+ # Are all letters in `self` lower-case ?
+ #
+ # assert "hello world".is_lower == true
+ # assert "%$&%!".is_lower == true
+ # assert "Hello World".is_lower == false
+ fun is_lower: Bool
+ do
+ for i in [0..length[ do
+ var char = chars[i]
+ if char.is_upper then return false
+ end
+ return true
+ end
+
+ # Removes the whitespaces at the beginning of self
+ #
+ # assert " \n\thello \n\t".l_trim == "hello \n\t"
+ #
+ # `Char::is_whitespace` determines what is a whitespace.
+ fun l_trim: SELFTYPE
+ do
+ var iter = self.chars.iterator
+ while iter.is_ok do
+ if not iter.item.is_whitespace then break
+ iter.next
+ end
+ if iter.index == length then return self.empty
+ return self.substring_from(iter.index)
+ end
+
+ # Removes the whitespaces at the end of self
+ #
+ # assert " \n\thello \n\t".r_trim == " \n\thello"
+ #
+ # `Char::is_whitespace` determines what is a whitespace.
+ fun r_trim: SELFTYPE
+ do
+ var iter = self.chars.reverse_iterator
+ while iter.is_ok do
+ if not iter.item.is_whitespace then break
+ iter.next
+ end
+ if iter.index < 0 then return self.empty
+ return self.substring(0, iter.index + 1)
+ end
+
+ # Trims trailing and preceding white spaces
+ #
+ # assert " Hello World ! ".trim == "Hello World !"
+ # assert "\na\nb\tc\t".trim == "a\nb\tc"
+ #
+ # `Char::is_whitespace` determines what is a whitespace.
+ fun trim: SELFTYPE do return (self.l_trim).r_trim
+
+ # Is the string non-empty but only made of whitespaces?
+ #
+ # assert " \n\t ".is_whitespace == true
+ # assert " hello ".is_whitespace == false
+ # assert "".is_whitespace == false
+ #
+ # `Char::is_whitespace` determines what is a whitespace.
+ fun is_whitespace: Bool
+ do
+ if is_empty then return false
+ for c in self.chars do
+ if not c.is_whitespace then return false
+ end
+ return true
+ end
+
+ # Returns `self` removed from its last line terminator (if any).
+ #
+ # assert "Hello\n".chomp == "Hello"
+ # assert "Hello".chomp == "Hello"
+ #
+ # assert "\n".chomp == ""
+ # assert "".chomp == ""
+ #
+ # Line terminators are `"\n"`, `"\r\n"` and `"\r"`.
+ # A single line terminator, the last one, is removed.
+ #
+ # assert "\r\n".chomp == ""
+ # assert "\r\n\n".chomp == "\r\n"
+ # assert "\r\n\r\n".chomp == "\r\n"
+ # assert "\r\n\r".chomp == "\r\n"
+ #
+ # Note: unlike with most IO methods like `Reader::read_line`,
+ # a single `\r` is considered here to be a line terminator and will be removed.
+ fun chomp: SELFTYPE
+ do
+ var len = length
+ if len == 0 then return self
+ var l = self.chars.last
+ if l == '\r' then
+ return substring(0, len-1)
+ else if l != '\n' then
+ return self
+ else if len > 1 and self.chars[len-2] == '\r' then
+ return substring(0, len-2)
+ else
+ return substring(0, len-1)
+ end
+ end
+
+ # Justify a self in a space of `length`
+ #
+ # `left` is the space ratio on the left side.
+ # * 0.0 for left-justified (no space at the left)
+ # * 1.0 for right-justified (all spaces at the left)
+ # * 0.5 for centered (half the spaces at the left)
+ #
+ # Examples
+ #
+ # assert "hello".justify(10, 0.0) == "hello "
+ # assert "hello".justify(10, 1.0) == " hello"
+ # assert "hello".justify(10, 0.5) == " hello "
+ #
+ # If `length` is not enough, `self` is returned as is.
+ #
+ # assert "hello".justify(2, 0.0) == "hello"
+ #
+ # REQUIRE: `left >= 0.0 and left <= 1.0`
+ # ENSURE: `self.length <= length implies result.length == length`
+ # ENSURE: `self.length >= length implies result == self`
+ fun justify(length: Int, left: Float): String
+ do
+ var diff = length - self.length
+ if diff <= 0 then return to_s
+ assert left >= 0.0 and left <= 1.0
+ var before = (diff.to_f * left).to_i
+ return " " * before + self + " " * (diff-before)
+ end
+
+ # Mangle a string to be a unique string only made of alphanumeric characters and underscores.
+ #
+ # This method is injective (two different inputs never produce the same
+ # output) and the returned string always respect the following rules:
+ #
+ # * Contains only US-ASCII letters, digits and underscores.
+ # * Never starts with a digit.
+ # * Never ends with an underscore.
+ # * Never contains two contiguous underscores.
+ #
+ # assert "42_is/The answer!".to_cmangle == "_52d2_is_47dThe_32danswer_33d"
+ # assert "__".to_cmangle == "_95d_95d"
+ # assert "__d".to_cmangle == "_95d_d"
+ # assert "_d_".to_cmangle == "_d_95d"
+ # assert "_42".to_cmangle == "_95d42"
+ # assert "foo".to_cmangle == "foo"
+ # assert "".to_cmangle == ""
+ fun to_cmangle: String
+ do
+ if is_empty then return ""
+ var res = new FlatBuffer
+ var underscore = false
+ var start = 0
+ var c = chars[0]
+
+ if c >= '0' and c <= '9' then
+ res.add('_')
+ res.append(c.ascii.to_s)
+ res.add('d')
+ start = 1
+ end
+ for i in [start..length[ do
+ c = chars[i]
+ if (c >= 'a' and c <= 'z') or (c >='A' and c <= 'Z') then
+ res.add(c)
+ underscore = false
+ continue
+ end
+ if underscore then
+ res.append('_'.ascii.to_s)
+ res.add('d')
+ end
+ if c >= '0' and c <= '9' then
+ res.add(c)
+ underscore = false
+ else if c == '_' then
+ res.add(c)
+ underscore = true
+ else
+ res.add('_')
+ res.append(c.ascii.to_s)
+ res.add('d')
+ underscore = false
+ end
+ end
+ if underscore then
+ res.append('_'.ascii.to_s)
+ res.add('d')
+ end
+ return res.to_s
+ end
+
+ # Escape " \ ' and non printable characters using the rules of literal C strings and characters
+ #
+ # assert "abAB12<>&".escape_to_c == "abAB12<>&"
+ # assert "\n\"'\\".escape_to_c == "\\n\\\"\\'\\\\"
+ fun escape_to_c: String
+ do
+ var b = new FlatBuffer
+ for i in [0..length[ do
+ var c = chars[i]
+ if c == '\n' then
+ b.append("\\n")
+ else if c == '\0' then
+ b.append("\\0")
+ else if c == '"' then
+ b.append("\\\"")
+ else if c == '\'' then
+ b.append("\\\'")
+ else if c == '\\' then
+ b.append("\\\\")
+ else if c.ascii < 32 then
+ b.append("\\{c.ascii.to_base(8, false)}")
+ else
+ b.add(c)
+ end
+ end
+ return b.to_s
+ end
+
+ # Escape additionnal characters
+ # The result might no be legal in C but be used in other languages
+ #
+ # assert "ab|\{\}".escape_more_to_c("|\{\}") == "ab\\|\\\{\\\}"
+ fun escape_more_to_c(chars: String): String
+ do
+ var b = new FlatBuffer
+ for c in escape_to_c.chars do
+ if chars.chars.has(c) then
+ b.add('\\')
+ end
+ b.add(c)
+ end
+ return b.to_s
+ end
+
+ # Escape to C plus braces
+ #
+ # assert "\n\"'\\\{\}".escape_to_nit == "\\n\\\"\\'\\\\\\\{\\\}"
+ fun escape_to_nit: String do return escape_more_to_c("\{\}")
+
+ # Escape to POSIX Shell (sh).
+ #
+ # Abort if the text contains a null byte.
+ #
+ # assert "\n\"'\\\{\}0".escape_to_sh == "'\n\"'\\''\\\{\}0'"
+ fun escape_to_sh: String do
+ var b = new FlatBuffer
+ b.chars.add '\''
+ for i in [0..length[ do
+ var c = chars[i]
+ if c == '\'' then
+ b.append("'\\''")
+ else
+ assert without_null_byte: c != '\0'
+ b.add(c)
+ end
+ end
+ b.chars.add '\''
+ return b.to_s
+ end
+
+ # Escape to include in a Makefile
+ #
+ # Unfortunately, some characters are not escapable in Makefile.
+ # These characters are `;`, `|`, `\`, and the non-printable ones.
+ # They will be rendered as `"?{hex}"`.
+ fun escape_to_mk: String do
+ var b = new FlatBuffer
+ for i in [0..length[ do
+ var c = chars[i]
+ if c == '$' then
+ b.append("$$")
+ else if c == ':' or c == ' ' or c == '#' then
+ b.add('\\')
+ b.add(c)
+ else if c.ascii < 32 or c == ';' or c == '|' or c == '\\' or c == '=' then
+ b.append("?{c.ascii.to_base(16, false)}")
+ else
+ b.add(c)
+ end
+ end
+ return b.to_s
+ end
+
+ # Return a string where Nit escape sequences are transformed.
+ #
+ # var s = "\\n"
+ # assert s.length == 2
+ # var u = s.unescape_nit
+ # assert u.length == 1
+ # assert u.chars[0].ascii == 10 # (the ASCII value of the "new line" character)
+ fun unescape_nit: String
+ do
+ var res = new FlatBuffer.with_capacity(self.length)
+ var was_slash = false
+ for i in [0..length[ do
+ var c = chars[i]
+ if not was_slash then
+ if c == '\\' then
+ was_slash = true
+ else
+ res.add(c)
+ end
+ continue
+ end
+ was_slash = false
+ if c == 'n' then
+ res.add('\n')
+ else if c == 'r' then
+ res.add('\r')
+ else if c == 't' then
+ res.add('\t')
+ else if c == '0' then
+ res.add('\0')
+ else
+ res.add(c)
+ end
+ end
+ return res.to_s
+ end
+
+ # Encode `self` to percent (or URL) encoding
+ #
+ # assert "aBc09-._~".to_percent_encoding == "aBc09-._~"
+ # assert "%()< >".to_percent_encoding == "%25%28%29%3c%20%3e"
+ # assert ".com/post?e=asdf&f=123".to_percent_encoding == ".com%2fpost%3fe%3dasdf%26f%3d123"
+ fun to_percent_encoding: String
+ do
+ var buf = new FlatBuffer
+
+ for i in [0..length[ do
+ var c = chars[i]
+ if (c >= '0' and c <= '9') or
+ (c >= 'a' and c <= 'z') or
+ (c >= 'A' and c <= 'Z') or
+ c == '-' or c == '.' or
+ c == '_' or c == '~'
+ then
+ buf.add c
+ else buf.append "%{c.ascii.to_hex}"
+ end
+
+ return buf.to_s
+ end
+
+ # Decode `self` from percent (or URL) encoding to a clear string
+ #
+ # Replace invalid use of '%' with '?'.
+ #
+ # assert "aBc09-._~".from_percent_encoding == "aBc09-._~"
+ # assert "%25%28%29%3c%20%3e".from_percent_encoding == "%()< >"
+ # assert ".com%2fpost%3fe%3dasdf%26f%3d123".from_percent_encoding == ".com/post?e=asdf&f=123"
+ # assert "%25%28%29%3C%20%3E".from_percent_encoding == "%()< >"
+ # assert "incomplete %".from_percent_encoding == "incomplete ?"
+ # assert "invalid % usage".from_percent_encoding == "invalid ? usage"
+ fun from_percent_encoding: String
+ do
+ var buf = new FlatBuffer
+
+ var i = 0
+ while i < length do
+ var c = chars[i]
+ if c == '%' then
+ if i + 2 >= length then
+ # What follows % has been cut off
+ buf.add '?'
+ else
+ i += 1
+ var hex_s = substring(i, 2)
+ if hex_s.is_hex then
+ var hex_i = hex_s.to_hex
+ buf.add hex_i.ascii
+ i += 1
+ else
+ # What follows a % is not Hex
+ buf.add '?'
+ i -= 1
+ end
+ end
+ else buf.add c
+
+ i += 1
+ end
+
+ return buf.to_s
+ end
+
+ # Escape the characters `<`, `>`, `&`, `"`, `'` and `/` as HTML/XML entity references.
+ #
+ # assert "a&b-<>\"x\"/'".html_escape == "a&b-<>"x"/'"
+ #
+ # 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>
+ fun html_escape: String
+ do
+ var buf = new FlatBuffer
+
+ for i in [0..length[ do
+ var c = chars[i]
+ if c == '&' then
+ buf.append "&"
+ else if c == '<' then
+ buf.append "<"
+ else if c == '>' then
+ buf.append ">"
+ else if c == '"' then
+ buf.append """
+ else if c == '\'' then
+ buf.append "'"
+ else if c == '/' then
+ buf.append "/"
+ else buf.add c
+ end
+
+ return buf.to_s
+ end
+
+ # Equality of text
+ # Two pieces of text are equals if thez have the same characters in the same order.
+ #
+ # assert "hello" == "hello"
+ # assert "hello" != "HELLO"
+ # assert "hello" == "hel"+"lo"
+ #
+ # Things that are not Text are not equal.
+ #
+ # assert "9" != '9'
+ # assert "9" != ['9']
+ # assert "9" != 9
+ #
+ # assert "9".chars.first == '9' # equality of Char
+ # assert "9".chars == ['9'] # equality of Sequence
+ # assert "9".to_i == 9 # equality of Int
+ redef fun ==(o)
+ do
+ if o == null then return false
+ if not o isa Text then return false
+ if self.is_same_instance(o) then return true
+ if self.length != o.length then return false
+ return self.chars == o.chars
+ end
+
+ # Lexicographical comparaison
+ #
+ # assert "abc" < "xy"
+ # assert "ABC" < "abc"
+ redef fun <(other)
+ do
+ var self_chars = self.chars.iterator
+ var other_chars = other.chars.iterator
+
+ while self_chars.is_ok and other_chars.is_ok do
+ if self_chars.item < other_chars.item then return true
+ if self_chars.item > other_chars.item then return false
+ self_chars.next
+ other_chars.next
+ end
+
+ if self_chars.is_ok then
+ return false
+ else
+ return true
+ end
+ end
+
+ # Escape string used in labels for graphviz
+ #
+ # assert ">><<".escape_to_dot == "\\>\\>\\<\\<"
+ fun escape_to_dot: String
+ do
+ return escape_more_to_c("|\{\}<>")
+ end
+
+ # Flat representation of self
+ fun flatten: FlatText is abstract
+
+ private var hash_cache: nullable Int = null
+
+ redef fun hash
+ do
+ if hash_cache == null then
+ # djb2 hash algorithm
+ var h = 5381
+
+ for i in [0..length[ do
+ var char = chars[i]
+ h = h.lshift(5) + h + char.ascii
+ end
+
+ hash_cache = h
+ end
+ return hash_cache.as(not null)
+ end
+
+end
+
+# All kinds of array-based text representations.
+abstract class FlatText
+ super Text
+
+ # Underlying C-String (`char*`)
+ #
+ # Warning : Might be void in some subclasses, be sure to check
+ # if set before using it.
+ private var items: NativeString is noinit
+
+ # Real items, used as cache for to_cstring is called
+ private var real_items: nullable NativeString = null
+
+ redef var length: Int = 0