X-Git-Url: http://nitlanguage.org diff --git a/lib/standard/string.nit b/lib/standard/string.nit index e1505df..e4c9fe4 100644 --- a/lib/standard/string.nit +++ b/lib/standard/string.nit @@ -25,18 +25,14 @@ intrude import collection # FIXME should be collection::array # String # ############################################################################### -# Common subclass for String and Buffer -abstract class AbstractString - super AbstractArrayRead[Char] - - readable private var _items: NativeString +# High-level abstraction for all text representations +abstract class Text + # Gets a view on the chars of the Text object fun chars: StringCharView is abstract - # Access a character at `index` in the string. - # - # assert "abcd"[2] == 'c' - redef fun [](index) do return _items[index] + # Number of characters contained in self. + fun length: Int is abstract # Create a substring. # @@ -48,24 +44,103 @@ abstract class AbstractString # 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`. - fun substring(from: Int, count: Int): String + fun substring(from: Int, count: Int): String is abstract + + # Is the current Text empty (== "") + # assert "".is_empty + # assert not "foo".is_empty + fun is_empty: Bool do return self.length == 0 + + # Gets the first char of the Text + # + # DEPRECATED : Use self.chars.first instead + fun first: Char do return self.chars[0] + + # Access a character at `index` in the string. + # + # assert "abcd"[2] == 'c' + # + # DEPRECATED : Use self.chars.[] instead + fun [](index: Int): Char do return self.chars[index] + + # Gets the index of the first occurence of 'c' + # + # Returns -1 if not found + # + # DEPRECATED : Use self.chars.index_of instead + fun index_of(c: Char): Int do - assert count >= 0 - count += from - if from < 0 then from = 0 - if count > length then count = length - if from < count then - var r = new Buffer.with_capacity(count - from) - while from < count do - r.chars.push(_items[from]) - from += 1 - end - return r.to_s - else - return "" + return index_of_from(c, 0) + end + + # Gets the last char of self + # + # DEPRECATED : Use self.chars.last instead + fun last: Char do return self.chars[length-1] + + # Gets the index of the first occurence of ´c´ starting from ´pos´ + # + # Returns -1 if not found + # + # DEPRECATED : Use self.chars.index_of_from instead + fun index_of_from(c: Char, pos: Int): Int + do + var iter = self.chars.iterator_from(pos) + while iter.is_ok do + if iter.item == c then return iter.index + end + return -1 + end + + # Gets the last index of char ´c´ + # + # Returns -1 if not found + # + # DEPRECATED : Use self.chars.last_index_of instead + fun last_index_of(c: Char): Int + do + return last_index_of_from(c, length - 1) + end + + # The index of the last occurrence of an element starting from pos (in reverse order). + # Example : + # assert "/etc/bin/test/test.nit".last_index_of_from('/', length-1) == 13 + # assert "/etc/bin/test/test.nit".last_index_of_from('/', 12) == 8 + # + # Returns -1 if not found + # + # DEPRECATED : Use self.chars.last_index_of_from instead + fun last_index_of_from(item: Char, pos: Int): Int + do + var iter = self.chars.reverse_iterator_from(pos) + while iter.is_ok do + if iter.item == item then return iter.index + iter.next end + return -1 end + # Gets an iterator on the chars of self + # + # DEPRECATED : Use self.chars.iterator instead + fun iterator: Iterator[Char] + do + return self.chars.iterator + end + + # Is 'c' contained in self ? + # + # DEPRECATED : Use self.chars.has instead + fun has(c: Char): Bool + do + return self.chars.has(c) + end + + # Gets an Array containing the chars of self + # + # DEPRECATED : Use self.chars.to_a instead + fun to_a: Array[Char] do return chars.to_a + # Create a substring from `self` beginning at the `from` position # # assert "abcd".substring_from(1) == "bcd" @@ -85,18 +160,14 @@ abstract class AbstractString # assert "abcd".has_substring("bc",2) == false 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 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 + var myiter = self.chars.iterator_from(pos) + var itsiter = str.iterator + while myiter.is_ok and itsiter.is_ok do + if myiter.item != itsiter.item then return false + myiter.next + itsiter.next end + if itsiter.is_ok then return false return true end @@ -196,7 +267,7 @@ abstract class AbstractString # assert "Hello World!".to_upper == "HELLO WORLD!" fun to_upper: String do - var s = new Buffer.with_capacity(length) + var s = new FlatBuffer.with_capacity(length) for i in self.chars do s.add(i.to_upper) return s.to_s end @@ -206,7 +277,7 @@ abstract class AbstractString # assert "Hello World!".to_lower == "hello world!" fun to_lower : String do - var s = new Buffer.with_capacity(length) + var s = new FlatBuffer.with_capacity(length) for i in self.chars do s.add(i.to_lower) return s.to_s end @@ -234,19 +305,10 @@ abstract class AbstractString return self.substring(start_pos, end_pos - start_pos + 1) end - redef fun output - do - var i = 0 - while i < length do - _items[i].output - i += 1 - end - end - # Mangle a string to be a unique string only made of alphanumeric characters fun to_cmangle: String do - var res = new Buffer + var res = new FlatBuffer var underscore = false for c in self.chars do if (c >= 'a' and c <= 'z') or (c >='A' and c <= 'Z') then @@ -280,7 +342,7 @@ abstract class AbstractString # assert "\n\"'\\".escape_to_c == "\\n\\\"\\'\\\\" fun escape_to_c: String do - var b = new Buffer + var b = new FlatBuffer for c in self.chars do if c == '\n' then b.append("\\n") @@ -307,9 +369,9 @@ abstract class AbstractString # assert "ab|\{\}".escape_more_to_c("|\{\}") == "ab\\|\\\{\\\}" fun escape_more_to_c(chars: String): String do - var b = new Buffer + var b = new FlatBuffer for c in escape_to_c do - if chars.has(c) then + if chars.chars.has(c) then b.add('\\') end b.add(c) @@ -332,7 +394,7 @@ abstract class AbstractString # assert u[0].ascii == 10 # (the ASCII value of the "new line" character) fun unescape_nit: String do - var res = new Buffer.with_capacity(self.length) + var res = new FlatBuffer.with_capacity(self.length) var was_slash = false for c in self do if not was_slash then @@ -358,6 +420,27 @@ abstract class AbstractString end return res.to_s end + +end + +# Common subclass for String and Buffer +abstract class AbstractString + super Text + + private var items: NativeString + + redef var length: Int + + init do end + + redef fun output + do + var i = 0 + while i < length do + items[i].output + i += 1 + end + end end # Abstract class for the SequenceRead compatible @@ -365,7 +448,7 @@ end abstract class StringCharView super SequenceRead[Char] - type SELFTYPE: AbstractString + type SELFTYPE: Text private var target: SELFTYPE @@ -378,6 +461,44 @@ abstract class StringCharView redef fun length do return target.length + redef fun iterator: IndexedIterator[Char] do return self.iterator_from(0) + + # Gets a new Iterator starting at position `pos` + # + # Ex : + # var iter = "abcd".iterator_from(2) + # while iter.is_ok do + # printn iter.item + # iter.next + # end + # + # Outputs : cd + fun iterator_from(pos: Int): IndexedIterator[Char] is abstract + + # Gets an iterator starting at the end and going backwards + # + # Ex : + # var reviter = "now step live...".reverse_iterator + # while reviter.is_ok do + # printn reviter.item + # reviter.next + # end + # + # Outputs : ...evil pets won + fun reverse_iterator: IndexedIterator[Char] do return self.reverse_iterator_from(self.length - 1) + + # Gets an iterator on the chars of self starting from `pos` + # + # Ex : + # var iter = "abcd".reverse_iterator_from(1) + # while iter.is_ok do + # printn iter.item + # iter.next + # end + # + # Outputs : ba + fun reverse_iterator_from(pos: Int): IndexedIterator[Char] is abstract + redef fun has(c: Char): Bool do for i in self do @@ -407,10 +528,10 @@ class String redef type OTHER: String # Index in _items of the start of the string - readable var _index_from: Int + private var index_from: Int # Indes in _items of the last item of the string - readable var _index_to: Int + private var index_to: Int redef var chars: StringCharView = new FlatStringCharView(self) @@ -422,8 +543,8 @@ class String 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] + assert (index + index_from) <= index_to + return items[index + index_from] end redef fun substring(from: Int, count: Int): String @@ -436,56 +557,25 @@ class String from = 0 end - var realFrom = _index_from + from + var realFrom = index_from + from - if (realFrom + count) > _index_to then return new String.with_infos(_items, _index_to - realFrom + 1, realFrom, _index_to) + if (realFrom + count) > index_to then return new String.with_infos(items, index_to - realFrom + 1, realFrom, index_to) if count == 0 then return "" var to = realFrom + count - 1 - return new String.with_infos(_items, to - realFrom + 1, realFrom, to) - end - - redef fun substring_from(from: Int): String - do - if from > _length then return "" - if from < 0 then from = 0 - return substring(from, _length) - end - - 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 + return new String.with_infos(items, to - realFrom + 1, realFrom, to) end redef fun to_upper: String do - var outstr = calloc_string(self._length + 1) + 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 + 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 @@ -495,17 +585,17 @@ class String outstr[self.length] = '\0' - return outstr.to_s_with_length(self._length) + return outstr.to_s_with_length(self.length) end redef fun to_lower : String do - var outstr = calloc_string(self._length + 1) + 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 + 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 @@ -515,23 +605,23 @@ class String outstr[self.length] = '\0' - return outstr.to_s_with_length(self._length) + return outstr.to_s_with_length(self.length) end redef fun trim: String do - if self._length == 0 then return self + 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 + var start_pos = self.index_from + while items[start_pos].ascii <= 32 do start_pos += 1 - if start_pos == _index_to + 1 then return "" + 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 + 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 + if end_pos == start_pos then return items[start_pos].to_s end start_pos -= index_from end_pos -= index_from @@ -540,10 +630,10 @@ class String redef fun output do - var i = self._index_from - var imax = self._index_to + var i = self.index_from + var imax = self.index_to while i <= imax do - _items[i].output + items[i].output i += 1 end end @@ -554,22 +644,22 @@ class String private init with_infos(items: NativeString, len: Int, from: Int, to: Int) do - self._items = items - _length = len - _index_from = from - _index_to = to + self.items = items + length = len + index_from = from + index_to = to end # Return a null terminated char * fun to_cstring: NativeString do - 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) + 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 + return items end redef fun ==(other) @@ -578,17 +668,17 @@ class String if self.object_id == other.object_id then return true - var my_length = _length + var my_length = length - if other._length != my_length then return false + if other.length != my_length then return false - var my_index = _index_from - var its_index = other._index_from + 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 + var itsitems = other.items + var myitems = self.items while my_index < last_iteration do if myitems[my_index] != itsitems[its_index] then return false @@ -609,14 +699,14 @@ class String 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 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_items = self.items + var its_items = other.items - var my_length = self._length - var its_length = other._length + var my_length = self.length + var its_length = other.length var max_iterations = curr_id_self + my_length @@ -641,15 +731,15 @@ class String # assert "hello " + "world!" == "hello world!" fun +(s: String): String do - var my_length = self._length - var its_length = s._length + var my_length = self.length + var its_length = s.length var total_length = my_length + its_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) + 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[total_length] = '\0' @@ -665,11 +755,11 @@ class String do assert i >= 0 - var my_length = self._length + var my_length = self.length var final_length = my_length * i - var my_items = self._items + var my_items = self.items var target_string = calloc_string((final_length) + 1) @@ -691,15 +781,15 @@ class String do # djb2 hash algorythm var h = 5381 - var i = _length - 1 + var i = length - 1 - var myitems = _items - var strStart = _index_from + var myitems = items + var strStart = index_from i += strStart while i >= strStart do - h = (h * 32) + h + self._items[i].ascii + h = (h * 32) + h + self.items[i].ascii i -= 1 end @@ -707,6 +797,32 @@ class String end end +private class FlatStringReverseIterator + super IndexedIterator[Char] + + var target: String + + var target_items: NativeString + + var curr_pos: Int + + init with_pos(tgt: String, pos: Int) + do + target = tgt + target_items = tgt.items + curr_pos = pos + tgt.index_from + end + + redef fun is_ok do return curr_pos >= 0 + + redef fun item do return target_items[curr_pos] + + redef fun next do curr_pos -= 1 + + redef fun index do return curr_pos - target.index_from + +end + private class FlatStringIterator super IndexedIterator[Char] @@ -743,24 +859,53 @@ private class FlatStringCharView # Check that the index (+ index_from) is not larger than indexTo # In other terms, if the index is valid assert index >= 0 - assert (index + target._index_from) <= target._index_to - return target._items[index + target._index_from] + assert (index + target.index_from) <= target.index_to + return target.items[index + target.index_from] end - redef fun iterator: IndexedIterator[Char] do return new FlatStringIterator.with_pos(target, 0) + redef fun iterator_from(start) do return new FlatStringIterator.with_pos(target, start) + + redef fun reverse_iterator_from(start) do return new FlatStringReverseIterator.with_pos(target, start) + +end + +abstract class Buffer + super Text + + # Modifies the char contained at pos `index` + # + # DEPRECATED : Use self.chars.[]= instead + fun []=(index: Int, item: Char) is abstract + + # Adds a char `c` at the end of self + # + # DEPRECATED : Use self.chars.add instead + fun add(c: Char) is abstract + + # Clears the buffer + fun clear is abstract + # Enlarges the subsequent array containing the chars of self + fun enlarge(cap: Int) is abstract + + # Adds the content of string `s` at the end of self + fun append(s: String) is abstract + + redef fun chars: BufferCharView is abstract end # Mutable strings of characters. -class Buffer +class FlatBuffer super AbstractString super Comparable super StringCapable - super AbstractArray[Char] + super Buffer redef type OTHER: String - redef var chars: BufferCharView = new FlatBufferCharView(self) + redef var chars: FlatBufferCharView = new FlatBufferCharView(self) + + var capacity: Int redef fun []=(index, item) do @@ -769,44 +914,35 @@ class Buffer return end assert index >= 0 and index < length - _items[index] = item + items[index] = item end redef fun add(c) do - if _capacity <= length then enlarge(length + 5) - _items[length] = c - _length += 1 + if capacity <= length then enlarge(length + 5) + items[length] = c + length += 1 end + redef fun clear do length = 0 + redef fun enlarge(cap) do - var c = _capacity + var c = capacity if cap <= c then return while c <= cap do c = c * 2 + 2 var a = calloc_string(c+1) - _items.copy_to(a, length, 0, 0) - _items = a - _capacity = c - end - - redef fun append(s) - do - if s isa String then - var sl = s.length - if _capacity < _length + sl then enlarge(_length + sl) - s.items.copy_to(_items, sl, s._index_from, _length) - _length += sl - else - super - end + items.copy_to(a, length, 0, 0) + items = a + capacity = c + items.copy_to(a, length, 0, 0) end redef fun to_s: String do var l = length var a = calloc_string(l+1) - _items.copy_to(a, l, 0, 0) + items.copy_to(a, l, 0, 0) # Ensure the afterlast byte is '\0' to nul-terminated char * a[length] = '\0' @@ -844,10 +980,10 @@ class Buffer init from(s: String) do - _capacity = s.length + 1 - _length = s.length - _items = calloc_string(_capacity) - s.items.copy_to(_items, _length, s._index_from, 0) + capacity = s.length + 1 + length = s.length + items = calloc_string(capacity) + s.items.copy_to(items, length, s.index_from, 0) end # Create a new empty string with a given capacity. @@ -855,19 +991,27 @@ class Buffer do assert cap >= 0 # _items = new NativeString.calloc(cap) - _items = calloc_string(cap+1) - _capacity = cap - _length = 0 + items = calloc_string(cap+1) + capacity = cap + length = 0 + end + + redef fun append(s) + do + var sl = s.length + if capacity < length + sl then enlarge(length + sl) + s.items.copy_to(items, sl, s.index_from, length) + length += sl end redef fun ==(o) do - if not o isa Buffer then return false + if not o isa FlatBuffer then return false var l = length if o.length != l then return false var i = 0 - var it = _items - var oit = o._items + var it = items + var oit = o.items while i < l do if it[i] != oit[i] then return false i += 1 @@ -875,16 +1019,68 @@ class Buffer return true end - readable private var _capacity: Int + # Copies the content of self in `dest` + fun copy(start: Int, len: Int, dest: Buffer, new_start: Int) + do + var self_chars = self.chars + var dest_chars = dest.chars + for i in [0..len-1] do + dest_chars[new_start+i] = self_chars[start+i] + end + end + + redef fun substring(from, count) + do + assert count >= 0 + count += from + if from < 0 then from = 0 + if count > length then count = length + if from < count then + var r = new FlatBuffer.with_capacity(count - from) + while from < count do + r.chars.push(items[from]) + from += 1 + end + return r.to_s + else + return "" + end + end +end + +private class FlatBufferReverseIterator + super IndexedIterator[Char] + + var target: FlatBuffer + + var target_items: NativeString + + var curr_pos: Int + + init with_pos(tgt: FlatBuffer, pos: Int) + do + target = tgt + target_items = tgt.items + curr_pos = pos + end + + redef fun index do return curr_pos + + redef fun is_ok do return curr_pos >= 0 + + redef fun item do return target_items[curr_pos] + + redef fun next do curr_pos -= 1 + end private class FlatBufferCharView super BufferCharView super StringCapable - redef type SELFTYPE: Buffer + redef type SELFTYPE: FlatBuffer - redef fun [](index) do return target._items[index] + redef fun [](index) do return target.items[index] redef fun []=(index, item) do @@ -893,7 +1089,7 @@ private class FlatBufferCharView add(item) return end - target._items[index] = item + target.items[index] = item end redef fun push(c) @@ -918,20 +1114,22 @@ private class FlatBufferCharView if target.capacity < s.length then enlarge(s_length + target.length) end - redef fun iterator: IndexedIterator[Char] do return new FlatBufferIterator.with_pos(target, 0) + redef fun iterator_from(pos) do return new FlatBufferIterator.with_pos(target, pos) + + redef fun reverse_iterator_from(pos) do return new FlatBufferReverseIterator.with_pos(target, pos) end private class FlatBufferIterator super IndexedIterator[Char] - var target: Buffer + var target: FlatBuffer var target_items: NativeString var curr_pos: Int - init with_pos(tgt: Buffer, pos: Int) + init with_pos(tgt: FlatBuffer, pos: Int) do target = tgt target_items = tgt.items @@ -1041,7 +1239,7 @@ redef class Int fun to_base(base: Int, signed: Bool): String do var l = digit_count(base) - var s = new Buffer.from(" " * l) + var s = new FlatBuffer.from(" " * l) fill_buffer(s, base, signed) return s.to_s end @@ -1116,7 +1314,7 @@ redef class Char # assert 'x'.to_s == "x" redef fun to_s do - var s = new Buffer.with_capacity(1) + var s = new FlatBuffer.with_capacity(1) s.chars[0] = self return s.to_s end @@ -1150,7 +1348,7 @@ redef class Collection[E] # Concatenate elements. redef fun to_s do - var s = new Buffer + var s = new FlatBuffer for e in self do if e != null then s.append(e.to_s) return s.to_s end @@ -1163,7 +1361,7 @@ redef class Collection[E] do if is_empty then return "" - var s = new Buffer # Result + var s = new FlatBuffer # Result # Concat first item var i = iterator @@ -1186,7 +1384,7 @@ redef class Array[E] # Fast implementation redef fun to_s do - var s = new Buffer + var s = new FlatBuffer var i = 0 var l = length while i < l do @@ -1211,7 +1409,7 @@ redef class Map[K,V] do if is_empty then return "" - var s = new Buffer # Result + var s = new FlatBuffer # Result # Concat first item var i = iterator