone or is a part of
14 # Basic manipulations of strings of characters
19 intrude import collection
::array
26 ###############################################################################
28 ###############################################################################
30 # High-level abstraction for all text representations
34 redef type OTHER: Text
36 # Type of self (used for factorization of several methods, ex : substring_from, empty...)
39 # Gets a view on the chars of the Text object
41 # assert "hello".chars.to_a == ['h', 'e', 'l', 'l', 'o']
42 fun chars
: SequenceRead[Char] is abstract
44 # Number of characters contained in self.
46 # assert "12345".length == 5
47 # assert "".length == 0
48 fun length
: Int is abstract
52 # assert "abcd".substring(1, 2) == "bc"
53 # assert "abcd".substring(-1, 2) == "a"
54 # assert "abcd".substring(1, 0) == ""
55 # assert "abcd".substring(2, 5) == "cd"
57 # A `from` index < 0 will be replaced by 0.
58 # Unless a `count` value is > 0 at the same time.
59 # In this case, `from += count` and `count -= from`.
60 fun substring
(from
: Int, count
: Int): SELFTYPE is abstract
62 # Iterates on the substrings of self if any
63 fun substrings
: Iterator[Text] is abstract
65 # Is the current Text empty (== "")
68 # assert not "foo".is_empty
69 fun is_empty
: Bool do return self.length
== 0
71 # Returns an empty Text of the right type
73 # This method is used internally to get the right
74 # implementation of an empty string.
75 protected fun empty
: SELFTYPE is abstract
77 # Gets the first char of the Text
79 # DEPRECATED : Use self.chars.first instead
80 fun first
: Char do return self.chars
[0]
82 # Access a character at `index` in the string.
84 # assert "abcd"[2] == 'c'
86 # DEPRECATED : Use self.chars.[] instead
87 fun [](index
: Int): Char do return self.chars
[index
]
89 # Gets the index of the first occurence of 'c'
91 # Returns -1 if not found
93 # DEPRECATED : Use self.chars.index_of instead
94 fun index_of
(c
: Char): Int
96 return index_of_from
(c
, 0)
99 # Gets the last char of self
101 # DEPRECATED : Use self.chars.last instead
102 fun last
: Char do return self.chars
[length-1
]
104 # Gets the index of the first occurence of ´c´ starting from ´pos´
106 # Returns -1 if not found
108 # DEPRECATED : Use self.chars.index_of_from instead
109 fun index_of_from
(c
: Char, pos
: Int): Int
111 var iter
= self.chars
.iterator_from
(pos
)
113 if iter
.item
== c
then return iter
.index
119 # Gets the last index of char ´c´
121 # Returns -1 if not found
123 # DEPRECATED : Use self.chars.last_index_of instead
124 fun last_index_of
(c
: Char): Int
126 return last_index_of_from
(c
, length
- 1)
129 # Return a null terminated char *
130 fun to_cstring
: NativeString do return flatten
.to_cstring
132 # The index of the last occurrence of an element starting from pos (in reverse order).
134 # var s = "/etc/bin/test/test.nit"
135 # assert s.last_index_of_from('/', s.length-1) == 13
136 # assert s.last_index_of_from('/', 12) == 8
138 # Returns -1 if not found
140 # DEPRECATED : Use self.chars.last_index_of_from instead
141 fun last_index_of_from
(item
: Char, pos
: Int): Int
143 var iter
= self.chars
.reverse_iterator_from
(pos
)
145 if iter
.item
== item
then return iter
.index
151 # Gets an iterator on the chars of self
153 # DEPRECATED : Use self.chars.iterator instead
154 fun iterator
: Iterator[Char]
156 return self.chars
.iterator
160 # Gets an Array containing the chars of self
162 # DEPRECATED : Use self.chars.to_a instead
163 fun to_a
: Array[Char] do return chars
.to_a
165 # Create a substring from `self` beginning at the `from` position
167 # assert "abcd".substring_from(1) == "bcd"
168 # assert "abcd".substring_from(-1) == "abcd"
169 # assert "abcd".substring_from(2) == "cd"
171 # As with substring, a `from` index < 0 will be replaced by 0
172 fun substring_from
(from
: Int): SELFTYPE
174 if from
>= self.length
then return empty
175 if from
< 0 then from
= 0
176 return substring
(from
, length
- from
)
179 # Does self have a substring `str` starting from position `pos`?
181 # assert "abcd".has_substring("bc",1) == true
182 # assert "abcd".has_substring("bc",2) == false
184 # Returns true iff all characters of `str` are presents
185 # at the expected index in `self.`
186 # The first character of `str` being at `pos`, the second
187 # character being at `pos+1` and so on...
189 # This means that all characters of `str` need to be inside `self`.
191 # assert "abcd".has_substring("xab", -1) == false
192 # assert "abcd".has_substring("cdx", 2) == false
194 # And that the empty string is always a valid substring.
196 # assert "abcd".has_substring("", 2) == true
197 # assert "abcd".has_substring("", 200) == true
198 fun has_substring
(str
: String, pos
: Int): Bool
200 if str
.is_empty
then return true
201 if pos
< 0 or pos
+ str
.length
> length
then return false
202 var myiter
= self.chars
.iterator_from
(pos
)
203 var itsiter
= str
.chars
.iterator
204 while myiter
.is_ok
and itsiter
.is_ok
do
205 if myiter
.item
!= itsiter
.item
then return false
209 if itsiter
.is_ok
then return false
213 # Is this string prefixed by `prefix`?
215 # assert "abcd".has_prefix("ab") == true
216 # assert "abcbc".has_prefix("bc") == false
217 # assert "ab".has_prefix("abcd") == false
218 fun has_prefix
(prefix
: String): Bool do return has_substring
(prefix
,0)
220 # Is this string suffixed by `suffix`?
222 # assert "abcd".has_suffix("abc") == false
223 # assert "abcd".has_suffix("bcd") == true
224 fun has_suffix
(suffix
: String): Bool do return has_substring
(suffix
, length
- suffix
.length
)
226 # If `self` contains only digits, return the corresponding integer
228 # assert "123".to_i == 123
229 # assert "-1".to_i == -1
233 return to_s
.to_cstring
.atoi
236 # If `self` contains a float, return the corresponding float
238 # assert "123".to_f == 123.0
239 # assert "-1".to_f == -1.0
240 # assert "-1.2e-3".to_f == -0.0012
244 return to_s
.to_cstring
.atof
247 # If `self` contains only digits and alpha <= 'f', return the corresponding integer.
249 # assert "ff".to_hex == 255
250 fun to_hex
: Int do return a_to
(16)
252 # If `self` contains only digits and letters, return the corresponding integer in a given base
254 # assert "120".a_to(3) == 15
255 fun a_to
(base
: Int) : Int
260 for j
in [0..length
[ do
282 # Returns `true` if the string contains only Numeric values (and one "," or one "." character)
284 # assert "123".is_numeric == true
285 # assert "1.2".is_numeric == true
286 # assert "1,2".is_numeric == true
287 # assert "1..2".is_numeric == false
290 var has_point_or_comma
= false
291 for i
in [0..length
[ do
293 if not c
.is_numeric
then
294 if (c
== '.' or c
== ',') and not has_point_or_comma
then
295 has_point_or_comma
= true
304 # Returns `true` if the string contains only Hex chars
306 # assert "048bf".is_hex == true
307 # assert "ABCDEF".is_hex == true
308 # assert "0G".is_hex == false
311 for i
in [0..length
[ do
313 if not (c
>= 'a' and c
<= 'f') and
314 not (c
>= 'A' and c
<= 'F') and
315 not (c
>= '0' and c
<= '9') then return false
320 # Are all letters in `self` upper-case ?
322 # assert "HELLO WORLD".is_upper == true
323 # assert "%$&%!".is_upper == true
324 # assert "hello world".is_upper == false
325 # assert "Hello World".is_upper == false
328 for i
in [0..length
[ do
330 if char
.is_lower
then return false
335 # Are all letters in `self` lower-case ?
337 # assert "hello world".is_lower == true
338 # assert "%$&%!".is_lower == true
339 # assert "Hello World".is_lower == false
342 for i
in [0..length
[ do
344 if char
.is_upper
then return false
349 # Removes the whitespaces at the beginning of self
351 # assert " \n\thello \n\t".l_trim == "hello \n\t"
353 # A whitespace is defined as any character which ascii value is less than or equal to 32
356 var iter
= self.chars
.iterator
358 if iter
.item
.ascii
> 32 then break
361 if iter
.index
== length
then return self.empty
362 return self.substring_from
(iter
.index
)
365 # Removes the whitespaces at the end of self
367 # assert " \n\thello \n\t".r_trim == " \n\thello"
369 # A whitespace is defined as any character which ascii value is less than or equal to 32
372 var iter
= self.chars
.reverse_iterator
374 if iter
.item
.ascii
> 32 then break
377 if iter
.index
< 0 then return self.empty
378 return self.substring
(0, iter
.index
+ 1)
381 # Trims trailing and preceding white spaces
382 # A whitespace is defined as any character which ascii value is less than or equal to 32
384 # assert " Hello World ! ".trim == "Hello World !"
385 # assert "\na\nb\tc\t".trim == "a\nb\tc"
386 fun trim
: SELFTYPE do return (self.l_trim
).r_trim
388 # Returns `self` removed from its last `\n` (if any).
390 # assert "Hello\n".chomp == "Hello"
391 # assert "Hello".chomp == "Hello"
392 # assert "\n\n\n".chomp == "\n\n"
394 # This method is mainly used to remove the LINE_FEED character from lines of text.
397 if self.chars
.last
!= '\n' then return self
398 return substring
(0, length-1
)
401 # Justify a self in a space of `length`
403 # `left` is the space ratio on the left side.
404 # * 0.0 for left-justified (no space at the left)
405 # * 1.0 for right-justified (all spaces at the left)
406 # * 0.5 for centered (half the spaces at the left)
410 # assert "hello".justify(10, 0.0) == "hello "
411 # assert "hello".justify(10, 1.0) == " hello"
412 # assert "hello".justify(10, 0.5) == " hello "
414 # If `length` is not enough, `self` is returned as is.
416 # assert "hello".justify(2, 0.0) == "hello"
418 # REQUIRE: `left >= 0.0 and left <= 1.0`
419 # ENSURE: `self.length <= length implies result.length == length`
420 # ENSURE: `self.length >= length implies result == self`
421 fun justify
(length
: Int, left
: Float): SELFTYPE
423 var diff
= length
- self.length
424 if diff
<= 0 then return self
425 assert left
>= 0.0 and left
<= 1.0
426 var before
= (diff
.to_f
* left
).to_i
427 return " " * before
+ self + " " * (diff-before
)
430 # Mangle a string to be a unique string only made of alphanumeric characters
431 fun to_cmangle
: String
433 var res
= new FlatBuffer
434 var underscore
= false
435 for i
in [0..length
[ do
437 if (c
>= 'a' and c
<= 'z') or (c
>='A' and c
<= 'Z') then
443 res
.append
('_'.ascii
.to_s
)
446 if c
>= '0' and c
<= '9' then
449 else if c
== '_' then
454 res
.append
(c
.ascii
.to_s
)
462 # Escape " \ ' and non printable characters using the rules of literal C strings and characters
464 # assert "abAB12<>&".escape_to_c == "abAB12<>&"
465 # assert "\n\"'\\".escape_to_c == "\\n\\\"\\'\\\\"
466 fun escape_to_c
: String
468 var b
= new FlatBuffer
469 for i
in [0..length
[ do
473 else if c
== '\0' then
475 else if c
== '"' then
477 else if c == '\'' then
479 else if c == '\\
' then
481 else if c.ascii < 32 then
482 b.append("\\{c.ascii.to_base(8, false)}")
490 # Escape additionnal characters
491 # The result might no be legal in C but be used in other languages
493 # assert "ab|\{\}".escape_more_to_c("|\{\}") == "ab\\|\\\{\\\}"
494 fun escape_more_to_c(chars: String): String
496 var b = new FlatBuffer
497 for c in escape_to_c.chars do
498 if chars.chars.has(c) then
506 # Escape to C plus braces
508 # assert "\n\"'\\\
{\}".escape_to_nit == "\\n\\\
"\\'\\\\\\\{\\\}"
509 fun escape_to_nit
: String do return escape_more_to_c
("\{\}")
511 # Escape to POSIX Shell (sh).
513 # Abort if the text contains a null byte.
515 # assert "\n\"'\\\{\}0".escape_to_sh == "'\n\"'\\''\\\{\}0'"
516 fun escape_to_sh
: String do
517 var b
= new FlatBuffer
519 for i in [0..length[ do
524 assert without_null_byte
: c
!= '\0'
532 # Escape to include in a Makefile
534 # Unfortunately, some characters are not escapable in Makefile.
535 # These characters are `;`, `|`, `\`, and the non-printable ones.
536 # They will be rendered as `"?{hex}"`.
537 fun escape_to_mk: String do
538 var b = new FlatBuffer
539 for i in [0..length[ do
543 else if c == ':' or c == ' ' or c == '#' then
546 else if c
.ascii
< 32 or c
== ';' or c
== '|' or c
== '\\' or c
== '=' then
547 b
.append
("?{c.ascii.to_base(16, false)}")
555 # Return a string where Nit escape sequences are transformed.
558 # assert s.length == 2
559 # var u = s.unescape_nit
560 # assert u.length == 1
561 # assert u.chars[0].ascii == 10 # (the ASCII value of the "new line" character)
562 fun unescape_nit
: String
564 var res
= new FlatBuffer.with_capacity
(self.length
)
565 var was_slash
= false
566 for i
in [0..length
[ do
568 if not was_slash
then
579 else if c
== 'r' then
581 else if c
== 't' then
583 else if c
== '0' then
592 # Encode `self` to percent (or URL) encoding
594 # assert "aBc09-._~".to_percent_encoding == "aBc09-._~"
595 # assert "%()< >".to_percent_encoding == "%25%28%29%3c%20%3e"
596 # assert ".com/post?e=asdf&f=123".to_percent_encoding == ".com%2fpost%3fe%3dasdf%26f%3d123"
597 fun to_percent_encoding
: String
599 var buf
= new FlatBuffer
601 for i
in [0..length
[ do
603 if (c
>= '0' and c
<= '9') or
604 (c
>= 'a' and c
<= 'z') or
605 (c
>= 'A' and c
<= 'Z') or
606 c
== '-' or c
== '.' or
610 else buf
.append
"%{c.ascii.to_hex}"
616 # Decode `self` from percent (or URL) encoding to a clear string
618 # Replace invalid use of '%' with '?'.
620 # assert "aBc09-._~".from_percent_encoding == "aBc09-._~"
621 # assert "%25%28%29%3c%20%3e".from_percent_encoding == "%()< >"
622 # assert ".com%2fpost%3fe%3dasdf%26f%3d123".from_percent_encoding == ".com/post?e=asdf&f=123"
623 # assert "%25%28%29%3C%20%3E".from_percent_encoding == "%()< >"
624 # assert "incomplete %".from_percent_encoding == "incomplete ?"
625 # assert "invalid % usage".from_percent_encoding == "invalid ? usage"
626 fun from_percent_encoding
: String
628 var buf
= new FlatBuffer
634 if i
+ 2 >= length
then
635 # What follows % has been cut off
639 var hex_s
= substring
(i
, 2)
641 var hex_i
= hex_s
.to_hex
645 # What follows a % is not Hex
658 # Escape the characters `<`, `>`, `&`, `"`, `'` and `/` as HTML/XML entity references.
660 # assert "a&b-<>\"x\"/'".html_escape == "a&b-<>"x"/'"
662 # 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>
663 fun html_escape
: SELFTYPE
665 var buf
= new FlatBuffer
667 for i
in [0..length
[ do
671 else if c
== '<' then
673 else if c
== '>' then
675 else if c
== '"' then
677 else if c
== '\'' then
679 else if c == '/' then
688 # Two pieces of text are equals if thez have the same characters in the same order.
690 # assert "hello" == "hello"
691 # assert "hello" != "HELLO"
692 # assert "hello" == "hel"+"lo"
694 # Things that are not Text are not equal.
697 # assert "9" != ['9']
700 # assert "9".chars.first == '9' # equality of Char
701 # assert "9".chars == ['9'] # equality of Sequence
702 # assert "9".to_i == 9 # equality of Int
705 if o == null then return false
706 if not o isa Text then return false
707 if self.is_same_instance(o) then return true
708 if self.length != o.length then return false
709 return self.chars == o.chars
712 # Lexicographical comparaison
714 # assert "abc" < "xy"
715 # assert "ABC" < "abc"
718 var self_chars = self.chars.iterator
719 var other_chars = other.chars.iterator
721 while self_chars.is_ok and other_chars.is_ok do
722 if self_chars.item < other_chars.item then return true
723 if self_chars.item > other_chars.item then return false
728 if self_chars.is_ok then
735 # Escape string used in labels for graphviz
737 # assert ">><<".escape_to_dot == "\\>\\>\\<\\<"
738 fun escape_to_dot: String
740 return escape_more_to_c("|\{\}<>")
743 # Flat representation of self
744 fun flatten: FlatText is abstract
746 private var hash_cache: nullable Int = null
750 if hash_cache == null then
751 # djb2 hash algorithm
754 for i in [0..length[ do
756 h = h.lshift(5) + h + char.ascii
761 return hash_cache.as(not null)
766 # All kinds of array-based text representations.
767 abstract class FlatText
770 # Underlying C-String (`char*`)
772 # Warning : Might be void in some subclasses, be sure to check
773 # if set before using it.
774 private var items: NativeString is noinit
776 # Real items, used as cache for to_cstring is called
777 private var real_items: nullable NativeString = null
779 redef var length: Int = 0
790 redef fun flatten do return self
793 # Abstract class for the SequenceRead compatible
794 # views on String and Buffer objects
795 private abstract class StringCharView
796 super SequenceRead[Char]
802 redef fun is_empty do return target.is_empty
804 redef fun length do return target.length
806 redef fun iterator: IndexedIterator[Char] do return self.iterator_from(0)
808 redef fun reverse_iterator do return self.reverse_iterator_from(self.length - 1)
811 # View on Buffer objects, extends Sequence
812 # for mutation operations
813 private abstract class BufferCharView
817 redef type SELFTYPE: Buffer
821 # A `String` holds and manipulates an arbitrary sequence of characters.
823 # String objects may be created using literals.
825 # assert "Hello World!" isa String
826 abstract class String
829 redef type SELFTYPE: String
831 redef fun to_s do return self
833 # Concatenates `o` to `self`
835 # assert "hello" + "world" == "helloworld"
836 # assert "" + "hello" + "" == "hello"
837 fun +(o: Text): SELFTYPE is abstract
839 # Concatenates self `i` times
841 # assert "abc" * 4 == "abcabcabcabc"
842 # assert "abc" * 1 == "abc"
843 # assert "abc" * 0 == ""
844 fun *(i: Int): SELFTYPE is abstract
846 # Insert `s` at `pos`.
848 # assert "helloworld".insert_at(" ", 5) == "hello world"
849 fun insert_at(s: String, pos: Int): SELFTYPE is abstract
851 redef fun substrings: Iterator[String] is abstract
853 # Returns a reversed version of self
855 # assert "hello".reversed == "olleh"
856 # assert "bob".reversed == "bob"
857 # assert "".reversed == ""
858 fun reversed: SELFTYPE is abstract
860 # A upper case version of `self`
862 # assert "Hello World!".to_upper == "HELLO WORLD!"
863 fun to_upper: SELFTYPE is abstract
865 # A lower case version of `self`
867 # assert "Hello World!".to_lower == "hello world!"
868 fun to_lower : SELFTYPE is abstract
870 # Takes a camel case `self` and converts it to snake case
872 # assert "randomMethodId".to_snake_case == "random_method_id"
874 # If `self` is upper, it is returned unchanged
876 # assert "RANDOM_METHOD_ID".to_snake_case == "RANDOM_METHOD_ID"
878 # If the identifier is prefixed by an underscore, the underscore is ignored
880 # assert "_privateField".to_snake_case == "_private_field"
881 fun to_snake_case: SELFTYPE
883 if self.is_upper then return self
885 var new_str = new FlatBuffer.with_capacity(self.length)
886 var is_first_char = true
888 for i in [0..length[ do
890 if is_first_char then
891 new_str.add(char.to_lower)
892 is_first_char = false
893 else if char.is_upper then
895 new_str.add(char.to_lower)
904 # Takes a snake case `self` and converts it to camel case
906 # assert "random_method_id".to_camel_case == "randomMethodId"
908 # If the identifier is prefixed by an underscore, the underscore is ignored
910 # assert "_private_field".to_camel_case == "_privateField"
912 # If `self` is upper, it is returned unchanged
914 # assert "RANDOM_ID".to_camel_case == "RANDOM_ID"
916 # If there are several consecutive underscores, they are considered as a single one
918 # assert "random__method_id".to_camel_case == "randomMethodId"
919 fun to_camel_case: SELFTYPE
921 if self.is_upper then return self
923 var new_str = new FlatBuffer
924 var is_first_char = true
925 var follows_us = false
927 for i in [0..length[ do
929 if is_first_char then
931 is_first_char = false
932 else if char == '_
' then
934 else if follows_us then
935 new_str.add(char.to_upper)
945 # Returns a capitalized `self`
947 # Letters that follow a letter are lowercased
948 # Letters that follow a non-letter are upcased.
950 # SEE : `Char::is_letter` for the definition of letter.
952 # assert "jAVASCRIPT".capitalized == "Javascript"
953 # assert "i am root".capitalized == "I Am Root"
954 # assert "ab_c -ab0c ab\nc".capitalized == "Ab_C -Ab0C Ab\nC"
955 fun capitalized: SELFTYPE do
956 if length == 0 then return self
958 var buf = new FlatBuffer.with_capacity(length)
960 var curr = chars[0].to_upper
964 for i in [1 .. length[ do
967 if prev.is_letter then
968 buf[i] = curr.to_lower
970 buf[i] = curr.to_upper
978 private class FlatSubstringsIter
979 super Iterator[FlatText]
981 var tgt: nullable FlatText
985 return tgt.as(not null)
988 redef fun is_ok do return tgt != null
990 redef fun next do tgt = null
993 # Immutable strings of characters.
998 # Index in _items of the start of the string
999 private var index_from: Int is noinit
1001 # Indes in _items of the last item of the string
1002 private var index_to: Int is noinit
1004 redef var chars: SequenceRead[Char] = new FlatStringCharView(self)
1008 # Check that the index (+ index_from) is not larger than indexTo
1009 # In other terms, if the index is valid
1011 assert (index + index_from) <= index_to
1012 return items[index + index_from]
1015 ################################################
1016 # AbstractString specific methods #
1017 ################################################
1021 var native = new NativeString(self.length + 1)
1022 var length = self.length
1023 var items = self.items
1026 while pos < length do
1027 native[pos] = items[ipos]
1031 return native.to_s_with_length(self.length)
1034 redef fun substring(from, count)
1040 if count < 0 then count = 0
1044 var realFrom = index_from + from
1046 if (realFrom + count) > index_to then return new FlatString.with_infos(items, index_to - realFrom + 1, realFrom, index_to)
1048 if count == 0 then return empty
1050 var to = realFrom + count - 1
1052 return new FlatString.with_infos(items, to - realFrom + 1, realFrom, to)
1055 redef fun empty do return "".as(FlatString)
1059 var outstr = new NativeString(self.length + 1)
1062 var myitems = self.items
1063 var index_from = self.index_from
1064 var max = self.index_to
1066 while index_from <= max do
1067 outstr[out_index] = myitems[index_from].to_upper
1072 outstr[self.length] = '\
0'
1074 return outstr.to_s_with_length(self.length)
1079 var outstr = new NativeString(self.length + 1)
1082 var myitems = self.items
1083 var index_from = self.index_from
1084 var max = self.index_to
1086 while index_from <= max do
1087 outstr[out_index] = myitems[index_from].to_lower
1092 outstr[self.length] = '\
0'
1094 return outstr.to_s_with_length(self.length)
1099 var i = self.index_from
1100 var imax = self.index_to
1107 ##################################################
1108 # String Specific Methods #
1109 ##################################################
1111 private init with_infos(items: NativeString, len: Int, from: Int, to: Int)
1119 redef fun to_cstring: NativeString
1121 if real_items != null then
1122 return real_items.as(not null)
1124 var newItems = new NativeString(length + 1)
1125 self.items.copy_to(newItems, length, index_from, 0)
1126 newItems[length] = '\
0'
1127 self.real_items = newItems
1134 if not other isa FlatString then return super
1136 if self.object_id == other.object_id then return true
1138 var my_length = length
1140 if other.length != my_length then return false
1142 var my_index = index_from
1143 var its_index = other.index_from
1145 var last_iteration = my_index + my_length
1147 var itsitems = other.items
1148 var myitems = self.items
1150 while my_index < last_iteration do
1151 if myitems[my_index] != itsitems[its_index] then return false
1161 if not other isa FlatString then return super
1163 if self.object_id == other.object_id then return false
1165 var my_curr_char : Char
1166 var its_curr_char : Char
1168 var curr_id_self = self.index_from
1169 var curr_id_other = other.index_from
1171 var my_items = self.items
1172 var its_items = other.items
1174 var my_length = self.length
1175 var its_length = other.length
1177 var max_iterations = curr_id_self + my_length
1179 while curr_id_self < max_iterations do
1180 my_curr_char = my_items[curr_id_self]
1181 its_curr_char = its_items[curr_id_other]
1183 if my_curr_char != its_curr_char then
1184 if my_curr_char < its_curr_char then return true
1192 return my_length < its_length
1197 var my_length = self.length
1198 var its_length = s.length
1200 var total_length = my_length + its_length
1202 var target_string = new NativeString(my_length + its_length + 1)
1204 self.items.copy_to(target_string, my_length, index_from, 0)
1205 if s isa FlatString then
1206 s.items.copy_to(target_string, its_length, s.index_from, my_length)
1207 else if s isa FlatBuffer then
1208 s.items.copy_to(target_string, its_length, 0, my_length)
1210 var curr_pos = my_length
1211 for i in [0..s.length[ do
1213 target_string[curr_pos] = c
1218 target_string[total_length] = '\
0'
1220 return target_string.to_s_with_length(total_length)
1227 var my_length = self.length
1229 var final_length = my_length * i
1231 var my_items = self.items
1233 var target_string = new NativeString(final_length + 1)
1235 target_string[final_length] = '\
0'
1237 var current_last = 0
1239 for iteration in [1 .. i] do
1240 my_items.copy_to(target_string, my_length, 0, current_last)
1241 current_last += my_length
1244 return target_string.to_s_with_length(final_length)
1249 if hash_cache == null then
1250 # djb2 hash algorithm
1256 while i <= index_to do
1257 h = h.lshift(5) + h + myitems[i].ascii
1264 return hash_cache.as(not null)
1267 redef fun substrings do return new FlatSubstringsIter(self)
1270 private class FlatStringReverseIterator
1271 super IndexedIterator[Char]
1273 var target: FlatString
1275 var target_items: NativeString
1279 init with_pos(tgt: FlatString, pos: Int)
1282 target_items = tgt.items
1283 curr_pos = pos + tgt.index_from
1286 redef fun is_ok do return curr_pos >= target.index_from
1288 redef fun item do return target_items[curr_pos]
1290 redef fun next do curr_pos -= 1
1292 redef fun index do return curr_pos - target.index_from
1296 private class FlatStringIterator
1297 super IndexedIterator[Char]
1299 var target: FlatString
1301 var target_items: NativeString
1305 init with_pos(tgt: FlatString, pos: Int)
1308 target_items = tgt.items
1309 curr_pos = pos + target.index_from
1312 redef fun is_ok do return curr_pos <= target.index_to
1314 redef fun item do return target_items[curr_pos]
1316 redef fun next do curr_pos += 1
1318 redef fun index do return curr_pos - target.index_from
1322 private class FlatStringCharView
1323 super StringCharView
1325 redef type SELFTYPE: FlatString
1329 # Check that the index (+ index_from) is not larger than indexTo
1330 # In other terms, if the index is valid
1332 var target = self.target
1333 assert (index + target.index_from) <= target.index_to
1334 return target.items[index + target.index_from]
1337 redef fun iterator_from(start) do return new FlatStringIterator.with_pos(target, start)
1339 redef fun reverse_iterator_from(start) do return new FlatStringReverseIterator.with_pos(target, start)
1343 # A mutable sequence of characters.
1344 abstract class Buffer
1347 redef type SELFTYPE: Buffer
1349 # Specific implementations MUST set this to `true` in order to invalidate caches
1350 protected var is_dirty = true
1352 # Copy-On-Write flag
1354 # If the `Buffer` was to_s'd
, the next in-place altering
1355 # operation will cause the current `Buffer` to be re-allocated.
1357 # The flag will then be set at `false`.
1358 protected var written
= false
1360 # Modifies the char contained at pos `index`
1362 # DEPRECATED : Use self.chars.[]= instead
1363 fun []=(index
: Int, item
: Char) is abstract
1365 # Adds a char `c` at the end of self
1367 # DEPRECATED : Use self.chars.add instead
1368 fun add
(c
: Char) is abstract
1372 # var b = new FlatBuffer
1374 # assert not b.is_empty
1377 fun clear
is abstract
1379 # Enlarges the subsequent array containing the chars of self
1380 fun enlarge
(cap
: Int) is abstract
1382 # Adds the content of text `s` at the end of self
1384 # var b = new FlatBuffer
1387 # assert b == "helloworld"
1388 fun append
(s
: Text) is abstract
1390 # `self` is appended in such a way that `self` is repeated `r` times
1392 # var b = new FlatBuffer
1395 # assert b == "hellohellohello"
1396 fun times
(r
: Int) is abstract
1398 # Reverses itself in-place
1400 # var b = new FlatBuffer
1403 # assert b == "olleh"
1404 fun reverse
is abstract
1406 # Changes each lower-case char in `self` by its upper-case variant
1408 # var b = new FlatBuffer
1409 # b.append("Hello World!")
1411 # assert b == "HELLO WORLD!"
1412 fun upper
is abstract
1414 # Changes each upper-case char in `self` by its lower-case variant
1416 # var b = new FlatBuffer
1417 # b.append("Hello World!")
1419 # assert b == "hello world!"
1420 fun lower
is abstract
1422 # Capitalizes each word in `self`
1424 # Letters that follow a letter are lowercased
1425 # Letters that follow a non-letter are upcased.
1427 # SEE: `Char::is_letter` for the definition of a letter.
1429 # var b = new FlatBuffer.from("jAVAsCriPt")
1431 # assert b == "Javascript"
1432 # b = new FlatBuffer.from("i am root")
1434 # assert b == "I Am Root"
1435 # b = new FlatBuffer.from("ab_c -ab0c ab\nc")
1437 # assert b == "Ab_C -Ab0C Ab\nC"
1439 if length
== 0 then return
1440 var c
= self[0].to_upper
1443 for i
in [1 .. length
[ do
1446 if prev
.is_letter
then
1447 self[i
] = c
.to_lower
1449 self[i
] = c
.to_upper
1456 if is_dirty
then hash_cache
= null
1460 # In Buffers, the internal sequence of character is mutable
1461 # Thus, `chars` can be used to modify the buffer.
1462 redef fun chars
: Sequence[Char] is abstract
1465 # Mutable strings of characters.
1470 redef type SELFTYPE: FlatBuffer
1472 redef var chars
: Sequence[Char] = new FlatBufferCharView(self)
1474 private var capacity
: Int = 0
1476 redef fun substrings
do return new FlatSubstringsIter(self)
1478 # Re-copies the `NativeString` into a new one and sets it as the new `Buffer`
1480 # This happens when an operation modifies the current `Buffer` and
1481 # the Copy-On-Write flag `written` is set at true.
1482 private fun reset
do
1483 var nns
= new NativeString(capacity
)
1484 items
.copy_to
(nns
, length
, 0, 0)
1492 assert index
< length
1496 redef fun []=(index
, item
)
1499 if index
== length
then
1503 if written
then reset
1504 assert index
>= 0 and index
< length
1511 if capacity
<= length
then enlarge
(length
+ 5)
1518 if written
then reset
1522 redef fun empty
do return new FlatBuffer
1524 redef fun enlarge
(cap
)
1527 if cap
<= c
then return
1528 while c
<= cap
do c
= c
* 2 + 2
1529 # The COW flag can be set at false here, since
1530 # it does a copy of the current `Buffer`
1532 var a
= new NativeString(c
+1)
1533 if length
> 0 then items
.copy_to
(a
, length
, 0, 0)
1538 redef fun to_s
: String
1541 if length
== 0 then items
= new NativeString(1)
1542 return new FlatString.with_infos
(items
, length
, 0, length
- 1)
1545 redef fun to_cstring
1548 var new_native
= new NativeString(length
+ 1)
1549 new_native
[length
] = '\0'
1550 if length
> 0 then items
.copy_to
(new_native
, length
, 0, 0)
1551 real_items
= new_native
1554 return real_items
.as(not null)
1557 # Create a new empty string.
1560 # Create a new string copied from `s`.
1563 capacity
= s
.length
+ 1
1565 items
= new NativeString(capacity
)
1566 if s
isa FlatString then
1567 s
.items
.copy_to
(items
, length
, s
.index_from
, 0)
1568 else if s
isa FlatBuffer then
1569 s
.items
.copy_to
(items
, length
, 0, 0)
1572 for i
in [0..s
.length
[ do
1580 # Create a new empty string with a given capacity.
1581 init with_capacity
(cap
: Int)
1584 # _items = new NativeString.calloc(cap)
1585 items
= new NativeString(cap
+1)
1592 if s
.is_empty
then return
1595 if capacity
< length
+ sl
then enlarge
(length
+ sl
)
1596 if s
isa FlatString then
1597 s
.items
.copy_to
(items
, sl
, s
.index_from
, length
)
1598 else if s
isa FlatBuffer then
1599 s
.items
.copy_to
(items
, sl
, 0, length
)
1601 var curr_pos
= self.length
1602 for i
in [0..s
.length
[ do
1611 # Copies the content of self in `dest`
1612 fun copy
(start
: Int, len
: Int, dest
: Buffer, new_start
: Int)
1614 var self_chars
= self.chars
1615 var dest_chars
= dest
.chars
1616 for i
in [0..len-1
] do
1617 dest_chars
[new_start
+i
] = self_chars
[start
+i
]
1621 redef fun substring
(from
, count
)
1625 if from
< 0 then from
= 0
1626 if count
> length
then count
= length
1627 if from
< count
then
1628 var r
= new FlatBuffer.with_capacity
(count
- from
)
1629 while from
< count
do
1630 r
.chars
.push
(items
[from
])
1635 return new FlatBuffer
1642 var ns
= new NativeString(capacity
)
1654 redef fun times
(repeats
)
1656 var x
= new FlatString.with_infos
(items
, length
, 0, length
- 1)
1657 for i
in [1..repeats
[ do
1664 if written
then reset
1668 it
[id
] = it
[id
].to_upper
1675 if written
then reset
1679 it
[id
] = it
[id
].to_lower
1685 private class FlatBufferReverseIterator
1686 super IndexedIterator[Char]
1688 var target
: FlatBuffer
1690 var target_items
: NativeString
1694 init with_pos
(tgt
: FlatBuffer, pos
: Int)
1697 if tgt
.length
> 0 then target_items
= tgt
.items
1701 redef fun index
do return curr_pos
1703 redef fun is_ok
do return curr_pos
>= 0
1705 redef fun item
do return target_items
[curr_pos
]
1707 redef fun next
do curr_pos
-= 1
1711 private class FlatBufferCharView
1712 super BufferCharView
1714 redef type SELFTYPE: FlatBuffer
1716 redef fun [](index
) do return target
.items
[index
]
1718 redef fun []=(index
, item
)
1720 assert index
>= 0 and index
<= length
1721 if index
== length
then
1725 target
.items
[index
] = item
1738 fun enlarge
(cap
: Int)
1745 var s_length
= s
.length
1746 if target
.capacity
< s
.length
then enlarge
(s_length
+ target
.length
)
1749 redef fun iterator_from
(pos
) do return new FlatBufferIterator.with_pos
(target
, pos
)
1751 redef fun reverse_iterator_from
(pos
) do return new FlatBufferReverseIterator.with_pos
(target
, pos
)
1755 private class FlatBufferIterator
1756 super IndexedIterator[Char]
1758 var target
: FlatBuffer
1760 var target_items
: NativeString
1764 init with_pos
(tgt
: FlatBuffer, pos
: Int)
1767 if tgt
.length
> 0 then target_items
= tgt
.items
1771 redef fun index
do return curr_pos
1773 redef fun is_ok
do return curr_pos
< target
.length
1775 redef fun item
do return target_items
[curr_pos
]
1777 redef fun next
do curr_pos
+= 1
1781 ###############################################################################
1783 ###############################################################################
1786 # User readable representation of `self`.
1787 fun to_s
: String do return inspect
1789 # The class name of the object in NativeString format.
1790 private fun native_class_name
: NativeString is intern
1792 # The class name of the object.
1794 # assert 5.class_name == "Int"
1795 fun class_name
: String do return native_class_name
.to_s
1797 # Developer readable representation of `self`.
1798 # Usually, it uses the form "<CLASSNAME:#OBJECTID bla bla bla>"
1801 return "<{inspect_head}>"
1804 # Return "CLASSNAME:#OBJECTID".
1805 # This function is mainly used with the redefinition of the inspect method
1806 protected fun inspect_head
: String
1808 return "{class_name}:#{object_id.to_hex}"
1813 # assert true.to_s == "true"
1814 # assert false.to_s == "false"
1827 # Wrapper of strerror C function
1828 private fun strerror_ext
: NativeString is extern `{
1829 return strerror(recv);
1832 # Returns a string describing error number
1833 fun strerror
: String do return strerror_ext
.to_s
1835 # Fill `s` with the digits in base `base` of `self` (and with the '-' sign if 'signed' and negative).
1836 # assume < to_c max const of char
1837 private fun fill_buffer
(s
: Buffer, base
: Int, signed
: Bool)
1844 else if self == 0 then
1851 var pos
= digit_count
(base
) - 1
1852 while pos
>= 0 and n
> 0 do
1853 s
.chars
[pos
] = (n
% base
).to_c
1859 # C function to convert an nit Int to a NativeString (char*)
1860 private fun native_int_to_s
: NativeString is extern "native_int_to_s"
1862 # return displayable int in base 10 and signed
1864 # assert 1.to_s == "1"
1865 # assert (-123).to_s == "-123"
1867 return native_int_to_s
.to_s
1870 # return displayable int in hexadecimal
1872 # assert 1.to_hex == "1"
1873 # assert (-255).to_hex == "-ff"
1874 fun to_hex
: String do return to_base
(16,false)
1876 # return displayable int in base base and signed
1877 fun to_base
(base
: Int, signed
: Bool): String
1879 var l
= digit_count
(base
)
1880 var s
= new FlatBuffer.from
(" " * l
)
1881 fill_buffer
(s
, base
, signed
)
1887 # Pretty representation of `self`, with decimals as needed from 1 to a maximum of 3
1889 # assert 12.34.to_s == "12.34"
1890 # assert (-0120.030).to_s == "-120.03"
1892 # see `to_precision` for a custom precision.
1894 var str
= to_precision
( 3 )
1895 if is_inf
!= 0 or is_nan
then return str
1896 var len
= str
.length
1897 for i
in [0..len-1
] do
1899 var c
= str
.chars
[j
]
1902 else if c
== '.' then
1903 return str
.substring
( 0, j
+2 )
1905 return str
.substring
( 0, j
+1 )
1911 # `String` representation of `self` with the given number of `decimals`
1913 # assert 12.345.to_precision(0) == "12"
1914 # assert 12.345.to_precision(3) == "12.345"
1915 # assert (-12.345).to_precision(3) == "-12.345"
1916 # assert (-0.123).to_precision(3) == "-0.123"
1917 # assert 0.999.to_precision(2) == "1.00"
1918 # assert 0.999.to_precision(4) == "0.9990"
1919 fun to_precision
(decimals
: Int): String
1921 if is_nan
then return "nan"
1923 var isinf
= self.is_inf
1926 else if isinf
== -1 then
1930 if decimals
== 0 then return self.to_i
.to_s
1932 for i
in [0..decimals
[ do f
= f
* 10.0
1939 if i
== 0 then return "0." + "0"*decimals
1941 # Prepare both parts of the float, before and after the "."
1946 if sl
> decimals
then
1947 # Has something before the "."
1948 p1
= s
.substring
(0, sl-decimals
)
1949 p2
= s
.substring
(sl-decimals
, decimals
)
1952 p2
= "0"*(decimals-sl
) + s
1955 if i
< 0 then p1
= "-" + p1
1957 return p1
+ "." + p2
1960 # `self` representation with `nb` digits after the '.'.
1962 # assert 12.345.to_precision_native(1) == "12.3"
1963 # assert 12.345.to_precision_native(2) == "12.35"
1964 # assert 12.345.to_precision_native(3) == "12.345"
1965 # assert 12.345.to_precision_native(4) == "12.3450"
1966 fun to_precision_native
(nb
: Int): String import NativeString.to_s
`{
1970 size = snprintf(NULL, 0, "%.*f", (int)nb, recv);
1971 str = malloc(size + 1);
1972 sprintf(str, "%.*f", (int)nb, recv );
1974 return NativeString_to_s( str );
1979 # assert 'x'.to_s == "x"
1982 var s
= new FlatBuffer.with_capacity
(1)
1987 # Returns true if the char is a numerical digit
1989 # assert '0'.is_numeric
1990 # assert '9'.is_numeric
1991 # assert not 'a'.is_numeric
1992 # assert not '?'.is_numeric
1993 fun is_numeric
: Bool
1995 return self >= '0' and self <= '9'
1998 # Returns true if the char is an alpha digit
2000 # assert 'a'.is_alpha
2001 # assert 'Z'.is_alpha
2002 # assert not '0'.is_alpha
2003 # assert not '?'.is_alpha
2006 return (self >= 'a' and self <= 'z') or (self >= 'A' and self <= 'Z')
2009 # Returns true if the char is an alpha or a numeric digit
2011 # assert 'a'.is_alphanumeric
2012 # assert 'Z'.is_alphanumeric
2013 # assert '0'.is_alphanumeric
2014 # assert '9'.is_alphanumeric
2015 # assert not '?'.is_alphanumeric
2016 fun is_alphanumeric
: Bool
2018 return self.is_numeric
or self.is_alpha
2022 redef class Collection[E
]
2023 # Concatenate elements.
2026 var s
= new FlatBuffer
2027 for e
in self do if e
!= null then s
.append
(e
.to_s
)
2031 # Concatenate and separate each elements with `sep`.
2033 # assert [1, 2, 3].join(":") == "1:2:3"
2034 # assert [1..3].join(":") == "1:2:3"
2035 fun join
(sep
: Text): String
2037 if is_empty
then return ""
2039 var s
= new FlatBuffer # Result
2044 if e
!= null then s
.append
(e
.to_s
)
2046 # Concat other items
2051 if e
!= null then s
.append
(e
.to_s
)
2058 redef class Array[E
]
2060 # Fast implementation
2064 if l
== 0 then return ""
2065 if l
== 1 then if self[0] == null then return "" else return self[0].to_s
2067 var na
= new NativeArray[String](l
)
2073 if itsi
== null then
2083 var ns
= new NativeString(sl
+ 1)
2089 var tpl
= tmp
.length
2090 if tmp
isa FlatString then
2091 tmp
.items
.copy_to
(ns
, tpl
, tmp
.index_from
, off
)
2094 for j
in tmp
.substrings
do
2095 var s
= j
.as(FlatString)
2097 s
.items
.copy_to
(ns
, slen
, s
.index_from
, off
)
2103 return ns
.to_s_with_length
(sl
)
2107 redef class Map[K
,V
]
2108 # Concatenate couple of 'key value'.
2109 # key and value are separated by `couple_sep`.
2110 # each couple is separated each couple with `sep`.
2112 # var m = new ArrayMap[Int, String]
2115 # assert m.join("; ", "=") == "1=one; 10=ten"
2116 fun join
(sep
: String, couple_sep
: String): String
2118 if is_empty
then return ""
2120 var s
= new FlatBuffer # Result
2126 s
.append
("{k}{couple_sep}{e or else "<null>"}")
2128 # Concat other items
2134 s
.append
("{k}{couple_sep}{e or else "<null>"}")
2141 ###############################################################################
2143 ###############################################################################
2145 # Native strings are simple C char *
2146 extern class NativeString `{ char* `}
2147 # Creates a new NativeString with a capacity of `length
`
2148 new(length: Int) is intern
2150 # Get char at `index
`.
2151 fun [](index: Int): Char is intern
2153 # Set char `item
` at index.
2154 fun []=(index: Int, item: Char) is intern
2156 # Copy `self` to `dest
`.
2157 fun copy_to(dest: NativeString, length: Int, from: Int, to: Int) is intern
2159 # Position of the first nul character.
2160 fun cstring_length: Int
2163 while self[l] != '\0' do l += 1
2167 # Parse `self` as an Int.
2168 fun atoi: Int is intern
2170 # Parse `self` as a Float.
2171 fun atof: Float is extern "atof"
2175 return to_s_with_length(cstring_length)
2178 # Returns `self` as a String of `length
`.
2179 fun to_s_with_length(length: Int): FlatString
2182 var str = new FlatString.with_infos(self, length, 0, length - 1)
2186 # Returns `self` as a new String.
2187 fun to_s_with_copy: FlatString
2189 var length = cstring_length
2190 var new_self = new NativeString(length + 1)
2191 copy_to(new_self, length, 0, 0)
2192 var str = new FlatString.with_infos(new_self, length, 0, length - 1)
2193 new_self[length] = '\0'
2194 str.real_items = new_self
2200 private var args_cache: nullable Sequence[String]
2202 # The arguments of the program as given by the OS
2203 fun program_args: Sequence[String]
2205 if _args_cache == null then init_args
2206 return _args_cache.as(not null)
2209 # The name of the program as given by the OS
2210 fun program_name: String
2212 return native_argv(0).to_s
2215 # Initialize `program_args
` with the contents of `native_argc
` and `native_argv
`.
2216 private fun init_args
2218 var argc = native_argc
2219 var args = new Array[String].with_capacity(0)
2222 args[i-1] = native_argv(i).to_s
2228 # First argument of the main C function.
2229 private fun native_argc: Int is intern
2231 # Second argument of the main C function.
2232 private fun native_argv(i: Int): NativeString is intern
2235 # Comparator that efficienlty use `to_s
` to compare things
2237 # The comparaison call `to_s
` on object and use the result to order things.
2239 # var a = [1, 2, 3, 10, 20]
2240 # (new CachedAlphaComparator).sort(a)
2241 # assert a == [1, 10, 2, 20, 3]
2243 # Internally the result of `to_s
` is cached in a HashMap to counter
2244 # uneficient implementation of `to_s
`.
2246 # Note: it caching is not usefull, see `alpha_comparator
`
2247 class CachedAlphaComparator
2249 redef type COMPARED: Object
2251 private var cache = new HashMap[Object, String]
2253 private fun do_to_s(a: Object): String do
2254 if cache.has_key(a) then return cache[a]
2260 redef fun compare(a, b) do
2261 return do_to_s(a) <=> do_to_s(b)
2265 # see `alpha_comparator
`
2266 private class AlphaComparator
2268 redef fun compare(a, b) do return a.to_s <=> b.to_s
2271 # Stateless comparator that naively use `to_s
` to compare things.
2273 # Note: the result of `to_s
` is not cached, thus can be invoked a lot
2274 # on a single instace. See `CachedAlphaComparator` as an alternative.
2276 # var a = [1, 2, 3, 10, 20]
2277 # alpha_comparator.sort(a)
2278 # assert a == [1, 10, 2, 20, 3]
2279 fun alpha_comparator: Comparator do return once new AlphaComparator
2281 # The arguments of the program as given by the OS
2282 fun args: Sequence[String]
2284 return sys.program_args