src/Pure/PIDE/yxml.scala
author wenzelm
Wed, 26 Jun 2024 15:22:20 +0200
changeset 80425 efa212a6699a
parent 75393 87ebf5a50283
child 80427 c92356464bb3
permissions -rw-r--r--
clarified signature;
Ignore whitespace changes - Everywhere: Within whitespace: At end of lines:
44698
0385292321a0 moved XML/YXML to src/Pure/PIDE;
wenzelm
parents: 44181
diff changeset
     1
/*  Title:      Pure/PIDE/yxml.scala
27930
2b44df907cc2 Efficient text representation of XML trees.
wenzelm
parents:
diff changeset
     2
    Author:     Makarius
2b44df907cc2 Efficient text representation of XML trees.
wenzelm
parents:
diff changeset
     3
44698
0385292321a0 moved XML/YXML to src/Pure/PIDE;
wenzelm
parents: 44181
diff changeset
     4
Efficient text representation of XML trees.  Suitable for direct
0385292321a0 moved XML/YXML to src/Pure/PIDE;
wenzelm
parents: 44181
diff changeset
     5
inlining into plain text.
27930
2b44df907cc2 Efficient text representation of XML trees.
wenzelm
parents:
diff changeset
     6
*/
2b44df907cc2 Efficient text representation of XML trees.
wenzelm
parents:
diff changeset
     7
2b44df907cc2 Efficient text representation of XML trees.
wenzelm
parents:
diff changeset
     8
package isabelle
2b44df907cc2 Efficient text representation of XML trees.
wenzelm
parents:
diff changeset
     9
2b44df907cc2 Efficient text representation of XML trees.
wenzelm
parents:
diff changeset
    10
40453
wenzelm
parents: 38475
diff changeset
    11
import scala.collection.mutable
36684
943f1ca7b375 misc tuning -- accumulate body via ListBuffer;
wenzelm
parents: 34201
diff changeset
    12
943f1ca7b375 misc tuning -- accumulate body via ListBuffer;
wenzelm
parents: 34201
diff changeset
    13
75393
87ebf5a50283 clarified formatting, for the sake of scala3;
wenzelm
parents: 73712
diff changeset
    14
object YXML {
27943
f34ff5e7728f replaced Pattern.split by chunks iterator (more efficient, resembles ML version more closely);
wenzelm
parents: 27930
diff changeset
    15
  /* chunk markers */
27930
2b44df907cc2 Efficient text representation of XML trees.
wenzelm
parents:
diff changeset
    16
80425
efa212a6699a clarified signature;
wenzelm
parents: 75393
diff changeset
    17
  val X_byte: Byte = 5
efa212a6699a clarified signature;
wenzelm
parents: 75393
diff changeset
    18
  val Y_byte: Byte = 6
efa212a6699a clarified signature;
wenzelm
parents: 75393
diff changeset
    19
efa212a6699a clarified signature;
wenzelm
parents: 75393
diff changeset
    20
  val X = X_byte.toChar
efa212a6699a clarified signature;
wenzelm
parents: 75393
diff changeset
    21
  val Y = Y_byte.toChar
56600
628e039cc34d more specific support for sequence of words;
wenzelm
parents: 55554
diff changeset
    22
73029
wenzelm
parents: 73028
diff changeset
    23
  val is_X: Char => Boolean = _ == X
wenzelm
parents: 73028
diff changeset
    24
  val is_Y: Char => Boolean = _ == Y
56600
628e039cc34d more specific support for sequence of words;
wenzelm
parents: 55554
diff changeset
    25
71601
97ccf48c2f0c misc tuning based on hints by IntelliJ IDEA;
wenzelm
parents: 71534
diff changeset
    26
  val X_string: String = X.toString
97ccf48c2f0c misc tuning based on hints by IntelliJ IDEA;
wenzelm
parents: 71534
diff changeset
    27
  val Y_string: String = Y.toString
97ccf48c2f0c misc tuning based on hints by IntelliJ IDEA;
wenzelm
parents: 71534
diff changeset
    28
  val XY_string: String = X_string + Y_string
97ccf48c2f0c misc tuning based on hints by IntelliJ IDEA;
wenzelm
parents: 71534
diff changeset
    29
  val XYX_string: String = XY_string + X_string
27930
2b44df907cc2 Efficient text representation of XML trees.
wenzelm
parents:
diff changeset
    30
43782
834de29572d5 clarified YXML.detect;
wenzelm
parents: 43774
diff changeset
    31
  def detect(s: String): Boolean = s.exists(c => c == X || c == Y)
67820
e30d6368c7c8 clarified argument formats: explicit Unit, allow XML.Elem as well;
wenzelm
parents: 64370
diff changeset
    32
  def detect_elem(s: String): Boolean = s.startsWith(XY_string)
43782
834de29572d5 clarified YXML.detect;
wenzelm
parents: 43774
diff changeset
    33
27930
2b44df907cc2 Efficient text representation of XML trees.
wenzelm
parents:
diff changeset
    34
73029
wenzelm
parents: 73028
diff changeset
    35
  /* string representation */
38268
beb86b805590 more uniform XML/YXML string_of_body/string_of_tree;
wenzelm
parents: 38267
diff changeset
    36
80425
efa212a6699a clarified signature;
wenzelm
parents: 75393
diff changeset
    37
  trait Output {
efa212a6699a clarified signature;
wenzelm
parents: 75393
diff changeset
    38
    def byte(b: Byte): Unit
efa212a6699a clarified signature;
wenzelm
parents: 75393
diff changeset
    39
    def string(s: String): Unit
38268
beb86b805590 more uniform XML/YXML string_of_body/string_of_tree;
wenzelm
parents: 38267
diff changeset
    40
    def tree(t: XML.Tree): Unit =
beb86b805590 more uniform XML/YXML string_of_body/string_of_tree;
wenzelm
parents: 38267
diff changeset
    41
      t match {
73556
192bcee4f8b8 more robust treatment of empty markup: it allows to produce formal chunks;
wenzelm
parents: 73340
diff changeset
    42
        case XML.Elem(markup @ Markup(name, atts), ts) =>
192bcee4f8b8 more robust treatment of empty markup: it allows to produce formal chunks;
wenzelm
parents: 73340
diff changeset
    43
          if (markup.is_empty) ts.foreach(tree)
192bcee4f8b8 more robust treatment of empty markup: it allows to produce formal chunks;
wenzelm
parents: 73340
diff changeset
    44
          else {
80425
efa212a6699a clarified signature;
wenzelm
parents: 75393
diff changeset
    45
            byte(X_byte)
efa212a6699a clarified signature;
wenzelm
parents: 75393
diff changeset
    46
            byte(Y_byte)
73556
192bcee4f8b8 more robust treatment of empty markup: it allows to produce formal chunks;
wenzelm
parents: 73340
diff changeset
    47
            string(name)
80425
efa212a6699a clarified signature;
wenzelm
parents: 75393
diff changeset
    48
            for ((a, b) <- atts) {
efa212a6699a clarified signature;
wenzelm
parents: 75393
diff changeset
    49
              byte(Y_byte)
efa212a6699a clarified signature;
wenzelm
parents: 75393
diff changeset
    50
              string(a)
efa212a6699a clarified signature;
wenzelm
parents: 75393
diff changeset
    51
              byte('='.toByte)
efa212a6699a clarified signature;
wenzelm
parents: 75393
diff changeset
    52
              string(b)
efa212a6699a clarified signature;
wenzelm
parents: 75393
diff changeset
    53
            }
efa212a6699a clarified signature;
wenzelm
parents: 75393
diff changeset
    54
            byte(X_byte)
73556
192bcee4f8b8 more robust treatment of empty markup: it allows to produce formal chunks;
wenzelm
parents: 73340
diff changeset
    55
            ts.foreach(tree)
80425
efa212a6699a clarified signature;
wenzelm
parents: 75393
diff changeset
    56
            byte(X_byte)
efa212a6699a clarified signature;
wenzelm
parents: 75393
diff changeset
    57
            byte(Y_byte)
efa212a6699a clarified signature;
wenzelm
parents: 75393
diff changeset
    58
            byte(X_byte)
73556
192bcee4f8b8 more robust treatment of empty markup: it allows to produce formal chunks;
wenzelm
parents: 73340
diff changeset
    59
          }
71534
f10bffaa2bae more scalable output of YXML files;
wenzelm
parents: 67820
diff changeset
    60
        case XML.Text(text) => string(text)
38268
beb86b805590 more uniform XML/YXML string_of_body/string_of_tree;
wenzelm
parents: 38267
diff changeset
    61
      }
80425
efa212a6699a clarified signature;
wenzelm
parents: 75393
diff changeset
    62
    def trees(ts: Iterable[XML.Tree]): Unit = ts.foreach(tree)
71534
f10bffaa2bae more scalable output of YXML files;
wenzelm
parents: 67820
diff changeset
    63
  }
f10bffaa2bae more scalable output of YXML files;
wenzelm
parents: 67820
diff changeset
    64
80425
efa212a6699a clarified signature;
wenzelm
parents: 75393
diff changeset
    65
  class Output_String(builder: StringBuilder) extends Output {
efa212a6699a clarified signature;
wenzelm
parents: 75393
diff changeset
    66
    override def byte(b: Byte): Unit = { builder += b.toChar }
efa212a6699a clarified signature;
wenzelm
parents: 75393
diff changeset
    67
    override def string(s: String): Unit = { builder ++= s }
efa212a6699a clarified signature;
wenzelm
parents: 75393
diff changeset
    68
    def result(ts: Iterable[XML.Tree]): String = { trees(ts); builder.toString }
efa212a6699a clarified signature;
wenzelm
parents: 75393
diff changeset
    69
    def result(t: XML.Tree): String = { tree(t); builder.toString }
38268
beb86b805590 more uniform XML/YXML string_of_body/string_of_tree;
wenzelm
parents: 38267
diff changeset
    70
  }
beb86b805590 more uniform XML/YXML string_of_body/string_of_tree;
wenzelm
parents: 38267
diff changeset
    71
80425
efa212a6699a clarified signature;
wenzelm
parents: 75393
diff changeset
    72
  def string_of_body(body: XML.Body): String =
efa212a6699a clarified signature;
wenzelm
parents: 75393
diff changeset
    73
    new Output_String(new StringBuilder).result(body)
efa212a6699a clarified signature;
wenzelm
parents: 75393
diff changeset
    74
efa212a6699a clarified signature;
wenzelm
parents: 75393
diff changeset
    75
  def string_of_tree(tree: XML.Tree): String =
efa212a6699a clarified signature;
wenzelm
parents: 75393
diff changeset
    76
    new Output_String(new StringBuilder).result(tree)
38268
beb86b805590 more uniform XML/YXML string_of_body/string_of_tree;
wenzelm
parents: 38267
diff changeset
    77
beb86b805590 more uniform XML/YXML string_of_body/string_of_tree;
wenzelm
parents: 38267
diff changeset
    78
27930
2b44df907cc2 Efficient text representation of XML trees.
wenzelm
parents:
diff changeset
    79
  /* parsing */
2b44df907cc2 Efficient text representation of XML trees.
wenzelm
parents:
diff changeset
    80
27993
6dd90ef9f927 simplified exceptions: use plain error function / RuntimeException;
wenzelm
parents: 27971
diff changeset
    81
  private def err(msg: String) = error("Malformed YXML: " + msg)
27930
2b44df907cc2 Efficient text representation of XML trees.
wenzelm
parents:
diff changeset
    82
  private def err_attribute() = err("bad attribute")
2b44df907cc2 Efficient text representation of XML trees.
wenzelm
parents:
diff changeset
    83
  private def err_element() = err("bad element")
2b44df907cc2 Efficient text representation of XML trees.
wenzelm
parents:
diff changeset
    84
  private def err_unbalanced(name: String) =
2b44df907cc2 Efficient text representation of XML trees.
wenzelm
parents:
diff changeset
    85
    if (name == "") err("unbalanced element")
44181
wenzelm
parents: 43841
diff changeset
    86
    else err("unbalanced element " + quote(name))
27930
2b44df907cc2 Efficient text representation of XML trees.
wenzelm
parents:
diff changeset
    87
73029
wenzelm
parents: 73028
diff changeset
    88
  private def parse_attrib(source: CharSequence): (String, String) =
73712
3eba8d4b624b clarified signature;
wenzelm
parents: 73556
diff changeset
    89
    Properties.Eq.unapply(source.toString) getOrElse err_attribute()
27930
2b44df907cc2 Efficient text representation of XML trees.
wenzelm
parents:
diff changeset
    90
2b44df907cc2 Efficient text representation of XML trees.
wenzelm
parents:
diff changeset
    91
75393
87ebf5a50283 clarified formatting, for the sake of scala3;
wenzelm
parents: 73712
diff changeset
    92
  def parse_body(source: CharSequence, cache: XML.Cache = XML.Cache.none): XML.Body = {
27930
2b44df907cc2 Efficient text representation of XML trees.
wenzelm
parents:
diff changeset
    93
    /* stack operations */
2b44df907cc2 Efficient text representation of XML trees.
wenzelm
parents:
diff changeset
    94
40453
wenzelm
parents: 38475
diff changeset
    95
    def buffer(): mutable.ListBuffer[XML.Tree] = new mutable.ListBuffer[XML.Tree]
wenzelm
parents: 38475
diff changeset
    96
    var stack: List[(Markup, mutable.ListBuffer[XML.Tree])] = List((Markup.Empty, buffer()))
27930
2b44df907cc2 Efficient text representation of XML trees.
wenzelm
parents:
diff changeset
    97
73340
0ffcad1f6130 tuned --- fewer warnings;
wenzelm
parents: 73030
diff changeset
    98
    def add(x: XML.Tree): Unit =
73029
wenzelm
parents: 73028
diff changeset
    99
      (stack: @unchecked) match { case (_, body) :: _ => body += x }
27930
2b44df907cc2 Efficient text representation of XML trees.
wenzelm
parents:
diff changeset
   100
73340
0ffcad1f6130 tuned --- fewer warnings;
wenzelm
parents: 73030
diff changeset
   101
    def push(name: String, atts: XML.Attributes): Unit =
27930
2b44df907cc2 Efficient text representation of XML trees.
wenzelm
parents:
diff changeset
   102
      if (name == "") err_element()
73030
72a8fdfa185d support more direct hash-consing via XML.Cache;
wenzelm
parents: 73029
diff changeset
   103
      else stack = (cache.markup(Markup(name, atts)), buffer()) :: stack
27930
2b44df907cc2 Efficient text representation of XML trees.
wenzelm
parents:
diff changeset
   104
73340
0ffcad1f6130 tuned --- fewer warnings;
wenzelm
parents: 73030
diff changeset
   105
    def pop(): Unit =
36684
943f1ca7b375 misc tuning -- accumulate body via ListBuffer;
wenzelm
parents: 34201
diff changeset
   106
      (stack: @unchecked) match {
73029
wenzelm
parents: 73028
diff changeset
   107
        case (Markup.Empty, _) :: _ => err_unbalanced("")
wenzelm
parents: 73028
diff changeset
   108
        case (markup, body) :: pending =>
38230
ed147003de4b simplified type XML.Tree: embed Markup directly, avoid slightly odd triple;
wenzelm
parents: 36685
diff changeset
   109
          stack = pending
73030
72a8fdfa185d support more direct hash-consing via XML.Cache;
wenzelm
parents: 73029
diff changeset
   110
          add(cache.tree0(XML.Elem(markup, body.toList)))
36684
943f1ca7b375 misc tuning -- accumulate body via ListBuffer;
wenzelm
parents: 34201
diff changeset
   111
      }
27930
2b44df907cc2 Efficient text representation of XML trees.
wenzelm
parents:
diff changeset
   112
2b44df907cc2 Efficient text representation of XML trees.
wenzelm
parents:
diff changeset
   113
2b44df907cc2 Efficient text representation of XML trees.
wenzelm
parents:
diff changeset
   114
    /* parse chunks */
2b44df907cc2 Efficient text representation of XML trees.
wenzelm
parents:
diff changeset
   115
56600
628e039cc34d more specific support for sequence of words;
wenzelm
parents: 55554
diff changeset
   116
    for (chunk <- Library.separated_chunks(is_X, source) if chunk.length != 0) {
34099
2541de190d92 added decode_chars, with raw character view on byte buffer and adhoc decoding via toString;
wenzelm
parents: 32450
diff changeset
   117
      if (chunk.length == 1 && chunk.charAt(0) == Y) pop()
27945
d2dc5a1903e8 tuned parse performance: avoid splitting terminal Y chunk;
wenzelm
parents: 27944
diff changeset
   118
      else {
56600
628e039cc34d more specific support for sequence of words;
wenzelm
parents: 55554
diff changeset
   119
        Library.separated_chunks(is_Y, chunk).toList match {
34099
2541de190d92 added decode_chars, with raw character view on byte buffer and adhoc decoding via toString;
wenzelm
parents: 32450
diff changeset
   120
          case ch :: name :: atts if ch.length == 0 =>
38234
4b610fbb2d83 YXML.parse: refrain from interning, let XML.Cache do it (partially);
wenzelm
parents: 38231
diff changeset
   121
            push(name.toString, atts.map(parse_attrib))
73030
72a8fdfa185d support more direct hash-consing via XML.Cache;
wenzelm
parents: 73029
diff changeset
   122
          case txts => for (txt <- txts) add(cache.tree0(XML.Text(cache.string(txt.toString))))
27945
d2dc5a1903e8 tuned parse performance: avoid splitting terminal Y chunk;
wenzelm
parents: 27944
diff changeset
   123
        }
27930
2b44df907cc2 Efficient text representation of XML trees.
wenzelm
parents:
diff changeset
   124
      }
2b44df907cc2 Efficient text representation of XML trees.
wenzelm
parents:
diff changeset
   125
    }
42719
wenzelm
parents: 40453
diff changeset
   126
    (stack: @unchecked) match {
38475
wenzelm
parents: 38268
diff changeset
   127
      case List((Markup.Empty, body)) => body.toList
38230
ed147003de4b simplified type XML.Tree: embed Markup directly, avoid slightly odd triple;
wenzelm
parents: 36685
diff changeset
   128
      case (Markup(name, _), _) :: _ => err_unbalanced(name)
27930
2b44df907cc2 Efficient text representation of XML trees.
wenzelm
parents:
diff changeset
   129
    }
2b44df907cc2 Efficient text representation of XML trees.
wenzelm
parents:
diff changeset
   130
  }
2b44df907cc2 Efficient text representation of XML trees.
wenzelm
parents:
diff changeset
   131
73030
72a8fdfa185d support more direct hash-consing via XML.Cache;
wenzelm
parents: 73029
diff changeset
   132
  def parse(source: CharSequence, cache: XML.Cache = XML.Cache.none): XML.Tree =
72a8fdfa185d support more direct hash-consing via XML.Cache;
wenzelm
parents: 73029
diff changeset
   133
    parse_body(source, cache = cache) match {
27930
2b44df907cc2 Efficient text representation of XML trees.
wenzelm
parents:
diff changeset
   134
      case List(result) => result
73028
95e0f014cd24 tuned signature;
wenzelm
parents: 71601
diff changeset
   135
      case Nil => XML.no_text
67820
e30d6368c7c8 clarified argument formats: explicit Unit, allow XML.Elem as well;
wenzelm
parents: 64370
diff changeset
   136
      case _ => err("multiple XML trees")
e30d6368c7c8 clarified argument formats: explicit Unit, allow XML.Elem as well;
wenzelm
parents: 64370
diff changeset
   137
    }
e30d6368c7c8 clarified argument formats: explicit Unit, allow XML.Elem as well;
wenzelm
parents: 64370
diff changeset
   138
73030
72a8fdfa185d support more direct hash-consing via XML.Cache;
wenzelm
parents: 73029
diff changeset
   139
  def parse_elem(source: CharSequence, cache: XML.Cache = XML.Cache.none): XML.Tree =
72a8fdfa185d support more direct hash-consing via XML.Cache;
wenzelm
parents: 73029
diff changeset
   140
    parse_body(source, cache = cache) match {
67820
e30d6368c7c8 clarified argument formats: explicit Unit, allow XML.Elem as well;
wenzelm
parents: 64370
diff changeset
   141
      case List(elem: XML.Elem) => elem
e30d6368c7c8 clarified argument formats: explicit Unit, allow XML.Elem as well;
wenzelm
parents: 64370
diff changeset
   142
      case _ => err("single XML element expected")
27930
2b44df907cc2 Efficient text representation of XML trees.
wenzelm
parents:
diff changeset
   143
    }
27960
65b10d8ef0c6 added parse_failsafe;
wenzelm
parents: 27946
diff changeset
   144
29521
736bf7117153 added parse_body_failsafe;
wenzelm
parents: 29180
diff changeset
   145
736bf7117153 added parse_body_failsafe;
wenzelm
parents: 29180
diff changeset
   146
  /* failsafe parsing */
736bf7117153 added parse_body_failsafe;
wenzelm
parents: 29180
diff changeset
   147
45666
d83797ef0d2d separate module for concrete Isabelle markup;
wenzelm
parents: 44698
diff changeset
   148
  private def markup_broken(source: CharSequence) =
55551
4a5f65df29fa tuned signature;
wenzelm
parents: 48996
diff changeset
   149
    XML.Elem(Markup.Broken, List(XML.Text(source.toString)))
29521
736bf7117153 added parse_body_failsafe;
wenzelm
parents: 29180
diff changeset
   150
75393
87ebf5a50283 clarified formatting, for the sake of scala3;
wenzelm
parents: 73712
diff changeset
   151
  def parse_body_failsafe(source: CharSequence, cache: XML.Cache = XML.Cache.none): XML.Body = {
73030
72a8fdfa185d support more direct hash-consing via XML.Cache;
wenzelm
parents: 73029
diff changeset
   152
    try { parse_body(source, cache = cache) }
45666
d83797ef0d2d separate module for concrete Isabelle markup;
wenzelm
parents: 44698
diff changeset
   153
    catch { case ERROR(_) => List(markup_broken(source)) }
29521
736bf7117153 added parse_body_failsafe;
wenzelm
parents: 29180
diff changeset
   154
  }
736bf7117153 added parse_body_failsafe;
wenzelm
parents: 29180
diff changeset
   155
75393
87ebf5a50283 clarified formatting, for the sake of scala3;
wenzelm
parents: 73712
diff changeset
   156
  def parse_failsafe(source: CharSequence, cache: XML.Cache = XML.Cache.none): XML.Tree = {
73030
72a8fdfa185d support more direct hash-consing via XML.Cache;
wenzelm
parents: 73029
diff changeset
   157
    try { parse(source, cache = cache) }
45666
d83797ef0d2d separate module for concrete Isabelle markup;
wenzelm
parents: 44698
diff changeset
   158
    catch { case ERROR(_) => markup_broken(source) }
27960
65b10d8ef0c6 added parse_failsafe;
wenzelm
parents: 27946
diff changeset
   159
  }
27930
2b44df907cc2 Efficient text representation of XML trees.
wenzelm
parents:
diff changeset
   160
}