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[FlatText] 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 is abstract
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 <= '7', return the corresponding integer.
254 # assert "714".to_oct == 460
255 fun to_oct
: Int do return a_to
(8)
257 # If `self` contains only '0' et '1', return the corresponding integer.
259 # assert "101101".to_bin == 45
260 fun to_bin
: Int do return a_to
(2)
262 # If `self` contains only digits and letters, return the corresponding integer in a given base
264 # assert "120".a_to(3) == 15
265 fun a_to
(base
: Int) : Int
270 for j
in [0..length
[ do
292 # Returns `true` if the string contains only Numeric values (and one "," or one "." character)
294 # assert "123".is_numeric == true
295 # assert "1.2".is_numeric == true
296 # assert "1,2".is_numeric == true
297 # assert "1..2".is_numeric == false
300 var has_point_or_comma
= false
301 for i
in [0..length
[ do
303 if not c
.is_numeric
then
304 if (c
== '.' or c
== ',') and not has_point_or_comma
then
305 has_point_or_comma
= true
314 # Returns `true` if the string contains only Hex chars
316 # assert "048bf".is_hex == true
317 # assert "ABCDEF".is_hex == true
318 # assert "0G".is_hex == false
321 for i
in [0..length
[ do
323 if not (c
>= 'a' and c
<= 'f') and
324 not (c
>= 'A' and c
<= 'F') and
325 not (c
>= '0' and c
<= '9') then return false
330 # Are all letters in `self` upper-case ?
332 # assert "HELLO WORLD".is_upper == true
333 # assert "%$&%!".is_upper == true
334 # assert "hello world".is_upper == false
335 # assert "Hello World".is_upper == false
338 for i
in [0..length
[ do
340 if char
.is_lower
then return false
345 # Are all letters in `self` lower-case ?
347 # assert "hello world".is_lower == true
348 # assert "%$&%!".is_lower == true
349 # assert "Hello World".is_lower == false
352 for i
in [0..length
[ do
354 if char
.is_upper
then return false
359 # Removes the whitespaces at the beginning of self
361 # assert " \n\thello \n\t".l_trim == "hello \n\t"
363 # `Char::is_whitespace` determines what is a whitespace.
366 var iter
= self.chars
.iterator
368 if not iter
.item
.is_whitespace
then break
371 if iter
.index
== length
then return self.empty
372 return self.substring_from
(iter
.index
)
375 # Removes the whitespaces at the end of self
377 # assert " \n\thello \n\t".r_trim == " \n\thello"
379 # `Char::is_whitespace` determines what is a whitespace.
382 var iter
= self.chars
.reverse_iterator
384 if not iter
.item
.is_whitespace
then break
387 if iter
.index
< 0 then return self.empty
388 return self.substring
(0, iter
.index
+ 1)
391 # Trims trailing and preceding white spaces
393 # assert " Hello World ! ".trim == "Hello World !"
394 # assert "\na\nb\tc\t".trim == "a\nb\tc"
396 # `Char::is_whitespace` determines what is a whitespace.
397 fun trim
: SELFTYPE do return (self.l_trim
).r_trim
399 # Is the string non-empty but only made of whitespaces?
401 # assert " \n\t ".is_whitespace == true
402 # assert " hello ".is_whitespace == false
403 # assert "".is_whitespace == false
405 # `Char::is_whitespace` determines what is a whitespace.
406 fun is_whitespace
: Bool
408 if is_empty
then return false
409 for c
in self.chars
do
410 if not c
.is_whitespace
then return false
415 # Returns `self` removed from its last line terminator (if any).
417 # assert "Hello\n".chomp == "Hello"
418 # assert "Hello".chomp == "Hello"
420 # assert "\n".chomp == ""
421 # assert "".chomp == ""
423 # Line terminators are `"\n"`, `"\r\n"` and `"\r"`.
424 # A single line terminator, the last one, is removed.
426 # assert "\r\n".chomp == ""
427 # assert "\r\n\n".chomp == "\r\n"
428 # assert "\r\n\r\n".chomp == "\r\n"
429 # assert "\r\n\r".chomp == "\r\n"
431 # Note: unlike with most IO methods like `Reader::read_line`,
432 # a single `\r` is considered here to be a line terminator and will be removed.
436 if len
== 0 then return self
437 var l
= self.chars
.last
439 return substring
(0, len-1
)
440 else if l
!= '\n' then
442 else if len
> 1 and self.chars
[len-2
] == '\r' then
443 return substring
(0, len-2
)
445 return substring
(0, len-1
)
449 # Justify a self in a space of `length`
451 # `left` is the space ratio on the left side.
452 # * 0.0 for left-justified (no space at the left)
453 # * 1.0 for right-justified (all spaces at the left)
454 # * 0.5 for centered (half the spaces at the left)
458 # assert "hello".justify(10, 0.0) == "hello "
459 # assert "hello".justify(10, 1.0) == " hello"
460 # assert "hello".justify(10, 0.5) == " hello "
462 # If `length` is not enough, `self` is returned as is.
464 # assert "hello".justify(2, 0.0) == "hello"
466 # REQUIRE: `left >= 0.0 and left <= 1.0`
467 # ENSURE: `self.length <= length implies result.length == length`
468 # ENSURE: `self.length >= length implies result == self`
469 fun justify
(length
: Int, left
: Float): String
471 var diff
= length
- self.length
472 if diff
<= 0 then return to_s
473 assert left
>= 0.0 and left
<= 1.0
474 var before
= (diff
.to_f
* left
).to_i
475 return " " * before
+ self + " " * (diff-before
)
478 # Mangle a string to be a unique string only made of alphanumeric characters and underscores.
480 # This method is injective (two different inputs never produce the same
481 # output) and the returned string always respect the following rules:
483 # * Contains only US-ASCII letters, digits and underscores.
484 # * Never starts with a digit.
485 # * Never ends with an underscore.
486 # * Never contains two contiguous underscores.
488 # assert "42_is/The answer!".to_cmangle == "_52d2_is_47dThe_32danswer_33d"
489 # assert "__".to_cmangle == "_95d_95d"
490 # assert "__d".to_cmangle == "_95d_d"
491 # assert "_d_".to_cmangle == "_d_95d"
492 # assert "_42".to_cmangle == "_95d42"
493 # assert "foo".to_cmangle == "foo"
494 # assert "".to_cmangle == ""
495 fun to_cmangle
: String
497 if is_empty
then return ""
499 var underscore
= false
503 if c
>= '0' and c
<= '9' then
505 res
.append
(c
.ascii
.to_s
)
509 for i
in [start
..length
[ do
511 if (c
>= 'a' and c
<= 'z') or (c
>='A' and c
<= 'Z') then
517 res
.append
('_'.ascii
.to_s
)
520 if c
>= '0' and c
<= '9' then
523 else if c
== '_' then
528 res
.append
(c
.ascii
.to_s
)
534 res
.append
('_'.ascii
.to_s
)
540 # Escape " \ ' and non printable characters using the rules of literal C strings and characters
542 # assert "abAB12<>&".escape_to_c == "abAB12<>&"
543 # assert "\n\"'\\".escape_to_c == "\\n\\\"\\'\\\\"
545 # Most non-printable characters (bellow ASCII 32) are escaped to an octal form `\nnn`.
546 # Three digits are always used to avoid following digits to be interpreted as an element
547 # of the octal sequence.
549 # assert "{0.ascii}{1.ascii}{8.ascii}{31.ascii}{32.ascii}".escape_to_c == "\\000\\001\\010\\037 "
551 # The exceptions are the common `\t` and `\n`.
552 fun escape_to_c
: String
555 for i
in [0..length
[ do
559 else if c
== '\t' then
561 else if c
== '\0' then
563 else if c
== '"' then
565 else if c == '\'' then
567 else if c == '\\
' then
569 else if c.ascii < 32 then
571 var oct = c.ascii.to_base(8, false)
572 # Force 3 octal digits since it is the
573 # maximum allowed in the C specification
574 if oct.length == 1 then
577 else if oct.length == 2 then
588 # Escape additionnal characters
589 # The result might no be legal in C but be used in other languages
591 # assert "ab|\{\}".escape_more_to_c("|\{\}") == "ab\\|\\\{\\\}"
592 fun escape_more_to_c(chars: String): String
595 for c in escape_to_c.chars do
596 if chars.chars.has(c) then
604 # Escape to C plus braces
606 # assert "\n\"'\\\
{\}".escape_to_nit == "\\n\\\
"\\'\\\\\\\{\\\}"
607 fun escape_to_nit
: String do return escape_more_to_c
("\{\}")
609 # Escape to POSIX Shell (sh).
611 # Abort if the text contains a null byte.
613 # assert "\n\"'\\\{\}0".escape_to_sh == "'\n\"'\\''\\\{\}0'"
614 fun escape_to_sh
: String do
617 for i in [0..length[ do
622 assert without_null_byte
: c
!= '\0'
630 # Escape to include in a Makefile
632 # Unfortunately, some characters are not escapable in Makefile.
633 # These characters are `;`, `|`, `\`, and the non-printable ones.
634 # They will be rendered as `"?{hex}"`.
635 fun escape_to_mk: String do
637 for i in [0..length[ do
641 else if c == ':' or c == ' ' or c == '#' then
644 else if c
.ascii
< 32 or c
== ';' or c
== '|' or c
== '\\' or c
== '=' then
645 b
.append
("?{c.ascii.to_base(16, false)}")
653 # Return a string where Nit escape sequences are transformed.
656 # assert s.length == 2
657 # var u = s.unescape_nit
658 # assert u.length == 1
659 # assert u.chars[0].ascii == 10 # (the ASCII value of the "new line" character)
660 fun unescape_nit
: String
662 var res
= new Buffer.with_cap
(self.length
)
663 var was_slash
= false
664 for i
in [0..length
[ do
666 if not was_slash
then
677 else if c
== 'r' then
679 else if c
== 't' then
681 else if c
== '0' then
690 # Encode `self` to percent (or URL) encoding
692 # assert "aBc09-._~".to_percent_encoding == "aBc09-._~"
693 # assert "%()< >".to_percent_encoding == "%25%28%29%3c%20%3e"
694 # assert ".com/post?e=asdf&f=123".to_percent_encoding == ".com%2fpost%3fe%3dasdf%26f%3d123"
695 fun to_percent_encoding
: String
699 for i
in [0..length
[ do
701 if (c
>= '0' and c
<= '9') or
702 (c
>= 'a' and c
<= 'z') or
703 (c
>= 'A' and c
<= 'Z') or
704 c
== '-' or c
== '.' or
708 else buf
.append
"%{c.ascii.to_hex}"
714 # Decode `self` from percent (or URL) encoding to a clear string
716 # Replace invalid use of '%' with '?'.
718 # assert "aBc09-._~".from_percent_encoding == "aBc09-._~"
719 # assert "%25%28%29%3c%20%3e".from_percent_encoding == "%()< >"
720 # assert ".com%2fpost%3fe%3dasdf%26f%3d123".from_percent_encoding == ".com/post?e=asdf&f=123"
721 # assert "%25%28%29%3C%20%3E".from_percent_encoding == "%()< >"
722 # assert "incomplete %".from_percent_encoding == "incomplete ?"
723 # assert "invalid % usage".from_percent_encoding == "invalid ? usage"
724 fun from_percent_encoding
: String
732 if i
+ 2 >= length
then
733 # What follows % has been cut off
737 var hex_s
= substring
(i
, 2)
739 var hex_i
= hex_s
.to_hex
743 # What follows a % is not Hex
756 # Escape the characters `<`, `>`, `&`, `"`, `'` and `/` as HTML/XML entity references.
758 # assert "a&b-<>\"x\"/'".html_escape == "a&b-<>"x"/'"
760 # 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>
761 fun html_escape
: String
765 for i
in [0..length
[ do
769 else if c
== '<' then
771 else if c
== '>' then
773 else if c
== '"' then
775 else if c
== '\'' then
777 else if c == '/' then
786 # Two pieces of text are equals if thez have the same characters in the same order.
788 # assert "hello" == "hello"
789 # assert "hello" != "HELLO"
790 # assert "hello" == "hel"+"lo"
792 # Things that are not Text are not equal.
795 # assert "9" != ['9']
798 # assert "9".chars.first == '9' # equality of Char
799 # assert "9".chars == ['9'] # equality of Sequence
800 # assert "9".to_i == 9 # equality of Int
803 if o == null then return false
804 if not o isa Text then return false
805 if self.is_same_instance(o) then return true
806 if self.length != o.length then return false
807 return self.chars == o.chars
810 # Lexicographical comparaison
812 # assert "abc" < "xy"
813 # assert "ABC" < "abc"
816 var self_chars = self.chars.iterator
817 var other_chars = other.chars.iterator
819 while self_chars.is_ok and other_chars.is_ok do
820 if self_chars.item < other_chars.item then return true
821 if self_chars.item > other_chars.item then return false
826 if self_chars.is_ok then
833 # Escape string used in labels for graphviz
835 # assert ">><<".escape_to_dot == "\\>\\>\\<\\<"
836 fun escape_to_dot: String
838 return escape_more_to_c("|\{\}<>")
841 private var hash_cache: nullable Int = null
845 if hash_cache == null then
846 # djb2 hash algorithm
849 for i in [0..length[ do
851 h = h.lshift(5) + h + char.ascii
856 return hash_cache.as(not null)
859 # Gives the formatted string back as a Nit string with `args` in place
861 # assert "This %1 is a %2.".format("String", "formatted String") == "This String is a formatted String."
862 # assert "\\%1 This string".format("String") == "\\%1 This string"
863 fun format(args: Object...): String do
864 var s = new Array[Text]
868 # Skip escaped characters
869 if self[i] == '\\
' then
872 else if self[i] == '%' then
876 while i < length and self[i].is_numeric do
881 var ciph_len = fmt_end - ciph_st + 1
882 s.push substring(curr_st, fmt_st - curr_st)
883 s.push args[substring(ciph_st, ciph_len).to_i - 1].to_s
888 s.push substring(curr_st, length - curr_st)
892 # Copies `n` bytes from `self` at `src_offset` into `dest` starting at `dest_offset`
894 # Basically a high-level synonym of NativeString::copy_to
896 # REQUIRE: `n` must be large enough to contain `len` bytes
898 # var ns = new NativeString(8)
899 # "Text is String".copy_to_native(ns, 8, 2, 0)
900 # assert ns.to_s_with_length(8) == "xt is St"
902 fun copy_to_native(dest: NativeString, n, src_offset, dest_offset: Int) do
903 var mypos = src_offset
904 var itspos = dest_offset
906 dest[itspos] = self.chars[mypos]
915 # All kinds of array-based text representations.
916 abstract class FlatText
919 # Underlying C-String (`char*`)
921 # Warning : Might be void in some subclasses, be sure to check
922 # if set before using it.
923 private var items: NativeString is noinit
925 # Real items, used as cache for to_cstring is called
926 private var real_items: nullable NativeString = null
928 # Returns a char* starting at position `index_from`
930 # WARNING: If you choose to use this service, be careful of the following.
932 # Strings and NativeString are *ideally* always allocated through a Garbage Collector.
933 # Since the GC tracks the use of the pointer for the beginning of the char*, it may be
934 # deallocated at any moment, rendering the pointer returned by this function invalid.
935 # Any access to freed memory may very likely cause undefined behaviour or a crash.
936 # (Failure to do so will most certainly result in long and painful debugging hours)
938 # The only safe use of this pointer is if it is ephemeral (e.g. read in a C function
939 # then immediately return).
941 # As always, do not modify the content of the String in C code, if this is what you want
942 # copy locally the char* as Nit Strings are immutable.
943 private fun fast_cstring: NativeString is abstract
956 redef fun copy_to_native(dest, n, src_offset, dest_offset) do
957 items.copy_to(dest, n, src_offset, dest_offset)
961 # Abstract class for the SequenceRead compatible
962 # views on String and Buffer objects
963 private abstract class StringCharView
964 super SequenceRead[Char]
970 redef fun is_empty do return target.is_empty
972 redef fun length do return target.length
974 redef fun iterator: IndexedIterator[Char] do return self.iterator_from(0)
976 redef fun reverse_iterator do return self.reverse_iterator_from(self.length - 1)
979 # View on Buffer objects, extends Sequence
980 # for mutation operations
981 private abstract class BufferCharView
985 redef type SELFTYPE: Buffer
989 # A `String` holds and manipulates an arbitrary sequence of characters.
991 # String objects may be created using literals.
993 # assert "Hello World!" isa String
994 abstract class String
997 redef type SELFTYPE: String is fixed
999 redef fun to_s do return self
1001 # Concatenates `o` to `self`
1003 # assert "hello" + "world" == "helloworld"
1004 # assert "" + "hello" + "" == "hello"
1005 fun +(o: Text): SELFTYPE is abstract
1007 # Concatenates self `i` times
1009 # assert "abc" * 4 == "abcabcabcabc"
1010 # assert "abc" * 1 == "abc"
1011 # assert "abc" * 0 == ""
1012 fun *(i: Int): SELFTYPE is abstract
1014 # Insert `s` at `pos`.
1016 # assert "helloworld".insert_at(" ", 5) == "hello world"
1017 fun insert_at(s: String, pos: Int): SELFTYPE is abstract
1019 redef fun substrings is abstract
1021 # Returns a reversed version of self
1023 # assert "hello".reversed == "olleh"
1024 # assert "bob".reversed == "bob"
1025 # assert "".reversed == ""
1026 fun reversed: SELFTYPE is abstract
1028 # A upper case version of `self`
1030 # assert "Hello World!".to_upper == "HELLO WORLD!"
1031 fun to_upper: SELFTYPE is abstract
1033 # A lower case version of `self`
1035 # assert "Hello World!".to_lower == "hello world!"
1036 fun to_lower : SELFTYPE is abstract
1038 # Takes a camel case `self` and converts it to snake case
1040 # assert "randomMethodId".to_snake_case == "random_method_id"
1042 # The rules are the following:
1044 # An uppercase is always converted to a lowercase
1046 # assert "HELLO_WORLD".to_snake_case == "hello_world"
1048 # An uppercase that follows a lowercase is prefixed with an underscore
1050 # assert "HelloTheWORLD".to_snake_case == "hello_the_world"
1052 # An uppercase that follows an uppercase and is followed by a lowercase, is prefixed with an underscore
1054 # assert "HelloTHEWorld".to_snake_case == "hello_the_world"
1056 # All other characters are kept as is; `self` does not need to be a proper CamelCased string.
1058 # assert "=-_H3ll0Th3W0rld_-=".to_snake_case == "=-_h3ll0th3w0rld_-="
1059 fun to_snake_case: SELFTYPE
1061 if self.is_lower then return self
1063 var new_str = new Buffer.with_cap(self.length)
1064 var prev_is_lower = false
1065 var prev_is_upper = false
1067 for i in [0..length[ do
1069 if char.is_lower then
1071 prev_is_lower = true
1072 prev_is_upper = false
1073 else if char.is_upper then
1074 if prev_is_lower then
1076 else if prev_is_upper and i+1 < length and chars[i+1].is_lower then
1079 new_str.add(char.to_lower)
1080 prev_is_lower = false
1081 prev_is_upper = true
1084 prev_is_lower = false
1085 prev_is_upper = false
1092 # Takes a snake case `self` and converts it to camel case
1094 # assert "random_method_id".to_camel_case == "randomMethodId"
1096 # If the identifier is prefixed by an underscore, the underscore is ignored
1098 # assert "_private_field".to_camel_case == "_privateField"
1100 # If `self` is upper, it is returned unchanged
1102 # assert "RANDOM_ID".to_camel_case == "RANDOM_ID"
1104 # If there are several consecutive underscores, they are considered as a single one
1106 # assert "random__method_id".to_camel_case == "randomMethodId"
1107 fun to_camel_case: SELFTYPE
1109 if self.is_upper then return self
1111 var new_str = new Buffer
1112 var is_first_char = true
1113 var follows_us = false
1115 for i in [0..length[ do
1117 if is_first_char then
1119 is_first_char = false
1120 else if char == '_
' then
1122 else if follows_us then
1123 new_str.add(char.to_upper)
1133 # Returns a capitalized `self`
1135 # Letters that follow a letter are lowercased
1136 # Letters that follow a non-letter are upcased.
1138 # SEE : `Char::is_letter` for the definition of letter.
1140 # assert "jAVASCRIPT".capitalized == "Javascript"
1141 # assert "i am root".capitalized == "I Am Root"
1142 # assert "ab_c -ab0c ab\nc".capitalized == "Ab_C -Ab0C Ab\nC"
1143 fun capitalized: SELFTYPE do
1144 if length == 0 then return self
1146 var buf = new Buffer.with_cap(length)
1148 var curr = chars[0].to_upper
1152 for i in [1 .. length[ do
1155 if prev.is_letter then
1156 buf[i] = curr.to_lower
1158 buf[i] = curr.to_upper
1166 private class FlatSubstringsIter
1167 super Iterator[FlatText]
1169 var tgt: nullable FlatText
1173 return tgt.as(not null)
1176 redef fun is_ok do return tgt != null
1178 redef fun next do tgt = null
1181 # Immutable strings of characters.
1186 # Index in _items of the start of the string
1187 private var index_from: Int is noinit
1189 # Indes in _items of the last item of the string
1190 private var index_to: Int is noinit
1192 redef var chars = new FlatStringCharView(self) is lazy
1196 # Check that the index (+ index_from) is not larger than indexTo
1197 # In other terms, if the index is valid
1199 assert (index + index_from) <= index_to
1200 return items[index + index_from]
1203 ################################################
1204 # AbstractString specific methods #
1205 ################################################
1209 var native = new NativeString(self.length + 1)
1210 var length = self.length
1211 var items = self.items
1214 while pos < length do
1215 native[pos] = items[ipos]
1219 return native.to_s_with_length(self.length)
1222 redef fun fast_cstring do return items.fast_cstring(index_from)
1224 redef fun substring(from, count)
1230 if count < 0 then count = 0
1234 var new_from = index_from + from
1236 if (new_from + count) > index_to then
1237 var new_len = index_to - new_from + 1
1238 if new_len <= 0 then return empty
1239 return new FlatString.with_infos(items, new_len, new_from, index_to)
1242 if count <= 0 then return empty
1244 var to = new_from + count - 1
1246 return new FlatString.with_infos(items, to - new_from + 1, new_from, to)
1249 redef fun empty do return "".as(FlatString)
1253 var outstr = new NativeString(self.length + 1)
1256 var myitems = self.items
1257 var index_from = self.index_from
1258 var max = self.index_to
1260 while index_from <= max do
1261 outstr[out_index] = myitems[index_from].to_upper
1266 outstr[self.length] = '\
0'
1268 return outstr.to_s_with_length(self.length)
1273 var outstr = new NativeString(self.length + 1)
1276 var myitems = self.items
1277 var index_from = self.index_from
1278 var max = self.index_to
1280 while index_from <= max do
1281 outstr[out_index] = myitems[index_from].to_lower
1286 outstr[self.length] = '\
0'
1288 return outstr.to_s_with_length(self.length)
1293 var i = self.index_from
1294 var imax = self.index_to
1301 ##################################################
1302 # String Specific Methods #
1303 ##################################################
1305 # Low-level creation of a new string with given data.
1307 # `items` will be used as is, without copy, to retrieve the characters of the string.
1308 # Aliasing issues is the responsibility of the caller.
1309 private init with_infos(items: NativeString, length: Int, from: Int, to: Int)
1312 self.length = length
1317 redef fun to_cstring do
1318 if real_items != null then
1319 return real_items.as(not null)
1321 var newItems = new NativeString(length + 1)
1322 self.items.copy_to(newItems, length, index_from, 0)
1323 newItems[length] = '\
0'
1324 self.real_items = newItems
1331 if not other isa FlatString then return super
1333 if self.object_id == other.object_id then return true
1335 var my_length = length
1337 if other.length != my_length then return false
1339 var my_index = index_from
1340 var its_index = other.index_from
1342 var last_iteration = my_index + my_length
1344 var itsitems = other.items
1345 var myitems = self.items
1347 while my_index < last_iteration do
1348 if myitems[my_index] != itsitems[its_index] then return false
1358 if not other isa FlatString then return super
1360 if self.object_id == other.object_id then return false
1362 var my_curr_char : Char
1363 var its_curr_char : Char
1365 var curr_id_self = self.index_from
1366 var curr_id_other = other.index_from
1368 var my_items = self.items
1369 var its_items = other.items
1371 var my_length = self.length
1372 var its_length = other.length
1374 var max_iterations = curr_id_self + my_length
1376 while curr_id_self < max_iterations do
1377 my_curr_char = my_items[curr_id_self]
1378 its_curr_char = its_items[curr_id_other]
1380 if my_curr_char != its_curr_char then
1381 if my_curr_char < its_curr_char then return true
1389 return my_length < its_length
1394 var my_length = self.length
1395 var its_length = s.length
1397 var total_length = my_length + its_length
1399 var target_string = new NativeString(my_length + its_length + 1)
1401 self.items.copy_to(target_string, my_length, index_from, 0)
1402 if s isa FlatString then
1403 s.items.copy_to(target_string, its_length, s.index_from, my_length)
1404 else if s isa FlatBuffer then
1405 s.items.copy_to(target_string, its_length, 0, my_length)
1407 var curr_pos = my_length
1408 for i in [0..s.length[ do
1410 target_string[curr_pos] = c
1415 target_string[total_length] = '\
0'
1417 return target_string.to_s_with_length(total_length)
1424 var my_length = self.length
1426 var final_length = my_length * i
1428 var my_items = self.items
1430 var target_string = new NativeString(final_length + 1)
1432 target_string[final_length] = '\
0'
1434 var current_last = 0
1436 for iteration in [1 .. i] do
1437 my_items.copy_to(target_string, my_length, 0, current_last)
1438 current_last += my_length
1441 return target_string.to_s_with_length(final_length)
1446 if hash_cache == null then
1447 # djb2 hash algorithm
1453 while i <= index_to do
1454 h = h.lshift(5) + h + myitems[i].ascii
1461 return hash_cache.as(not null)
1464 redef fun substrings do return new FlatSubstringsIter(self)
1467 private class FlatStringReverseIterator
1468 super IndexedIterator[Char]
1470 var target: FlatString
1472 var target_items: NativeString
1476 init with_pos(tgt: FlatString, pos: Int)
1479 target_items = tgt.items
1480 curr_pos = pos + tgt.index_from
1483 redef fun is_ok do return curr_pos >= target.index_from
1485 redef fun item do return target_items[curr_pos]
1487 redef fun next do curr_pos -= 1
1489 redef fun index do return curr_pos - target.index_from
1493 private class FlatStringIterator
1494 super IndexedIterator[Char]
1496 var target: FlatString
1498 var target_items: NativeString
1502 init with_pos(tgt: FlatString, pos: Int)
1505 target_items = tgt.items
1506 curr_pos = pos + target.index_from
1509 redef fun is_ok do return curr_pos <= target.index_to
1511 redef fun item do return target_items[curr_pos]
1513 redef fun next do curr_pos += 1
1515 redef fun index do return curr_pos - target.index_from
1519 private class FlatStringCharView
1520 super StringCharView
1522 redef type SELFTYPE: FlatString
1526 # Check that the index (+ index_from) is not larger than indexTo
1527 # In other terms, if the index is valid
1529 var target = self.target
1530 assert (index + target.index_from) <= target.index_to
1531 return target.items[index + target.index_from]
1534 redef fun iterator_from(start) do return new FlatStringIterator.with_pos(target, start)
1536 redef fun reverse_iterator_from(start) do return new FlatStringReverseIterator.with_pos(target, start)
1540 # A mutable sequence of characters.
1541 abstract class Buffer
1544 # New `Buffer` factory, will return a concrete `Buffer` type with default capacity
1545 new do return new FlatBuffer
1547 # New `Buffer` factory, returns a concrete `Buffer` with a capacity of `i`
1548 new with_cap(i: Int) do return new FlatBuffer.with_capacity(i)
1550 redef type SELFTYPE: Buffer is fixed
1552 # Specific implementations MUST set this to `true` in order to invalidate caches
1553 protected var is_dirty = true
1555 # Copy-On-Write flag
1557 # If the `Buffer` was to_s'd
, the next in-place altering
1558 # operation will cause the current `Buffer` to be re-allocated.
1560 # The flag will then be set at `false`.
1561 protected var written
= false
1563 # Modifies the char contained at pos `index`
1565 # DEPRECATED : Use self.chars.[]= instead
1566 fun []=(index
: Int, item
: Char) is abstract
1568 # Adds a char `c` at the end of self
1570 # DEPRECATED : Use self.chars.add instead
1571 fun add
(c
: Char) is abstract
1575 # var b = new Buffer
1577 # assert not b.is_empty
1580 fun clear
is abstract
1582 # Enlarges the subsequent array containing the chars of self
1583 fun enlarge
(cap
: Int) is abstract
1585 # Adds the content of text `s` at the end of self
1587 # var b = new Buffer
1590 # assert b == "helloworld"
1591 fun append
(s
: Text) is abstract
1593 # `self` is appended in such a way that `self` is repeated `r` times
1595 # var b = new Buffer
1598 # assert b == "hellohellohello"
1599 fun times
(r
: Int) is abstract
1601 # Reverses itself in-place
1603 # var b = new Buffer
1606 # assert b == "olleh"
1607 fun reverse
is abstract
1609 # Changes each lower-case char in `self` by its upper-case variant
1611 # var b = new Buffer
1612 # b.append("Hello World!")
1614 # assert b == "HELLO WORLD!"
1615 fun upper
is abstract
1617 # Changes each upper-case char in `self` by its lower-case variant
1619 # var b = new Buffer
1620 # b.append("Hello World!")
1622 # assert b == "hello world!"
1623 fun lower
is abstract
1625 # Capitalizes each word in `self`
1627 # Letters that follow a letter are lowercased
1628 # Letters that follow a non-letter are upcased.
1630 # SEE: `Char::is_letter` for the definition of a letter.
1632 # var b = new FlatBuffer.from("jAVAsCriPt")
1634 # assert b == "Javascript"
1635 # b = new FlatBuffer.from("i am root")
1637 # assert b == "I Am Root"
1638 # b = new FlatBuffer.from("ab_c -ab0c ab\nc")
1640 # assert b == "Ab_C -Ab0C Ab\nC"
1642 if length
== 0 then return
1643 var c
= self[0].to_upper
1646 for i
in [1 .. length
[ do
1649 if prev
.is_letter
then
1650 self[i
] = c
.to_lower
1652 self[i
] = c
.to_upper
1659 if is_dirty
then hash_cache
= null
1663 # In Buffers, the internal sequence of character is mutable
1664 # Thus, `chars` can be used to modify the buffer.
1665 redef fun chars
: Sequence[Char] is abstract
1668 # Mutable strings of characters.
1673 redef var chars
: Sequence[Char] = new FlatBufferCharView(self) is lazy
1675 private var capacity
: Int = 0
1677 redef fun fast_cstring
do return items
.fast_cstring
(0)
1679 redef fun substrings
do return new FlatSubstringsIter(self)
1681 # Re-copies the `NativeString` into a new one and sets it as the new `Buffer`
1683 # This happens when an operation modifies the current `Buffer` and
1684 # the Copy-On-Write flag `written` is set at true.
1685 private fun reset
do
1686 var nns
= new NativeString(capacity
)
1687 items
.copy_to
(nns
, length
, 0, 0)
1695 assert index
< length
1699 redef fun []=(index
, item
)
1702 if index
== length
then
1706 if written
then reset
1707 assert index
>= 0 and index
< length
1714 if capacity
<= length
then enlarge
(length
+ 5)
1721 if written
then reset
1725 redef fun empty
do return new Buffer
1727 redef fun enlarge
(cap
)
1730 if cap
<= c
then return
1731 while c
<= cap
do c
= c
* 2 + 2
1732 # The COW flag can be set at false here, since
1733 # it does a copy of the current `Buffer`
1735 var a
= new NativeString(c
+1)
1736 if length
> 0 then items
.copy_to
(a
, length
, 0, 0)
1743 if length
== 0 then items
= new NativeString(1)
1744 return new FlatString.with_infos
(items
, length
, 0, length
- 1)
1747 redef fun to_cstring
1750 var new_native
= new NativeString(length
+ 1)
1751 new_native
[length
] = '\0'
1752 if length
> 0 then items
.copy_to
(new_native
, length
, 0, 0)
1753 real_items
= new_native
1756 return real_items
.as(not null)
1759 # Create a new empty string.
1762 # Low-level creation a new buffer with given data.
1764 # `items` will be used as is, without copy, to store the characters of the buffer.
1765 # Aliasing issues is the responsibility of the caller.
1767 # If `items` is shared, `written` should be set to true after the creation
1768 # so that a modification will do a copy-on-write.
1769 private init with_infos
(items
: NativeString, capacity
, length
: Int)
1772 self.length
= length
1773 self.capacity
= capacity
1776 # Create a new string copied from `s`.
1779 capacity
= s
.length
+ 1
1781 items
= new NativeString(capacity
)
1782 if s
isa FlatString then
1783 s
.items
.copy_to
(items
, length
, s
.index_from
, 0)
1784 else if s
isa FlatBuffer then
1785 s
.items
.copy_to
(items
, length
, 0, 0)
1788 for i
in [0..s
.length
[ do
1796 # Create a new empty string with a given capacity.
1797 init with_capacity
(cap
: Int)
1800 items
= new NativeString(cap
+1)
1807 if s
.is_empty
then return
1810 if capacity
< length
+ sl
then enlarge
(length
+ sl
)
1811 if s
isa FlatString then
1812 s
.items
.copy_to
(items
, sl
, s
.index_from
, length
)
1813 else if s
isa FlatBuffer then
1814 s
.items
.copy_to
(items
, sl
, 0, length
)
1816 var curr_pos
= self.length
1817 for i
in [0..s
.length
[ do
1826 # Copies the content of self in `dest`
1827 fun copy
(start
: Int, len
: Int, dest
: Buffer, new_start
: Int)
1829 var self_chars
= self.chars
1830 var dest_chars
= dest
.chars
1831 for i
in [0..len-1
] do
1832 dest_chars
[new_start
+i
] = self_chars
[start
+i
]
1836 redef fun substring
(from
, count
)
1840 if from
< 0 then from
= 0
1841 if count
> length
then count
= length
1842 if from
< count
then
1843 var len
= count
- from
1844 var r_items
= new NativeString(len
)
1845 items
.copy_to
(r_items
, len
, from
, 0)
1846 var r
= new FlatBuffer.with_infos
(r_items
, len
, len
)
1856 var ns
= new NativeString(capacity
)
1868 redef fun times
(repeats
)
1870 var x
= new FlatString.with_infos
(items
, length
, 0, length
- 1)
1871 for i
in [1..repeats
[ do
1878 if written
then reset
1882 it
[id
] = it
[id
].to_upper
1889 if written
then reset
1893 it
[id
] = it
[id
].to_lower
1899 private class FlatBufferReverseIterator
1900 super IndexedIterator[Char]
1902 var target
: FlatBuffer
1904 var target_items
: NativeString
1908 init with_pos
(tgt
: FlatBuffer, pos
: Int)
1911 if tgt
.length
> 0 then target_items
= tgt
.items
1915 redef fun index
do return curr_pos
1917 redef fun is_ok
do return curr_pos
>= 0
1919 redef fun item
do return target_items
[curr_pos
]
1921 redef fun next
do curr_pos
-= 1
1925 private class FlatBufferCharView
1926 super BufferCharView
1928 redef type SELFTYPE: FlatBuffer
1930 redef fun [](index
) do return target
.items
[index
]
1932 redef fun []=(index
, item
)
1934 assert index
>= 0 and index
<= length
1935 if index
== length
then
1939 target
.items
[index
] = item
1952 fun enlarge
(cap
: Int)
1959 var s_length
= s
.length
1960 if target
.capacity
< s
.length
then enlarge
(s_length
+ target
.length
)
1963 redef fun iterator_from
(pos
) do return new FlatBufferIterator.with_pos
(target
, pos
)
1965 redef fun reverse_iterator_from
(pos
) do return new FlatBufferReverseIterator.with_pos
(target
, pos
)
1969 private class FlatBufferIterator
1970 super IndexedIterator[Char]
1972 var target
: FlatBuffer
1974 var target_items
: NativeString
1978 init with_pos
(tgt
: FlatBuffer, pos
: Int)
1981 if tgt
.length
> 0 then target_items
= tgt
.items
1985 redef fun index
do return curr_pos
1987 redef fun is_ok
do return curr_pos
< target
.length
1989 redef fun item
do return target_items
[curr_pos
]
1991 redef fun next
do curr_pos
+= 1
1995 ###############################################################################
1997 ###############################################################################
2000 # User readable representation of `self`.
2001 fun to_s
: String do return inspect
2003 # The class name of the object in NativeString format.
2004 private fun native_class_name
: NativeString is intern
2006 # The class name of the object.
2008 # assert 5.class_name == "Int"
2009 fun class_name
: String do return native_class_name
.to_s
2011 # Developer readable representation of `self`.
2012 # Usually, it uses the form "<CLASSNAME:#OBJECTID bla bla bla>"
2015 return "<{inspect_head}>"
2018 # Return "CLASSNAME:#OBJECTID".
2019 # This function is mainly used with the redefinition of the inspect method
2020 protected fun inspect_head
: String
2022 return "{class_name}:#{object_id.to_hex}"
2027 # assert true.to_s == "true"
2028 # assert false.to_s == "false"
2040 # C function to calculate the length of the `NativeString` to receive `self`
2041 private fun byte_to_s_len
: Int is extern "native_byte_length_str"
2043 # C function to convert an nit Int to a NativeString (char*)
2044 private fun native_byte_to_s
(nstr
: NativeString, strlen
: Int) is extern "native_byte_to_s"
2046 # Displayable byte in its hexadecimal form (0x..)
2048 # assert 1.to_b.to_s == "0x01"
2049 # assert (-123).to_b.to_s == "0x85"
2051 var nslen
= byte_to_s_len
2052 var ns
= new NativeString(nslen
+ 1)
2054 native_byte_to_s
(ns
, nslen
+ 1)
2055 return ns
.to_s_with_length
(nslen
)
2061 # Wrapper of strerror C function
2062 private fun strerror_ext
: NativeString is extern "strerror"
2064 # Returns a string describing error number
2065 fun strerror
: String do return strerror_ext
.to_s
2067 # Fill `s` with the digits in base `base` of `self` (and with the '-' sign if 'signed' and negative).
2068 # assume < to_c max const of char
2069 private fun fill_buffer
(s
: Buffer, base
: Int, signed
: Bool)
2076 else if self == 0 then
2083 var pos
= digit_count
(base
) - 1
2084 while pos
>= 0 and n
> 0 do
2085 s
.chars
[pos
] = (n
% base
).to_c
2091 # C function to calculate the length of the `NativeString` to receive `self`
2092 private fun int_to_s_len
: Int is extern "native_int_length_str"
2094 # C function to convert an nit Int to a NativeString (char*)
2095 private fun native_int_to_s
(nstr
: NativeString, strlen
: Int) is extern "native_int_to_s"
2097 # return displayable int in base 10 and signed
2099 # assert 1.to_s == "1"
2100 # assert (-123).to_s == "-123"
2102 # Fast case for common numbers
2103 if self == 0 then return "0"
2104 if self == 1 then return "1"
2106 var nslen
= int_to_s_len
2107 var ns
= new NativeString(nslen
+ 1)
2109 native_int_to_s
(ns
, nslen
+ 1)
2110 return ns
.to_s_with_length
(nslen
)
2113 # return displayable int in hexadecimal
2115 # assert 1.to_hex == "1"
2116 # assert (-255).to_hex == "-ff"
2117 fun to_hex
: String do return to_base
(16,false)
2119 # return displayable int in base base and signed
2120 fun to_base
(base
: Int, signed
: Bool): String
2122 var l
= digit_count
(base
)
2123 var s
= new FlatBuffer.from
(" " * l
)
2124 fill_buffer
(s
, base
, signed
)
2130 # Pretty representation of `self`, with decimals as needed from 1 to a maximum of 3
2132 # assert 12.34.to_s == "12.34"
2133 # assert (-0120.030).to_s == "-120.03"
2135 # see `to_precision` for a custom precision.
2137 var str
= to_precision
( 3 )
2138 if is_inf
!= 0 or is_nan
then return str
2139 var len
= str
.length
2140 for i
in [0..len-1
] do
2142 var c
= str
.chars
[j
]
2145 else if c
== '.' then
2146 return str
.substring
( 0, j
+2 )
2148 return str
.substring
( 0, j
+1 )
2154 # `String` representation of `self` with the given number of `decimals`
2156 # assert 12.345.to_precision(0) == "12"
2157 # assert 12.345.to_precision(3) == "12.345"
2158 # assert (-12.345).to_precision(3) == "-12.345"
2159 # assert (-0.123).to_precision(3) == "-0.123"
2160 # assert 0.999.to_precision(2) == "1.00"
2161 # assert 0.999.to_precision(4) == "0.9990"
2162 fun to_precision
(decimals
: Int): String
2164 if is_nan
then return "nan"
2166 var isinf
= self.is_inf
2169 else if isinf
== -1 then
2173 if decimals
== 0 then return self.to_i
.to_s
2175 for i
in [0..decimals
[ do f
= f
* 10.0
2182 if i
== 0 then return "0." + "0"*decimals
2184 # Prepare both parts of the float, before and after the "."
2189 if sl
> decimals
then
2190 # Has something before the "."
2191 p1
= s
.substring
(0, sl-decimals
)
2192 p2
= s
.substring
(sl-decimals
, decimals
)
2195 p2
= "0"*(decimals-sl
) + s
2198 if i
< 0 then p1
= "-" + p1
2200 return p1
+ "." + p2
2205 # assert 'x'.to_s == "x"
2208 var s
= new Buffer.with_cap
(1)
2213 # Returns true if the char is a numerical digit
2215 # assert '0'.is_numeric
2216 # assert '9'.is_numeric
2217 # assert not 'a'.is_numeric
2218 # assert not '?'.is_numeric
2219 fun is_numeric
: Bool
2221 return self >= '0' and self <= '9'
2224 # Returns true if the char is an alpha digit
2226 # assert 'a'.is_alpha
2227 # assert 'Z'.is_alpha
2228 # assert not '0'.is_alpha
2229 # assert not '?'.is_alpha
2232 return (self >= 'a' and self <= 'z') or (self >= 'A' and self <= 'Z')
2235 # Returns true if the char is an alpha or a numeric digit
2237 # assert 'a'.is_alphanumeric
2238 # assert 'Z'.is_alphanumeric
2239 # assert '0'.is_alphanumeric
2240 # assert '9'.is_alphanumeric
2241 # assert not '?'.is_alphanumeric
2242 fun is_alphanumeric
: Bool
2244 return self.is_numeric
or self.is_alpha
2248 redef class Collection[E
]
2249 # Concatenate elements.
2255 # Concatenate element without separators
2256 fun plain_to_s
: String
2259 for e
in self do if e
!= null then s
.append
(e
.to_s
)
2263 # Concatenate and separate each elements with `sep`.
2265 # assert [1, 2, 3].join(":") == "1:2:3"
2266 # assert [1..3].join(":") == "1:2:3"
2267 fun join
(sep
: Text): String
2269 if is_empty
then return ""
2271 var s
= new Buffer # Result
2276 if e
!= null then s
.append
(e
.to_s
)
2278 # Concat other items
2283 if e
!= null then s
.append
(e
.to_s
)
2290 redef class Array[E
]
2292 # Fast implementation
2293 redef fun plain_to_s
2296 if l
== 0 then return ""
2297 if l
== 1 then if self[0] == null then return "" else return self[0].to_s
2299 var na
= new NativeArray[String](l
)
2305 if itsi
== null then
2315 var ns
= new NativeString(sl
+ 1)
2321 var tpl
= tmp
.length
2322 if tmp
isa FlatString then
2323 tmp
.items
.copy_to
(ns
, tpl
, tmp
.index_from
, off
)
2326 for j
in tmp
.substrings
do
2327 var s
= j
.as(FlatString)
2329 s
.items
.copy_to
(ns
, slen
, s
.index_from
, off
)
2335 return ns
.to_s_with_length
(sl
)
2339 redef class NativeArray[E
]
2340 # Join all the elements using `to_s`
2342 # REQUIRE: `self isa NativeArray[String]`
2343 # REQUIRE: all elements are initialized
2344 fun native_to_s
: String
2346 assert self isa NativeArray[String]
2357 var ns
= new NativeString(sl
+ 1)
2363 var tpl
= tmp
.length
2364 if tmp
isa FlatString then
2365 tmp
.items
.copy_to
(ns
, tpl
, tmp
.index_from
, off
)
2368 for j
in tmp
.substrings
do
2369 var s
= j
.as(FlatString)
2371 s
.items
.copy_to
(ns
, slen
, s
.index_from
, off
)
2377 return ns
.to_s_with_length
(sl
)
2381 redef class Map[K
,V
]
2382 # Concatenate couple of 'key value'.
2383 # key and value are separated by `couple_sep`.
2384 # each couple is separated each couple with `sep`.
2386 # var m = new ArrayMap[Int, String]
2389 # assert m.join("; ", "=") == "1=one; 10=ten"
2390 fun join
(sep
: String, couple_sep
: String): String
2392 if is_empty
then return ""
2394 var s
= new Buffer # Result
2400 s
.append
("{k or else "<null>"}{couple_sep}{e or else "<null>"}")
2402 # Concat other items
2408 s
.append
("{k or else "<null>"}{couple_sep}{e or else "<null>"}")
2415 ###############################################################################
2417 ###############################################################################
2419 # Native strings are simple C char *
2420 extern class NativeString `{ char* `}
2421 # Creates a new NativeString with a capacity of `length
`
2422 new(length: Int) is intern
2424 # Returns a char* starting at `index
`.
2426 # WARNING: Unsafe for extern code, use only for temporary
2427 # pointer manipulation purposes (e.g. write to file or such)
2428 fun fast_cstring(index: Int): NativeString is intern
2430 # Get char at `index
`.
2431 fun [](index: Int): Char is intern
2433 # Set char `item
` at index.
2434 fun []=(index: Int, item: Char) is intern
2436 # Copy `self` to `dest
`.
2437 fun copy_to(dest: NativeString, length: Int, from: Int, to: Int) is intern
2439 # Position of the first nul character.
2440 fun cstring_length: Int
2443 while self[l] != '\0' do l += 1
2447 # Parse `self` as an Int.
2448 fun atoi: Int is intern
2450 # Parse `self` as a Float.
2451 fun atof: Float is extern "atof"
2455 return to_s_with_length(cstring_length)
2458 # Returns `self` as a String of `length
`.
2459 fun to_s_with_length(length: Int): FlatString
2462 var str = new FlatString.with_infos(self, length, 0, length - 1)
2466 # Returns `self` as a new String.
2467 fun to_s_with_copy: FlatString
2469 var length = cstring_length
2470 var new_self = new NativeString(length + 1)
2471 copy_to(new_self, length, 0, 0)
2472 var str = new FlatString.with_infos(new_self, length, 0, length - 1)
2473 new_self[length] = '\0'
2474 str.real_items = new_self
2480 private var args_cache: nullable Sequence[String] = null
2482 # The arguments of the program as given by the OS
2483 fun program_args: Sequence[String]
2485 if _args_cache == null then init_args
2486 return _args_cache.as(not null)
2489 # The name of the program as given by the OS
2490 fun program_name: String
2492 return native_argv(0).to_s
2495 # Initialize `program_args
` with the contents of `native_argc
` and `native_argv
`.
2496 private fun init_args
2498 var argc = native_argc
2499 var args = new Array[String].with_capacity(0)
2502 args[i-1] = native_argv(i).to_s
2508 # First argument of the main C function.
2509 private fun native_argc: Int is intern
2511 # Second argument of the main C function.
2512 private fun native_argv(i: Int): NativeString is intern
2515 # Comparator that efficienlty use `to_s
` to compare things
2517 # The comparaison call `to_s
` on object and use the result to order things.
2519 # var a = [1, 2, 3, 10, 20]
2520 # (new CachedAlphaComparator).sort(a)
2521 # assert a == [1, 10, 2, 20, 3]
2523 # Internally the result of `to_s
` is cached in a HashMap to counter
2524 # uneficient implementation of `to_s
`.
2526 # Note: it caching is not usefull, see `alpha_comparator
`
2527 class CachedAlphaComparator
2529 redef type COMPARED: Object
2531 private var cache = new HashMap[Object, String]
2533 private fun do_to_s(a: Object): String do
2534 if cache.has_key(a) then return cache[a]
2540 redef fun compare(a, b) do
2541 return do_to_s(a) <=> do_to_s(b)
2545 # see `alpha_comparator
`
2546 private class AlphaComparator
2548 redef fun compare(a, b) do return a.to_s <=> b.to_s
2551 # Stateless comparator that naively use `to_s
` to compare things.
2553 # Note: the result of `to_s
` is not cached, thus can be invoked a lot
2554 # on a single instace. See `CachedAlphaComparator` as an alternative.
2556 # var a = [1, 2, 3, 10, 20]
2557 # alpha_comparator.sort(a)
2558 # assert a == [1, 10, 2, 20, 3]
2559 fun alpha_comparator: Comparator do return once new AlphaComparator
2561 # The arguments of the program as given by the OS
2562 fun args: Sequence[String]
2564 return sys.program_args