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
18 intrude import collection
# FIXME should be collection::array
25 ###############################################################################
27 ###############################################################################
29 # 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
159 # Is 'c' contained in self ?
161 # DEPRECATED : Use self.chars.has instead
162 fun has
(c
: Char): Bool
164 return self.chars
.has
(c
)
167 # Gets an Array containing the chars of self
169 # DEPRECATED : Use self.chars.to_a instead
170 fun to_a
: Array[Char] do return chars
.to_a
172 # Create a substring from `self` beginning at the `from` position
174 # assert "abcd".substring_from(1) == "bcd"
175 # assert "abcd".substring_from(-1) == "abcd"
176 # assert "abcd".substring_from(2) == "cd"
178 # As with substring, a `from` index < 0 will be replaced by 0
179 fun substring_from
(from
: Int): SELFTYPE
181 if from
> self.length
then return empty
182 if from
< 0 then from
= 0
183 return substring
(from
, length
- from
)
186 # Does self have a substring `str` starting from position `pos`?
188 # assert "abcd".has_substring("bc",1) == true
189 # assert "abcd".has_substring("bc",2) == false
190 fun has_substring
(str
: String, pos
: Int): Bool
192 var myiter
= self.chars
.iterator_from
(pos
)
193 var itsiter
= str
.chars
.iterator
194 while myiter
.is_ok
and itsiter
.is_ok
do
195 if myiter
.item
!= itsiter
.item
then return false
199 if itsiter
.is_ok
then return false
203 # Is this string prefixed by `prefix`?
205 # assert "abcd".has_prefix("ab") == true
206 # assert "abcbc".has_prefix("bc") == false
207 # assert "ab".has_prefix("abcd") == false
208 fun has_prefix
(prefix
: String): Bool do return has_substring
(prefix
,0)
210 # Is this string suffixed by `suffix`?
212 # assert "abcd".has_suffix("abc") == false
213 # assert "abcd".has_suffix("bcd") == true
214 fun has_suffix
(suffix
: String): Bool do return has_substring
(suffix
, length
- suffix
.length
)
216 # If `self` contains only digits, return the corresponding integer
218 # assert "123".to_i == 123
219 # assert "-1".to_i == -1
223 return to_s
.to_cstring
.atoi
226 # If `self` contains a float, return the corresponding float
228 # assert "123".to_f == 123.0
229 # assert "-1".to_f == -1.0
230 # assert "-1.2e-3".to_f == -0.0012
234 return to_s
.to_cstring
.atof
237 # If `self` contains only digits and alpha <= 'f', return the corresponding integer.
239 # assert "ff".to_hex == 255
240 fun to_hex
: Int do return a_to
(16)
242 # If `self` contains only digits and letters, return the corresponding integer in a given base
244 # assert "120".a_to(3) == 15
245 fun a_to
(base
: Int) : Int
272 # Returns `true` if the string contains only Numeric values (and one "," or one "." character)
274 # assert "123".is_numeric == true
275 # assert "1.2".is_numeric == true
276 # assert "1,2".is_numeric == true
277 # assert "1..2".is_numeric == false
280 var has_point_or_comma
= false
285 if (i
== '.' or i
== ',') and not has_point_or_comma
287 has_point_or_comma
= true
296 # Returns `true` if the string contains only Hex chars
298 # assert "048bf".is_hex == true
299 # assert "ABCDEF".is_hex == true
300 # assert "0G".is_hex == false
303 for c
in self.chars
do
304 if not (c
>= 'a' and c
<= 'f') and
305 not (c
>= 'A' and c
<= 'F') and
306 not (c
>= '0' and c
<= '9') then return false
311 # Are all letters in `self` upper-case ?
313 # assert "HELLO WORLD".is_upper == true
314 # assert "%$&%!".is_upper == true
315 # assert "hello world".is_upper == false
316 # assert "Hello World".is_upper == false
319 for char
in self.chars
do
320 if char
.is_lower
then return false
325 # Are all letters in `self` lower-case ?
327 # assert "hello world".is_lower == true
328 # assert "%$&%!".is_lower == true
329 # assert "Hello World".is_lower == false
332 for char
in self.chars
do
333 if char
.is_upper
then return false
338 # Removes the whitespaces at the beginning of self
340 # assert " \n\thello \n\t".l_trim == "hello \n\t"
342 # A whitespace is defined as any character which ascii value is less than or equal to 32
345 var iter
= self.chars
.iterator
347 if iter
.item
.ascii
> 32 then break
350 if iter
.index
== length
then return self.empty
351 return self.substring_from
(iter
.index
)
354 # Removes the whitespaces at the end of self
356 # assert " \n\thello \n\t".r_trim == " \n\thello"
358 # A whitespace is defined as any character which ascii value is less than or equal to 32
361 var iter
= self.chars
.reverse_iterator
363 if iter
.item
.ascii
> 32 then break
366 if iter
.index
== length
then return self.empty
367 return self.substring
(0, iter
.index
+ 1)
370 # Trims trailing and preceding white spaces
371 # A whitespace is defined as any character which ascii value is less than or equal to 32
373 # assert " Hello World ! ".trim == "Hello World !"
374 # assert "\na\nb\tc\t".trim == "a\nb\tc"
375 fun trim
: SELFTYPE do return (self.l_trim
).r_trim
377 # Mangle a string to be a unique string only made of alphanumeric characters
378 fun to_cmangle
: String
380 var res
= new FlatBuffer
381 var underscore
= false
382 for c
in self.chars
do
383 if (c
>= 'a' and c
<= 'z') or (c
>='A' and c
<= 'Z') then
389 res
.append
('_'.ascii
.to_s
)
392 if c
>= '0' and c
<= '9' then
395 else if c
== '_' then
400 res
.append
(c
.ascii
.to_s
)
408 # Escape " \ ' and non printable characters using the rules of literal C strings and characters
410 # assert "abAB12<>&".escape_to_c == "abAB12<>&"
411 # assert "\n\"'\\".escape_to_c == "\\n\\\"\\'\\\\"
412 fun escape_to_c
: String
414 var b
= new FlatBuffer
415 for c
in self.chars
do
418 else if c
== '\0' then
420 else if c
== '"' then
422 else if c == '\'' then
424 else if c == '\\
' then
426 else if c.ascii < 32 then
427 b.append("\\{c.ascii.to_base(8, false)}")
435 # Escape additionnal characters
436 # The result might no be legal in C but be used in other languages
438 # assert "ab|\{\}".escape_more_to_c("|\{\}") == "ab\\|\\\{\\\}"
439 fun escape_more_to_c(chars: String): String
441 var b = new FlatBuffer
442 for c in escape_to_c.chars do
443 if chars.chars.has(c) then
451 # Escape to C plus braces
453 # assert "\n\"'\\\
{\}".escape_to_nit == "\\n\\\
"\\'\\\\\\\{\\\}"
454 fun escape_to_nit
: String do return escape_more_to_c
("\{\}")
456 # Return a string where Nit escape sequences are transformed.
459 # assert s.length == 2
460 # var u = s.unescape_nit
461 # assert u.length == 1
462 # assert u.chars[0].ascii == 10 # (the ASCII value of the "new line" character)
463 fun unescape_nit
: String
465 var res
= new FlatBuffer.with_capacity
(self.length
)
466 var was_slash
= false
468 if not was_slash
then
479 else if c
== 'r' then
481 else if c
== 't' then
483 else if c
== '0' then
492 # Encode `self` to percent (or URL) encoding
494 # assert "aBc09-._~".to_percent_encoding == "aBc09-._~"
495 # assert "%()< >".to_percent_encoding == "%25%28%29%3c%20%3e"
496 # assert ".com/post?e=asdf&f=123".to_percent_encoding == ".com%2fpost%3fe%3dasdf%26f%3d123"
497 fun to_percent_encoding
: String
499 var buf
= new FlatBuffer
501 for c
in self.chars
do
502 if (c
>= '0' and c
<= '9') or
503 (c
>= 'a' and c
<= 'z') or
504 (c
>= 'A' and c
<= 'Z') or
505 c
== '-' or c
== '.' or
509 else buf
.append
"%{c.ascii.to_hex}"
515 # Decode `self` from percent (or URL) encoding to a clear string
517 # Replace invalid use of '%' with '?'.
519 # assert "aBc09-._~".from_percent_encoding == "aBc09-._~"
520 # assert "%25%28%29%3c%20%3e".from_percent_encoding == "%()< >"
521 # assert ".com%2fpost%3fe%3dasdf%26f%3d123".from_percent_encoding == ".com/post?e=asdf&f=123"
522 # assert "%25%28%29%3C%20%3E".from_percent_encoding == "%()< >"
523 # assert "incomplete %".from_percent_encoding == "incomplete ?"
524 # assert "invalid % usage".from_percent_encoding == "invalid ? usage"
525 fun from_percent_encoding
: String
527 var buf
= new FlatBuffer
533 if i
+ 2 >= length
then
534 # What follows % has been cut off
538 var hex_s
= substring
(i
, 2)
540 var hex_i
= hex_s
.to_hex
544 # What follows a % is not Hex
558 # Two pieces of text are equals if thez have the same characters in the same order.
560 # assert "hello" == "hello"
561 # assert "hello" != "HELLO"
562 # assert "hello" == "hel"+"lo"
564 # Things that are not Text are not equal.
567 # assert "9" != ['9']
570 # assert "9".chars.first == '9' # equality of Char
571 # assert "9".chars == ['9'] # equality of Sequence
572 # assert "9".to_i == 9 # equality of Int
575 if o
== null then return false
576 if not o
isa Text then return false
577 if self.is_same_instance
(o
) then return true
578 if self.length
!= o
.length
then return false
579 return self.chars
== o
.chars
582 # Lexicographical comparaison
584 # assert "abc" < "xy"
585 # assert "ABC" < "abc"
588 var self_chars
= self.chars
.iterator
589 var other_chars
= other
.chars
.iterator
591 while self_chars
.is_ok
and other_chars
.is_ok
do
592 if self_chars
.item
< other_chars
.item
then return true
593 if self_chars
.item
> other_chars
.item
then return false
598 if self_chars
.is_ok
then
605 # Flat representation of self
606 fun flatten
: FlatText is abstract
608 private var hash_cache
: nullable Int = null
612 if hash_cache
== null then
613 # djb2 hash algorithm
616 for char
in self.chars
do
617 h
= h
.lshift
(5) + h
+ char
.ascii
622 return hash_cache
.as(not null)
627 # All kinds of array-based text representations.
628 abstract class FlatText
631 private var items
: NativeString
633 # Real items, used as cache for to_cstring is called
634 private var real_items
: nullable NativeString = null
636 redef var length
: Int = 0
649 redef fun flatten
do return self
652 # Abstract class for the SequenceRead compatible
653 # views on String and Buffer objects
654 private abstract class StringCharView
655 super SequenceRead[Char]
659 private var target
: SELFTYPE
661 private init(tgt
: SELFTYPE)
666 redef fun is_empty
do return target
.is_empty
668 redef fun length
do return target
.length
670 redef fun iterator
: IndexedIterator[Char] do return self.iterator_from
(0)
672 redef fun reverse_iterator
do return self.reverse_iterator_from
(self.length
- 1)
675 # View on Buffer objects, extends Sequence
676 # for mutation operations
677 private abstract class BufferCharView
681 redef type SELFTYPE: Buffer
685 abstract class String
688 redef type SELFTYPE: String
690 redef fun to_s
do return self
692 # Concatenates `o` to `self`
694 # assert "hello" + "world" == "helloworld"
695 # assert "" + "hello" + "" == "hello"
696 fun +(o
: Text): SELFTYPE is abstract
698 # Concatenates self `i` times
700 # assert "abc" * 4 == "abcabcabcabc"
701 # assert "abc" * 1 == "abc"
702 # assert "abc" * 0 == ""
703 fun *(i
: Int): SELFTYPE is abstract
705 fun insert_at
(s
: String, pos
: Int): SELFTYPE is abstract
707 # Returns a reversed version of self
709 # assert "hello".reversed == "olleh"
710 # assert "bob".reversed == "bob"
711 # assert "".reversed == ""
712 fun reversed
: SELFTYPE is abstract
714 # A upper case version of `self`
716 # assert "Hello World!".to_upper == "HELLO WORLD!"
717 fun to_upper
: SELFTYPE is abstract
719 # A lower case version of `self`
721 # assert "Hello World!".to_lower == "hello world!"
722 fun to_lower
: SELFTYPE is abstract
724 # Takes a camel case `self` and converts it to snake case
726 # assert "randomMethodId".to_snake_case == "random_method_id"
728 # If `self` is upper, it is returned unchanged
730 # assert "RANDOM_METHOD_ID".to_snake_case == "RANDOM_METHOD_ID"
732 # If the identifier is prefixed by an underscore, the underscore is ignored
734 # assert "_privateField".to_snake_case == "_private_field"
735 fun to_snake_case
: SELFTYPE
737 if self.is_upper
then return self
739 var new_str
= new FlatBuffer.with_capacity
(self.length
)
740 var is_first_char
= true
742 for char
in self.chars
do
743 if is_first_char
then
744 new_str
.add
(char
.to_lower
)
745 is_first_char
= false
746 else if char
.is_upper
then
748 new_str
.add
(char
.to_lower
)
757 # Takes a snake case `self` and converts it to camel case
759 # assert "random_method_id".to_camel_case == "randomMethodId"
761 # If the identifier is prefixed by an underscore, the underscore is ignored
763 # assert "_private_field".to_camel_case == "_privateField"
765 # If `self` is upper, it is returned unchanged
767 # assert "RANDOM_ID".to_camel_case == "RANDOM_ID"
769 # If there are several consecutive underscores, they are considered as a single one
771 # assert "random__method_id".to_camel_case == "randomMethodId"
772 fun to_camel_case
: SELFTYPE
774 if self.is_upper
then return self
776 var new_str
= new FlatBuffer
777 var is_first_char
= true
778 var follows_us
= false
780 for char
in self.chars
do
781 if is_first_char
then
783 is_first_char
= false
784 else if char
== '_' then
786 else if follows_us
then
787 new_str
.add
(char
.to_upper
)
798 private class FlatSubstringsIter
799 super Iterator[FlatText]
801 var tgt
: nullable FlatText
803 init(tgt
: FlatText) do self.tgt
= tgt
807 return tgt
.as(not null)
810 redef fun is_ok
do return tgt
!= null
812 redef fun next
do tgt
= null
815 # Immutable strings of characters.
820 # Index in _items of the start of the string
821 private var index_from
: Int
823 # Indes in _items of the last item of the string
824 private var index_to
: Int
826 redef var chars
: SequenceRead[Char] = new FlatStringCharView(self)
828 ################################################
829 # AbstractString specific methods #
830 ################################################
834 var native
= calloc_string
(self.length
+ 1)
835 var length
= self.length
836 var items
= self.items
839 while pos
< length
do
840 native
[pos
] = items
[ipos
]
844 return native
.to_s_with_length
(self.length
)
847 redef fun substring
(from
, count
)
853 if count
< 0 then count
= 0
857 var realFrom
= index_from
+ from
859 if (realFrom
+ count
) > index_to
then return new FlatString.with_infos
(items
, index_to
- realFrom
+ 1, realFrom
, index_to
)
861 if count
== 0 then return empty
863 var to
= realFrom
+ count
- 1
865 return new FlatString.with_infos
(items
, to
- realFrom
+ 1, realFrom
, to
)
868 redef fun empty
do return "".as(FlatString)
872 var outstr
= calloc_string
(self.length
+ 1)
875 var myitems
= self.items
876 var index_from
= self.index_from
877 var max
= self.index_to
879 while index_from
<= max
do
880 outstr
[out_index
] = myitems
[index_from
].to_upper
885 outstr
[self.length
] = '\0'
887 return outstr
.to_s_with_length
(self.length
)
892 var outstr
= calloc_string
(self.length
+ 1)
895 var myitems
= self.items
896 var index_from
= self.index_from
897 var max
= self.index_to
899 while index_from
<= max
do
900 outstr
[out_index
] = myitems
[index_from
].to_lower
905 outstr
[self.length
] = '\0'
907 return outstr
.to_s_with_length
(self.length
)
912 var i
= self.index_from
913 var imax
= self.index_to
920 ##################################################
921 # String Specific Methods #
922 ##################################################
924 private init with_infos
(items
: NativeString, len
: Int, from
: Int, to
: Int)
932 redef fun to_cstring
: NativeString
934 if real_items
!= null then return real_items
.as(not null)
935 if index_from
> 0 or index_to
!= items
.cstring_length
- 1 then
936 var newItems
= calloc_string
(length
+ 1)
937 self.items
.copy_to
(newItems
, length
, index_from
, 0)
938 newItems
[length
] = '\0'
939 self.real_items
= newItems
947 if not other
isa FlatString then return super
949 if self.object_id
== other
.object_id
then return true
951 var my_length
= length
953 if other
.length
!= my_length
then return false
955 var my_index
= index_from
956 var its_index
= other
.index_from
958 var last_iteration
= my_index
+ my_length
960 var itsitems
= other
.items
961 var myitems
= self.items
963 while my_index
< last_iteration
do
964 if myitems
[my_index
] != itsitems
[its_index
] then return false
974 if not other
isa FlatString then return super
976 if self.object_id
== other
.object_id
then return false
978 var my_curr_char
: Char
979 var its_curr_char
: Char
981 var curr_id_self
= self.index_from
982 var curr_id_other
= other
.index_from
984 var my_items
= self.items
985 var its_items
= other
.items
987 var my_length
= self.length
988 var its_length
= other
.length
990 var max_iterations
= curr_id_self
+ my_length
992 while curr_id_self
< max_iterations
do
993 my_curr_char
= my_items
[curr_id_self
]
994 its_curr_char
= its_items
[curr_id_other
]
996 if my_curr_char
!= its_curr_char
then
997 if my_curr_char
< its_curr_char
then return true
1005 return my_length
< its_length
1010 var my_length
= self.length
1011 var its_length
= s
.length
1013 var total_length
= my_length
+ its_length
1015 var target_string
= calloc_string
(my_length
+ its_length
+ 1)
1017 self.items
.copy_to
(target_string
, my_length
, index_from
, 0)
1018 if s
isa FlatString then
1019 s
.items
.copy_to
(target_string
, its_length
, s
.index_from
, my_length
)
1020 else if s
isa FlatBuffer then
1021 s
.items
.copy_to
(target_string
, its_length
, 0, my_length
)
1023 var curr_pos
= my_length
1025 target_string
[curr_pos
] = i
1030 target_string
[total_length
] = '\0'
1032 return target_string
.to_s_with_length
(total_length
)
1039 var my_length
= self.length
1041 var final_length
= my_length
* i
1043 var my_items
= self.items
1045 var target_string
= calloc_string
((final_length
) + 1)
1047 target_string
[final_length
] = '\0'
1049 var current_last
= 0
1051 for iteration
in [1 .. i
] do
1052 my_items
.copy_to
(target_string
, my_length
, 0, current_last
)
1053 current_last
+= my_length
1056 return target_string
.to_s_with_length
(final_length
)
1061 if hash_cache
== null then
1062 # djb2 hash algorithm
1068 while i
<= index_to
do
1069 h
= h
.lshift
(5) + h
+ myitems
[i
].ascii
1076 return hash_cache
.as(not null)
1079 redef fun substrings
do return new FlatSubstringsIter(self)
1082 private class FlatStringReverseIterator
1083 super IndexedIterator[Char]
1085 var target
: FlatString
1087 var target_items
: NativeString
1091 init with_pos
(tgt
: FlatString, pos
: Int)
1094 target_items
= tgt
.items
1095 curr_pos
= pos
+ tgt
.index_from
1098 redef fun is_ok
do return curr_pos
>= 0
1100 redef fun item
do return target_items
[curr_pos
]
1102 redef fun next
do curr_pos
-= 1
1104 redef fun index
do return curr_pos
- target
.index_from
1108 private class FlatStringIterator
1109 super IndexedIterator[Char]
1111 var target
: FlatString
1113 var target_items
: NativeString
1117 init with_pos
(tgt
: FlatString, pos
: Int)
1120 target_items
= tgt
.items
1121 curr_pos
= pos
+ target
.index_from
1124 redef fun is_ok
do return curr_pos
<= target
.index_to
1126 redef fun item
do return target_items
[curr_pos
]
1128 redef fun next
do curr_pos
+= 1
1130 redef fun index
do return curr_pos
- target
.index_from
1134 private class FlatStringCharView
1135 super StringCharView
1137 redef type SELFTYPE: FlatString
1141 # Check that the index (+ index_from) is not larger than indexTo
1142 # In other terms, if the index is valid
1144 var target
= self.target
1145 assert (index
+ target
.index_from
) <= target
.index_to
1146 return target
.items
[index
+ target
.index_from
]
1149 redef fun iterator_from
(start
) do return new FlatStringIterator.with_pos
(target
, start
)
1151 redef fun reverse_iterator_from
(start
) do return new FlatStringReverseIterator.with_pos
(target
, start
)
1155 abstract class Buffer
1158 redef type SELFTYPE: Buffer
1160 # Specific implementations MUST set this to `true` in order to invalidate caches
1161 protected var is_dirty
= true
1163 # Modifies the char contained at pos `index`
1165 # DEPRECATED : Use self.chars.[]= instead
1166 fun []=(index
: Int, item
: Char) is abstract
1168 # Adds a char `c` at the end of self
1170 # DEPRECATED : Use self.chars.add instead
1171 fun add
(c
: Char) is abstract
1175 # var b = new FlatBuffer
1177 # assert not b.is_empty
1180 fun clear
is abstract
1182 # Enlarges the subsequent array containing the chars of self
1183 fun enlarge
(cap
: Int) is abstract
1185 # Adds the content of text `s` at the end of self
1187 # var b = new FlatBuffer
1190 # assert b == "helloworld"
1191 fun append
(s
: Text) is abstract
1193 # `self` is appended in such a way that `self` is repeated `r` times
1195 # var b = new FlatBuffer
1198 # assert b == "hellohellohello"
1199 fun times
(r
: Int) is abstract
1201 # Reverses itself in-place
1203 # var b = new FlatBuffer
1206 # assert b == "olleh"
1207 fun reverse
is abstract
1209 # Changes each lower-case char in `self` by its upper-case variant
1211 # var b = new FlatBuffer
1212 # b.append("Hello World!")
1214 # assert b == "HELLO WORLD!"
1215 fun upper
is abstract
1217 # Changes each upper-case char in `self` by its lower-case variant
1219 # var b = new FlatBuffer
1220 # b.append("Hello World!")
1222 # assert b == "hello world!"
1223 fun lower
is abstract
1227 if is_dirty
then hash_cache
= null
1231 # In Buffers, the internal sequence of character is mutable
1232 # Thus, `chars` can be used to modify the buffer.
1233 redef fun chars
: Sequence[Char] is abstract
1236 # Mutable strings of characters.
1241 redef type SELFTYPE: FlatBuffer
1243 redef var chars
: Sequence[Char] = new FlatBufferCharView(self)
1245 private var capacity
: Int = 0
1247 redef fun substrings
do return new FlatSubstringsIter(self)
1249 redef fun []=(index
, item
)
1252 if index
== length
then
1256 assert index
>= 0 and index
< length
1263 if capacity
<= length
then enlarge
(length
+ 5)
1273 redef fun empty
do return new FlatBuffer
1275 redef fun enlarge
(cap
)
1278 if cap
<= c
then return
1279 while c
<= cap
do c
= c
* 2 + 2
1280 var a
= calloc_string
(c
+1)
1281 if length
> 0 then items
.copy_to
(a
, length
, 0, 0)
1286 redef fun to_s
: String
1288 return to_cstring
.to_s_with_length
(length
)
1291 redef fun to_cstring
1294 var new_native
= calloc_string
(length
+ 1)
1295 new_native
[length
] = '\0'
1296 if length
> 0 then items
.copy_to
(new_native
, length
, 0, 0)
1297 real_items
= new_native
1300 return real_items
.as(not null)
1303 # Create a new empty string.
1308 capacity
= s
.length
+ 1
1310 items
= calloc_string
(capacity
)
1311 if s
isa FlatString then
1312 s
.items
.copy_to
(items
, length
, s
.index_from
, 0)
1313 else if s
isa FlatBuffer then
1314 s
.items
.copy_to
(items
, length
, 0, 0)
1324 # Create a new empty string with a given capacity.
1325 init with_capacity
(cap
: Int)
1328 # _items = new NativeString.calloc(cap)
1329 items
= calloc_string
(cap
+1)
1336 if s
.is_empty
then return
1339 if capacity
< length
+ sl
then enlarge
(length
+ sl
)
1340 if s
isa FlatString then
1341 s
.items
.copy_to
(items
, sl
, s
.index_from
, length
)
1342 else if s
isa FlatBuffer then
1343 s
.items
.copy_to
(items
, sl
, 0, length
)
1345 var curr_pos
= self.length
1354 # Copies the content of self in `dest`
1355 fun copy
(start
: Int, len
: Int, dest
: Buffer, new_start
: Int)
1357 var self_chars
= self.chars
1358 var dest_chars
= dest
.chars
1359 for i
in [0..len-1
] do
1360 dest_chars
[new_start
+i
] = self_chars
[start
+i
]
1364 redef fun substring
(from
, count
)
1368 if from
< 0 then from
= 0
1369 if count
> length
then count
= length
1370 if from
< count
then
1371 var r
= new FlatBuffer.with_capacity
(count
- from
)
1372 while from
< count
do
1373 r
.chars
.push
(items
[from
])
1378 return new FlatBuffer
1384 var ns
= calloc_string
(capacity
)
1396 redef fun times
(repeats
)
1398 var x
= new FlatString.with_infos
(items
, length
, 0, length
- 1)
1399 for i
in [1..repeats
[ do
1409 it
[id
] = it
[id
].to_upper
1419 it
[id
] = it
[id
].to_lower
1425 private class FlatBufferReverseIterator
1426 super IndexedIterator[Char]
1428 var target
: FlatBuffer
1430 var target_items
: NativeString
1434 init with_pos
(tgt
: FlatBuffer, pos
: Int)
1437 if tgt
.length
> 0 then target_items
= tgt
.items
1441 redef fun index
do return curr_pos
1443 redef fun is_ok
do return curr_pos
>= 0
1445 redef fun item
do return target_items
[curr_pos
]
1447 redef fun next
do curr_pos
-= 1
1451 private class FlatBufferCharView
1452 super BufferCharView
1455 redef type SELFTYPE: FlatBuffer
1457 redef fun [](index
) do return target
.items
[index
]
1459 redef fun []=(index
, item
)
1461 assert index
>= 0 and index
<= length
1462 if index
== length
then
1466 target
.items
[index
] = item
1479 fun enlarge
(cap
: Int)
1486 var my_items
= target
.items
1487 var s_length
= s
.length
1488 if target
.capacity
< s
.length
then enlarge
(s_length
+ target
.length
)
1491 redef fun iterator_from
(pos
) do return new FlatBufferIterator.with_pos
(target
, pos
)
1493 redef fun reverse_iterator_from
(pos
) do return new FlatBufferReverseIterator.with_pos
(target
, pos
)
1497 private class FlatBufferIterator
1498 super IndexedIterator[Char]
1500 var target
: FlatBuffer
1502 var target_items
: NativeString
1506 init with_pos
(tgt
: FlatBuffer, pos
: Int)
1509 if tgt
.length
> 0 then target_items
= tgt
.items
1513 redef fun index
do return curr_pos
1515 redef fun is_ok
do return curr_pos
< target
.length
1517 redef fun item
do return target_items
[curr_pos
]
1519 redef fun next
do curr_pos
+= 1
1523 ###############################################################################
1525 ###############################################################################
1528 # User readable representation of `self`.
1529 fun to_s
: String do return inspect
1531 # The class name of the object in NativeString format.
1532 private fun native_class_name
: NativeString is intern
1534 # The class name of the object.
1536 # assert 5.class_name == "Int"
1537 fun class_name
: String do return native_class_name
.to_s
1539 # Developer readable representation of `self`.
1540 # Usually, it uses the form "<CLASSNAME:#OBJECTID bla bla bla>"
1543 return "<{inspect_head}>"
1546 # Return "CLASSNAME:#OBJECTID".
1547 # This function is mainly used with the redefinition of the inspect method
1548 protected fun inspect_head
: String
1550 return "{class_name}:#{object_id.to_hex}"
1555 # assert true.to_s == "true"
1556 # assert false.to_s == "false"
1569 # Wrapper of strerror C function
1570 private fun strerror_ext
: NativeString is extern `{
1571 return strerror(recv);
1574 # Returns a string describing error number
1575 fun strerror
: String do return strerror_ext
.to_s
1577 # Fill `s` with the digits in base `base` of `self` (and with the '-' sign if 'signed' and negative).
1578 # assume < to_c max const of char
1579 private fun fill_buffer
(s
: Buffer, base
: Int, signed
: Bool)
1586 else if self == 0 then
1593 var pos
= digit_count
(base
) - 1
1594 while pos
>= 0 and n
> 0 do
1595 s
.chars
[pos
] = (n
% base
).to_c
1601 # C function to convert an nit Int to a NativeString (char*)
1602 private fun native_int_to_s
: NativeString is extern "native_int_to_s"
1604 # return displayable int in base 10 and signed
1606 # assert 1.to_s == "1"
1607 # assert (-123).to_s == "-123"
1609 return native_int_to_s
.to_s
1612 # return displayable int in hexadecimal
1614 # assert 1.to_hex == "1"
1615 # assert (-255).to_hex == "-ff"
1616 fun to_hex
: String do return to_base
(16,false)
1618 # return displayable int in base base and signed
1619 fun to_base
(base
: Int, signed
: Bool): String
1621 var l
= digit_count
(base
)
1622 var s
= new FlatBuffer.from
(" " * l
)
1623 fill_buffer
(s
, base
, signed
)
1629 # Pretty print self, print needoed decimals up to a max of 3.
1631 # assert 12.34.to_s == "12.34"
1632 # assert (-0120.03450).to_s == "-120.035"
1634 # see `to_precision` for a different precision.
1636 var str
= to_precision
( 3 )
1637 if is_inf
!= 0 or is_nan
then return str
1638 var len
= str
.length
1639 for i
in [0..len-1
] do
1641 var c
= str
.chars
[j
]
1644 else if c
== '.' then
1645 return str
.substring
( 0, j
+2 )
1647 return str
.substring
( 0, j
+1 )
1653 # `self` representation with `nb` digits after the '.'.
1655 # assert 12.345.to_precision(1) == "12.3"
1656 # assert 12.345.to_precision(2) == "12.35"
1657 # assert 12.345.to_precision(3) == "12.345"
1658 # assert 12.345.to_precision(4) == "12.3450"
1659 fun to_precision
(nb
: Int): String
1661 if is_nan
then return "nan"
1663 var isinf
= self.is_inf
1666 else if isinf
== -1 then
1670 if nb
== 0 then return self.to_i
.to_s
1672 for i
in [0..nb
[ do f
= f
* 10.0
1679 if i
== 0 then return "0.0"
1683 var p1
= s
.substring
(0, s
.length-nb
)
1684 var p2
= s
.substring
(s
.length-nb
, nb
)
1685 return p1
+ "." + p2
1687 return "0." + ("0"*(nb-sl
)) + s
1691 # `self` representation with `nb` digits after the '.'.
1693 # assert 12.345.to_precision_native(1) == "12.3"
1694 # assert 12.345.to_precision_native(2) == "12.35"
1695 # assert 12.345.to_precision_native(3) == "12.345"
1696 # assert 12.345.to_precision_native(4) == "12.3450"
1697 fun to_precision_native
(nb
: Int): String import NativeString.to_s
`{
1701 size = snprintf(NULL, 0, "%.*f", (int)nb, recv);
1702 str = malloc(size + 1);
1703 sprintf(str, "%.*f", (int)nb, recv );
1705 return NativeString_to_s( str );
1710 # assert 'x'.to_s == "x"
1713 var s
= new FlatBuffer.with_capacity
(1)
1718 # Returns true if the char is a numerical digit
1720 # assert '0'.is_numeric
1721 # assert '9'.is_numeric
1722 # assert not 'a'.is_numeric
1723 # assert not '?'.is_numeric
1724 fun is_numeric
: Bool
1726 return self >= '0' and self <= '9'
1729 # Returns true if the char is an alpha digit
1731 # assert 'a'.is_alpha
1732 # assert 'Z'.is_alpha
1733 # assert not '0'.is_alpha
1734 # assert not '?'.is_alpha
1737 return (self >= 'a' and self <= 'z') or (self >= 'A' and self <= 'Z')
1740 # Returns true if the char is an alpha or a numeric digit
1742 # assert 'a'.is_alphanumeric
1743 # assert 'Z'.is_alphanumeric
1744 # assert '0'.is_alphanumeric
1745 # assert '9'.is_alphanumeric
1746 # assert not '?'.is_alphanumeric
1747 fun is_alphanumeric
: Bool
1749 return self.is_numeric
or self.is_alpha
1753 redef class Collection[E
]
1754 # Concatenate elements.
1757 var s
= new FlatBuffer
1758 for e
in self do if e
!= null then s
.append
(e
.to_s
)
1762 # Concatenate and separate each elements with `sep`.
1764 # assert [1, 2, 3].join(":") == "1:2:3"
1765 # assert [1..3].join(":") == "1:2:3"
1766 fun join
(sep
: Text): String
1768 if is_empty
then return ""
1770 var s
= new FlatBuffer # Result
1775 if e
!= null then s
.append
(e
.to_s
)
1777 # Concat other items
1782 if e
!= null then s
.append
(e
.to_s
)
1789 redef class Array[E
]
1790 # Fast implementation
1793 var s
= new FlatBuffer
1798 if e
!= null then s
.append
(e
.to_s
)
1805 redef class Map[K
,V
]
1806 # Concatenate couple of 'key value'.
1807 # key and value are separated by `couple_sep`.
1808 # each couple is separated each couple with `sep`.
1810 # var m = new ArrayMap[Int, String]
1813 # assert m.join("; ", "=") == "1=one; 10=ten"
1814 fun join
(sep
: String, couple_sep
: String): String
1816 if is_empty
then return ""
1818 var s
= new FlatBuffer # Result
1824 s
.append
("{k}{couple_sep}{e or else "<null>"}")
1826 # Concat other items
1832 s
.append
("{k}{couple_sep}{e or else "<null>"}")
1839 ###############################################################################
1841 ###############################################################################
1843 # Native strings are simple C char *
1844 extern class NativeString `{ char* `}
1847 fun [](index: Int): Char is intern
1848 fun []=(index: Int, item: Char) is intern
1849 fun copy_to(dest: NativeString, length: Int, from: Int, to: Int) is intern
1851 # Position of the first nul character.
1852 fun cstring_length: Int
1855 while self[l] != '\0' do l += 1
1858 fun atoi: Int is intern
1859 fun atof: Float is extern "atof"
1863 return to_s_with_length(cstring_length)
1866 fun to_s_with_length(length: Int): FlatString
1869 return new FlatString.with_infos(self, length, 0, length - 1)
1872 fun to_s_with_copy: FlatString
1874 var length = cstring_length
1875 var new_self = calloc_string(length + 1)
1876 copy_to(new_self, length, 0, 0)
1877 return new FlatString.with_infos(new_self, length, 0, length - 1)
1881 # StringCapable objects can create native strings
1882 interface StringCapable
1883 protected fun calloc_string(size: Int): NativeString is intern
1887 var _args_cache: nullable Sequence[String]
1889 # The arguments of the program as given by the OS
1890 fun program_args: Sequence[String]
1892 if _args_cache == null then init_args
1893 return _args_cache.as(not null)
1896 # The name of the program as given by the OS
1897 fun program_name: String
1899 return native_argv(0).to_s
1902 # Initialize `program_args
` with the contents of `native_argc
` and `native_argv
`.
1903 private fun init_args
1905 var argc = native_argc
1906 var args = new Array[String].with_capacity(0)
1909 args[i-1] = native_argv(i).to_s
1915 # First argument of the main C function.
1916 private fun native_argc: Int is intern
1918 # Second argument of the main C function.
1919 private fun native_argv(i: Int): NativeString is intern
1922 # Comparator that efficienlty use `to_s
` to compare things
1924 # The comparaison call `to_s
` on object and use the result to order things.
1926 # var a = [1, 2, 3, 10, 20]
1927 # (new CachedAlphaComparator).sort(a)
1928 # assert a == [1, 10, 2, 20, 3]
1930 # Internally the result of `to_s
` is cached in a HashMap to counter
1931 # uneficient implementation of `to_s
`.
1933 # Note: it caching is not usefull, see `alpha_comparator
`
1934 class CachedAlphaComparator
1935 super Comparator[Object]
1937 private var cache = new HashMap[Object, String]
1939 private fun do_to_s(a: Object): String do
1940 if cache.has_key(a) then return cache[a]
1946 redef fun compare(a, b) do
1947 return do_to_s(a) <=> do_to_s(b)
1951 # see `alpha_comparator
`
1952 private class AlphaComparator
1953 super Comparator[Object]
1954 redef fun compare(a, b) do return a.to_s <=> b.to_s
1957 # Stateless comparator that naively use `to_s
` to compare things.
1959 # Note: the result of `to_s
` is not cached, thus can be invoked a lot
1960 # on a single instace. See `CachedAlphaComparator` as an alternative.
1962 # var a = [1, 2, 3, 10, 20]
1963 # alpha_comparator.sort(a)
1964 # assert a == [1, 10, 2, 20, 3]
1965 fun alpha_comparator: Comparator[Object] do return once new AlphaComparator
1967 # The arguments of the program as given by the OS
1968 fun args: Sequence[String]
1970 return sys.program_args