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 line terminator (if any).
390 # assert "Hello\n".chomp == "Hello"
391 # assert "Hello".chomp == "Hello"
393 # assert "\n".chomp == ""
394 # assert "".chomp == ""
396 # Line terminators are `"\n"`, `"\r\n"` and `"\r"`.
397 # A single line terminator, the last one, is removed.
399 # assert "\r\n".chomp == ""
400 # assert "\r\n\n".chomp == "\r\n"
401 # assert "\r\n\r\n".chomp == "\r\n"
402 # assert "\r\n\r".chomp == "\r\n"
404 # Note: unlike with most IO methods like `IStream::read_line`,
405 # a single `\r` is considered here to be a line terminator and will be removed.
409 if len
== 0 then return self
410 var l
= self.chars
.last
412 return substring
(0, len-1
)
413 else if l
!= '\n' then
415 else if len
> 1 and self.chars
[len-2
] == '\r' then
416 return substring
(0, len-2
)
418 return substring
(0, len-1
)
422 # Justify a self in a space of `length`
424 # `left` is the space ratio on the left side.
425 # * 0.0 for left-justified (no space at the left)
426 # * 1.0 for right-justified (all spaces at the left)
427 # * 0.5 for centered (half the spaces at the left)
431 # assert "hello".justify(10, 0.0) == "hello "
432 # assert "hello".justify(10, 1.0) == " hello"
433 # assert "hello".justify(10, 0.5) == " hello "
435 # If `length` is not enough, `self` is returned as is.
437 # assert "hello".justify(2, 0.0) == "hello"
439 # REQUIRE: `left >= 0.0 and left <= 1.0`
440 # ENSURE: `self.length <= length implies result.length == length`
441 # ENSURE: `self.length >= length implies result == self`
442 fun justify
(length
: Int, left
: Float): SELFTYPE
444 var diff
= length
- self.length
445 if diff
<= 0 then return self
446 assert left
>= 0.0 and left
<= 1.0
447 var before
= (diff
.to_f
* left
).to_i
448 return " " * before
+ self + " " * (diff-before
)
451 # Mangle a string to be a unique string only made of alphanumeric characters
452 fun to_cmangle
: String
454 var res
= new FlatBuffer
455 var underscore
= false
456 for i
in [0..length
[ do
458 if (c
>= 'a' and c
<= 'z') or (c
>='A' and c
<= 'Z') then
464 res
.append
('_'.ascii
.to_s
)
467 if c
>= '0' and c
<= '9' then
470 else if c
== '_' then
475 res
.append
(c
.ascii
.to_s
)
483 # Escape " \ ' and non printable characters using the rules of literal C strings and characters
485 # assert "abAB12<>&".escape_to_c == "abAB12<>&"
486 # assert "\n\"'\\".escape_to_c == "\\n\\\"\\'\\\\"
487 fun escape_to_c
: String
489 var b
= new FlatBuffer
490 for i
in [0..length
[ do
494 else if c
== '\0' then
496 else if c
== '"' then
498 else if c == '\'' then
500 else if c == '\\
' then
502 else if c.ascii < 32 then
503 b.append("\\{c.ascii.to_base(8, false)}")
511 # Escape additionnal characters
512 # The result might no be legal in C but be used in other languages
514 # assert "ab|\{\}".escape_more_to_c("|\{\}") == "ab\\|\\\{\\\}"
515 fun escape_more_to_c(chars: String): String
517 var b = new FlatBuffer
518 for c in escape_to_c.chars do
519 if chars.chars.has(c) then
527 # Escape to C plus braces
529 # assert "\n\"'\\\
{\}".escape_to_nit == "\\n\\\
"\\'\\\\\\\{\\\}"
530 fun escape_to_nit
: String do return escape_more_to_c
("\{\}")
532 # Escape to POSIX Shell (sh).
534 # Abort if the text contains a null byte.
536 # assert "\n\"'\\\{\}0".escape_to_sh == "'\n\"'\\''\\\{\}0'"
537 fun escape_to_sh
: String do
538 var b
= new FlatBuffer
540 for i in [0..length[ do
545 assert without_null_byte
: c
!= '\0'
553 # Escape to include in a Makefile
555 # Unfortunately, some characters are not escapable in Makefile.
556 # These characters are `;`, `|`, `\`, and the non-printable ones.
557 # They will be rendered as `"?{hex}"`.
558 fun escape_to_mk: String do
559 var b = new FlatBuffer
560 for i in [0..length[ do
564 else if c == ':' or c == ' ' or c == '#' then
567 else if c
.ascii
< 32 or c
== ';' or c
== '|' or c
== '\\' or c
== '=' then
568 b
.append
("?{c.ascii.to_base(16, false)}")
576 # Return a string where Nit escape sequences are transformed.
579 # assert s.length == 2
580 # var u = s.unescape_nit
581 # assert u.length == 1
582 # assert u.chars[0].ascii == 10 # (the ASCII value of the "new line" character)
583 fun unescape_nit
: String
585 var res
= new FlatBuffer.with_capacity
(self.length
)
586 var was_slash
= false
587 for i
in [0..length
[ do
589 if not was_slash
then
600 else if c
== 'r' then
602 else if c
== 't' then
604 else if c
== '0' then
613 # Encode `self` to percent (or URL) encoding
615 # assert "aBc09-._~".to_percent_encoding == "aBc09-._~"
616 # assert "%()< >".to_percent_encoding == "%25%28%29%3c%20%3e"
617 # assert ".com/post?e=asdf&f=123".to_percent_encoding == ".com%2fpost%3fe%3dasdf%26f%3d123"
618 fun to_percent_encoding
: String
620 var buf
= new FlatBuffer
622 for i
in [0..length
[ do
624 if (c
>= '0' and c
<= '9') or
625 (c
>= 'a' and c
<= 'z') or
626 (c
>= 'A' and c
<= 'Z') or
627 c
== '-' or c
== '.' or
631 else buf
.append
"%{c.ascii.to_hex}"
637 # Decode `self` from percent (or URL) encoding to a clear string
639 # Replace invalid use of '%' with '?'.
641 # assert "aBc09-._~".from_percent_encoding == "aBc09-._~"
642 # assert "%25%28%29%3c%20%3e".from_percent_encoding == "%()< >"
643 # assert ".com%2fpost%3fe%3dasdf%26f%3d123".from_percent_encoding == ".com/post?e=asdf&f=123"
644 # assert "%25%28%29%3C%20%3E".from_percent_encoding == "%()< >"
645 # assert "incomplete %".from_percent_encoding == "incomplete ?"
646 # assert "invalid % usage".from_percent_encoding == "invalid ? usage"
647 fun from_percent_encoding
: String
649 var buf
= new FlatBuffer
655 if i
+ 2 >= length
then
656 # What follows % has been cut off
660 var hex_s
= substring
(i
, 2)
662 var hex_i
= hex_s
.to_hex
666 # What follows a % is not Hex
679 # Escape the characters `<`, `>`, `&`, `"`, `'` and `/` as HTML/XML entity references.
681 # assert "a&b-<>\"x\"/'".html_escape == "a&b-<>"x"/'"
683 # 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>
684 fun html_escape
: SELFTYPE
686 var buf
= new FlatBuffer
688 for i
in [0..length
[ do
692 else if c
== '<' then
694 else if c
== '>' then
696 else if c
== '"' then
698 else if c
== '\'' then
700 else if c == '/' then
709 # Two pieces of text are equals if thez have the same characters in the same order.
711 # assert "hello" == "hello"
712 # assert "hello" != "HELLO"
713 # assert "hello" == "hel"+"lo"
715 # Things that are not Text are not equal.
718 # assert "9" != ['9']
721 # assert "9".chars.first == '9' # equality of Char
722 # assert "9".chars == ['9'] # equality of Sequence
723 # assert "9".to_i == 9 # equality of Int
726 if o == null then return false
727 if not o isa Text then return false
728 if self.is_same_instance(o) then return true
729 if self.length != o.length then return false
730 return self.chars == o.chars
733 # Lexicographical comparaison
735 # assert "abc" < "xy"
736 # assert "ABC" < "abc"
739 var self_chars = self.chars.iterator
740 var other_chars = other.chars.iterator
742 while self_chars.is_ok and other_chars.is_ok do
743 if self_chars.item < other_chars.item then return true
744 if self_chars.item > other_chars.item then return false
749 if self_chars.is_ok then
756 # Escape string used in labels for graphviz
758 # assert ">><<".escape_to_dot == "\\>\\>\\<\\<"
759 fun escape_to_dot: String
761 return escape_more_to_c("|\{\}<>")
764 # Flat representation of self
765 fun flatten: FlatText is abstract
767 private var hash_cache: nullable Int = null
771 if hash_cache == null then
772 # djb2 hash algorithm
775 for i in [0..length[ do
777 h = h.lshift(5) + h + char.ascii
782 return hash_cache.as(not null)
787 # All kinds of array-based text representations.
788 abstract class FlatText
791 # Underlying C-String (`char*`)
793 # Warning : Might be void in some subclasses, be sure to check
794 # if set before using it.
795 private var items: NativeString is noinit
797 # Real items, used as cache for to_cstring is called
798 private var real_items: nullable NativeString = null
800 redef var length: Int = 0
811 redef fun flatten do return self
814 # Abstract class for the SequenceRead compatible
815 # views on String and Buffer objects
816 private abstract class StringCharView
817 super SequenceRead[Char]
823 redef fun is_empty do return target.is_empty
825 redef fun length do return target.length
827 redef fun iterator: IndexedIterator[Char] do return self.iterator_from(0)
829 redef fun reverse_iterator do return self.reverse_iterator_from(self.length - 1)
832 # View on Buffer objects, extends Sequence
833 # for mutation operations
834 private abstract class BufferCharView
838 redef type SELFTYPE: Buffer
842 # A `String` holds and manipulates an arbitrary sequence of characters.
844 # String objects may be created using literals.
846 # assert "Hello World!" isa String
847 abstract class String
850 redef type SELFTYPE: String
852 redef fun to_s do return self
854 # Concatenates `o` to `self`
856 # assert "hello" + "world" == "helloworld"
857 # assert "" + "hello" + "" == "hello"
858 fun +(o: Text): SELFTYPE is abstract
860 # Concatenates self `i` times
862 # assert "abc" * 4 == "abcabcabcabc"
863 # assert "abc" * 1 == "abc"
864 # assert "abc" * 0 == ""
865 fun *(i: Int): SELFTYPE is abstract
867 # Insert `s` at `pos`.
869 # assert "helloworld".insert_at(" ", 5) == "hello world"
870 fun insert_at(s: String, pos: Int): SELFTYPE is abstract
872 redef fun substrings: Iterator[String] is abstract
874 # Returns a reversed version of self
876 # assert "hello".reversed == "olleh"
877 # assert "bob".reversed == "bob"
878 # assert "".reversed == ""
879 fun reversed: SELFTYPE is abstract
881 # A upper case version of `self`
883 # assert "Hello World!".to_upper == "HELLO WORLD!"
884 fun to_upper: SELFTYPE is abstract
886 # A lower case version of `self`
888 # assert "Hello World!".to_lower == "hello world!"
889 fun to_lower : SELFTYPE is abstract
891 # Takes a camel case `self` and converts it to snake case
893 # assert "randomMethodId".to_snake_case == "random_method_id"
895 # If `self` is upper, it is returned unchanged
897 # assert "RANDOM_METHOD_ID".to_snake_case == "RANDOM_METHOD_ID"
899 # If the identifier is prefixed by an underscore, the underscore is ignored
901 # assert "_privateField".to_snake_case == "_private_field"
902 fun to_snake_case: SELFTYPE
904 if self.is_upper then return self
906 var new_str = new FlatBuffer.with_capacity(self.length)
907 var is_first_char = true
909 for i in [0..length[ do
911 if is_first_char then
912 new_str.add(char.to_lower)
913 is_first_char = false
914 else if char.is_upper then
916 new_str.add(char.to_lower)
925 # Takes a snake case `self` and converts it to camel case
927 # assert "random_method_id".to_camel_case == "randomMethodId"
929 # If the identifier is prefixed by an underscore, the underscore is ignored
931 # assert "_private_field".to_camel_case == "_privateField"
933 # If `self` is upper, it is returned unchanged
935 # assert "RANDOM_ID".to_camel_case == "RANDOM_ID"
937 # If there are several consecutive underscores, they are considered as a single one
939 # assert "random__method_id".to_camel_case == "randomMethodId"
940 fun to_camel_case: SELFTYPE
942 if self.is_upper then return self
944 var new_str = new FlatBuffer
945 var is_first_char = true
946 var follows_us = false
948 for i in [0..length[ do
950 if is_first_char then
952 is_first_char = false
953 else if char == '_
' then
955 else if follows_us then
956 new_str.add(char.to_upper)
966 # Returns a capitalized `self`
968 # Letters that follow a letter are lowercased
969 # Letters that follow a non-letter are upcased.
971 # SEE : `Char::is_letter` for the definition of letter.
973 # assert "jAVASCRIPT".capitalized == "Javascript"
974 # assert "i am root".capitalized == "I Am Root"
975 # assert "ab_c -ab0c ab\nc".capitalized == "Ab_C -Ab0C Ab\nC"
976 fun capitalized: SELFTYPE do
977 if length == 0 then return self
979 var buf = new FlatBuffer.with_capacity(length)
981 var curr = chars[0].to_upper
985 for i in [1 .. length[ do
988 if prev.is_letter then
989 buf[i] = curr.to_lower
991 buf[i] = curr.to_upper
999 private class FlatSubstringsIter
1000 super Iterator[FlatText]
1002 var tgt: nullable FlatText
1006 return tgt.as(not null)
1009 redef fun is_ok do return tgt != null
1011 redef fun next do tgt = null
1014 # Immutable strings of characters.
1019 # Index in _items of the start of the string
1020 private var index_from: Int is noinit
1022 # Indes in _items of the last item of the string
1023 private var index_to: Int is noinit
1025 redef var chars: SequenceRead[Char] = new FlatStringCharView(self)
1029 # Check that the index (+ index_from) is not larger than indexTo
1030 # In other terms, if the index is valid
1032 assert (index + index_from) <= index_to
1033 return items[index + index_from]
1036 ################################################
1037 # AbstractString specific methods #
1038 ################################################
1042 var native = new NativeString(self.length + 1)
1043 var length = self.length
1044 var items = self.items
1047 while pos < length do
1048 native[pos] = items[ipos]
1052 return native.to_s_with_length(self.length)
1055 redef fun substring(from, count)
1061 if count < 0 then count = 0
1065 var realFrom = index_from + from
1067 if (realFrom + count) > index_to then return new FlatString.with_infos(items, index_to - realFrom + 1, realFrom, index_to)
1069 if count == 0 then return empty
1071 var to = realFrom + count - 1
1073 return new FlatString.with_infos(items, to - realFrom + 1, realFrom, to)
1076 redef fun empty do return "".as(FlatString)
1080 var outstr = new NativeString(self.length + 1)
1083 var myitems = self.items
1084 var index_from = self.index_from
1085 var max = self.index_to
1087 while index_from <= max do
1088 outstr[out_index] = myitems[index_from].to_upper
1093 outstr[self.length] = '\
0'
1095 return outstr.to_s_with_length(self.length)
1100 var outstr = new NativeString(self.length + 1)
1103 var myitems = self.items
1104 var index_from = self.index_from
1105 var max = self.index_to
1107 while index_from <= max do
1108 outstr[out_index] = myitems[index_from].to_lower
1113 outstr[self.length] = '\
0'
1115 return outstr.to_s_with_length(self.length)
1120 var i = self.index_from
1121 var imax = self.index_to
1128 ##################################################
1129 # String Specific Methods #
1130 ##################################################
1132 private init with_infos(items: NativeString, len: Int, from: Int, to: Int)
1140 redef fun to_cstring: NativeString
1142 if real_items != null then
1143 return real_items.as(not null)
1145 var newItems = new NativeString(length + 1)
1146 self.items.copy_to(newItems, length, index_from, 0)
1147 newItems[length] = '\
0'
1148 self.real_items = newItems
1155 if not other isa FlatString then return super
1157 if self.object_id == other.object_id then return true
1159 var my_length = length
1161 if other.length != my_length then return false
1163 var my_index = index_from
1164 var its_index = other.index_from
1166 var last_iteration = my_index + my_length
1168 var itsitems = other.items
1169 var myitems = self.items
1171 while my_index < last_iteration do
1172 if myitems[my_index] != itsitems[its_index] then return false
1182 if not other isa FlatString then return super
1184 if self.object_id == other.object_id then return false
1186 var my_curr_char : Char
1187 var its_curr_char : Char
1189 var curr_id_self = self.index_from
1190 var curr_id_other = other.index_from
1192 var my_items = self.items
1193 var its_items = other.items
1195 var my_length = self.length
1196 var its_length = other.length
1198 var max_iterations = curr_id_self + my_length
1200 while curr_id_self < max_iterations do
1201 my_curr_char = my_items[curr_id_self]
1202 its_curr_char = its_items[curr_id_other]
1204 if my_curr_char != its_curr_char then
1205 if my_curr_char < its_curr_char then return true
1213 return my_length < its_length
1218 var my_length = self.length
1219 var its_length = s.length
1221 var total_length = my_length + its_length
1223 var target_string = new NativeString(my_length + its_length + 1)
1225 self.items.copy_to(target_string, my_length, index_from, 0)
1226 if s isa FlatString then
1227 s.items.copy_to(target_string, its_length, s.index_from, my_length)
1228 else if s isa FlatBuffer then
1229 s.items.copy_to(target_string, its_length, 0, my_length)
1231 var curr_pos = my_length
1232 for i in [0..s.length[ do
1234 target_string[curr_pos] = c
1239 target_string[total_length] = '\
0'
1241 return target_string.to_s_with_length(total_length)
1248 var my_length = self.length
1250 var final_length = my_length * i
1252 var my_items = self.items
1254 var target_string = new NativeString(final_length + 1)
1256 target_string[final_length] = '\
0'
1258 var current_last = 0
1260 for iteration in [1 .. i] do
1261 my_items.copy_to(target_string, my_length, 0, current_last)
1262 current_last += my_length
1265 return target_string.to_s_with_length(final_length)
1270 if hash_cache == null then
1271 # djb2 hash algorithm
1277 while i <= index_to do
1278 h = h.lshift(5) + h + myitems[i].ascii
1285 return hash_cache.as(not null)
1288 redef fun substrings do return new FlatSubstringsIter(self)
1291 private class FlatStringReverseIterator
1292 super IndexedIterator[Char]
1294 var target: FlatString
1296 var target_items: NativeString
1300 init with_pos(tgt: FlatString, pos: Int)
1303 target_items = tgt.items
1304 curr_pos = pos + tgt.index_from
1307 redef fun is_ok do return curr_pos >= target.index_from
1309 redef fun item do return target_items[curr_pos]
1311 redef fun next do curr_pos -= 1
1313 redef fun index do return curr_pos - target.index_from
1317 private class FlatStringIterator
1318 super IndexedIterator[Char]
1320 var target: FlatString
1322 var target_items: NativeString
1326 init with_pos(tgt: FlatString, pos: Int)
1329 target_items = tgt.items
1330 curr_pos = pos + target.index_from
1333 redef fun is_ok do return curr_pos <= target.index_to
1335 redef fun item do return target_items[curr_pos]
1337 redef fun next do curr_pos += 1
1339 redef fun index do return curr_pos - target.index_from
1343 private class FlatStringCharView
1344 super StringCharView
1346 redef type SELFTYPE: FlatString
1350 # Check that the index (+ index_from) is not larger than indexTo
1351 # In other terms, if the index is valid
1353 var target = self.target
1354 assert (index + target.index_from) <= target.index_to
1355 return target.items[index + target.index_from]
1358 redef fun iterator_from(start) do return new FlatStringIterator.with_pos(target, start)
1360 redef fun reverse_iterator_from(start) do return new FlatStringReverseIterator.with_pos(target, start)
1364 # A mutable sequence of characters.
1365 abstract class Buffer
1368 redef type SELFTYPE: Buffer
1370 # Specific implementations MUST set this to `true` in order to invalidate caches
1371 protected var is_dirty = true
1373 # Copy-On-Write flag
1375 # If the `Buffer` was to_s'd
, the next in-place altering
1376 # operation will cause the current `Buffer` to be re-allocated.
1378 # The flag will then be set at `false`.
1379 protected var written
= false
1381 # Modifies the char contained at pos `index`
1383 # DEPRECATED : Use self.chars.[]= instead
1384 fun []=(index
: Int, item
: Char) is abstract
1386 # Adds a char `c` at the end of self
1388 # DEPRECATED : Use self.chars.add instead
1389 fun add
(c
: Char) is abstract
1393 # var b = new FlatBuffer
1395 # assert not b.is_empty
1398 fun clear
is abstract
1400 # Enlarges the subsequent array containing the chars of self
1401 fun enlarge
(cap
: Int) is abstract
1403 # Adds the content of text `s` at the end of self
1405 # var b = new FlatBuffer
1408 # assert b == "helloworld"
1409 fun append
(s
: Text) is abstract
1411 # `self` is appended in such a way that `self` is repeated `r` times
1413 # var b = new FlatBuffer
1416 # assert b == "hellohellohello"
1417 fun times
(r
: Int) is abstract
1419 # Reverses itself in-place
1421 # var b = new FlatBuffer
1424 # assert b == "olleh"
1425 fun reverse
is abstract
1427 # Changes each lower-case char in `self` by its upper-case variant
1429 # var b = new FlatBuffer
1430 # b.append("Hello World!")
1432 # assert b == "HELLO WORLD!"
1433 fun upper
is abstract
1435 # Changes each upper-case char in `self` by its lower-case variant
1437 # var b = new FlatBuffer
1438 # b.append("Hello World!")
1440 # assert b == "hello world!"
1441 fun lower
is abstract
1443 # Capitalizes each word in `self`
1445 # Letters that follow a letter are lowercased
1446 # Letters that follow a non-letter are upcased.
1448 # SEE: `Char::is_letter` for the definition of a letter.
1450 # var b = new FlatBuffer.from("jAVAsCriPt")
1452 # assert b == "Javascript"
1453 # b = new FlatBuffer.from("i am root")
1455 # assert b == "I Am Root"
1456 # b = new FlatBuffer.from("ab_c -ab0c ab\nc")
1458 # assert b == "Ab_C -Ab0C Ab\nC"
1460 if length
== 0 then return
1461 var c
= self[0].to_upper
1464 for i
in [1 .. length
[ do
1467 if prev
.is_letter
then
1468 self[i
] = c
.to_lower
1470 self[i
] = c
.to_upper
1477 if is_dirty
then hash_cache
= null
1481 # In Buffers, the internal sequence of character is mutable
1482 # Thus, `chars` can be used to modify the buffer.
1483 redef fun chars
: Sequence[Char] is abstract
1486 # Mutable strings of characters.
1491 redef type SELFTYPE: FlatBuffer
1493 redef var chars
: Sequence[Char] = new FlatBufferCharView(self)
1495 private var capacity
: Int = 0
1497 redef fun substrings
do return new FlatSubstringsIter(self)
1499 # Re-copies the `NativeString` into a new one and sets it as the new `Buffer`
1501 # This happens when an operation modifies the current `Buffer` and
1502 # the Copy-On-Write flag `written` is set at true.
1503 private fun reset
do
1504 var nns
= new NativeString(capacity
)
1505 items
.copy_to
(nns
, length
, 0, 0)
1513 assert index
< length
1517 redef fun []=(index
, item
)
1520 if index
== length
then
1524 if written
then reset
1525 assert index
>= 0 and index
< length
1532 if capacity
<= length
then enlarge
(length
+ 5)
1539 if written
then reset
1543 redef fun empty
do return new FlatBuffer
1545 redef fun enlarge
(cap
)
1548 if cap
<= c
then return
1549 while c
<= cap
do c
= c
* 2 + 2
1550 # The COW flag can be set at false here, since
1551 # it does a copy of the current `Buffer`
1553 var a
= new NativeString(c
+1)
1554 if length
> 0 then items
.copy_to
(a
, length
, 0, 0)
1559 redef fun to_s
: String
1562 if length
== 0 then items
= new NativeString(1)
1563 return new FlatString.with_infos
(items
, length
, 0, length
- 1)
1566 redef fun to_cstring
1569 var new_native
= new NativeString(length
+ 1)
1570 new_native
[length
] = '\0'
1571 if length
> 0 then items
.copy_to
(new_native
, length
, 0, 0)
1572 real_items
= new_native
1575 return real_items
.as(not null)
1578 # Create a new empty string.
1581 # Create a new string copied from `s`.
1584 capacity
= s
.length
+ 1
1586 items
= new NativeString(capacity
)
1587 if s
isa FlatString then
1588 s
.items
.copy_to
(items
, length
, s
.index_from
, 0)
1589 else if s
isa FlatBuffer then
1590 s
.items
.copy_to
(items
, length
, 0, 0)
1593 for i
in [0..s
.length
[ do
1601 # Create a new empty string with a given capacity.
1602 init with_capacity
(cap
: Int)
1605 # _items = new NativeString.calloc(cap)
1606 items
= new NativeString(cap
+1)
1613 if s
.is_empty
then return
1616 if capacity
< length
+ sl
then enlarge
(length
+ sl
)
1617 if s
isa FlatString then
1618 s
.items
.copy_to
(items
, sl
, s
.index_from
, length
)
1619 else if s
isa FlatBuffer then
1620 s
.items
.copy_to
(items
, sl
, 0, length
)
1622 var curr_pos
= self.length
1623 for i
in [0..s
.length
[ do
1632 # Copies the content of self in `dest`
1633 fun copy
(start
: Int, len
: Int, dest
: Buffer, new_start
: Int)
1635 var self_chars
= self.chars
1636 var dest_chars
= dest
.chars
1637 for i
in [0..len-1
] do
1638 dest_chars
[new_start
+i
] = self_chars
[start
+i
]
1642 redef fun substring
(from
, count
)
1646 if from
< 0 then from
= 0
1647 if count
> length
then count
= length
1648 if from
< count
then
1649 var r
= new FlatBuffer.with_capacity
(count
- from
)
1650 while from
< count
do
1651 r
.chars
.push
(items
[from
])
1656 return new FlatBuffer
1663 var ns
= new NativeString(capacity
)
1675 redef fun times
(repeats
)
1677 var x
= new FlatString.with_infos
(items
, length
, 0, length
- 1)
1678 for i
in [1..repeats
[ do
1685 if written
then reset
1689 it
[id
] = it
[id
].to_upper
1696 if written
then reset
1700 it
[id
] = it
[id
].to_lower
1706 private class FlatBufferReverseIterator
1707 super IndexedIterator[Char]
1709 var target
: FlatBuffer
1711 var target_items
: NativeString
1715 init with_pos
(tgt
: FlatBuffer, pos
: Int)
1718 if tgt
.length
> 0 then target_items
= tgt
.items
1722 redef fun index
do return curr_pos
1724 redef fun is_ok
do return curr_pos
>= 0
1726 redef fun item
do return target_items
[curr_pos
]
1728 redef fun next
do curr_pos
-= 1
1732 private class FlatBufferCharView
1733 super BufferCharView
1735 redef type SELFTYPE: FlatBuffer
1737 redef fun [](index
) do return target
.items
[index
]
1739 redef fun []=(index
, item
)
1741 assert index
>= 0 and index
<= length
1742 if index
== length
then
1746 target
.items
[index
] = item
1759 fun enlarge
(cap
: Int)
1766 var s_length
= s
.length
1767 if target
.capacity
< s
.length
then enlarge
(s_length
+ target
.length
)
1770 redef fun iterator_from
(pos
) do return new FlatBufferIterator.with_pos
(target
, pos
)
1772 redef fun reverse_iterator_from
(pos
) do return new FlatBufferReverseIterator.with_pos
(target
, pos
)
1776 private class FlatBufferIterator
1777 super IndexedIterator[Char]
1779 var target
: FlatBuffer
1781 var target_items
: NativeString
1785 init with_pos
(tgt
: FlatBuffer, pos
: Int)
1788 if tgt
.length
> 0 then target_items
= tgt
.items
1792 redef fun index
do return curr_pos
1794 redef fun is_ok
do return curr_pos
< target
.length
1796 redef fun item
do return target_items
[curr_pos
]
1798 redef fun next
do curr_pos
+= 1
1802 ###############################################################################
1804 ###############################################################################
1807 # User readable representation of `self`.
1808 fun to_s
: String do return inspect
1810 # The class name of the object in NativeString format.
1811 private fun native_class_name
: NativeString is intern
1813 # The class name of the object.
1815 # assert 5.class_name == "Int"
1816 fun class_name
: String do return native_class_name
.to_s
1818 # Developer readable representation of `self`.
1819 # Usually, it uses the form "<CLASSNAME:#OBJECTID bla bla bla>"
1822 return "<{inspect_head}>"
1825 # Return "CLASSNAME:#OBJECTID".
1826 # This function is mainly used with the redefinition of the inspect method
1827 protected fun inspect_head
: String
1829 return "{class_name}:#{object_id.to_hex}"
1834 # assert true.to_s == "true"
1835 # assert false.to_s == "false"
1848 # Wrapper of strerror C function
1849 private fun strerror_ext
: NativeString is extern `{
1850 return strerror(recv);
1853 # Returns a string describing error number
1854 fun strerror
: String do return strerror_ext
.to_s
1856 # Fill `s` with the digits in base `base` of `self` (and with the '-' sign if 'signed' and negative).
1857 # assume < to_c max const of char
1858 private fun fill_buffer
(s
: Buffer, base
: Int, signed
: Bool)
1865 else if self == 0 then
1872 var pos
= digit_count
(base
) - 1
1873 while pos
>= 0 and n
> 0 do
1874 s
.chars
[pos
] = (n
% base
).to_c
1880 # C function to convert an nit Int to a NativeString (char*)
1881 private fun native_int_to_s
: NativeString is extern "native_int_to_s"
1883 # return displayable int in base 10 and signed
1885 # assert 1.to_s == "1"
1886 # assert (-123).to_s == "-123"
1888 return native_int_to_s
.to_s
1891 # return displayable int in hexadecimal
1893 # assert 1.to_hex == "1"
1894 # assert (-255).to_hex == "-ff"
1895 fun to_hex
: String do return to_base
(16,false)
1897 # return displayable int in base base and signed
1898 fun to_base
(base
: Int, signed
: Bool): String
1900 var l
= digit_count
(base
)
1901 var s
= new FlatBuffer.from
(" " * l
)
1902 fill_buffer
(s
, base
, signed
)
1908 # Pretty representation of `self`, with decimals as needed from 1 to a maximum of 3
1910 # assert 12.34.to_s == "12.34"
1911 # assert (-0120.030).to_s == "-120.03"
1913 # see `to_precision` for a custom precision.
1915 var str
= to_precision
( 3 )
1916 if is_inf
!= 0 or is_nan
then return str
1917 var len
= str
.length
1918 for i
in [0..len-1
] do
1920 var c
= str
.chars
[j
]
1923 else if c
== '.' then
1924 return str
.substring
( 0, j
+2 )
1926 return str
.substring
( 0, j
+1 )
1932 # `String` representation of `self` with the given number of `decimals`
1934 # assert 12.345.to_precision(0) == "12"
1935 # assert 12.345.to_precision(3) == "12.345"
1936 # assert (-12.345).to_precision(3) == "-12.345"
1937 # assert (-0.123).to_precision(3) == "-0.123"
1938 # assert 0.999.to_precision(2) == "1.00"
1939 # assert 0.999.to_precision(4) == "0.9990"
1940 fun to_precision
(decimals
: Int): String
1942 if is_nan
then return "nan"
1944 var isinf
= self.is_inf
1947 else if isinf
== -1 then
1951 if decimals
== 0 then return self.to_i
.to_s
1953 for i
in [0..decimals
[ do f
= f
* 10.0
1960 if i
== 0 then return "0." + "0"*decimals
1962 # Prepare both parts of the float, before and after the "."
1967 if sl
> decimals
then
1968 # Has something before the "."
1969 p1
= s
.substring
(0, sl-decimals
)
1970 p2
= s
.substring
(sl-decimals
, decimals
)
1973 p2
= "0"*(decimals-sl
) + s
1976 if i
< 0 then p1
= "-" + p1
1978 return p1
+ "." + p2
1981 # `self` representation with `nb` digits after the '.'.
1983 # assert 12.345.to_precision_native(1) == "12.3"
1984 # assert 12.345.to_precision_native(2) == "12.35"
1985 # assert 12.345.to_precision_native(3) == "12.345"
1986 # assert 12.345.to_precision_native(4) == "12.3450"
1987 fun to_precision_native
(nb
: Int): String import NativeString.to_s
`{
1991 size = snprintf(NULL, 0, "%.*f", (int)nb, recv);
1992 str = malloc(size + 1);
1993 sprintf(str, "%.*f", (int)nb, recv );
1995 return NativeString_to_s( str );
2000 # assert 'x'.to_s == "x"
2003 var s
= new FlatBuffer.with_capacity
(1)
2008 # Returns true if the char is a numerical digit
2010 # assert '0'.is_numeric
2011 # assert '9'.is_numeric
2012 # assert not 'a'.is_numeric
2013 # assert not '?'.is_numeric
2014 fun is_numeric
: Bool
2016 return self >= '0' and self <= '9'
2019 # Returns true if the char is an alpha digit
2021 # assert 'a'.is_alpha
2022 # assert 'Z'.is_alpha
2023 # assert not '0'.is_alpha
2024 # assert not '?'.is_alpha
2027 return (self >= 'a' and self <= 'z') or (self >= 'A' and self <= 'Z')
2030 # Returns true if the char is an alpha or a numeric digit
2032 # assert 'a'.is_alphanumeric
2033 # assert 'Z'.is_alphanumeric
2034 # assert '0'.is_alphanumeric
2035 # assert '9'.is_alphanumeric
2036 # assert not '?'.is_alphanumeric
2037 fun is_alphanumeric
: Bool
2039 return self.is_numeric
or self.is_alpha
2043 redef class Collection[E
]
2044 # Concatenate elements.
2047 var s
= new FlatBuffer
2048 for e
in self do if e
!= null then s
.append
(e
.to_s
)
2052 # Concatenate and separate each elements with `sep`.
2054 # assert [1, 2, 3].join(":") == "1:2:3"
2055 # assert [1..3].join(":") == "1:2:3"
2056 fun join
(sep
: Text): String
2058 if is_empty
then return ""
2060 var s
= new FlatBuffer # Result
2065 if e
!= null then s
.append
(e
.to_s
)
2067 # Concat other items
2072 if e
!= null then s
.append
(e
.to_s
)
2079 redef class Array[E
]
2081 # Fast implementation
2085 if l
== 0 then return ""
2086 if l
== 1 then if self[0] == null then return "" else return self[0].to_s
2088 var na
= new NativeArray[String](l
)
2094 if itsi
== null then
2104 var ns
= new NativeString(sl
+ 1)
2110 var tpl
= tmp
.length
2111 if tmp
isa FlatString then
2112 tmp
.items
.copy_to
(ns
, tpl
, tmp
.index_from
, off
)
2115 for j
in tmp
.substrings
do
2116 var s
= j
.as(FlatString)
2118 s
.items
.copy_to
(ns
, slen
, s
.index_from
, off
)
2124 return ns
.to_s_with_length
(sl
)
2128 redef class Map[K
,V
]
2129 # Concatenate couple of 'key value'.
2130 # key and value are separated by `couple_sep`.
2131 # each couple is separated each couple with `sep`.
2133 # var m = new ArrayMap[Int, String]
2136 # assert m.join("; ", "=") == "1=one; 10=ten"
2137 fun join
(sep
: String, couple_sep
: String): String
2139 if is_empty
then return ""
2141 var s
= new FlatBuffer # Result
2147 s
.append
("{k}{couple_sep}{e or else "<null>"}")
2149 # Concat other items
2155 s
.append
("{k}{couple_sep}{e or else "<null>"}")
2162 ###############################################################################
2164 ###############################################################################
2166 # Native strings are simple C char *
2167 extern class NativeString `{ char* `}
2168 # Creates a new NativeString with a capacity of `length
`
2169 new(length: Int) is intern
2171 # Get char at `index
`.
2172 fun [](index: Int): Char is intern
2174 # Set char `item
` at index.
2175 fun []=(index: Int, item: Char) is intern
2177 # Copy `self` to `dest
`.
2178 fun copy_to(dest: NativeString, length: Int, from: Int, to: Int) is intern
2180 # Position of the first nul character.
2181 fun cstring_length: Int
2184 while self[l] != '\0' do l += 1
2188 # Parse `self` as an Int.
2189 fun atoi: Int is intern
2191 # Parse `self` as a Float.
2192 fun atof: Float is extern "atof"
2196 return to_s_with_length(cstring_length)
2199 # Returns `self` as a String of `length
`.
2200 fun to_s_with_length(length: Int): FlatString
2203 var str = new FlatString.with_infos(self, length, 0, length - 1)
2207 # Returns `self` as a new String.
2208 fun to_s_with_copy: FlatString
2210 var length = cstring_length
2211 var new_self = new NativeString(length + 1)
2212 copy_to(new_self, length, 0, 0)
2213 var str = new FlatString.with_infos(new_self, length, 0, length - 1)
2214 new_self[length] = '\0'
2215 str.real_items = new_self
2221 private var args_cache: nullable Sequence[String]
2223 # The arguments of the program as given by the OS
2224 fun program_args: Sequence[String]
2226 if _args_cache == null then init_args
2227 return _args_cache.as(not null)
2230 # The name of the program as given by the OS
2231 fun program_name: String
2233 return native_argv(0).to_s
2236 # Initialize `program_args
` with the contents of `native_argc
` and `native_argv
`.
2237 private fun init_args
2239 var argc = native_argc
2240 var args = new Array[String].with_capacity(0)
2243 args[i-1] = native_argv(i).to_s
2249 # First argument of the main C function.
2250 private fun native_argc: Int is intern
2252 # Second argument of the main C function.
2253 private fun native_argv(i: Int): NativeString is intern
2256 # Comparator that efficienlty use `to_s
` to compare things
2258 # The comparaison call `to_s
` on object and use the result to order things.
2260 # var a = [1, 2, 3, 10, 20]
2261 # (new CachedAlphaComparator).sort(a)
2262 # assert a == [1, 10, 2, 20, 3]
2264 # Internally the result of `to_s
` is cached in a HashMap to counter
2265 # uneficient implementation of `to_s
`.
2267 # Note: it caching is not usefull, see `alpha_comparator
`
2268 class CachedAlphaComparator
2270 redef type COMPARED: Object
2272 private var cache = new HashMap[Object, String]
2274 private fun do_to_s(a: Object): String do
2275 if cache.has_key(a) then return cache[a]
2281 redef fun compare(a, b) do
2282 return do_to_s(a) <=> do_to_s(b)
2286 # see `alpha_comparator
`
2287 private class AlphaComparator
2289 redef fun compare(a, b) do return a.to_s <=> b.to_s
2292 # Stateless comparator that naively use `to_s
` to compare things.
2294 # Note: the result of `to_s
` is not cached, thus can be invoked a lot
2295 # on a single instace. See `CachedAlphaComparator` as an alternative.
2297 # var a = [1, 2, 3, 10, 20]
2298 # alpha_comparator.sort(a)
2299 # assert a == [1, 10, 2, 20, 3]
2300 fun alpha_comparator: Comparator do return once new AlphaComparator
2302 # The arguments of the program as given by the OS
2303 fun args: Sequence[String]
2305 return sys.program_args