# You are allowed to redistribute it and sell it, alone or is a part of
# another product.
-# This module is about character strings.
+# Basic manipulations of strings of characters
package string
intrude import collection # FIXME should be collection::array
# String #
###############################################################################
+# Common subclass for String and Buffer
abstract class AbstractString
-special AbstractArrayRead[Char]
+ super AbstractArrayRead[Char]
+
readable private var _items: NativeString
redef fun [](index) do return _items[index]
end
end
- # Create a substring with the string beginning at the 'from' position
+ # Create a substring from `self' beginning at the 'from' position
#
# "abcd".substring(1) # --> "bcd"
# "abcd".substring(-1) # --> "abcd"
return substring(from, length - from)
end
- # is this string a substring of the 'of' string from pos 'pos'
+ # Does self have a substring 'str' starting from position 'pos
#
- # "bc".is_substring("abcd",1) # --> true
- # "bc".is_substring("abcd",2) # --> false
+ # "abcd".has_substring("bc",1) # --> true
+ # "abcd".has_substring("bc",2) # --> false
fun has_substring(str: String, pos: Int): Bool
do
var itsindex = str.length - 1
var myitems = _items
var itsitems = str._items
if myindex > length or itsindex > myindex then return false
- while itsindex >= 0 do
+ var its_index_from = str._index_from
+ itsindex += its_index_from
+ while itsindex >= its_index_from do
if myitems[myindex] != itsitems[itsindex] then return false
myindex -= 1
itsindex -= 1
return to_s.to_cstring.atoi
end
+ # If `self' contains a float, return the corresponding float
+ fun to_f: Float
+ do
+ # Shortcut
+ return to_s.to_cstring.atof
+ end
+
# If `self' contains only digits and alpha <= 'f', return the corresponding integer.
fun to_hex: Int do return a_to(16)
end
end
- # String to upper case
+ # Returns true if the string contains only Numeric values (and one "," or one "." character)
+ fun is_numeric: Bool
+ do
+ var has_point_or_comma = false
+ for i in self
+ do
+ if not i.is_numeric
+ then
+ if (i == '.' or i == ',') and not has_point_or_comma
+ then
+ has_point_or_comma = true
+ else
+ return false
+ end
+ end
+ end
+ return true
+ end
+
+ # A upper case version of `self'
fun to_upper: String
do
var s = new Buffer.with_capacity(length)
return s.to_s
end
- # String to lower case
+ # A lower case version of `self'
fun to_lower : String
do
var s = new Buffer.with_capacity(length)
return s.to_s
end
+ # Trims trailing and preceding white spaces
+ # A whitespace is defined as any character which ascii value is less than or equal to 32
+ fun trim: String
+ do
+ if self._length == 0 then return self.to_s
+ # find position of the first non white space char (ascii < 32) from the start of the string
+ var start_pos = 0
+ while self[start_pos].ascii <= 32 do
+ start_pos += 1
+ if start_pos == _length then return ""
+ end
+ # find position of the first non white space char from the end of the string
+ var end_pos = length - 1
+ while self[end_pos].ascii <= 32 do
+ end_pos -= 1
+ if end_pos == start_pos then return self[start_pos].to_s
+ end
+ return self.substring(start_pos, end_pos - start_pos + 1)
+ end
redef fun output
do
end
end
-
+# Immutable strings of characters.
class String
-special Comparable
-special AbstractString
+ super Comparable
+ super AbstractString
+ super StringCapable
+
redef type OTHER: String
+ # Index in _items of the start of the string
+ readable var _index_from: Int
+
+ # Indes in _items of the last item of the string
+ readable var _index_to: Int
+
+ ################################################
+ # AbstractString specific methods #
+ ################################################
+
+ # Access a character at index in String
+ #
+ redef fun [](index) do
+ assert index >= 0
+ # Check that the index (+ index_from) is not larger than indexTo
+ # In other terms, if the index is valid
+ assert (index + _index_from) <= _index_to
+ return _items[index + _index_from]
+ end
+
+ # Create a substring.
+ #
+ # "abcd".substring(1, 2) # --> "bc"
+ # "abcd".substring(-1, 2) # --> "a"
+ # "abcd".substring(1, 0) # --> ""
+ # "abcd".substring(2, 5) # --> "cd"
+ #
+ # A "from" index < 0 will be replaced by 0
+ # Unless a count value is > 0 at the same time
+ # In this case, from += count and count -= from
+ #
+ redef fun substring(from: Int, count: Int): String
+ do
+ assert count >= 0
+
+ if from < 0 then
+ count += from
+ if count < 0 then count = 0
+ from = 0
+ end
+
+ var realFrom = _index_from + from
+
+ if (realFrom + count) > _index_to then return new String.from_substring(realFrom, _index_to, _items)
+
+ if count == 0 then return ""
+
+ return new String.from_substring(realFrom, realFrom + count - 1, _items)
+ end
+
+ # Create a substring from `self' beginning at the 'from' position
+ #
+ # "abcd".substring_from(1) # --> "bcd"
+ # "abcd".substring_from(-1) # --> "abcd"
+ # "abcd".substring_from(2) # --> "cd"
+ #
+ # As with substring, a "from" index < 0 will be replaced by 0
+ #
+ redef fun substring_from(from: Int): String
+ do
+ if from > _length then return ""
+ if from < 0 then from = 0
+ return substring(from, _length)
+ end
+
+ # Does self have a substring 'str' starting from position 'pos
+ #
+ # "abcd".has_substring("bc",1) # --> true
+ # "abcd".has_substring("bc",2) # --> false
+ redef fun has_substring(str: String, pos: Int): Bool
+ do
+ var itsindex = str._length - 1
+
+ var myindex = pos + itsindex
+ var myitems = _items
+
+ var itsitems = str._items
+
+ if myindex > _length or itsindex > myindex then return false
+
+ var itsindexfrom = str.index_from
+ itsindex += itsindexfrom
+ myindex += index_from
+
+ while itsindex >= itsindexfrom do
+ if myitems[myindex] != itsitems[itsindex] then return false
+ myindex -= 1
+ itsindex -= 1
+ end
+
+ return true
+ end
+
+ # A upper case version of `self'
+ redef fun to_upper: String
+ do
+ var outstr = calloc_string(self._length + 1)
+ var out_index = 0
+
+ var myitems = self._items
+ var index_from = self._index_from
+ var max = self._index_to
+
+ while index_from <= max do
+ outstr[out_index] = myitems[index_from].to_upper
+ out_index += 1
+ index_from += 1
+ end
+
+ outstr[self.length] = '\0'
+
+ return new String.with_native(outstr, self._length)
+ end
+
+ # A lower case version of `self'
+ redef fun to_lower : String
+ do
+ var outstr = calloc_string(self._length + 1)
+ var out_index = 0
+
+ var myitems = self._items
+ var index_from = self._index_from
+ var max = self._index_to
+
+ while index_from <= max do
+ outstr[out_index] = myitems[index_from].to_lower
+ out_index += 1
+ index_from += 1
+ end
+
+ outstr[self.length] = '\0'
+
+ return new String.with_native(outstr, self._length)
+ end
+
+ redef fun trim: String
+ do
+ if self._length == 0 then return self
+ # find position of the first non white space char (ascii < 32) from the start of the string
+ var start_pos = self._index_from
+ while _items[start_pos].ascii <= 32 do
+ start_pos += 1
+ if start_pos == _index_to + 1 then return ""
+ end
+ # find position of the first non white space char from the end of the string
+ var end_pos = _index_to
+ while _items[end_pos].ascii <= 32 do
+ end_pos -= 1
+ if end_pos == start_pos then return _items[start_pos].to_s
+ end
+ start_pos -= index_from
+ end_pos -= index_from
+ return self.substring(start_pos, end_pos - start_pos + 1)
+ end
+
+ redef fun output
+ do
+ var i = self._index_from
+ var imax = self._index_to
+ while i <= imax do
+ _items[i].output
+ i += 1
+ end
+ end
+
+ ##################################################
+ # String Specific Methods #
+ ##################################################
+
+ # Creates a String object as a substring of another String
+ #
+ # From : index to start at
+ #
+ # To : Index to stop at (from + count -1)
+ #
+ private init from_substring(from: Int, to: Int, internalString: NativeString)
+ do
+ _items = internalString
+ _index_from = from
+ _index_to = to
+ _length = to - from + 1
+ end
+
# Create a new string from a given char *.
init with_native(nat: NativeString, size: Int)
do
assert size >= 0
_items = nat
_length = size
+ _index_from = 0
+ _index_to = _length - 1
end
# Create a new string from a null terminated char *.
init from_cstring(str: NativeString)
do
- var size = str.cstring_length
- _items = str
- _length = size
+ with_native(str,str.cstring_length)
+ end
+
+ # Creates a new Nit String from an existing CString
+ # Pretty much equals to from_cstring but copies instead
+ # of passing a reference
+ # Avoids manual/automatic dealloc problems when dealing with native C code
+ init copy_from_native(str: NativeString)
+ do
+ var temp_length = str.cstring_length
+ var new_str = calloc_string(temp_length + 1)
+ str.copy_to(new_str, temp_length, 0, 0)
+ new_str[temp_length] = '\0'
+ with_native(new_str, temp_length)
end
# Return a null terminated char *
fun to_cstring: NativeString
do
+ #return items
+ if _index_from > 0 or _index_to != items.cstring_length - 1 then
+ var newItems = calloc_string(_length + 1)
+ self.items.copy_to(newItems, _length, _index_from, 0)
+ newItems[length] = '\0'
+ return newItems
+ end
return _items
end
- redef fun ==(o)
+ redef fun ==(other)
do
- if not o isa String or o is null then return false
- var l = length
- if o.length != l then return false
- var i = 0
- var it = _items
- var oit = o._items
- while i < l do
- if it[i] != oit[i] then return false
- i += 1
+ if not other isa String or other is null then return false
+
+ if self.object_id == other.object_id then return true
+
+ var my_length = _length
+
+ if other._length != my_length then return false
+
+ var my_index = _index_from
+ var its_index = other._index_from
+
+ var last_iteration = my_index + my_length
+
+ var itsitems = other._items
+ var myitems = self._items
+
+ while my_index < last_iteration do
+ if myitems[my_index] != itsitems[its_index] then return false
+ my_index += 1
+ its_index += 1
end
+
return true
end
- redef fun <(s)
+ # The comparison between two strings is done on a lexicographical basis
+ # Eg : "aa" < "b" => true
+ redef fun <(other)
do
- var i = 0
- var l1 = length
- var l2 = s.length
- var n1 = _items
- var n2 = s._items
- while i < l1 and i < l2 do
- var c1 = n1[i].ascii
- var c2 = n2[i].ascii
- if c1 < c2 then
- return true
- else if c2 < c1 then
+ if self.object_id == other.object_id then return false
+
+ var my_curr_char : Char
+ var its_curr_char : Char
+
+ var curr_id_self = self._index_from
+ var curr_id_other = other._index_from
+
+ var my_items = self._items
+ var its_items = other._items
+
+ var my_length = self._length
+ var its_length = other._length
+
+ var max_iterations = curr_id_self + my_length
+
+ while curr_id_self < max_iterations do
+ my_curr_char = my_items[curr_id_self]
+ its_curr_char = its_items[curr_id_other]
+
+ if my_curr_char != its_curr_char then
+ if my_curr_char < its_curr_char then return true
return false
end
- i += 1
- end
- if l1 < l2 then
- return true
- else
- return false
+
+ curr_id_self += 1
+ curr_id_other += 1
end
+
+ return my_length < its_length
end
# The concatenation of `self' with `r'
fun +(s: String): String
do
- var r = new Buffer.with_capacity(length + s.length)
- r.append(self)
- r.append(s)
- return r.to_s
+ var my_length = self._length
+ var its_length = s._length
+
+ var target_string = calloc_string(my_length + its_length + 1)
+
+ self._items.copy_to(target_string, my_length, _index_from, 0)
+ s._items.copy_to(target_string, its_length, s._index_from, my_length)
+
+ target_string[my_length + its_length] = '\0'
+
+ return new String.with_native(target_string, my_length + its_length)
end
# i repetitions of self
fun *(i: Int): String
do
assert i >= 0
- var r = new Buffer.with_capacity(length * i)
- while i > 0 do
- r.append(self)
- i -= 1
+
+ var my_length = self._length
+
+ var final_length = my_length * i
+
+ var my_items = self._items
+
+ var target_string = calloc_string((final_length) + 1)
+
+ target_string[final_length] = '\0'
+
+ var current_last = 0
+
+ for iteration in [1 .. i] do
+ my_items.copy_to(target_string, my_length, 0, current_last)
+ current_last += my_length
end
- return r.to_s
+
+ return new String.with_native(target_string, final_length)
end
redef fun to_s do return self
# djb2 hash algorythm
var h = 5381
var i = _length - 1
- var it = _items
- while i >= 0 do
- h = (h * 32) + h + it[i].ascii
+
+ var myitems = _items
+ var strStart = _index_from
+
+ i += strStart
+
+ while i >= strStart do
+ h = (h * 32) + h + self._items[i].ascii
i -= 1
end
- return h
+ return h
end
end
-# Strings are arrays of characters.
+# Mutable strings of characters.
class Buffer
-special AbstractString
-special Comparable
-special StringCapable
-special AbstractArray[Char]
+ super AbstractString
+ super Comparable
+ super StringCapable
+ super AbstractArray[Char]
redef type OTHER: String
do
if s isa String then
var sl = s.length
- if _capacity < length + sl then enlarge(length + sl)
- s.items.copy_to(_items, sl, 0, length)
+ if _capacity < _length + sl then enlarge(_length + sl)
+ s.items.copy_to(_items, sl, s._index_from, _length)
_length += sl
else
super
_capacity = s.length + 1
_length = s.length
_items = calloc_string(_capacity)
- s.items.copy_to(_items, _length, 0, 0)
+ s.items.copy_to(_items, _length, s._index_from, 0)
end
# Create a new empty string with a given capacity.
###############################################################################
redef class Object
- # fun class_name: String is extern intern # The name of the class
-
- # User redeable representation of `self'.
+ # User readable representation of `self'.
fun to_s: String do return inspect
- # Developper readable representation of `self'.
- # Usualy, it uses the form "<CLASSNAME:#OBJECTID bla bla bla>"
+ # The class name of the object in NativeString format.
+ private fun native_class_name: NativeString is intern
+
+ # The class name of the object.
+ # FIXME: real type information is not available at runtime.
+ # Therefore, for instance, an instance of List[Bool] has just
+ # "List" for class_name
+ fun class_name: String do return new String.from_cstring(native_class_name)
+
+ # Developer readable representation of `self'.
+ # Usually, it uses the form "<CLASSNAME:#OBJECTID bla bla bla>"
fun inspect: String
do
- var r = inspect_head
- # r.add('>')
- return r
+ return "<{inspect_head}>"
end
- # Return "<CLASSNAME:#OBJECTID".
- # This fuction is mainly used with the redefinition of the inspect(0) method
+ # Return "CLASSNAME:#OBJECTID".
+ # This function is mainly used with the redefinition of the inspect method
protected fun inspect_head: String
do
- return "<{object_id.to_hex}"
+ return "{class_name}:#{object_id.to_hex}"
end
protected fun args: Sequence[String]
end
redef class Float
- redef fun to_s do return to_precision(6)
+ # Pretty print self, print needed decimals up to a max of 6.
+ redef fun to_s do
+ var str = to_precision( 3 )
+ var len = str.length
+ for i in [0..len-1] do
+ var j = len-1-i
+ var c = str[j]
+ if c == '0' then
+ continue
+ else if c == '.' then
+ return str.substring( 0, j+2 )
+ else
+ return str.substring( 0, j+1 )
+ end
+ end
+ return str
+ end
# `self' representation with `nb' digits after the '.'.
fun to_precision(nb: Int): String
do
- if nb == 0 then return to_i.to_s
-
- var i = to_i
- var dec = 1.0
- while nb > 0 do
- dec = dec * 10.0
- nb -= 1
+ if nb == 0 then return self.to_i.to_s
+ var f = self
+ for i in [0..nb[ do f = f * 10.0
+ if self > 0.0 then
+ f = f + 0.5
+ else
+ f = f - 0.5
+ end
+ var i = f.to_i
+ if i == 0 then return "0.0"
+ var s = i.to_s
+ var sl = s.length
+ if sl > nb then
+ var p1 = s.substring(0, s.length-nb)
+ var p2 = s.substring(s.length-nb, nb)
+ return p1 + "." + p2
+ else
+ return "0." + ("0"*(nb-sl)) + s
end
- var d = ((self-i.to_f)*dec).to_i
- return "{i}.{d}"
end
+
+ fun to_precision_native(nb: Int): String import String::from_cstring `{
+ int size;
+ char *str;
+
+ size = snprintf(NULL, 0, "%.*f", (int)nb, recv);
+ str = malloc(size + 1);
+ sprintf(str, "%.*f", (int)nb, recv );
+
+ return new_String_from_cstring( str );
+ `}
end
redef class Char
s[0] = self
return s.to_s
end
+
+ # Returns true if the char is a numerical digit
+ fun is_numeric: Bool
+ do
+ if self >= '0' and self <= '9'
+ then
+ return true
+ end
+ return false
+ end
+
+ # Returns true if the char is an alpha digit
+ fun is_alpha: Bool
+ do
+ if (self >= 'a' and self <= 'z') or (self >= 'A' and self <= 'Z') then return true
+ return false
+ end
+
+ # Returns true if the char is an alpha or a numeric digit
+ fun is_alphanumeric: Bool
+ do
+ if self.is_numeric or self.is_alpha then return true
+ return false
+ end
end
redef class Collection[E]
end
redef class Map[K,V]
- # Concatenate couple of 'key value' separate by 'couple_sep' and separate each couple with `sep'.
- fun map_join(sep: String, couple_sep: String): String
+ # Concatenate couple of 'key value'.
+ # key and value are separated by 'couple_sep'.
+ # each couple is separated each couple with `sep'.
+ fun join(sep: String, couple_sep: String): String
do
if is_empty then return ""
end
###############################################################################
-# Native classe #
+# Native classes #
###############################################################################
# Native strings are simple C char *
return l
end
fun atoi: Int is intern
+ fun atof: Float is extern "atof"
end
# StringCapable objects can create native strings
-class StringCapable
+interface StringCapable
protected fun calloc_string(size: Int): NativeString is intern
end
_args_cache = args
end
- private fun native_argc: Int is extern "kernel_Sys_Sys_native_argc_0" # First argument of the main C function.
-
- private fun native_argv(i: Int): NativeString is extern "kernel_Sys_Sys_native_argv_1" # Second argument of the main C function.
+ # First argument of the main C function.
+ private fun native_argc: Int is intern
+
+ # Second argument of the main C function.
+ private fun native_argv(i: Int): NativeString is intern
end