wenzelm@38425
|
1 |
/* Title: Pure/PIDE/text.scala
|
wenzelm@45673
|
2 |
Module: PIDE
|
wenzelm@34276
|
3 |
Author: Fabian Immler, TU Munich
|
wenzelm@34276
|
4 |
Author: Makarius
|
wenzelm@34276
|
5 |
|
wenzelm@38425
|
6 |
Basic operations on plain text.
|
wenzelm@34276
|
7 |
*/
|
wenzelm@34276
|
8 |
|
wenzelm@34276
|
9 |
package isabelle
|
wenzelm@34276
|
10 |
|
wenzelm@34276
|
11 |
|
wenzelm@44379
|
12 |
import scala.collection.mutable
|
wenzelm@44379
|
13 |
import scala.util.Sorting
|
wenzelm@44379
|
14 |
|
wenzelm@44379
|
15 |
|
wenzelm@38425
|
16 |
object Text
|
wenzelm@34276
|
17 |
{
|
wenzelm@38477
|
18 |
/* offset */
|
wenzelm@38426
|
19 |
|
wenzelm@38426
|
20 |
type Offset = Int
|
wenzelm@38426
|
21 |
|
wenzelm@38477
|
22 |
|
wenzelm@38565
|
23 |
/* range -- with total quasi-ordering */
|
wenzelm@38477
|
24 |
|
wenzelm@38568
|
25 |
object Range
|
wenzelm@38568
|
26 |
{
|
wenzelm@38568
|
27 |
def apply(start: Offset): Range = Range(start, start)
|
wenzelm@44379
|
28 |
|
wenzelm@56172
|
29 |
val offside: Range = apply(-1)
|
wenzelm@56172
|
30 |
|
wenzelm@44379
|
31 |
object Ordering extends scala.math.Ordering[Text.Range]
|
wenzelm@44379
|
32 |
{
|
wenzelm@44379
|
33 |
def compare(r1: Text.Range, r2: Text.Range): Int = r1 compare r2
|
wenzelm@44379
|
34 |
}
|
wenzelm@38568
|
35 |
}
|
wenzelm@38568
|
36 |
|
wenzelm@38426
|
37 |
sealed case class Range(val start: Offset, val stop: Offset)
|
wenzelm@38427
|
38 |
{
|
wenzelm@38565
|
39 |
// denotation: {start} Un {i. start < i & i < stop}
|
wenzelm@43425
|
40 |
if (start > stop)
|
wenzelm@43425
|
41 |
error("Bad range: [" + start.toString + ":" + stop.toString + "]")
|
wenzelm@38477
|
42 |
|
wenzelm@38563
|
43 |
override def toString = "[" + start.toString + ":" + stop.toString + "]"
|
wenzelm@38563
|
44 |
|
wenzelm@47542
|
45 |
def length: Int = stop - start
|
wenzelm@47542
|
46 |
|
wenzelm@38427
|
47 |
def map(f: Offset => Offset): Range = Range(f(start), f(stop))
|
wenzelm@56308
|
48 |
def +(i: Offset): Range = if (i == 0) this else map(_ + i)
|
wenzelm@56308
|
49 |
def -(i: Offset): Range = if (i == 0) this else map(_ - i)
|
wenzelm@38662
|
50 |
|
wenzelm@38725
|
51 |
def is_singularity: Boolean = start == stop
|
wenzelm@38662
|
52 |
|
wenzelm@38565
|
53 |
def contains(i: Offset): Boolean = start == i || start < i && i < stop
|
wenzelm@38565
|
54 |
def contains(that: Range): Boolean = this.contains(that.start) && that.stop <= this.stop
|
wenzelm@38565
|
55 |
def overlaps(that: Range): Boolean = this.contains(that.start) || that.contains(this.start)
|
wenzelm@38565
|
56 |
def compare(that: Range): Int = if (overlaps(that)) 0 else this.start compare that.start
|
wenzelm@38485
|
57 |
|
wenzelm@45240
|
58 |
def apart(that: Range): Boolean =
|
wenzelm@45240
|
59 |
(this.start max that.start) > (this.stop min that.stop)
|
wenzelm@45240
|
60 |
|
wenzelm@38564
|
61 |
def restrict(that: Range): Range =
|
wenzelm@38485
|
62 |
Range(this.start max that.start, this.stop min that.stop)
|
wenzelm@43428
|
63 |
|
wenzelm@43428
|
64 |
def try_restrict(that: Range): Option[Range] =
|
wenzelm@45240
|
65 |
if (this apart that) None
|
wenzelm@45240
|
66 |
else Some(restrict(that))
|
wenzelm@45240
|
67 |
|
wenzelm@45240
|
68 |
def try_join(that: Range): Option[Range] =
|
wenzelm@45240
|
69 |
if (this apart that) None
|
wenzelm@45240
|
70 |
else Some(Range(this.start min that.start, this.stop max that.stop))
|
wenzelm@38427
|
71 |
}
|
wenzelm@38426
|
72 |
|
wenzelm@38426
|
73 |
|
wenzelm@56468
|
74 |
/* named chunks */
|
wenzelm@56468
|
75 |
|
wenzelm@56468
|
76 |
abstract class Chunk
|
wenzelm@56468
|
77 |
{
|
wenzelm@56468
|
78 |
def name: Chunk.Name
|
wenzelm@56468
|
79 |
def range: Range
|
wenzelm@56468
|
80 |
def symbol_index: Symbol.Index
|
wenzelm@56468
|
81 |
|
wenzelm@56468
|
82 |
private lazy val hash: Int = (name, range, symbol_index).hashCode
|
wenzelm@56468
|
83 |
override def hashCode: Int = hash
|
wenzelm@56468
|
84 |
override def equals(that: Any): Boolean =
|
wenzelm@56468
|
85 |
that match {
|
wenzelm@56468
|
86 |
case other: Chunk =>
|
wenzelm@56468
|
87 |
hash == other.hash &&
|
wenzelm@56468
|
88 |
name == other.name &&
|
wenzelm@56468
|
89 |
range == other.range &&
|
wenzelm@56468
|
90 |
symbol_index == other.symbol_index
|
wenzelm@56468
|
91 |
case _ => false
|
wenzelm@56468
|
92 |
}
|
wenzelm@56468
|
93 |
override def toString: String = "Text.Chunk(" + name + ")"
|
wenzelm@56468
|
94 |
|
wenzelm@56468
|
95 |
def decode(symbol_offset: Symbol.Offset): Offset = symbol_index.decode(symbol_offset)
|
wenzelm@56468
|
96 |
def decode(symbol_range: Symbol.Range): Range = symbol_index.decode(symbol_range)
|
wenzelm@56468
|
97 |
def incorporate(symbol_range: Symbol.Range): Option[Range] =
|
wenzelm@56468
|
98 |
{
|
wenzelm@56468
|
99 |
def in(r: Symbol.Range): Option[Range] =
|
wenzelm@56468
|
100 |
range.try_restrict(decode(r)) match {
|
wenzelm@56468
|
101 |
case Some(r1) if !r1.is_singularity => Some(r1)
|
wenzelm@56468
|
102 |
case _ => None
|
wenzelm@56468
|
103 |
}
|
wenzelm@56468
|
104 |
in(symbol_range) orElse in(symbol_range - 1)
|
wenzelm@56468
|
105 |
}
|
wenzelm@56468
|
106 |
}
|
wenzelm@56468
|
107 |
|
wenzelm@56468
|
108 |
object Chunk
|
wenzelm@56468
|
109 |
{
|
wenzelm@56468
|
110 |
sealed abstract class Name
|
wenzelm@56468
|
111 |
case object Default extends Name
|
wenzelm@56468
|
112 |
case class File_Name(file_name: String) extends Name
|
wenzelm@56468
|
113 |
|
wenzelm@56468
|
114 |
class File(file_name: String, text: CharSequence) extends Chunk
|
wenzelm@56468
|
115 |
{
|
wenzelm@56468
|
116 |
val name = File_Name(file_name)
|
wenzelm@56468
|
117 |
val range = Range(0, text.length)
|
wenzelm@56468
|
118 |
val symbol_index = Symbol.Index(text)
|
wenzelm@56468
|
119 |
}
|
wenzelm@56468
|
120 |
}
|
wenzelm@56468
|
121 |
|
wenzelm@56468
|
122 |
|
wenzelm@44379
|
123 |
/* perspective */
|
wenzelm@44379
|
124 |
|
wenzelm@44473
|
125 |
object Perspective
|
wenzelm@44379
|
126 |
{
|
wenzelm@44474
|
127 |
val empty: Perspective = Perspective(Nil)
|
wenzelm@44379
|
128 |
|
wenzelm@46576
|
129 |
def full: Perspective = Perspective(List(Range(0, Integer.MAX_VALUE / 2)))
|
wenzelm@46576
|
130 |
|
wenzelm@44473
|
131 |
def apply(ranges: Seq[Range]): Perspective =
|
wenzelm@44379
|
132 |
{
|
wenzelm@44473
|
133 |
val result = new mutable.ListBuffer[Text.Range]
|
wenzelm@44473
|
134 |
var last: Option[Text.Range] = None
|
wenzelm@45240
|
135 |
def ship(next: Option[Range]) { result ++= last; last = next }
|
wenzelm@45240
|
136 |
|
wenzelm@45240
|
137 |
for (range <- ranges.sortBy(_.start))
|
wenzelm@44473
|
138 |
{
|
wenzelm@44473
|
139 |
last match {
|
wenzelm@45240
|
140 |
case None => ship(Some(range))
|
wenzelm@45240
|
141 |
case Some(last_range) =>
|
wenzelm@45240
|
142 |
last_range.try_join(range) match {
|
wenzelm@45240
|
143 |
case None => ship(Some(range))
|
wenzelm@45240
|
144 |
case joined => last = joined
|
wenzelm@45240
|
145 |
}
|
wenzelm@44473
|
146 |
}
|
wenzelm@44379
|
147 |
}
|
wenzelm@45240
|
148 |
ship(None)
|
wenzelm@44473
|
149 |
new Perspective(result.toList)
|
wenzelm@44379
|
150 |
}
|
wenzelm@44473
|
151 |
}
|
wenzelm@44473
|
152 |
|
wenzelm@46712
|
153 |
final class Perspective private(
|
wenzelm@46712
|
154 |
val ranges: List[Range]) // visible text partitioning in canonical order
|
wenzelm@44473
|
155 |
{
|
wenzelm@44473
|
156 |
def is_empty: Boolean = ranges.isEmpty
|
wenzelm@44473
|
157 |
def range: Range =
|
wenzelm@44473
|
158 |
if (is_empty) Range(0)
|
wenzelm@44473
|
159 |
else Range(ranges.head.start, ranges.last.stop)
|
wenzelm@45631
|
160 |
|
wenzelm@45631
|
161 |
override def hashCode: Int = ranges.hashCode
|
wenzelm@45631
|
162 |
override def equals(that: Any): Boolean =
|
wenzelm@45631
|
163 |
that match {
|
wenzelm@45631
|
164 |
case other: Perspective => ranges == other.ranges
|
wenzelm@45631
|
165 |
case _ => false
|
wenzelm@45631
|
166 |
}
|
wenzelm@45240
|
167 |
override def toString = ranges.toString
|
wenzelm@44379
|
168 |
}
|
wenzelm@44379
|
169 |
|
wenzelm@44379
|
170 |
|
wenzelm@38577
|
171 |
/* information associated with text range */
|
wenzelm@38577
|
172 |
|
wenzelm@43714
|
173 |
sealed case class Info[A](val range: Text.Range, val info: A)
|
wenzelm@38577
|
174 |
{
|
wenzelm@38577
|
175 |
def restrict(r: Text.Range): Info[A] = Info(range.restrict(r), info)
|
wenzelm@46207
|
176 |
def try_restrict(r: Text.Range): Option[Info[A]] = range.try_restrict(r).map(Info(_, info))
|
wenzelm@38577
|
177 |
}
|
wenzelm@38577
|
178 |
|
wenzelm@45470
|
179 |
type Markup = Info[XML.Elem]
|
wenzelm@45455
|
180 |
|
wenzelm@38577
|
181 |
|
wenzelm@38426
|
182 |
/* editing */
|
wenzelm@34286
|
183 |
|
wenzelm@38425
|
184 |
object Edit
|
wenzelm@38425
|
185 |
{
|
wenzelm@38426
|
186 |
def insert(start: Offset, text: String): Edit = new Edit(true, start, text)
|
wenzelm@38426
|
187 |
def remove(start: Offset, text: String): Edit = new Edit(false, start, text)
|
wenzelm@38425
|
188 |
}
|
wenzelm@34286
|
189 |
|
wenzelm@46712
|
190 |
final class Edit private(val is_insert: Boolean, val start: Offset, val text: String)
|
wenzelm@38425
|
191 |
{
|
wenzelm@38425
|
192 |
override def toString =
|
wenzelm@38425
|
193 |
(if (is_insert) "Insert(" else "Remove(") + (start, text).toString + ")"
|
wenzelm@34286
|
194 |
|
wenzelm@34286
|
195 |
|
wenzelm@38425
|
196 |
/* transform offsets */
|
wenzelm@34286
|
197 |
|
wenzelm@38426
|
198 |
private def transform(do_insert: Boolean, i: Offset): Offset =
|
wenzelm@38426
|
199 |
if (i < start) i
|
wenzelm@43425
|
200 |
else if (do_insert) i + text.length
|
wenzelm@38426
|
201 |
else (i - text.length) max start
|
wenzelm@34286
|
202 |
|
wenzelm@43425
|
203 |
def convert(i: Offset): Offset = transform(is_insert, i)
|
wenzelm@43425
|
204 |
def revert(i: Offset): Offset = transform(!is_insert, i)
|
wenzelm@43425
|
205 |
def convert(range: Range): Range = range.map(convert)
|
wenzelm@43425
|
206 |
def revert(range: Range): Range = range.map(revert)
|
wenzelm@38425
|
207 |
|
wenzelm@34286
|
208 |
|
wenzelm@38425
|
209 |
/* edit strings */
|
wenzelm@38425
|
210 |
|
wenzelm@38426
|
211 |
private def insert(i: Offset, string: String): String =
|
wenzelm@38426
|
212 |
string.substring(0, i) + text + string.substring(i)
|
wenzelm@34276
|
213 |
|
wenzelm@38426
|
214 |
private def remove(i: Offset, count: Int, string: String): String =
|
wenzelm@38426
|
215 |
string.substring(0, i) + string.substring(i + count)
|
wenzelm@38425
|
216 |
|
wenzelm@38425
|
217 |
def can_edit(string: String, shift: Int): Boolean =
|
wenzelm@38425
|
218 |
shift <= start && start < shift + string.length
|
wenzelm@38425
|
219 |
|
wenzelm@38425
|
220 |
def edit(string: String, shift: Int): (Option[Edit], String) =
|
wenzelm@38425
|
221 |
if (!can_edit(string, shift)) (Some(this), string)
|
wenzelm@38425
|
222 |
else if (is_insert) (None, insert(start - shift, string))
|
wenzelm@38425
|
223 |
else {
|
wenzelm@38426
|
224 |
val i = start - shift
|
wenzelm@38426
|
225 |
val count = text.length min (string.length - i)
|
wenzelm@38425
|
226 |
val rest =
|
wenzelm@38425
|
227 |
if (count == text.length) None
|
wenzelm@38425
|
228 |
else Some(Edit.remove(start, text.substring(count)))
|
wenzelm@38426
|
229 |
(rest, remove(i, count, string))
|
wenzelm@38425
|
230 |
}
|
wenzelm@38425
|
231 |
}
|
wenzelm@34276
|
232 |
}
|