src/Pure/Isar/outer_syntax.scala
author wenzelm
Tue, 02 Aug 2016 18:45:34 +0200
changeset 63584 68751fe1c036
parent 63579 73939a9b70a3
child 63587 881e8e2cfec2
permissions -rw-r--r--
tuned signature -- prover-independence is presently theoretical;
Ignore whitespace changes - Everywhere: Within whitespace: At end of lines:
34166
446a33b874b3 renamed class Outer_Keyword to Outer_Syntax;
wenzelm
parents:
diff changeset
     1
/*  Title:      Pure/Isar/outer_syntax.scala
446a33b874b3 renamed class Outer_Keyword to Outer_Syntax;
wenzelm
parents:
diff changeset
     2
    Author:     Makarius
446a33b874b3 renamed class Outer_Keyword to Outer_Syntax;
wenzelm
parents:
diff changeset
     3
446a33b874b3 renamed class Outer_Keyword to Outer_Syntax;
wenzelm
parents:
diff changeset
     4
Isabelle/Isar outer syntax.
446a33b874b3 renamed class Outer_Keyword to Outer_Syntax;
wenzelm
parents:
diff changeset
     5
*/
446a33b874b3 renamed class Outer_Keyword to Outer_Syntax;
wenzelm
parents:
diff changeset
     6
446a33b874b3 renamed class Outer_Keyword to Outer_Syntax;
wenzelm
parents:
diff changeset
     7
package isabelle
446a33b874b3 renamed class Outer_Keyword to Outer_Syntax;
wenzelm
parents:
diff changeset
     8
446a33b874b3 renamed class Outer_Keyword to Outer_Syntax;
wenzelm
parents:
diff changeset
     9
43411
0206466ee473 some support for partial scans with explicit context;
wenzelm
parents: 40533
diff changeset
    10
import scala.collection.mutable
58706
70a947611792 tuned signature and modules;
wenzelm
parents: 58703
diff changeset
    11
import scala.annotation.tailrec
34166
446a33b874b3 renamed class Outer_Keyword to Outer_Syntax;
wenzelm
parents:
diff changeset
    12
446a33b874b3 renamed class Outer_Keyword to Outer_Syntax;
wenzelm
parents:
diff changeset
    13
43774
6dfdb70496fe added Outer_Syntax.quote_string, which is conceptually a bit different from Token.unparse;
wenzelm
parents: 43695
diff changeset
    14
object Outer_Syntax
6dfdb70496fe added Outer_Syntax.quote_string, which is conceptually a bit different from Token.unparse;
wenzelm
parents: 43695
diff changeset
    15
{
58706
70a947611792 tuned signature and modules;
wenzelm
parents: 58703
diff changeset
    16
  /* syntax */
70a947611792 tuned signature and modules;
wenzelm
parents: 58703
diff changeset
    17
70a947611792 tuned signature and modules;
wenzelm
parents: 58703
diff changeset
    18
  val empty: Outer_Syntax = new Outer_Syntax()
70a947611792 tuned signature and modules;
wenzelm
parents: 58703
diff changeset
    19
70a947611792 tuned signature and modules;
wenzelm
parents: 58703
diff changeset
    20
  def init(): Outer_Syntax = new Outer_Syntax(completion = Completion.init())
70a947611792 tuned signature and modules;
wenzelm
parents: 58703
diff changeset
    21
70a947611792 tuned signature and modules;
wenzelm
parents: 58703
diff changeset
    22
70a947611792 tuned signature and modules;
wenzelm
parents: 58703
diff changeset
    23
  /* string literals */
70a947611792 tuned signature and modules;
wenzelm
parents: 58703
diff changeset
    24
43774
6dfdb70496fe added Outer_Syntax.quote_string, which is conceptually a bit different from Token.unparse;
wenzelm
parents: 43695
diff changeset
    25
  def quote_string(str: String): String =
6dfdb70496fe added Outer_Syntax.quote_string, which is conceptually a bit different from Token.unparse;
wenzelm
parents: 43695
diff changeset
    26
  {
6dfdb70496fe added Outer_Syntax.quote_string, which is conceptually a bit different from Token.unparse;
wenzelm
parents: 43695
diff changeset
    27
    val result = new StringBuilder(str.length + 10)
6dfdb70496fe added Outer_Syntax.quote_string, which is conceptually a bit different from Token.unparse;
wenzelm
parents: 43695
diff changeset
    28
    result += '"'
6dfdb70496fe added Outer_Syntax.quote_string, which is conceptually a bit different from Token.unparse;
wenzelm
parents: 43695
diff changeset
    29
    for (s <- Symbol.iterator(str)) {
6dfdb70496fe added Outer_Syntax.quote_string, which is conceptually a bit different from Token.unparse;
wenzelm
parents: 43695
diff changeset
    30
      if (s.length == 1) {
6dfdb70496fe added Outer_Syntax.quote_string, which is conceptually a bit different from Token.unparse;
wenzelm
parents: 43695
diff changeset
    31
        val c = s(0)
6dfdb70496fe added Outer_Syntax.quote_string, which is conceptually a bit different from Token.unparse;
wenzelm
parents: 43695
diff changeset
    32
        if (c < 32 && c != YXML.X && c != YXML.Y || c == '\\' || c == '"') {
6dfdb70496fe added Outer_Syntax.quote_string, which is conceptually a bit different from Token.unparse;
wenzelm
parents: 43695
diff changeset
    33
          result += '\\'
6dfdb70496fe added Outer_Syntax.quote_string, which is conceptually a bit different from Token.unparse;
wenzelm
parents: 43695
diff changeset
    34
          if (c < 10) result += '0'
6dfdb70496fe added Outer_Syntax.quote_string, which is conceptually a bit different from Token.unparse;
wenzelm
parents: 43695
diff changeset
    35
          if (c < 100) result += '0'
60215
5fb4990dfc73 misc tuning, based on warnings by IntelliJ IDEA;
wenzelm
parents: 59939
diff changeset
    36
          result ++= c.asInstanceOf[Int].toString
43774
6dfdb70496fe added Outer_Syntax.quote_string, which is conceptually a bit different from Token.unparse;
wenzelm
parents: 43695
diff changeset
    37
        }
6dfdb70496fe added Outer_Syntax.quote_string, which is conceptually a bit different from Token.unparse;
wenzelm
parents: 43695
diff changeset
    38
        else result += c
6dfdb70496fe added Outer_Syntax.quote_string, which is conceptually a bit different from Token.unparse;
wenzelm
parents: 43695
diff changeset
    39
      }
6dfdb70496fe added Outer_Syntax.quote_string, which is conceptually a bit different from Token.unparse;
wenzelm
parents: 43695
diff changeset
    40
      else result ++= s
6dfdb70496fe added Outer_Syntax.quote_string, which is conceptually a bit different from Token.unparse;
wenzelm
parents: 43695
diff changeset
    41
    }
6dfdb70496fe added Outer_Syntax.quote_string, which is conceptually a bit different from Token.unparse;
wenzelm
parents: 43695
diff changeset
    42
    result += '"'
6dfdb70496fe added Outer_Syntax.quote_string, which is conceptually a bit different from Token.unparse;
wenzelm
parents: 43695
diff changeset
    43
    result.toString
6dfdb70496fe added Outer_Syntax.quote_string, which is conceptually a bit different from Token.unparse;
wenzelm
parents: 43695
diff changeset
    44
  }
46626
a02115865bcc streamlined abstract datatype;
wenzelm
parents: 46624
diff changeset
    45
58696
6b7445774ce3 more explicit Line_Nesting;
wenzelm
parents: 58695
diff changeset
    46
58697
5bc1d6c4a499 tuned signature;
wenzelm
parents: 58696
diff changeset
    47
  /* line-oriented structure */
58696
6b7445774ce3 more explicit Line_Nesting;
wenzelm
parents: 58695
diff changeset
    48
58697
5bc1d6c4a499 tuned signature;
wenzelm
parents: 58696
diff changeset
    49
  object Line_Structure
58696
6b7445774ce3 more explicit Line_Nesting;
wenzelm
parents: 58695
diff changeset
    50
  {
58700
4717d18cc619 clarified Line_Structure wrt. command span;
wenzelm
parents: 58697
diff changeset
    51
    val init = Line_Structure()
58696
6b7445774ce3 more explicit Line_Nesting;
wenzelm
parents: 58695
diff changeset
    52
  }
6b7445774ce3 more explicit Line_Nesting;
wenzelm
parents: 58695
diff changeset
    53
58700
4717d18cc619 clarified Line_Structure wrt. command span;
wenzelm
parents: 58697
diff changeset
    54
  sealed case class Line_Structure(
4717d18cc619 clarified Line_Structure wrt. command span;
wenzelm
parents: 58697
diff changeset
    55
    improper: Boolean = true,
4717d18cc619 clarified Line_Structure wrt. command span;
wenzelm
parents: 58697
diff changeset
    56
    command: Boolean = false,
4717d18cc619 clarified Line_Structure wrt. command span;
wenzelm
parents: 58697
diff changeset
    57
    depth: Int = 0,
4717d18cc619 clarified Line_Structure wrt. command span;
wenzelm
parents: 58697
diff changeset
    58
    span_depth: Int = 0,
4717d18cc619 clarified Line_Structure wrt. command span;
wenzelm
parents: 58697
diff changeset
    59
    after_span_depth: Int = 0)
58706
70a947611792 tuned signature and modules;
wenzelm
parents: 58703
diff changeset
    60
70a947611792 tuned signature and modules;
wenzelm
parents: 58703
diff changeset
    61
70a947611792 tuned signature and modules;
wenzelm
parents: 58703
diff changeset
    62
  /* overall document structure */
70a947611792 tuned signature and modules;
wenzelm
parents: 58703
diff changeset
    63
70a947611792 tuned signature and modules;
wenzelm
parents: 58703
diff changeset
    64
  sealed abstract class Document { def length: Int }
58747
c680f181b32e tuned rendering;
wenzelm
parents: 58743
diff changeset
    65
  case class Document_Block(name: String, text: String, body: List[Document]) extends Document
58706
70a947611792 tuned signature and modules;
wenzelm
parents: 58703
diff changeset
    66
  {
70a947611792 tuned signature and modules;
wenzelm
parents: 58703
diff changeset
    67
    val length: Int = (0 /: body)(_ + _.length)
70a947611792 tuned signature and modules;
wenzelm
parents: 58703
diff changeset
    68
  }
58747
c680f181b32e tuned rendering;
wenzelm
parents: 58743
diff changeset
    69
  case class Document_Atom(command: Command) extends Document
58706
70a947611792 tuned signature and modules;
wenzelm
parents: 58703
diff changeset
    70
  {
70a947611792 tuned signature and modules;
wenzelm
parents: 58703
diff changeset
    71
    def length: Int = command.length
70a947611792 tuned signature and modules;
wenzelm
parents: 58703
diff changeset
    72
  }
43774
6dfdb70496fe added Outer_Syntax.quote_string, which is conceptually a bit different from Token.unparse;
wenzelm
parents: 43695
diff changeset
    73
}
6dfdb70496fe added Outer_Syntax.quote_string, which is conceptually a bit different from Token.unparse;
wenzelm
parents: 43695
diff changeset
    74
46712
8650d9a95736 prefer final ADTs -- prevent ooddities;
wenzelm
parents: 46626
diff changeset
    75
final class Outer_Syntax private(
58900
1435cc20b022 explicit type Keyword.Keywords;
wenzelm
parents: 58899
diff changeset
    76
  val keywords: Keyword.Keywords = Keyword.Keywords.empty,
53280
c63a016805b9 explicit indication of outer syntax with no tokens;
wenzelm
parents: 52439
diff changeset
    77
  val completion: Completion = Completion.empty,
55749
75a48dc4383e tuned signature;
wenzelm
parents: 55666
diff changeset
    78
  val language_context: Completion.Language_Context = Completion.Language_Context.outer,
63584
68751fe1c036 tuned signature -- prover-independence is presently theoretical;
wenzelm
parents: 63579
diff changeset
    79
  val has_tokens: Boolean = true)
34166
446a33b874b3 renamed class Outer_Keyword to Outer_Syntax;
wenzelm
parents:
diff changeset
    80
{
58706
70a947611792 tuned signature and modules;
wenzelm
parents: 58703
diff changeset
    81
  /** syntax content **/
70a947611792 tuned signature and modules;
wenzelm
parents: 58703
diff changeset
    82
58900
1435cc20b022 explicit type Keyword.Keywords;
wenzelm
parents: 58899
diff changeset
    83
  override def toString: String = keywords.toString
56393
22f533e6a049 more abstract Prover.Syntax, as proposed by Carst Tankink;
wenzelm
parents: 56314
diff changeset
    84
58695
91839729224e tuned comments;
wenzelm
parents: 58694
diff changeset
    85
91839729224e tuned comments;
wenzelm
parents: 58694
diff changeset
    86
  /* add keywords */
91839729224e tuned comments;
wenzelm
parents: 58694
diff changeset
    87
63579
73939a9b70a3 support 'abbrevs' within theory header;
wenzelm
parents: 63528
diff changeset
    88
  def + (name: String, kind: String = "", tags: List[String] = Nil): Outer_Syntax =
53280
c63a016805b9 explicit indication of outer syntax with no tokens;
wenzelm
parents: 52439
diff changeset
    89
  {
63429
baedd4724f08 tuned signature: more uniform Keyword.spec;
wenzelm
parents: 63424
diff changeset
    90
    val keywords1 = keywords + (name, kind, tags)
53280
c63a016805b9 explicit indication of outer syntax with no tokens;
wenzelm
parents: 52439
diff changeset
    91
    val completion1 =
63579
73939a9b70a3 support 'abbrevs' within theory header;
wenzelm
parents: 63528
diff changeset
    92
      completion.add_keyword(name).add_abbrevs(
73939a9b70a3 support 'abbrevs' within theory header;
wenzelm
parents: 63528
diff changeset
    93
        if (Keyword.theory_block.contains(kind))
73939a9b70a3 support 'abbrevs' within theory header;
wenzelm
parents: 63528
diff changeset
    94
          List((name, name + "\nbegin\n\u0007\nend"), (name, name))
73939a9b70a3 support 'abbrevs' within theory header;
wenzelm
parents: 63528
diff changeset
    95
        else List((name, name)))
58900
1435cc20b022 explicit type Keyword.Keywords;
wenzelm
parents: 58899
diff changeset
    96
    new Outer_Syntax(keywords1, completion1, language_context, true)
53280
c63a016805b9 explicit indication of outer syntax with no tokens;
wenzelm
parents: 52439
diff changeset
    97
  }
48706
e2b512024eab tuned signature;
wenzelm
parents: 48671
diff changeset
    98
48873
18b17f15bc62 more direct cumulation of (sparse) keywords;
wenzelm
parents: 48872
diff changeset
    99
  def add_keywords(keywords: Thy_Header.Keywords): Outer_Syntax =
18b17f15bc62 more direct cumulation of (sparse) keywords;
wenzelm
parents: 48872
diff changeset
   100
    (this /: keywords) {
63579
73939a9b70a3 support 'abbrevs' within theory header;
wenzelm
parents: 63528
diff changeset
   101
      case (syntax, (name, ((kind, tags), _))) =>
73939a9b70a3 support 'abbrevs' within theory header;
wenzelm
parents: 63528
diff changeset
   102
        syntax + (Symbol.decode(name), kind, tags) + (Symbol.encode(name), kind, tags)
73939a9b70a3 support 'abbrevs' within theory header;
wenzelm
parents: 63528
diff changeset
   103
    }
73939a9b70a3 support 'abbrevs' within theory header;
wenzelm
parents: 63528
diff changeset
   104
73939a9b70a3 support 'abbrevs' within theory header;
wenzelm
parents: 63528
diff changeset
   105
  def add_abbrevs(abbrevs: Thy_Header.Abbrevs): Outer_Syntax =
73939a9b70a3 support 'abbrevs' within theory header;
wenzelm
parents: 63528
diff changeset
   106
    if (abbrevs.isEmpty) this
73939a9b70a3 support 'abbrevs' within theory header;
wenzelm
parents: 63528
diff changeset
   107
    else {
73939a9b70a3 support 'abbrevs' within theory header;
wenzelm
parents: 63528
diff changeset
   108
      val completion1 =
73939a9b70a3 support 'abbrevs' within theory header;
wenzelm
parents: 63528
diff changeset
   109
        completion.add_abbrevs(
73939a9b70a3 support 'abbrevs' within theory header;
wenzelm
parents: 63528
diff changeset
   110
          (for ((a, b) <- abbrevs) yield {
73939a9b70a3 support 'abbrevs' within theory header;
wenzelm
parents: 63528
diff changeset
   111
            val a1 = Symbol.decode(a)
73939a9b70a3 support 'abbrevs' within theory header;
wenzelm
parents: 63528
diff changeset
   112
            val a2 = Symbol.encode(a)
73939a9b70a3 support 'abbrevs' within theory header;
wenzelm
parents: 63528
diff changeset
   113
            val b1 = Symbol.decode(b)
73939a9b70a3 support 'abbrevs' within theory header;
wenzelm
parents: 63528
diff changeset
   114
            List((a1, b1), (a2, b1))
73939a9b70a3 support 'abbrevs' within theory header;
wenzelm
parents: 63528
diff changeset
   115
          }).flatten)
73939a9b70a3 support 'abbrevs' within theory header;
wenzelm
parents: 63528
diff changeset
   116
      new Outer_Syntax(keywords, completion1, language_context, has_tokens)
46940
a40be2f10ca9 explicit Outer_Syntax.Decl;
wenzelm
parents: 46712
diff changeset
   117
    }
34166
446a33b874b3 renamed class Outer_Keyword to Outer_Syntax;
wenzelm
parents:
diff changeset
   118
58695
91839729224e tuned comments;
wenzelm
parents: 58694
diff changeset
   119
59073
dcecfcc56dce more merge operations;
wenzelm
parents: 58938
diff changeset
   120
  /* merge */
dcecfcc56dce more merge operations;
wenzelm
parents: 58938
diff changeset
   121
63584
68751fe1c036 tuned signature -- prover-independence is presently theoretical;
wenzelm
parents: 63579
diff changeset
   122
  def ++ (other: Outer_Syntax): Outer_Syntax =
59073
dcecfcc56dce more merge operations;
wenzelm
parents: 58938
diff changeset
   123
    if (this eq other) this
dcecfcc56dce more merge operations;
wenzelm
parents: 58938
diff changeset
   124
    else {
59077
7e0d3da6e6d8 node-specific syntax, with base_syntax as default;
wenzelm
parents: 59073
diff changeset
   125
      val keywords1 = keywords ++ other.asInstanceOf[Outer_Syntax].keywords
7e0d3da6e6d8 node-specific syntax, with base_syntax as default;
wenzelm
parents: 59073
diff changeset
   126
      val completion1 = completion ++ other.asInstanceOf[Outer_Syntax].completion
7e0d3da6e6d8 node-specific syntax, with base_syntax as default;
wenzelm
parents: 59073
diff changeset
   127
      if ((keywords eq keywords1) && (completion eq completion1)) this
7e0d3da6e6d8 node-specific syntax, with base_syntax as default;
wenzelm
parents: 59073
diff changeset
   128
      else new Outer_Syntax(keywords1, completion1, language_context, has_tokens)
59073
dcecfcc56dce more merge operations;
wenzelm
parents: 58938
diff changeset
   129
    }
dcecfcc56dce more merge operations;
wenzelm
parents: 58938
diff changeset
   130
dcecfcc56dce more merge operations;
wenzelm
parents: 58938
diff changeset
   131
59735
24bee1b11fce misc tuning and simplification;
wenzelm
parents: 59702
diff changeset
   132
  /* load commands */
58900
1435cc20b022 explicit type Keyword.Keywords;
wenzelm
parents: 58899
diff changeset
   133
63441
4c3fa4dba79f explicit kind "before_command";
wenzelm
parents: 63429
diff changeset
   134
  def load_command(name: String): Option[List[String]] = keywords.load_commands.get(name)
58900
1435cc20b022 explicit type Keyword.Keywords;
wenzelm
parents: 58899
diff changeset
   135
  def load_commands_in(text: String): Boolean = keywords.load_commands_in(text)
1435cc20b022 explicit type Keyword.Keywords;
wenzelm
parents: 58899
diff changeset
   136
1435cc20b022 explicit type Keyword.Keywords;
wenzelm
parents: 58899
diff changeset
   137
58706
70a947611792 tuned signature and modules;
wenzelm
parents: 58703
diff changeset
   138
  /* language context */
34166
446a33b874b3 renamed class Outer_Keyword to Outer_Syntax;
wenzelm
parents:
diff changeset
   139
58706
70a947611792 tuned signature and modules;
wenzelm
parents: 58703
diff changeset
   140
  def set_language_context(context: Completion.Language_Context): Outer_Syntax =
58900
1435cc20b022 explicit type Keyword.Keywords;
wenzelm
parents: 58899
diff changeset
   141
    new Outer_Syntax(keywords, completion, context, has_tokens)
58706
70a947611792 tuned signature and modules;
wenzelm
parents: 58703
diff changeset
   142
70a947611792 tuned signature and modules;
wenzelm
parents: 58703
diff changeset
   143
  def no_tokens: Outer_Syntax =
46969
481b7d9ad6fe more abstract heading level;
wenzelm
parents: 46941
diff changeset
   144
  {
58900
1435cc20b022 explicit type Keyword.Keywords;
wenzelm
parents: 58899
diff changeset
   145
    require(keywords.is_empty)
58706
70a947611792 tuned signature and modules;
wenzelm
parents: 58703
diff changeset
   146
    new Outer_Syntax(
70a947611792 tuned signature and modules;
wenzelm
parents: 58703
diff changeset
   147
      completion = completion,
70a947611792 tuned signature and modules;
wenzelm
parents: 58703
diff changeset
   148
      language_context = language_context,
70a947611792 tuned signature and modules;
wenzelm
parents: 58703
diff changeset
   149
      has_tokens = false)
46969
481b7d9ad6fe more abstract heading level;
wenzelm
parents: 46941
diff changeset
   150
  }
40454
2516ea25a54b some support for nested source structure, based on section headings;
wenzelm
parents: 38471
diff changeset
   151
58706
70a947611792 tuned signature and modules;
wenzelm
parents: 58703
diff changeset
   152
40454
2516ea25a54b some support for nested source structure, based on section headings;
wenzelm
parents: 38471
diff changeset
   153
58706
70a947611792 tuned signature and modules;
wenzelm
parents: 58703
diff changeset
   154
  /** parsing **/
34166
446a33b874b3 renamed class Outer_Keyword to Outer_Syntax;
wenzelm
parents:
diff changeset
   155
58697
5bc1d6c4a499 tuned signature;
wenzelm
parents: 58696
diff changeset
   156
  /* line-oriented structure */
58696
6b7445774ce3 more explicit Line_Nesting;
wenzelm
parents: 58695
diff changeset
   157
63460
f41070510341 clarified;
wenzelm
parents: 63459
diff changeset
   158
  private val close_structure =
f41070510341 clarified;
wenzelm
parents: 63459
diff changeset
   159
    Set(Keyword.NEXT_BLOCK, Keyword.QED_BLOCK, Keyword.PRF_CLOSE)
f41070510341 clarified;
wenzelm
parents: 63459
diff changeset
   160
59924
801b979ec0c2 more general notion of command span: command keyword not necessarily at start;
wenzelm
parents: 59735
diff changeset
   161
  def line_structure(tokens: List[Token], structure: Outer_Syntax.Line_Structure)
58700
4717d18cc619 clarified Line_Structure wrt. command span;
wenzelm
parents: 58697
diff changeset
   162
    : Outer_Syntax.Line_Structure =
58696
6b7445774ce3 more explicit Line_Nesting;
wenzelm
parents: 58695
diff changeset
   163
  {
58700
4717d18cc619 clarified Line_Structure wrt. command span;
wenzelm
parents: 58697
diff changeset
   164
    val improper1 = tokens.forall(_.is_improper)
4717d18cc619 clarified Line_Structure wrt. command span;
wenzelm
parents: 58697
diff changeset
   165
    val command1 = tokens.exists(_.is_command)
4717d18cc619 clarified Line_Structure wrt. command span;
wenzelm
parents: 58697
diff changeset
   166
63458
723f9c673c1c closing 'qed' or '}' is outside of fold;
wenzelm
parents: 63441
diff changeset
   167
    val command_depth =
723f9c673c1c closing 'qed' or '}' is outside of fold;
wenzelm
parents: 63441
diff changeset
   168
      tokens.iterator.filter(_.is_proper).toStream.headOption match {
723f9c673c1c closing 'qed' or '}' is outside of fold;
wenzelm
parents: 63441
diff changeset
   169
        case Some(tok) =>
63460
f41070510341 clarified;
wenzelm
parents: 63459
diff changeset
   170
          if (keywords.is_command(tok, close_structure))
63458
723f9c673c1c closing 'qed' or '}' is outside of fold;
wenzelm
parents: 63441
diff changeset
   171
            Some(structure.after_span_depth - 1)
723f9c673c1c closing 'qed' or '}' is outside of fold;
wenzelm
parents: 63441
diff changeset
   172
          else None
723f9c673c1c closing 'qed' or '}' is outside of fold;
wenzelm
parents: 63441
diff changeset
   173
        case None => None
723f9c673c1c closing 'qed' or '}' is outside of fold;
wenzelm
parents: 63441
diff changeset
   174
      }
723f9c673c1c closing 'qed' or '}' is outside of fold;
wenzelm
parents: 63441
diff changeset
   175
58696
6b7445774ce3 more explicit Line_Nesting;
wenzelm
parents: 58695
diff changeset
   176
    val depth1 =
63459
8d68204d97d7 clarified;
wenzelm
parents: 63458
diff changeset
   177
      if (tokens.exists(tok =>
8d68204d97d7 clarified;
wenzelm
parents: 63458
diff changeset
   178
            keywords.is_before_command(tok) || keywords.is_command(tok, Keyword.theory))) 0
63458
723f9c673c1c closing 'qed' or '}' is outside of fold;
wenzelm
parents: 63441
diff changeset
   179
      else if (command_depth.isDefined) command_depth.get
59924
801b979ec0c2 more general notion of command span: command keyword not necessarily at start;
wenzelm
parents: 59735
diff changeset
   180
      else if (command1) structure.after_span_depth
801b979ec0c2 more general notion of command span: command keyword not necessarily at start;
wenzelm
parents: 59735
diff changeset
   181
      else structure.span_depth
58700
4717d18cc619 clarified Line_Structure wrt. command span;
wenzelm
parents: 58697
diff changeset
   182
4717d18cc619 clarified Line_Structure wrt. command span;
wenzelm
parents: 58697
diff changeset
   183
    val (span_depth1, after_span_depth1) =
59924
801b979ec0c2 more general notion of command span: command keyword not necessarily at start;
wenzelm
parents: 59735
diff changeset
   184
      ((structure.span_depth, structure.after_span_depth) /: tokens) {
58703
883efcc7a50d more folds;
wenzelm
parents: 58700
diff changeset
   185
        case ((x, y), tok) =>
883efcc7a50d more folds;
wenzelm
parents: 58700
diff changeset
   186
          if (tok.is_command) {
63424
e4e15bbfb3e2 clarified signature;
wenzelm
parents: 62453
diff changeset
   187
            if (keywords.is_command(tok, Keyword.theory_goal)) (2, 1)
e4e15bbfb3e2 clarified signature;
wenzelm
parents: 62453
diff changeset
   188
            else if (keywords.is_command(tok, Keyword.theory)) (1, 0)
e4e15bbfb3e2 clarified signature;
wenzelm
parents: 62453
diff changeset
   189
            else if (keywords.is_command(tok, Keyword.proof_open)) (y + 2, y + 1)
63479
wenzelm
parents: 63460
diff changeset
   190
            else if (keywords.is_command(tok, Set(Keyword.PRF_BLOCK))) (y + 2, y + 1)
wenzelm
parents: 63460
diff changeset
   191
            else if (keywords.is_command(tok, Set(Keyword.QED_BLOCK))) (y - 1, y - 2)
wenzelm
parents: 63460
diff changeset
   192
            else if (keywords.is_command(tok, Set(Keyword.PRF_CLOSE))) (y, y - 1)
63424
e4e15bbfb3e2 clarified signature;
wenzelm
parents: 62453
diff changeset
   193
            else if (keywords.is_command(tok, Keyword.proof_close)) (y + 1, y - 1)
e4e15bbfb3e2 clarified signature;
wenzelm
parents: 62453
diff changeset
   194
            else if (keywords.is_command(tok, Keyword.qed_global)) (1, 0)
58703
883efcc7a50d more folds;
wenzelm
parents: 58700
diff changeset
   195
            else (x, y)
883efcc7a50d more folds;
wenzelm
parents: 58700
diff changeset
   196
          }
883efcc7a50d more folds;
wenzelm
parents: 58700
diff changeset
   197
          else (x, y)
58696
6b7445774ce3 more explicit Line_Nesting;
wenzelm
parents: 58695
diff changeset
   198
      }
58700
4717d18cc619 clarified Line_Structure wrt. command span;
wenzelm
parents: 58697
diff changeset
   199
4717d18cc619 clarified Line_Structure wrt. command span;
wenzelm
parents: 58697
diff changeset
   200
    Outer_Syntax.Line_Structure(improper1, command1, depth1, span_depth1, after_span_depth1)
58696
6b7445774ce3 more explicit Line_Nesting;
wenzelm
parents: 58695
diff changeset
   201
  }
6b7445774ce3 more explicit Line_Nesting;
wenzelm
parents: 58695
diff changeset
   202
6b7445774ce3 more explicit Line_Nesting;
wenzelm
parents: 58695
diff changeset
   203
58706
70a947611792 tuned signature and modules;
wenzelm
parents: 58703
diff changeset
   204
  /* command spans */
57905
c0c5652e796e separate module Command_Span: mostly syntactic representation;
wenzelm
parents: 57901
diff changeset
   205
c0c5652e796e separate module Command_Span: mostly syntactic representation;
wenzelm
parents: 57901
diff changeset
   206
  def parse_spans(toks: List[Token]): List[Command_Span.Span] =
c0c5652e796e separate module Command_Span: mostly syntactic representation;
wenzelm
parents: 57901
diff changeset
   207
  {
c0c5652e796e separate module Command_Span: mostly syntactic representation;
wenzelm
parents: 57901
diff changeset
   208
    val result = new mutable.ListBuffer[Command_Span.Span]
c0c5652e796e separate module Command_Span: mostly syntactic representation;
wenzelm
parents: 57901
diff changeset
   209
    val content = new mutable.ListBuffer[Token]
c0c5652e796e separate module Command_Span: mostly syntactic representation;
wenzelm
parents: 57901
diff changeset
   210
    val improper = new mutable.ListBuffer[Token]
c0c5652e796e separate module Command_Span: mostly syntactic representation;
wenzelm
parents: 57901
diff changeset
   211
c0c5652e796e separate module Command_Span: mostly syntactic representation;
wenzelm
parents: 57901
diff changeset
   212
    def ship(span: List[Token])
c0c5652e796e separate module Command_Span: mostly syntactic representation;
wenzelm
parents: 57901
diff changeset
   213
    {
c0c5652e796e separate module Command_Span: mostly syntactic representation;
wenzelm
parents: 57901
diff changeset
   214
      val kind =
59924
801b979ec0c2 more general notion of command span: command keyword not necessarily at start;
wenzelm
parents: 59735
diff changeset
   215
        if (span.forall(_.is_improper)) Command_Span.Ignored_Span
801b979ec0c2 more general notion of command span: command keyword not necessarily at start;
wenzelm
parents: 59735
diff changeset
   216
        else if (span.exists(_.is_error)) Command_Span.Malformed_Span
801b979ec0c2 more general notion of command span: command keyword not necessarily at start;
wenzelm
parents: 59735
diff changeset
   217
        else
801b979ec0c2 more general notion of command span: command keyword not necessarily at start;
wenzelm
parents: 59735
diff changeset
   218
          span.find(_.is_command) match {
801b979ec0c2 more general notion of command span: command keyword not necessarily at start;
wenzelm
parents: 59735
diff changeset
   219
            case None => Command_Span.Malformed_Span
801b979ec0c2 more general notion of command span: command keyword not necessarily at start;
wenzelm
parents: 59735
diff changeset
   220
            case Some(cmd) =>
801b979ec0c2 more general notion of command span: command keyword not necessarily at start;
wenzelm
parents: 59735
diff changeset
   221
              val name = cmd.source
801b979ec0c2 more general notion of command span: command keyword not necessarily at start;
wenzelm
parents: 59735
diff changeset
   222
              val offset =
801b979ec0c2 more general notion of command span: command keyword not necessarily at start;
wenzelm
parents: 59735
diff changeset
   223
                (0 /: span.takeWhile(_ != cmd)) {
801b979ec0c2 more general notion of command span: command keyword not necessarily at start;
wenzelm
parents: 59735
diff changeset
   224
                  case (i, tok) => i + Symbol.iterator(tok.source).length }
801b979ec0c2 more general notion of command span: command keyword not necessarily at start;
wenzelm
parents: 59735
diff changeset
   225
              val end_offset = offset + Symbol.iterator(name).length
801b979ec0c2 more general notion of command span: command keyword not necessarily at start;
wenzelm
parents: 59735
diff changeset
   226
              val pos = Position.Range(Text.Range(offset, end_offset) + 1)
801b979ec0c2 more general notion of command span: command keyword not necessarily at start;
wenzelm
parents: 59735
diff changeset
   227
              Command_Span.Command_Span(name, pos)
801b979ec0c2 more general notion of command span: command keyword not necessarily at start;
wenzelm
parents: 59735
diff changeset
   228
          }
57905
c0c5652e796e separate module Command_Span: mostly syntactic representation;
wenzelm
parents: 57901
diff changeset
   229
      result += Command_Span.Span(kind, span)
c0c5652e796e separate module Command_Span: mostly syntactic representation;
wenzelm
parents: 57901
diff changeset
   230
    }
c0c5652e796e separate module Command_Span: mostly syntactic representation;
wenzelm
parents: 57901
diff changeset
   231
c0c5652e796e separate module Command_Span: mostly syntactic representation;
wenzelm
parents: 57901
diff changeset
   232
    def flush()
c0c5652e796e separate module Command_Span: mostly syntactic representation;
wenzelm
parents: 57901
diff changeset
   233
    {
59319
wenzelm
parents: 59122
diff changeset
   234
      if (content.nonEmpty) { ship(content.toList); content.clear }
wenzelm
parents: 59122
diff changeset
   235
      if (improper.nonEmpty) { ship(improper.toList); improper.clear }
57905
c0c5652e796e separate module Command_Span: mostly syntactic representation;
wenzelm
parents: 57901
diff changeset
   236
    }
c0c5652e796e separate module Command_Span: mostly syntactic representation;
wenzelm
parents: 57901
diff changeset
   237
c0c5652e796e separate module Command_Span: mostly syntactic representation;
wenzelm
parents: 57901
diff changeset
   238
    for (tok <- toks) {
59924
801b979ec0c2 more general notion of command span: command keyword not necessarily at start;
wenzelm
parents: 59735
diff changeset
   239
      if (tok.is_improper) improper += tok
63441
4c3fa4dba79f explicit kind "before_command";
wenzelm
parents: 63429
diff changeset
   240
      else if (keywords.is_before_command(tok) ||
4c3fa4dba79f explicit kind "before_command";
wenzelm
parents: 63429
diff changeset
   241
        tok.is_command &&
4c3fa4dba79f explicit kind "before_command";
wenzelm
parents: 63429
diff changeset
   242
          (!content.exists(keywords.is_before_command(_)) || content.exists(_.is_command)))
59924
801b979ec0c2 more general notion of command span: command keyword not necessarily at start;
wenzelm
parents: 59735
diff changeset
   243
      { flush(); content += tok }
57905
c0c5652e796e separate module Command_Span: mostly syntactic representation;
wenzelm
parents: 57901
diff changeset
   244
      else { content ++= improper; improper.clear; content += tok }
c0c5652e796e separate module Command_Span: mostly syntactic representation;
wenzelm
parents: 57901
diff changeset
   245
    }
c0c5652e796e separate module Command_Span: mostly syntactic representation;
wenzelm
parents: 57901
diff changeset
   246
    flush()
c0c5652e796e separate module Command_Span: mostly syntactic representation;
wenzelm
parents: 57901
diff changeset
   247
c0c5652e796e separate module Command_Span: mostly syntactic representation;
wenzelm
parents: 57901
diff changeset
   248
    result.toList
c0c5652e796e separate module Command_Span: mostly syntactic representation;
wenzelm
parents: 57901
diff changeset
   249
  }
c0c5652e796e separate module Command_Span: mostly syntactic representation;
wenzelm
parents: 57901
diff changeset
   250
57906
020df63dd0a9 tuned signature;
wenzelm
parents: 57905
diff changeset
   251
  def parse_spans(input: CharSequence): List[Command_Span.Span] =
59083
88b0b1f28adc tuned signature;
wenzelm
parents: 59077
diff changeset
   252
    parse_spans(Token.explode(keywords, input))
57906
020df63dd0a9 tuned signature;
wenzelm
parents: 57905
diff changeset
   253
57905
c0c5652e796e separate module Command_Span: mostly syntactic representation;
wenzelm
parents: 57901
diff changeset
   254
58706
70a947611792 tuned signature and modules;
wenzelm
parents: 58703
diff changeset
   255
  /* overall document structure */
55616
25a7a998852a default completion context via outer syntax;
wenzelm
parents: 55510
diff changeset
   256
58706
70a947611792 tuned signature and modules;
wenzelm
parents: 58703
diff changeset
   257
  def heading_level(command: Command): Option[Int] =
70a947611792 tuned signature and modules;
wenzelm
parents: 58703
diff changeset
   258
  {
59735
24bee1b11fce misc tuning and simplification;
wenzelm
parents: 59702
diff changeset
   259
    val name = command.span.name
24bee1b11fce misc tuning and simplification;
wenzelm
parents: 59702
diff changeset
   260
    name match {
24bee1b11fce misc tuning and simplification;
wenzelm
parents: 59702
diff changeset
   261
      case Thy_Header.CHAPTER => Some(0)
62453
b93cc7d73431 discontinued old 'header';
wenzelm
parents: 62244
diff changeset
   262
      case Thy_Header.SECTION => Some(1)
59735
24bee1b11fce misc tuning and simplification;
wenzelm
parents: 59702
diff changeset
   263
      case Thy_Header.SUBSECTION => Some(2)
24bee1b11fce misc tuning and simplification;
wenzelm
parents: 59702
diff changeset
   264
      case Thy_Header.SUBSUBSECTION => Some(3)
61463
8e46cea6a45a added 'paragraph', 'subparagraph';
wenzelm
parents: 60694
diff changeset
   265
      case Thy_Header.PARAGRAPH => Some(4)
8e46cea6a45a added 'paragraph', 'subparagraph';
wenzelm
parents: 60694
diff changeset
   266
      case Thy_Header.SUBPARAGRAPH => Some(5)
58868
c5e1cce7ace3 uniform heading commands work in any context, even in theory header;
wenzelm
parents: 58853
diff changeset
   267
      case _ =>
63441
4c3fa4dba79f explicit kind "before_command";
wenzelm
parents: 63429
diff changeset
   268
        keywords.kinds.get(name) match {
61463
8e46cea6a45a added 'paragraph', 'subparagraph';
wenzelm
parents: 60694
diff changeset
   269
          case Some(kind) if Keyword.theory(kind) && !Keyword.theory_end(kind) => Some(6)
58868
c5e1cce7ace3 uniform heading commands work in any context, even in theory header;
wenzelm
parents: 58853
diff changeset
   270
          case _ => None
c5e1cce7ace3 uniform heading commands work in any context, even in theory header;
wenzelm
parents: 58853
diff changeset
   271
        }
58706
70a947611792 tuned signature and modules;
wenzelm
parents: 58703
diff changeset
   272
    }
70a947611792 tuned signature and modules;
wenzelm
parents: 58703
diff changeset
   273
  }
70a947611792 tuned signature and modules;
wenzelm
parents: 58703
diff changeset
   274
58743
c07a59140fee clarified tree root;
wenzelm
parents: 58706
diff changeset
   275
  def parse_document(node_name: Document.Node.Name, text: CharSequence):
c07a59140fee clarified tree root;
wenzelm
parents: 58706
diff changeset
   276
    List[Outer_Syntax.Document] =
58706
70a947611792 tuned signature and modules;
wenzelm
parents: 58703
diff changeset
   277
  {
70a947611792 tuned signature and modules;
wenzelm
parents: 58703
diff changeset
   278
    /* stack operations */
70a947611792 tuned signature and modules;
wenzelm
parents: 58703
diff changeset
   279
70a947611792 tuned signature and modules;
wenzelm
parents: 58703
diff changeset
   280
    def buffer(): mutable.ListBuffer[Outer_Syntax.Document] =
70a947611792 tuned signature and modules;
wenzelm
parents: 58703
diff changeset
   281
      new mutable.ListBuffer[Outer_Syntax.Document]
70a947611792 tuned signature and modules;
wenzelm
parents: 58703
diff changeset
   282
58747
c680f181b32e tuned rendering;
wenzelm
parents: 58743
diff changeset
   283
    var stack: List[(Int, Command, mutable.ListBuffer[Outer_Syntax.Document])] =
c680f181b32e tuned rendering;
wenzelm
parents: 58743
diff changeset
   284
      List((0, Command.empty, buffer()))
55616
25a7a998852a default completion context via outer syntax;
wenzelm
parents: 55510
diff changeset
   285
58706
70a947611792 tuned signature and modules;
wenzelm
parents: 58703
diff changeset
   286
    @tailrec def close(level: Int => Boolean)
70a947611792 tuned signature and modules;
wenzelm
parents: 58703
diff changeset
   287
    {
70a947611792 tuned signature and modules;
wenzelm
parents: 58703
diff changeset
   288
      stack match {
58747
c680f181b32e tuned rendering;
wenzelm
parents: 58743
diff changeset
   289
        case (lev, command, body) :: (_, _, body2) :: rest if level(lev) =>
59735
24bee1b11fce misc tuning and simplification;
wenzelm
parents: 59702
diff changeset
   290
          body2 += Outer_Syntax.Document_Block(command.span.name, command.source, body.toList)
58706
70a947611792 tuned signature and modules;
wenzelm
parents: 58703
diff changeset
   291
          stack = stack.tail
70a947611792 tuned signature and modules;
wenzelm
parents: 58703
diff changeset
   292
          close(level)
70a947611792 tuned signature and modules;
wenzelm
parents: 58703
diff changeset
   293
        case _ =>
70a947611792 tuned signature and modules;
wenzelm
parents: 58703
diff changeset
   294
      }
70a947611792 tuned signature and modules;
wenzelm
parents: 58703
diff changeset
   295
    }
70a947611792 tuned signature and modules;
wenzelm
parents: 58703
diff changeset
   296
58743
c07a59140fee clarified tree root;
wenzelm
parents: 58706
diff changeset
   297
    def result(): List[Outer_Syntax.Document] =
58706
70a947611792 tuned signature and modules;
wenzelm
parents: 58703
diff changeset
   298
    {
70a947611792 tuned signature and modules;
wenzelm
parents: 58703
diff changeset
   299
      close(_ => true)
58743
c07a59140fee clarified tree root;
wenzelm
parents: 58706
diff changeset
   300
      stack.head._3.toList
58706
70a947611792 tuned signature and modules;
wenzelm
parents: 58703
diff changeset
   301
    }
70a947611792 tuned signature and modules;
wenzelm
parents: 58703
diff changeset
   302
70a947611792 tuned signature and modules;
wenzelm
parents: 58703
diff changeset
   303
    def add(command: Command)
70a947611792 tuned signature and modules;
wenzelm
parents: 58703
diff changeset
   304
    {
70a947611792 tuned signature and modules;
wenzelm
parents: 58703
diff changeset
   305
      heading_level(command) match {
70a947611792 tuned signature and modules;
wenzelm
parents: 58703
diff changeset
   306
        case Some(i) =>
70a947611792 tuned signature and modules;
wenzelm
parents: 58703
diff changeset
   307
          close(_ > i)
58747
c680f181b32e tuned rendering;
wenzelm
parents: 58743
diff changeset
   308
          stack = (i + 1, command, buffer()) :: stack
58706
70a947611792 tuned signature and modules;
wenzelm
parents: 58703
diff changeset
   309
        case None =>
70a947611792 tuned signature and modules;
wenzelm
parents: 58703
diff changeset
   310
      }
70a947611792 tuned signature and modules;
wenzelm
parents: 58703
diff changeset
   311
      stack.head._3 += Outer_Syntax.Document_Atom(command)
70a947611792 tuned signature and modules;
wenzelm
parents: 58703
diff changeset
   312
    }
70a947611792 tuned signature and modules;
wenzelm
parents: 58703
diff changeset
   313
70a947611792 tuned signature and modules;
wenzelm
parents: 58703
diff changeset
   314
70a947611792 tuned signature and modules;
wenzelm
parents: 58703
diff changeset
   315
    /* result structure */
70a947611792 tuned signature and modules;
wenzelm
parents: 58703
diff changeset
   316
70a947611792 tuned signature and modules;
wenzelm
parents: 58703
diff changeset
   317
    val spans = parse_spans(text)
59702
58dfaa369c11 hybrid use of command blobs: inlined errors and auxiliary files;
wenzelm
parents: 59700
diff changeset
   318
    spans.foreach(span => add(Command(Document_ID.none, node_name, Command.no_blobs, span)))
58706
70a947611792 tuned signature and modules;
wenzelm
parents: 58703
diff changeset
   319
    result()
55616
25a7a998852a default completion context via outer syntax;
wenzelm
parents: 55510
diff changeset
   320
  }
34166
446a33b874b3 renamed class Outer_Keyword to Outer_Syntax;
wenzelm
parents:
diff changeset
   321
}