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 # Justify a self in a space of `length`
390 # `left` is the space ratio on the left side.
391 # * 0.0 for left-justified (no space at the left)
392 # * 1.0 for right-justified (all spaces at the left)
393 # * 0.5 for centered (half the spaces at the left)
395 # assert "hello".justify(10, 0.0) == "hello "
396 # assert "hello".justify(10, 1.0) == " hello"
397 # assert "hello".justify(10, 0.5) == " hello "
399 # If `length` is not enough, `self` is returned as is.
401 # assert "hello".justify(2, 0.0) == "hello"
403 # REQUIRE: left >= 0.0 and left <= 1.0
404 # ENSURE: `self.length <= length implies result.length == length`
405 # ENSURE: `self.length >= length implies result == self
406 fun justify
(length
: Int, left
: Float): SELFTYPE
408 var diff
= length
- self.length
409 if diff
<= 0 then return self
410 assert left
>= 0.0 and left
<= 1.0
411 var before
= (diff
.to_f
* left
).to_i
412 return " " * before
+ self + " " * (diff-before
)
415 # Mangle a string to be a unique string only made of alphanumeric characters
416 fun to_cmangle
: String
418 var res
= new FlatBuffer
419 var underscore
= false
420 for i
in [0..length
[ do
422 if (c
>= 'a' and c
<= 'z') or (c
>='A' and c
<= 'Z') then
428 res
.append
('_'.ascii
.to_s
)
431 if c
>= '0' and c
<= '9' then
434 else if c
== '_' then
439 res
.append
(c
.ascii
.to_s
)
447 # Escape " \ ' and non printable characters using the rules of literal C strings and characters
449 # assert "abAB12<>&".escape_to_c == "abAB12<>&"
450 # assert "\n\"'\\".escape_to_c == "\\n\\\"\\'\\\\"
451 fun escape_to_c
: String
453 var b
= new FlatBuffer
454 for i
in [0..length
[ do
458 else if c
== '\0' then
460 else if c
== '"' then
462 else if c == '\'' then
464 else if c == '\\
' then
466 else if c.ascii < 32 then
467 b.append("\\{c.ascii.to_base(8, false)}")
475 # Escape additionnal characters
476 # The result might no be legal in C but be used in other languages
478 # assert "ab|\{\}".escape_more_to_c("|\{\}") == "ab\\|\\\{\\\}"
479 fun escape_more_to_c(chars: String): String
481 var b = new FlatBuffer
482 for c in escape_to_c.chars do
483 if chars.chars.has(c) then
491 # Escape to C plus braces
493 # assert "\n\"'\\\
{\}".escape_to_nit == "\\n\\\
"\\'\\\\\\\{\\\}"
494 fun escape_to_nit
: String do return escape_more_to_c
("\{\}")
496 # Escape to POSIX Shell (sh).
498 # Abort if the text contains a null byte.
500 # assert "\n\"'\\\{\}0".escape_to_sh == "'\n\"'\\''\\\{\}0'"
501 fun escape_to_sh
: String do
502 var b
= new FlatBuffer
504 for i in [0..length[ do
509 assert without_null_byte
: c
!= '\0'
517 # Escape to include in a Makefile
519 # Unfortunately, some characters are not escapable in Makefile.
520 # These characters are `;`, `|`, `\`, and the non-printable ones.
521 # They will be rendered as `"?{hex}"`.
522 fun escape_to_mk: String do
523 var b = new FlatBuffer
524 for i in [0..length[ do
528 else if c == ':' or c == ' ' or c == '#' then
531 else if c
.ascii
< 32 or c
== ';' or c
== '|' or c
== '\\' or c
== '=' then
532 b
.append
("?{c.ascii.to_base(16, false)}")
540 # Return a string where Nit escape sequences are transformed.
543 # assert s.length == 2
544 # var u = s.unescape_nit
545 # assert u.length == 1
546 # assert u.chars[0].ascii == 10 # (the ASCII value of the "new line" character)
547 fun unescape_nit
: String
549 var res
= new FlatBuffer.with_capacity
(self.length
)
550 var was_slash
= false
551 for i
in [0..length
[ do
553 if not was_slash
then
564 else if c
== 'r' then
566 else if c
== 't' then
568 else if c
== '0' then
577 # Encode `self` to percent (or URL) encoding
579 # assert "aBc09-._~".to_percent_encoding == "aBc09-._~"
580 # assert "%()< >".to_percent_encoding == "%25%28%29%3c%20%3e"
581 # assert ".com/post?e=asdf&f=123".to_percent_encoding == ".com%2fpost%3fe%3dasdf%26f%3d123"
582 fun to_percent_encoding
: String
584 var buf
= new FlatBuffer
586 for i
in [0..length
[ do
588 if (c
>= '0' and c
<= '9') or
589 (c
>= 'a' and c
<= 'z') or
590 (c
>= 'A' and c
<= 'Z') or
591 c
== '-' or c
== '.' or
595 else buf
.append
"%{c.ascii.to_hex}"
601 # Decode `self` from percent (or URL) encoding to a clear string
603 # Replace invalid use of '%' with '?'.
605 # assert "aBc09-._~".from_percent_encoding == "aBc09-._~"
606 # assert "%25%28%29%3c%20%3e".from_percent_encoding == "%()< >"
607 # assert ".com%2fpost%3fe%3dasdf%26f%3d123".from_percent_encoding == ".com/post?e=asdf&f=123"
608 # assert "%25%28%29%3C%20%3E".from_percent_encoding == "%()< >"
609 # assert "incomplete %".from_percent_encoding == "incomplete ?"
610 # assert "invalid % usage".from_percent_encoding == "invalid ? usage"
611 fun from_percent_encoding
: String
613 var buf
= new FlatBuffer
619 if i
+ 2 >= length
then
620 # What follows % has been cut off
624 var hex_s
= substring
(i
, 2)
626 var hex_i
= hex_s
.to_hex
630 # What follows a % is not Hex
643 # Escape the four characters `<`, `>`, `&`, and `"` with their html counterpart
645 # assert "a&b->\"x\"".html_escape == "a&b->"x""
646 fun html_escape
: SELFTYPE
648 var buf
= new FlatBuffer
650 for i
in [0..length
[ do
654 else if c
== '<' then
656 else if c
== '>' then
658 else if c
== '"' then
667 # Two pieces of text are equals if thez have the same characters in the same order.
669 # assert "hello" == "hello"
670 # assert "hello" != "HELLO"
671 # assert "hello" == "hel"+"lo"
673 # Things that are not Text are not equal.
676 # assert "9" != ['9']
679 # assert "9".chars.first == '9' # equality of Char
680 # assert "9".chars == ['9'] # equality of Sequence
681 # assert "9".to_i == 9 # equality of Int
684 if o
== null then return false
685 if not o
isa Text then return false
686 if self.is_same_instance
(o
) then return true
687 if self.length
!= o
.length
then return false
688 return self.chars
== o
.chars
691 # Lexicographical comparaison
693 # assert "abc" < "xy"
694 # assert "ABC" < "abc"
697 var self_chars
= self.chars
.iterator
698 var other_chars
= other
.chars
.iterator
700 while self_chars
.is_ok
and other_chars
.is_ok
do
701 if self_chars
.item
< other_chars
.item
then return true
702 if self_chars
.item
> other_chars
.item
then return false
707 if self_chars
.is_ok
then
714 # Escape string used in labels for graphviz
716 # assert ">><<".escape_to_dot == "\\>\\>\\<\\<"
717 fun escape_to_dot
: String
719 return escape_more_to_c
("|\{\}<>")
722 # Flat representation of self
723 fun flatten
: FlatText is abstract
725 private var hash_cache
: nullable Int = null
729 if hash_cache
== null then
730 # djb2 hash algorithm
733 for i
in [0..length
[ do
735 h
= h
.lshift
(5) + h
+ char
.ascii
740 return hash_cache
.as(not null)
745 # All kinds of array-based text representations.
746 abstract class FlatText
749 # Underlying C-String (`char*`)
751 # Warning : Might be void in some subclasses, be sure to check
752 # if set before using it.
753 private var items
: NativeString is noinit
755 # Real items, used as cache for to_cstring is called
756 private var real_items
: nullable NativeString = null
758 redef var length
: Int = 0
769 redef fun flatten
do return self
772 # Abstract class for the SequenceRead compatible
773 # views on String and Buffer objects
774 private abstract class StringCharView
775 super SequenceRead[Char]
781 redef fun is_empty
do return target
.is_empty
783 redef fun length
do return target
.length
785 redef fun iterator
: IndexedIterator[Char] do return self.iterator_from
(0)
787 redef fun reverse_iterator
do return self.reverse_iterator_from
(self.length
- 1)
790 # View on Buffer objects, extends Sequence
791 # for mutation operations
792 private abstract class BufferCharView
796 redef type SELFTYPE: Buffer
800 # A `String` holds and manipulates an arbitrary sequence of characters.
802 # String objects may be created using literals.
804 # assert "Hello World!" isa String
805 abstract class String
808 redef type SELFTYPE: String
810 redef fun to_s
do return self
812 # Concatenates `o` to `self`
814 # assert "hello" + "world" == "helloworld"
815 # assert "" + "hello" + "" == "hello"
816 fun +(o
: Text): SELFTYPE is abstract
818 # Concatenates self `i` times
820 # assert "abc" * 4 == "abcabcabcabc"
821 # assert "abc" * 1 == "abc"
822 # assert "abc" * 0 == ""
823 fun *(i
: Int): SELFTYPE is abstract
825 # Insert `s` at `pos`.
827 # assert "helloworld".insert_at(" ", 5) == "hello world"
828 fun insert_at
(s
: String, pos
: Int): SELFTYPE is abstract
830 redef fun substrings
: Iterator[String] is abstract
832 # Returns a reversed version of self
834 # assert "hello".reversed == "olleh"
835 # assert "bob".reversed == "bob"
836 # assert "".reversed == ""
837 fun reversed
: SELFTYPE is abstract
839 # A upper case version of `self`
841 # assert "Hello World!".to_upper == "HELLO WORLD!"
842 fun to_upper
: SELFTYPE is abstract
844 # A lower case version of `self`
846 # assert "Hello World!".to_lower == "hello world!"
847 fun to_lower
: SELFTYPE is abstract
849 # Takes a camel case `self` and converts it to snake case
851 # assert "randomMethodId".to_snake_case == "random_method_id"
853 # If `self` is upper, it is returned unchanged
855 # assert "RANDOM_METHOD_ID".to_snake_case == "RANDOM_METHOD_ID"
857 # If the identifier is prefixed by an underscore, the underscore is ignored
859 # assert "_privateField".to_snake_case == "_private_field"
860 fun to_snake_case
: SELFTYPE
862 if self.is_upper
then return self
864 var new_str
= new FlatBuffer.with_capacity
(self.length
)
865 var is_first_char
= true
867 for i
in [0..length
[ do
869 if is_first_char
then
870 new_str
.add
(char
.to_lower
)
871 is_first_char
= false
872 else if char
.is_upper
then
874 new_str
.add
(char
.to_lower
)
883 # Takes a snake case `self` and converts it to camel case
885 # assert "random_method_id".to_camel_case == "randomMethodId"
887 # If the identifier is prefixed by an underscore, the underscore is ignored
889 # assert "_private_field".to_camel_case == "_privateField"
891 # If `self` is upper, it is returned unchanged
893 # assert "RANDOM_ID".to_camel_case == "RANDOM_ID"
895 # If there are several consecutive underscores, they are considered as a single one
897 # assert "random__method_id".to_camel_case == "randomMethodId"
898 fun to_camel_case
: SELFTYPE
900 if self.is_upper
then return self
902 var new_str
= new FlatBuffer
903 var is_first_char
= true
904 var follows_us
= false
906 for i
in [0..length
[ do
908 if is_first_char
then
910 is_first_char
= false
911 else if char
== '_' then
913 else if follows_us
then
914 new_str
.add
(char
.to_upper
)
924 # Returns a capitalized `self`
926 # Letters that follow a letter are lowercased
927 # Letters that follow a non-letter are upcased.
929 # SEE : `Char::is_letter` for the definition of letter.
931 # assert "jAVASCRIPT".capitalized == "Javascript"
932 # assert "i am root".capitalized == "I Am Root"
933 # assert "ab_c -ab0c ab\nc".capitalized == "Ab_C -Ab0C Ab\nC"
934 fun capitalized
: SELFTYPE do
935 if length
== 0 then return self
937 var buf
= new FlatBuffer.with_capacity
(length
)
939 var curr
= chars
[0].to_upper
943 for i
in [1 .. length
[ do
946 if prev
.is_letter
then
947 buf
[i
] = curr
.to_lower
949 buf
[i
] = curr
.to_upper
957 private class FlatSubstringsIter
958 super Iterator[FlatText]
960 var tgt
: nullable FlatText
964 return tgt
.as(not null)
967 redef fun is_ok
do return tgt
!= null
969 redef fun next
do tgt
= null
972 # Immutable strings of characters.
977 # Index in _items of the start of the string
978 private var index_from
: Int is noinit
980 # Indes in _items of the last item of the string
981 private var index_to
: Int is noinit
983 redef var chars
: SequenceRead[Char] = new FlatStringCharView(self)
987 # Check that the index (+ index_from) is not larger than indexTo
988 # In other terms, if the index is valid
990 assert (index
+ index_from
) <= index_to
991 return items
[index
+ index_from
]
994 ################################################
995 # AbstractString specific methods #
996 ################################################
1000 var native
= new NativeString(self.length
+ 1)
1001 var length
= self.length
1002 var items
= self.items
1005 while pos
< length
do
1006 native
[pos
] = items
[ipos
]
1010 return native
.to_s_with_length
(self.length
)
1013 redef fun substring
(from
, count
)
1019 if count
< 0 then count
= 0
1023 var realFrom
= index_from
+ from
1025 if (realFrom
+ count
) > index_to
then return new FlatString.with_infos
(items
, index_to
- realFrom
+ 1, realFrom
, index_to
)
1027 if count
== 0 then return empty
1029 var to
= realFrom
+ count
- 1
1031 return new FlatString.with_infos
(items
, to
- realFrom
+ 1, realFrom
, to
)
1034 redef fun empty
do return "".as(FlatString)
1038 var outstr
= new NativeString(self.length
+ 1)
1041 var myitems
= self.items
1042 var index_from
= self.index_from
1043 var max
= self.index_to
1045 while index_from
<= max
do
1046 outstr
[out_index
] = myitems
[index_from
].to_upper
1051 outstr
[self.length
] = '\0'
1053 return outstr
.to_s_with_length
(self.length
)
1058 var outstr
= new NativeString(self.length
+ 1)
1061 var myitems
= self.items
1062 var index_from
= self.index_from
1063 var max
= self.index_to
1065 while index_from
<= max
do
1066 outstr
[out_index
] = myitems
[index_from
].to_lower
1071 outstr
[self.length
] = '\0'
1073 return outstr
.to_s_with_length
(self.length
)
1078 var i
= self.index_from
1079 var imax
= self.index_to
1086 ##################################################
1087 # String Specific Methods #
1088 ##################################################
1090 private init with_infos
(items
: NativeString, len
: Int, from
: Int, to
: Int)
1098 redef fun to_cstring
: NativeString
1100 if real_items
!= null then
1101 return real_items
.as(not null)
1103 var newItems
= new NativeString(length
+ 1)
1104 self.items
.copy_to
(newItems
, length
, index_from
, 0)
1105 newItems
[length
] = '\0'
1106 self.real_items
= newItems
1113 if not other
isa FlatString then return super
1115 if self.object_id
== other
.object_id
then return true
1117 var my_length
= length
1119 if other
.length
!= my_length
then return false
1121 var my_index
= index_from
1122 var its_index
= other
.index_from
1124 var last_iteration
= my_index
+ my_length
1126 var itsitems
= other
.items
1127 var myitems
= self.items
1129 while my_index
< last_iteration
do
1130 if myitems
[my_index
] != itsitems
[its_index
] then return false
1140 if not other
isa FlatString then return super
1142 if self.object_id
== other
.object_id
then return false
1144 var my_curr_char
: Char
1145 var its_curr_char
: Char
1147 var curr_id_self
= self.index_from
1148 var curr_id_other
= other
.index_from
1150 var my_items
= self.items
1151 var its_items
= other
.items
1153 var my_length
= self.length
1154 var its_length
= other
.length
1156 var max_iterations
= curr_id_self
+ my_length
1158 while curr_id_self
< max_iterations
do
1159 my_curr_char
= my_items
[curr_id_self
]
1160 its_curr_char
= its_items
[curr_id_other
]
1162 if my_curr_char
!= its_curr_char
then
1163 if my_curr_char
< its_curr_char
then return true
1171 return my_length
< its_length
1176 var my_length
= self.length
1177 var its_length
= s
.length
1179 var total_length
= my_length
+ its_length
1181 var target_string
= new NativeString(my_length
+ its_length
+ 1)
1183 self.items
.copy_to
(target_string
, my_length
, index_from
, 0)
1184 if s
isa FlatString then
1185 s
.items
.copy_to
(target_string
, its_length
, s
.index_from
, my_length
)
1186 else if s
isa FlatBuffer then
1187 s
.items
.copy_to
(target_string
, its_length
, 0, my_length
)
1189 var curr_pos
= my_length
1190 for i
in [0..s
.length
[ do
1192 target_string
[curr_pos
] = c
1197 target_string
[total_length
] = '\0'
1199 return target_string
.to_s_with_length
(total_length
)
1206 var my_length
= self.length
1208 var final_length
= my_length
* i
1210 var my_items
= self.items
1212 var target_string
= new NativeString(final_length
+ 1)
1214 target_string
[final_length
] = '\0'
1216 var current_last
= 0
1218 for iteration
in [1 .. i
] do
1219 my_items
.copy_to
(target_string
, my_length
, 0, current_last
)
1220 current_last
+= my_length
1223 return target_string
.to_s_with_length
(final_length
)
1228 if hash_cache
== null then
1229 # djb2 hash algorithm
1235 while i
<= index_to
do
1236 h
= h
.lshift
(5) + h
+ myitems
[i
].ascii
1243 return hash_cache
.as(not null)
1246 redef fun substrings
do return new FlatSubstringsIter(self)
1249 private class FlatStringReverseIterator
1250 super IndexedIterator[Char]
1252 var target
: FlatString
1254 var target_items
: NativeString
1258 init with_pos
(tgt
: FlatString, pos
: Int)
1261 target_items
= tgt
.items
1262 curr_pos
= pos
+ tgt
.index_from
1265 redef fun is_ok
do return curr_pos
>= target
.index_from
1267 redef fun item
do return target_items
[curr_pos
]
1269 redef fun next
do curr_pos
-= 1
1271 redef fun index
do return curr_pos
- target
.index_from
1275 private class FlatStringIterator
1276 super IndexedIterator[Char]
1278 var target
: FlatString
1280 var target_items
: NativeString
1284 init with_pos
(tgt
: FlatString, pos
: Int)
1287 target_items
= tgt
.items
1288 curr_pos
= pos
+ target
.index_from
1291 redef fun is_ok
do return curr_pos
<= target
.index_to
1293 redef fun item
do return target_items
[curr_pos
]
1295 redef fun next
do curr_pos
+= 1
1297 redef fun index
do return curr_pos
- target
.index_from
1301 private class FlatStringCharView
1302 super StringCharView
1304 redef type SELFTYPE: FlatString
1308 # Check that the index (+ index_from) is not larger than indexTo
1309 # In other terms, if the index is valid
1311 var target
= self.target
1312 assert (index
+ target
.index_from
) <= target
.index_to
1313 return target
.items
[index
+ target
.index_from
]
1316 redef fun iterator_from
(start
) do return new FlatStringIterator.with_pos
(target
, start
)
1318 redef fun reverse_iterator_from
(start
) do return new FlatStringReverseIterator.with_pos
(target
, start
)
1322 # A mutable sequence of characters.
1323 abstract class Buffer
1326 redef type SELFTYPE: Buffer
1328 # Specific implementations MUST set this to `true` in order to invalidate caches
1329 protected var is_dirty
= true
1331 # Copy-On-Write flag
1333 # If the `Buffer` was to_s'd, the next in-place altering
1334 # operation will cause the current `Buffer` to be re-allocated.
1336 # The flag will then be set at `false`.
1337 protected var written
= false
1339 # Modifies the char contained at pos `index`
1341 # DEPRECATED : Use self.chars.[]= instead
1342 fun []=(index
: Int, item
: Char) is abstract
1344 # Adds a char `c` at the end of self
1346 # DEPRECATED : Use self.chars.add instead
1347 fun add
(c
: Char) is abstract
1351 # var b = new FlatBuffer
1353 # assert not b.is_empty
1356 fun clear
is abstract
1358 # Enlarges the subsequent array containing the chars of self
1359 fun enlarge
(cap
: Int) is abstract
1361 # Adds the content of text `s` at the end of self
1363 # var b = new FlatBuffer
1366 # assert b == "helloworld"
1367 fun append
(s
: Text) is abstract
1369 # `self` is appended in such a way that `self` is repeated `r` times
1371 # var b = new FlatBuffer
1374 # assert b == "hellohellohello"
1375 fun times
(r
: Int) is abstract
1377 # Reverses itself in-place
1379 # var b = new FlatBuffer
1382 # assert b == "olleh"
1383 fun reverse
is abstract
1385 # Changes each lower-case char in `self` by its upper-case variant
1387 # var b = new FlatBuffer
1388 # b.append("Hello World!")
1390 # assert b == "HELLO WORLD!"
1391 fun upper
is abstract
1393 # Changes each upper-case char in `self` by its lower-case variant
1395 # var b = new FlatBuffer
1396 # b.append("Hello World!")
1398 # assert b == "hello world!"
1399 fun lower
is abstract
1401 # Capitalizes each word in `self`
1403 # Letters that follow a letter are lowercased
1404 # Letters that follow a non-letter are upcased.
1406 # SEE: `Char::is_letter` for the definition of a letter.
1408 # var b = new FlatBuffer.from("jAVAsCriPt")
1410 # assert b == "Javascript"
1411 # b = new FlatBuffer.from("i am root")
1413 # assert b == "I Am Root"
1414 # b = new FlatBuffer.from("ab_c -ab0c ab\nc")
1416 # assert b == "Ab_C -Ab0C Ab\nC"
1418 if length
== 0 then return
1419 var c
= self[0].to_upper
1422 for i
in [1 .. length
[ do
1425 if prev
.is_letter
then
1426 self[i
] = c
.to_lower
1428 self[i
] = c
.to_upper
1435 if is_dirty
then hash_cache
= null
1439 # In Buffers, the internal sequence of character is mutable
1440 # Thus, `chars` can be used to modify the buffer.
1441 redef fun chars
: Sequence[Char] is abstract
1444 # Mutable strings of characters.
1449 redef type SELFTYPE: FlatBuffer
1451 redef var chars
: Sequence[Char] = new FlatBufferCharView(self)
1453 private var capacity
: Int = 0
1455 redef fun substrings
do return new FlatSubstringsIter(self)
1457 # Re-copies the `NativeString` into a new one and sets it as the new `Buffer`
1459 # This happens when an operation modifies the current `Buffer` and
1460 # the Copy-On-Write flag `written` is set at true.
1461 private fun reset
do
1462 var nns
= new NativeString(capacity
)
1463 items
.copy_to
(nns
, length
, 0, 0)
1471 assert index
< length
1475 redef fun []=(index
, item
)
1478 if index
== length
then
1482 if written
then reset
1483 assert index
>= 0 and index
< length
1490 if capacity
<= length
then enlarge
(length
+ 5)
1497 if written
then reset
1501 redef fun empty
do return new FlatBuffer
1503 redef fun enlarge
(cap
)
1506 if cap
<= c
then return
1507 while c
<= cap
do c
= c
* 2 + 2
1508 # The COW flag can be set at false here, since
1509 # it does a copy of the current `Buffer`
1511 var a
= new NativeString(c
+1)
1512 if length
> 0 then items
.copy_to
(a
, length
, 0, 0)
1517 redef fun to_s
: String
1520 if length
== 0 then items
= new NativeString(1)
1521 return new FlatString.with_infos
(items
, length
, 0, length
- 1)
1524 redef fun to_cstring
1527 var new_native
= new NativeString(length
+ 1)
1528 new_native
[length
] = '\0'
1529 if length
> 0 then items
.copy_to
(new_native
, length
, 0, 0)
1530 real_items
= new_native
1533 return real_items
.as(not null)
1536 # Create a new empty string.
1539 # Create a new string copied from `s`.
1542 capacity
= s
.length
+ 1
1544 items
= new NativeString(capacity
)
1545 if s
isa FlatString then
1546 s
.items
.copy_to
(items
, length
, s
.index_from
, 0)
1547 else if s
isa FlatBuffer then
1548 s
.items
.copy_to
(items
, length
, 0, 0)
1551 for i
in [0..s
.length
[ do
1559 # Create a new empty string with a given capacity.
1560 init with_capacity
(cap
: Int)
1563 # _items = new NativeString.calloc(cap)
1564 items
= new NativeString(cap
+1)
1571 if s
.is_empty
then return
1574 if capacity
< length
+ sl
then enlarge
(length
+ sl
)
1575 if s
isa FlatString then
1576 s
.items
.copy_to
(items
, sl
, s
.index_from
, length
)
1577 else if s
isa FlatBuffer then
1578 s
.items
.copy_to
(items
, sl
, 0, length
)
1580 var curr_pos
= self.length
1581 for i
in [0..s
.length
[ do
1590 # Copies the content of self in `dest`
1591 fun copy
(start
: Int, len
: Int, dest
: Buffer, new_start
: Int)
1593 var self_chars
= self.chars
1594 var dest_chars
= dest
.chars
1595 for i
in [0..len-1
] do
1596 dest_chars
[new_start
+i
] = self_chars
[start
+i
]
1600 redef fun substring
(from
, count
)
1604 if from
< 0 then from
= 0
1605 if count
> length
then count
= length
1606 if from
< count
then
1607 var r
= new FlatBuffer.with_capacity
(count
- from
)
1608 while from
< count
do
1609 r
.chars
.push
(items
[from
])
1614 return new FlatBuffer
1621 var ns
= new NativeString(capacity
)
1633 redef fun times
(repeats
)
1635 var x
= new FlatString.with_infos
(items
, length
, 0, length
- 1)
1636 for i
in [1..repeats
[ do
1643 if written
then reset
1647 it
[id
] = it
[id
].to_upper
1654 if written
then reset
1658 it
[id
] = it
[id
].to_lower
1664 private class FlatBufferReverseIterator
1665 super IndexedIterator[Char]
1667 var target
: FlatBuffer
1669 var target_items
: NativeString
1673 init with_pos
(tgt
: FlatBuffer, pos
: Int)
1676 if tgt
.length
> 0 then target_items
= tgt
.items
1680 redef fun index
do return curr_pos
1682 redef fun is_ok
do return curr_pos
>= 0
1684 redef fun item
do return target_items
[curr_pos
]
1686 redef fun next
do curr_pos
-= 1
1690 private class FlatBufferCharView
1691 super BufferCharView
1693 redef type SELFTYPE: FlatBuffer
1695 redef fun [](index
) do return target
.items
[index
]
1697 redef fun []=(index
, item
)
1699 assert index
>= 0 and index
<= length
1700 if index
== length
then
1704 target
.items
[index
] = item
1717 fun enlarge
(cap
: Int)
1724 var s_length
= s
.length
1725 if target
.capacity
< s
.length
then enlarge
(s_length
+ target
.length
)
1728 redef fun iterator_from
(pos
) do return new FlatBufferIterator.with_pos
(target
, pos
)
1730 redef fun reverse_iterator_from
(pos
) do return new FlatBufferReverseIterator.with_pos
(target
, pos
)
1734 private class FlatBufferIterator
1735 super IndexedIterator[Char]
1737 var target
: FlatBuffer
1739 var target_items
: NativeString
1743 init with_pos
(tgt
: FlatBuffer, pos
: Int)
1746 if tgt
.length
> 0 then target_items
= tgt
.items
1750 redef fun index
do return curr_pos
1752 redef fun is_ok
do return curr_pos
< target
.length
1754 redef fun item
do return target_items
[curr_pos
]
1756 redef fun next
do curr_pos
+= 1
1760 ###############################################################################
1762 ###############################################################################
1765 # User readable representation of `self`.
1766 fun to_s
: String do return inspect
1768 # The class name of the object in NativeString format.
1769 private fun native_class_name
: NativeString is intern
1771 # The class name of the object.
1773 # assert 5.class_name == "Int"
1774 fun class_name
: String do return native_class_name
.to_s
1776 # Developer readable representation of `self`.
1777 # Usually, it uses the form "<CLASSNAME:#OBJECTID bla bla bla>"
1780 return "<{inspect_head}>"
1783 # Return "CLASSNAME:#OBJECTID".
1784 # This function is mainly used with the redefinition of the inspect method
1785 protected fun inspect_head
: String
1787 return "{class_name}:#{object_id.to_hex}"
1792 # assert true.to_s == "true"
1793 # assert false.to_s == "false"
1806 # Wrapper of strerror C function
1807 private fun strerror_ext
: NativeString is extern `{
1808 return strerror(recv);
1811 # Returns a string describing error number
1812 fun strerror
: String do return strerror_ext
.to_s
1814 # Fill `s` with the digits in base `base` of `self` (and with the '-' sign if 'signed' and negative).
1815 # assume < to_c max const of char
1816 private fun fill_buffer
(s
: Buffer, base
: Int, signed
: Bool)
1823 else if self == 0 then
1830 var pos
= digit_count
(base
) - 1
1831 while pos
>= 0 and n
> 0 do
1832 s
.chars
[pos
] = (n
% base
).to_c
1838 # C function to convert an nit Int to a NativeString (char*)
1839 private fun native_int_to_s
: NativeString is extern "native_int_to_s"
1841 # return displayable int in base 10 and signed
1843 # assert 1.to_s == "1"
1844 # assert (-123).to_s == "-123"
1846 return native_int_to_s
.to_s
1849 # return displayable int in hexadecimal
1851 # assert 1.to_hex == "1"
1852 # assert (-255).to_hex == "-ff"
1853 fun to_hex
: String do return to_base
(16,false)
1855 # return displayable int in base base and signed
1856 fun to_base
(base
: Int, signed
: Bool): String
1858 var l
= digit_count
(base
)
1859 var s
= new FlatBuffer.from
(" " * l
)
1860 fill_buffer
(s
, base
, signed
)
1866 # Pretty representation of `self`, with decimals as needed from 1 to a maximum of 3
1868 # assert 12.34.to_s == "12.34"
1869 # assert (-0120.030).to_s == "-120.03"
1871 # see `to_precision` for a custom precision.
1873 var str
= to_precision
( 3 )
1874 if is_inf
!= 0 or is_nan
then return str
1875 var len
= str
.length
1876 for i
in [0..len-1
] do
1878 var c
= str
.chars
[j
]
1881 else if c
== '.' then
1882 return str
.substring
( 0, j
+2 )
1884 return str
.substring
( 0, j
+1 )
1890 # `String` representation of `self` with the given number of `decimals`
1892 # assert 12.345.to_precision(0) == "12"
1893 # assert 12.345.to_precision(3) == "12.345"
1894 # assert (-12.345).to_precision(3) == "-12.345"
1895 # assert (-0.123).to_precision(3) == "-0.123"
1896 # assert 0.999.to_precision(2) == "1.00"
1897 # assert 0.999.to_precision(4) == "0.9990"
1898 fun to_precision
(decimals
: Int): String
1900 if is_nan
then return "nan"
1902 var isinf
= self.is_inf
1905 else if isinf
== -1 then
1909 if decimals
== 0 then return self.to_i
.to_s
1911 for i
in [0..decimals
[ do f
= f
* 10.0
1918 if i
== 0 then return "0." + "0"*decimals
1920 # Prepare both parts of the float, before and after the "."
1925 if sl
> decimals
then
1926 # Has something before the "."
1927 p1
= s
.substring
(0, sl-decimals
)
1928 p2
= s
.substring
(sl-decimals
, decimals
)
1931 p2
= "0"*(decimals-sl
) + s
1934 if i
< 0 then p1
= "-" + p1
1936 return p1
+ "." + p2
1939 # `self` representation with `nb` digits after the '.'.
1941 # assert 12.345.to_precision_native(1) == "12.3"
1942 # assert 12.345.to_precision_native(2) == "12.35"
1943 # assert 12.345.to_precision_native(3) == "12.345"
1944 # assert 12.345.to_precision_native(4) == "12.3450"
1945 fun to_precision_native
(nb
: Int): String import NativeString.to_s
`{
1949 size = snprintf(NULL, 0, "%.*f", (int)nb, recv);
1950 str = malloc(size + 1);
1951 sprintf(str, "%.*f", (int)nb, recv );
1953 return NativeString_to_s( str );
1958 # assert 'x'.to_s == "x"
1961 var s
= new FlatBuffer.with_capacity
(1)
1966 # Returns true if the char is a numerical digit
1968 # assert '0'.is_numeric
1969 # assert '9'.is_numeric
1970 # assert not 'a'.is_numeric
1971 # assert not '?'.is_numeric
1972 fun is_numeric
: Bool
1974 return self >= '0' and self <= '9'
1977 # Returns true if the char is an alpha digit
1979 # assert 'a'.is_alpha
1980 # assert 'Z'.is_alpha
1981 # assert not '0'.is_alpha
1982 # assert not '?'.is_alpha
1985 return (self >= 'a' and self <= 'z') or (self >= 'A' and self <= 'Z')
1988 # Returns true if the char is an alpha or a numeric digit
1990 # assert 'a'.is_alphanumeric
1991 # assert 'Z'.is_alphanumeric
1992 # assert '0'.is_alphanumeric
1993 # assert '9'.is_alphanumeric
1994 # assert not '?'.is_alphanumeric
1995 fun is_alphanumeric
: Bool
1997 return self.is_numeric
or self.is_alpha
2001 redef class Collection[E
]
2002 # Concatenate elements.
2005 var s
= new FlatBuffer
2006 for e
in self do if e
!= null then s
.append
(e
.to_s
)
2010 # Concatenate and separate each elements with `sep`.
2012 # assert [1, 2, 3].join(":") == "1:2:3"
2013 # assert [1..3].join(":") == "1:2:3"
2014 fun join
(sep
: Text): String
2016 if is_empty
then return ""
2018 var s
= new FlatBuffer # Result
2023 if e
!= null then s
.append
(e
.to_s
)
2025 # Concat other items
2030 if e
!= null then s
.append
(e
.to_s
)
2037 redef class Array[E
]
2039 # Fast implementation
2043 if l
== 0 then return ""
2044 if l
== 1 then if self[0] == null then return "" else return self[0].to_s
2046 var na
= new NativeArray[String](l
)
2052 if itsi
== null then
2062 var ns
= new NativeString(sl
+ 1)
2068 var tpl
= tmp
.length
2069 if tmp
isa FlatString then
2070 tmp
.items
.copy_to
(ns
, tpl
, tmp
.index_from
, off
)
2073 for j
in tmp
.substrings
do
2074 var s
= j
.as(FlatString)
2076 s
.items
.copy_to
(ns
, slen
, s
.index_from
, off
)
2082 return ns
.to_s_with_length
(sl
)
2086 redef class Map[K
,V
]
2087 # Concatenate couple of 'key value'.
2088 # key and value are separated by `couple_sep`.
2089 # each couple is separated each couple with `sep`.
2091 # var m = new ArrayMap[Int, String]
2094 # assert m.join("; ", "=") == "1=one; 10=ten"
2095 fun join
(sep
: String, couple_sep
: String): String
2097 if is_empty
then return ""
2099 var s
= new FlatBuffer # Result
2105 s
.append
("{k}{couple_sep}{e or else "<null>"}")
2107 # Concat other items
2113 s
.append
("{k}{couple_sep}{e or else "<null>"}")
2120 ###############################################################################
2122 ###############################################################################
2124 # Native strings are simple C char *
2125 extern class NativeString `{ char* `}
2126 # Creates a new NativeString with a capacity of `length
`
2127 new(length: Int) is intern
2129 # Get char at `index
`.
2130 fun [](index: Int): Char is intern
2132 # Set char `item
` at index.
2133 fun []=(index: Int, item: Char) is intern
2135 # Copy `self` to `dest
`.
2136 fun copy_to(dest: NativeString, length: Int, from: Int, to: Int) is intern
2138 # Position of the first nul character.
2139 fun cstring_length: Int
2142 while self[l] != '\0' do l += 1
2146 # Parse `self` as an Int.
2147 fun atoi: Int is intern
2149 # Parse `self` as a Float.
2150 fun atof: Float is extern "atof"
2154 return to_s_with_length(cstring_length)
2157 # Returns `self` as a String of `length
`.
2158 fun to_s_with_length(length: Int): FlatString
2161 var str = new FlatString.with_infos(self, length, 0, length - 1)
2165 # Returns `self` as a new String.
2166 fun to_s_with_copy: FlatString
2168 var length = cstring_length
2169 var new_self = new NativeString(length + 1)
2170 copy_to(new_self, length, 0, 0)
2171 var str = new FlatString.with_infos(new_self, length, 0, length - 1)
2172 new_self[length] = '\0'
2173 str.real_items = new_self
2179 private var args_cache: nullable Sequence[String]
2181 # The arguments of the program as given by the OS
2182 fun program_args: Sequence[String]
2184 if _args_cache == null then init_args
2185 return _args_cache.as(not null)
2188 # The name of the program as given by the OS
2189 fun program_name: String
2191 return native_argv(0).to_s
2194 # Initialize `program_args
` with the contents of `native_argc
` and `native_argv
`.
2195 private fun init_args
2197 var argc = native_argc
2198 var args = new Array[String].with_capacity(0)
2201 args[i-1] = native_argv(i).to_s
2207 # First argument of the main C function.
2208 private fun native_argc: Int is intern
2210 # Second argument of the main C function.
2211 private fun native_argv(i: Int): NativeString is intern
2214 # Comparator that efficienlty use `to_s
` to compare things
2216 # The comparaison call `to_s
` on object and use the result to order things.
2218 # var a = [1, 2, 3, 10, 20]
2219 # (new CachedAlphaComparator).sort(a)
2220 # assert a == [1, 10, 2, 20, 3]
2222 # Internally the result of `to_s
` is cached in a HashMap to counter
2223 # uneficient implementation of `to_s
`.
2225 # Note: it caching is not usefull, see `alpha_comparator
`
2226 class CachedAlphaComparator
2228 redef type COMPARED: Object
2230 private var cache = new HashMap[Object, String]
2232 private fun do_to_s(a: Object): String do
2233 if cache.has_key(a) then return cache[a]
2239 redef fun compare(a, b) do
2240 return do_to_s(a) <=> do_to_s(b)
2244 # see `alpha_comparator
`
2245 private class AlphaComparator
2247 redef fun compare(a, b) do return a.to_s <=> b.to_s
2250 # Stateless comparator that naively use `to_s
` to compare things.
2252 # Note: the result of `to_s
` is not cached, thus can be invoked a lot
2253 # on a single instace. See `CachedAlphaComparator` as an alternative.
2255 # var a = [1, 2, 3, 10, 20]
2256 # alpha_comparator.sort(a)
2257 # assert a == [1, 10, 2, 20, 3]
2258 fun alpha_comparator: Comparator do return once new AlphaComparator
2260 # The arguments of the program as given by the OS
2261 fun args: Sequence[String]
2263 return sys.program_args