# High-level abstraction for all text representations
abstract class Text
+ super Comparable
+ super StringCapable
+
+ redef type OTHER: Text
+
+ # Type of the view on self (.chars)
+ type SELFVIEW: StringCharView
+
+ # Type of self (used for factorization of several methods, ex : substring_from, empty...)
+ type SELFTYPE: Text
+
+ var hash_cache: nullable Int = null
# Gets a view on the chars of the Text object
- fun chars: StringCharView is abstract
+ fun chars: SELFVIEW is abstract
# Number of characters contained in self.
fun length: Int is abstract
# 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 is abstract
+ fun substring(from: Int, count: Int): SELFTYPE is abstract
+
+ # Concatenates `o` to `self`
+ fun +(o: Text): SELFTYPE is abstract
+
+ # Auto-concatenates self `i` times
+ fun *(i: Int): SELFTYPE is abstract
# Is the current Text empty (== "")
# assert "".is_empty
# assert not "foo".is_empty
fun is_empty: Bool do return self.length == 0
+ # Returns an empty Text of the right type
+ fun empty: SELFTYPE is abstract
+
# Gets the first char of the Text
#
# DEPRECATED : Use self.chars.first instead
return last_index_of_from(c, length - 1)
end
+ # Return a null terminated char *
+ fun to_cstring: NativeString do return flatten.to_cstring
+
# 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 "abcd".substring_from(2) == "cd"
#
# As with substring, a `from` index < 0 will be replaced by 0
- fun substring_from(from: Int): String
+ fun substring_from(from: Int): SELFTYPE
do
- assert from < length
+ if from > self.length then return empty
+ if from < 0 then from = 0
return substring(from, length - from)
end
+ # Returns a reversed version of self
+ fun reversed: SELFTYPE is abstract
+
# Does self have a substring `str` starting from position `pos`?
#
# assert "abcd".has_substring("bc",1) == true
# A upper case version of `self`
#
# assert "Hello World!".to_upper == "HELLO WORLD!"
- fun to_upper: String
- do
- var s = new FlatBuffer.with_capacity(length)
- for i in self.chars do s.add(i.to_upper)
- return s.to_s
- end
+ fun to_upper: SELFTYPE is abstract
# A lower case version of `self`
#
# assert "Hello World!".to_lower == "hello world!"
- fun to_lower : String
+ fun to_lower : SELFTYPE is abstract
+
+ # Removes the whitespaces at the beginning of self
+ fun l_trim: SELFTYPE
do
- var s = new FlatBuffer.with_capacity(length)
- for i in self.chars do s.add(i.to_lower)
- return s.to_s
+ var iter = self.chars.iterator
+ while iter.is_ok do
+ if iter.item.ascii > 32 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
+ fun r_trim: SELFTYPE
+ do
+ var iter = self.chars.reverse_iterator
+ while iter.is_ok do
+ if iter.item.ascii > 32 then break
+ iter.next
+ end
+ if iter.index == length 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"
- 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.chars[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.chars[end_pos].ascii <= 32 do
- end_pos -= 1
- if end_pos == start_pos then return self.chars[start_pos].to_s
- end
- return self.substring(start_pos, end_pos - start_pos + 1)
- end
+ fun trim: SELFTYPE do return (self.l_trim).r_trim
# Mangle a string to be a unique string only made of alphanumeric characters
fun to_cmangle: String
return res.to_s
end
+ 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
+
+ redef fun <(o)
+ do
+ return self.chars < o.chars
+ end
+
+ # Flat representation of self
+ fun flatten: FlatText is abstract
+
+ redef fun hash
+ do
+ if hash_cache == null then
+ # djb2 hash algorithm
+ var h = 5381
+ var i = length - 1
+
+ for char in self.chars do
+ h = (h * 32) + h + char.ascii
+ i -= 1
+ end
+
+ hash_cache = h
+ end
+ return hash_cache.as(not null)
+ end
+
end
# All kinds of array-based text representations.
private var items: NativeString
+ # Real items, used as cache for to_cstring is called
+ private var real_items: nullable NativeString = null
+
redef var length: Int
init do end
i += 1
end
end
+
+ redef fun flatten do return self
end
# Abstract class for the SequenceRead compatible
# views on String and Buffer objects
abstract class StringCharView
super SequenceRead[Char]
+ super Comparable
type SELFTYPE: Text
+ redef type OTHER: StringCharView
+
private var target: SELFTYPE
private init(tgt: SELFTYPE)
return false
end
+ redef fun ==(other)
+ do
+ if other == null then return false
+ if not other isa StringCharView then return false
+ var other_chars = other.iterator
+ for i in self do
+ if i != other_chars.item then return false
+ other_chars.next
+ end
+ return true
+ end
+
+ redef fun <(other)
+ do
+ var self_chars = self.iterator
+ var other_chars = other.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
end
# View on Buffer objects, extends Sequence
end
+abstract class String
+ super Text
+
+ redef type SELFTYPE: String
+
+ redef fun to_s do return self
+
+end
+
# Immutable strings of characters.
-class String
- super Comparable
+class FlatString
super FlatText
- super StringCapable
+ super String
- redef type OTHER: String
+ redef type SELFTYPE: FlatString
+ redef type SELFVIEW: FlatStringCharView
# Index in _items of the start of the string
private var index_from: Int
# Indes in _items of the last item of the string
private var index_to: Int
- redef var chars: StringCharView = new FlatStringCharView(self)
+ redef var chars: SELFVIEW = new FlatStringCharView(self)
################################################
# AbstractString specific methods #
return items[index + index_from]
end
- redef fun substring(from: Int, count: Int): String
+ redef fun reversed
+ do
+ var native = calloc_string(self.length + 1)
+ var reviter = chars.reverse_iterator
+ var pos = 0
+ while reviter.is_ok do
+ native[pos] = reviter.item
+ pos += 1
+ reviter.next
+ end
+ return native.to_s_with_length(self.length)
+ end
+
+ redef fun substring(from, count)
do
assert count >= 0
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 FlatString.with_infos(items, index_to - realFrom + 1, realFrom, index_to)
- if count == 0 then return ""
+ if count == 0 then return empty
var to = realFrom + count - 1
- return new String.with_infos(items, to - realFrom + 1, realFrom, to)
+ return new FlatString.with_infos(items, to - realFrom + 1, realFrom, to)
end
- redef fun to_upper: String
+ redef fun empty do return "".as(FlatString)
+
+ redef fun to_upper
do
var outstr = calloc_string(self.length + 1)
var out_index = 0
return outstr.to_s_with_length(self.length)
end
- redef fun to_lower : String
+ redef fun to_lower
do
var outstr = calloc_string(self.length + 1)
var out_index = 0
return outstr.to_s_with_length(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
end
# Return a null terminated char *
- fun to_cstring: NativeString
+ redef fun to_cstring: NativeString
do
+ if real_items != null then return real_items.as(not null)
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'
+ self.real_items = newItems
return newItems
end
return items
redef fun ==(other)
do
- if not other isa String then return false
+ if not other isa FlatString then return super
if self.object_id == other.object_id then return true
# assert ("aa" < "b") == true
redef fun <(other)
do
+ if not other isa FlatString then return super
+
if self.object_id == other.object_id then return false
var my_curr_char : Char
# The concatenation of `self` with `s`
#
# assert "hello " + "world!" == "hello world!"
- fun +(s: String): String
+ redef fun +(s)
do
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)
+ if s isa FlatString then
+ s.items.copy_to(target_string, its_length, s.index_from, my_length)
+ else if s isa FlatBuffer then
+ s.items.copy_to(target_string, its_length, 0, my_length)
+ else
+ var curr_pos = my_length
+ for i in s.chars do
+ target_string[curr_pos] = i
+ curr_pos += 1
+ end
+ end
target_string[total_length] = '\0'
return target_string.to_s_with_length(total_length)
end
- # `i` repetitions of `self`
- #
# assert "abc"*3 == "abcabcabc"
# assert "abc"*1 == "abc"
# assert "abc"*0 == ""
- fun *(i: Int): String
+ redef fun *(i)
do
assert i >= 0
return target_string.to_s_with_length(final_length)
end
- redef fun to_s do return self
-
redef fun hash
do
- # djb2 hash algorythm
- var h = 5381
- var i = length - 1
+ if hash_cache == null then
+ # djb2 hash algorythm
+ var h = 5381
+ var i = length - 1
- var myitems = items
- var strStart = index_from
+ var myitems = items
+ var strStart = index_from
- i += strStart
+ i += strStart
- while i >= strStart do
- h = (h * 32) + h + self.items[i].ascii
- i -= 1
+ while i >= strStart do
+ h = (h * 32) + h + self.items[i].ascii
+ i -= 1
+ end
+
+ hash_cache = h
end
- return h
+ return hash_cache.as(not null)
end
end
private class FlatStringReverseIterator
super IndexedIterator[Char]
- var target: String
+ var target: FlatString
var target_items: NativeString
var curr_pos: Int
- init with_pos(tgt: String, pos: Int)
+ init with_pos(tgt: FlatString, pos: Int)
do
target = tgt
target_items = tgt.items
private class FlatStringIterator
super IndexedIterator[Char]
- var target: String
+ var target: FlatString
var target_items: NativeString
var curr_pos: Int
- init with_pos(tgt: String, pos: Int)
+ init with_pos(tgt: FlatString, pos: Int)
do
target = tgt
target_items = tgt.items
private class FlatStringCharView
super StringCharView
- redef type SELFTYPE: String
+ redef type SELFTYPE: FlatString
redef fun [](index)
do
abstract class Buffer
super Text
+ redef type SELFVIEW: BufferCharView
+ redef type SELFTYPE: Buffer
+
# Modifies the char contained at pos `index`
#
# DEPRECATED : Use self.chars.[]= instead
# 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
+ # Adds the content of text `s` at the end of self
+ fun append(s: Text) is abstract
- redef fun chars: BufferCharView is abstract
end
# Mutable strings of characters.
class FlatBuffer
super FlatText
- super Comparable
- super StringCapable
super Buffer
- redef type OTHER: String
+ redef type SELFVIEW: FlatBufferCharView
+ redef type SELFTYPE: FlatBuffer
- redef var chars: FlatBufferCharView = new FlatBufferCharView(self)
+ redef var chars: SELFVIEW = new FlatBufferCharView(self)
var capacity: Int
redef fun clear do length = 0
+ redef fun empty do return new FlatBuffer
+
redef fun enlarge(cap)
do
var c = capacity
return a.to_s_with_length(length)
end
- redef fun <(s)
+ redef fun to_cstring
do
- var i = 0
- var l1 = length
- var l2 = s.length
- while i < l1 and i < l2 do
- var c1 = self.chars[i].ascii
- var c2 = s.chars[i].ascii
- if c1 < c2 then
- return true
- else if c2 < c1 then
- return false
- end
- i += 1
- end
- if l1 < l2 then
- return true
- else
- return false
+ if real_items == null then
+ var new_native = calloc_string(length + 1)
+ new_native[length] = '\0'
+ items.copy_to(new_native, length, 0, 0)
+ real_items = new_native
end
+ return real_items.as(not null)
end
# Create a new empty string.
with_capacity(5)
end
- init from(s: String)
+ init from(s: Text)
do
capacity = s.length + 1
length = s.length
items = calloc_string(capacity)
- s.items.copy_to(items, length, s.index_from, 0)
+ if s isa FlatString then
+ s.items.copy_to(items, length, s.index_from, 0)
+ else if s isa FlatBuffer then
+ s.items.copy_to(items, length, 0, 0)
+ else
+ var curr_pos = 0
+ for i in s.chars do
+ items[curr_pos] = i
+ curr_pos += 1
+ end
+ end
end
# Create a new empty string with a given capacity.
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 FlatBuffer 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 s isa FlatString then
+ s.items.copy_to(items, sl, s.index_from, length)
+ else if s isa FlatBuffer then
+ s.items.copy_to(items, sl, 0, length)
+ else
+ var curr_pos = self.length
+ for i in s.chars do
+ items[curr_pos] = i
+ curr_pos += 1
+ end
end
- return true
+ length += sl
end
# Copies the content of self in `dest`
r.chars.push(items[from])
from += 1
end
- return r.to_s
+ return r
else
- return ""
+ return new FlatBuffer
+ end
+ end
+
+ redef fun reversed
+ do
+ var new_buf = new FlatBuffer.with_capacity(self.length)
+ var reviter = self.chars.reverse_iterator
+ while reviter.is_ok do
+ new_buf.add(reviter.item)
+ reviter.next
+ end
+ return new_buf
+ end
+
+ redef fun +(other)
+ do
+ var new_buf = new FlatBuffer.with_capacity(self.length + other.length)
+ new_buf.append(self)
+ new_buf.append(other)
+ return new_buf
+ end
+
+ redef fun *(repeats)
+ do
+ var new_buf = new FlatBuffer.with_capacity(self.length * repeats)
+ for i in [0..repeats[ do
+ new_buf.append(self)
end
+ return new_buf
end
end
return to_s_with_length(cstring_length)
end
- fun to_s_with_length(length: Int): String
+ fun to_s_with_length(length: Int): FlatString
do
assert length >= 0
- return new String.with_infos(self, length, 0, length - 1)
+ return new FlatString.with_infos(self, length, 0, length - 1)
end
- fun to_s_with_copy: String
+ fun to_s_with_copy: FlatString
do
var length = cstring_length
var new_self = calloc_string(length + 1)
copy_to(new_self, length, 0, 0)
- return new String.with_infos(new_self, length, 0, length - 1)
+ return new FlatString.with_infos(new_self, length, 0, length - 1)
end
end