src/Pure/Isar/outer_syntax.scala
author wenzelm
Tue, 21 Aug 2012 12:15:25 +0200
changeset 48870 4accee106f0f
parent 48864 3ee314ae1e0a
child 48872 6124e0d1120a
permissions -rw-r--r--
clarified initialization of Thy_Load, Thy_Info, Session;
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
446a33b874b3 renamed class Outer_Keyword to Outer_Syntax;
wenzelm
parents:
diff changeset
    10
import scala.util.parsing.input.{Reader, CharSequenceReader}
43411
0206466ee473 some support for partial scans with explicit context;
wenzelm
parents: 40533
diff changeset
    11
import scala.collection.mutable
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
{
6dfdb70496fe added Outer_Syntax.quote_string, which is conceptually a bit different from Token.unparse;
wenzelm
parents: 43695
diff changeset
    16
  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
    17
  {
6dfdb70496fe added Outer_Syntax.quote_string, which is conceptually a bit different from Token.unparse;
wenzelm
parents: 43695
diff changeset
    18
    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
    19
    result += '"'
6dfdb70496fe added Outer_Syntax.quote_string, which is conceptually a bit different from Token.unparse;
wenzelm
parents: 43695
diff changeset
    20
    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
    21
      if (s.length == 1) {
6dfdb70496fe added Outer_Syntax.quote_string, which is conceptually a bit different from Token.unparse;
wenzelm
parents: 43695
diff changeset
    22
        val c = s(0)
6dfdb70496fe added Outer_Syntax.quote_string, which is conceptually a bit different from Token.unparse;
wenzelm
parents: 43695
diff changeset
    23
        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
    24
          result += '\\'
6dfdb70496fe added Outer_Syntax.quote_string, which is conceptually a bit different from Token.unparse;
wenzelm
parents: 43695
diff changeset
    25
          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
    26
          if (c < 100) result += '0'
6dfdb70496fe added Outer_Syntax.quote_string, which is conceptually a bit different from Token.unparse;
wenzelm
parents: 43695
diff changeset
    27
          result ++= (c.asInstanceOf[Int].toString)
6dfdb70496fe added Outer_Syntax.quote_string, which is conceptually a bit different from Token.unparse;
wenzelm
parents: 43695
diff changeset
    28
        }
6dfdb70496fe added Outer_Syntax.quote_string, which is conceptually a bit different from Token.unparse;
wenzelm
parents: 43695
diff changeset
    29
        else result += c
6dfdb70496fe added Outer_Syntax.quote_string, which is conceptually a bit different from Token.unparse;
wenzelm
parents: 43695
diff changeset
    30
      }
6dfdb70496fe added Outer_Syntax.quote_string, which is conceptually a bit different from Token.unparse;
wenzelm
parents: 43695
diff changeset
    31
      else result ++= s
6dfdb70496fe added Outer_Syntax.quote_string, which is conceptually a bit different from Token.unparse;
wenzelm
parents: 43695
diff changeset
    32
    }
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
    result.toString
6dfdb70496fe added Outer_Syntax.quote_string, which is conceptually a bit different from Token.unparse;
wenzelm
parents: 43695
diff changeset
    35
  }
46626
a02115865bcc streamlined abstract datatype;
wenzelm
parents: 46624
diff changeset
    36
46941
c0f776b661fa maintain Version.syntax within document state;
wenzelm
parents: 46940
diff changeset
    37
  val empty: Outer_Syntax = new Outer_Syntax()
48870
4accee106f0f clarified initialization of Thy_Load, Thy_Info, Session;
wenzelm
parents: 48864
diff changeset
    38
46941
c0f776b661fa maintain Version.syntax within document state;
wenzelm
parents: 46940
diff changeset
    39
  def init(): Outer_Syntax = new Outer_Syntax(completion = Completion.init())
48870
4accee106f0f clarified initialization of Thy_Load, Thy_Info, Session;
wenzelm
parents: 48864
diff changeset
    40
4accee106f0f clarified initialization of Thy_Load, Thy_Info, Session;
wenzelm
parents: 48864
diff changeset
    41
  def init_pure(): Outer_Syntax =
4accee106f0f clarified initialization of Thy_Load, Thy_Info, Session;
wenzelm
parents: 48864
diff changeset
    42
    init() + ("theory", Keyword.THY_BEGIN) + ("ML_file", Keyword.THY_LOAD)
43774
6dfdb70496fe added Outer_Syntax.quote_string, which is conceptually a bit different from Token.unparse;
wenzelm
parents: 43695
diff changeset
    43
}
6dfdb70496fe added Outer_Syntax.quote_string, which is conceptually a bit different from Token.unparse;
wenzelm
parents: 43695
diff changeset
    44
46712
8650d9a95736 prefer final ADTs -- prevent ooddities;
wenzelm
parents: 46626
diff changeset
    45
final class Outer_Syntax private(
48864
3ee314ae1e0a added keyword kind "thy_load" (with optional list of file extensions);
wenzelm
parents: 48708
diff changeset
    46
  keywords: Map[String, (String, List[String])] = Map.empty,
46626
a02115865bcc streamlined abstract datatype;
wenzelm
parents: 46624
diff changeset
    47
  lexicon: Scan.Lexicon = Scan.Lexicon.empty,
46941
c0f776b661fa maintain Version.syntax within document state;
wenzelm
parents: 46940
diff changeset
    48
  val completion: Completion = Completion.empty)
34166
446a33b874b3 renamed class Outer_Keyword to Outer_Syntax;
wenzelm
parents:
diff changeset
    49
{
48660
730ca503e955 static outer syntax based on session specifications;
wenzelm
parents: 47469
diff changeset
    50
  override def toString: String =
48864
3ee314ae1e0a added keyword kind "thy_load" (with optional list of file extensions);
wenzelm
parents: 48708
diff changeset
    51
    (for ((name, (kind, files)) <- keywords) yield {
48660
730ca503e955 static outer syntax based on session specifications;
wenzelm
parents: 47469
diff changeset
    52
      if (kind == Keyword.MINOR) quote(name)
48864
3ee314ae1e0a added keyword kind "thy_load" (with optional list of file extensions);
wenzelm
parents: 48708
diff changeset
    53
      else
3ee314ae1e0a added keyword kind "thy_load" (with optional list of file extensions);
wenzelm
parents: 48708
diff changeset
    54
        quote(name) + " :: " + quote(kind) +
3ee314ae1e0a added keyword kind "thy_load" (with optional list of file extensions);
wenzelm
parents: 48708
diff changeset
    55
        (if (files.isEmpty) "" else " (" + commas_quote(files) + ")")
48671
951bc4c3ee17 refined outer syntax;
wenzelm
parents: 48660
diff changeset
    56
    }).toList.sorted.mkString("keywords\n  ", " and\n  ", "")
48660
730ca503e955 static outer syntax based on session specifications;
wenzelm
parents: 47469
diff changeset
    57
48864
3ee314ae1e0a added keyword kind "thy_load" (with optional list of file extensions);
wenzelm
parents: 48708
diff changeset
    58
  def keyword_kind_files(name: String): Option[(String, List[String])] = keywords.get(name)
3ee314ae1e0a added keyword kind "thy_load" (with optional list of file extensions);
wenzelm
parents: 48708
diff changeset
    59
  def keyword_kind(name: String): Option[String] = keyword_kind_files(name).map(_._1)
38471
0924654b8163 report command token name instead of kind, which can be retrieved later via Outer_Syntax.keyword_kind;
wenzelm
parents: 36956
diff changeset
    60
48864
3ee314ae1e0a added keyword kind "thy_load" (with optional list of file extensions);
wenzelm
parents: 48708
diff changeset
    61
  def + (name: String, kind: (String, List[String]), replace: String): Outer_Syntax =
46626
a02115865bcc streamlined abstract datatype;
wenzelm
parents: 46624
diff changeset
    62
    new Outer_Syntax(
a02115865bcc streamlined abstract datatype;
wenzelm
parents: 46624
diff changeset
    63
      keywords + (name -> kind),
a02115865bcc streamlined abstract datatype;
wenzelm
parents: 46624
diff changeset
    64
      lexicon + name,
48864
3ee314ae1e0a added keyword kind "thy_load" (with optional list of file extensions);
wenzelm
parents: 48708
diff changeset
    65
      if (Keyword.control(kind._1)) completion else completion + (name, replace))
34166
446a33b874b3 renamed class Outer_Keyword to Outer_Syntax;
wenzelm
parents:
diff changeset
    66
48864
3ee314ae1e0a added keyword kind "thy_load" (with optional list of file extensions);
wenzelm
parents: 48708
diff changeset
    67
  def + (name: String, kind: (String, List[String])): Outer_Syntax = this + (name, kind, name)
3ee314ae1e0a added keyword kind "thy_load" (with optional list of file extensions);
wenzelm
parents: 48708
diff changeset
    68
  def + (name: String, kind: String): Outer_Syntax = this + (name, (kind, Nil), name)
36947
285b39022372 renamed Outer_Keyword to Keyword (in Scala);
wenzelm
parents: 34264
diff changeset
    69
  def + (name: String): Outer_Syntax = this + (name, Keyword.MINOR)
48706
e2b512024eab tuned signature;
wenzelm
parents: 48671
diff changeset
    70
48707
ba531af91148 simplified Document.Node.Header -- internalized errors;
wenzelm
parents: 48706
diff changeset
    71
  def add_keywords(header: Document.Node.Header): Outer_Syntax =
ba531af91148 simplified Document.Node.Header -- internalized errors;
wenzelm
parents: 48706
diff changeset
    72
    (this /: header.keywords) {
48708
189ece4b4ff1 permissive outer syntax wrt. symbol recoding;
wenzelm
parents: 48707
diff changeset
    73
      case (syntax, ((name, Some((kind, _))))) =>
189ece4b4ff1 permissive outer syntax wrt. symbol recoding;
wenzelm
parents: 48707
diff changeset
    74
        syntax + (Symbol.decode(name), kind) + (Symbol.encode(name), kind)
189ece4b4ff1 permissive outer syntax wrt. symbol recoding;
wenzelm
parents: 48707
diff changeset
    75
      case (syntax, ((name, None))) =>
189ece4b4ff1 permissive outer syntax wrt. symbol recoding;
wenzelm
parents: 48707
diff changeset
    76
        syntax + Symbol.decode(name) + Symbol.encode(name)
46940
a40be2f10ca9 explicit Outer_Syntax.Decl;
wenzelm
parents: 46712
diff changeset
    77
    }
34166
446a33b874b3 renamed class Outer_Keyword to Outer_Syntax;
wenzelm
parents:
diff changeset
    78
446a33b874b3 renamed class Outer_Keyword to Outer_Syntax;
wenzelm
parents:
diff changeset
    79
  def is_command(name: String): Boolean =
40458
12c8c64203b3 treat main theory commands like headings, and nest anything else inside;
wenzelm
parents: 40455
diff changeset
    80
    keyword_kind(name) match {
36947
285b39022372 renamed Outer_Keyword to Keyword (in Scala);
wenzelm
parents: 34264
diff changeset
    81
      case Some(kind) => kind != Keyword.MINOR
34166
446a33b874b3 renamed class Outer_Keyword to Outer_Syntax;
wenzelm
parents:
diff changeset
    82
      case None => false
446a33b874b3 renamed class Outer_Keyword to Outer_Syntax;
wenzelm
parents:
diff changeset
    83
    }
446a33b874b3 renamed class Outer_Keyword to Outer_Syntax;
wenzelm
parents:
diff changeset
    84
40454
2516ea25a54b some support for nested source structure, based on section headings;
wenzelm
parents: 38471
diff changeset
    85
  def heading_level(name: String): Option[Int] =
46969
481b7d9ad6fe more abstract heading level;
wenzelm
parents: 46941
diff changeset
    86
  {
481b7d9ad6fe more abstract heading level;
wenzelm
parents: 46941
diff changeset
    87
    keyword_kind(name) match {
481b7d9ad6fe more abstract heading level;
wenzelm
parents: 46941
diff changeset
    88
      case _ if name == "header" => Some(0)
481b7d9ad6fe more abstract heading level;
wenzelm
parents: 46941
diff changeset
    89
      case Some(Keyword.THY_HEADING1) => Some(1)
481b7d9ad6fe more abstract heading level;
wenzelm
parents: 46941
diff changeset
    90
      case Some(Keyword.THY_HEADING2) | Some(Keyword.PRF_HEADING2) => Some(2)
481b7d9ad6fe more abstract heading level;
wenzelm
parents: 46941
diff changeset
    91
      case Some(Keyword.THY_HEADING3) | Some(Keyword.PRF_HEADING3) => Some(3)
481b7d9ad6fe more abstract heading level;
wenzelm
parents: 46941
diff changeset
    92
      case Some(Keyword.THY_HEADING4) | Some(Keyword.PRF_HEADING4) => Some(4)
481b7d9ad6fe more abstract heading level;
wenzelm
parents: 46941
diff changeset
    93
      case Some(kind) if Keyword.theory(kind) => Some(5)
481b7d9ad6fe more abstract heading level;
wenzelm
parents: 46941
diff changeset
    94
      case _ => None
40454
2516ea25a54b some support for nested source structure, based on section headings;
wenzelm
parents: 38471
diff changeset
    95
    }
46969
481b7d9ad6fe more abstract heading level;
wenzelm
parents: 46941
diff changeset
    96
  }
40454
2516ea25a54b some support for nested source structure, based on section headings;
wenzelm
parents: 38471
diff changeset
    97
2516ea25a54b some support for nested source structure, based on section headings;
wenzelm
parents: 38471
diff changeset
    98
  def heading_level(command: Command): Option[Int] =
2516ea25a54b some support for nested source structure, based on section headings;
wenzelm
parents: 38471
diff changeset
    99
    heading_level(command.name)
2516ea25a54b some support for nested source structure, based on section headings;
wenzelm
parents: 38471
diff changeset
   100
34166
446a33b874b3 renamed class Outer_Keyword to Outer_Syntax;
wenzelm
parents:
diff changeset
   101
446a33b874b3 renamed class Outer_Keyword to Outer_Syntax;
wenzelm
parents:
diff changeset
   102
  /* tokenize */
446a33b874b3 renamed class Outer_Keyword to Outer_Syntax;
wenzelm
parents:
diff changeset
   103
36956
21be4832c362 renamed class Outer_Lex to Token and Token_Kind to Token.Kind;
wenzelm
parents: 36947
diff changeset
   104
  def scan(input: Reader[Char]): List[Token] =
34166
446a33b874b3 renamed class Outer_Keyword to Outer_Syntax;
wenzelm
parents:
diff changeset
   105
  {
446a33b874b3 renamed class Outer_Keyword to Outer_Syntax;
wenzelm
parents:
diff changeset
   106
    import lexicon._
446a33b874b3 renamed class Outer_Keyword to Outer_Syntax;
wenzelm
parents:
diff changeset
   107
43695
5130dfe1b7be simplified Symbol based on lazy Symbol.Interpretation -- reduced odd "functorial style";
wenzelm
parents: 43455
diff changeset
   108
    parseAll(rep(token(is_command)), input) match {
34166
446a33b874b3 renamed class Outer_Keyword to Outer_Syntax;
wenzelm
parents:
diff changeset
   109
      case Success(tokens, _) => tokens
34264
b5025782a4ed tuned message;
wenzelm
parents: 34166
diff changeset
   110
      case _ => error("Unexpected failure of tokenizing input:\n" + input.source.toString)
34166
446a33b874b3 renamed class Outer_Keyword to Outer_Syntax;
wenzelm
parents:
diff changeset
   111
    }
446a33b874b3 renamed class Outer_Keyword to Outer_Syntax;
wenzelm
parents:
diff changeset
   112
  }
446a33b874b3 renamed class Outer_Keyword to Outer_Syntax;
wenzelm
parents:
diff changeset
   113
36956
21be4832c362 renamed class Outer_Lex to Token and Token_Kind to Token.Kind;
wenzelm
parents: 36947
diff changeset
   114
  def scan(input: CharSequence): List[Token] =
34166
446a33b874b3 renamed class Outer_Keyword to Outer_Syntax;
wenzelm
parents:
diff changeset
   115
    scan(new CharSequenceReader(input))
43411
0206466ee473 some support for partial scans with explicit context;
wenzelm
parents: 40533
diff changeset
   116
0206466ee473 some support for partial scans with explicit context;
wenzelm
parents: 40533
diff changeset
   117
  def scan_context(input: CharSequence, context: Scan.Context): (List[Token], Scan.Context) =
0206466ee473 some support for partial scans with explicit context;
wenzelm
parents: 40533
diff changeset
   118
  {
0206466ee473 some support for partial scans with explicit context;
wenzelm
parents: 40533
diff changeset
   119
    import lexicon._
0206466ee473 some support for partial scans with explicit context;
wenzelm
parents: 40533
diff changeset
   120
0206466ee473 some support for partial scans with explicit context;
wenzelm
parents: 40533
diff changeset
   121
    var in: Reader[Char] = new CharSequenceReader(input)
0206466ee473 some support for partial scans with explicit context;
wenzelm
parents: 40533
diff changeset
   122
    val toks = new mutable.ListBuffer[Token]
0206466ee473 some support for partial scans with explicit context;
wenzelm
parents: 40533
diff changeset
   123
    var ctxt = context
0206466ee473 some support for partial scans with explicit context;
wenzelm
parents: 40533
diff changeset
   124
    while (!in.atEnd) {
43695
5130dfe1b7be simplified Symbol based on lazy Symbol.Interpretation -- reduced odd "functorial style";
wenzelm
parents: 43455
diff changeset
   125
      parse(token_context(is_command, ctxt), in) match {
43411
0206466ee473 some support for partial scans with explicit context;
wenzelm
parents: 40533
diff changeset
   126
        case Success((x, c), rest) => { toks += x; ctxt = c; in = rest }
0206466ee473 some support for partial scans with explicit context;
wenzelm
parents: 40533
diff changeset
   127
        case NoSuccess(_, rest) =>
0206466ee473 some support for partial scans with explicit context;
wenzelm
parents: 40533
diff changeset
   128
          error("Unexpected failure of tokenizing input:\n" + rest.source.toString)
0206466ee473 some support for partial scans with explicit context;
wenzelm
parents: 40533
diff changeset
   129
      }
0206466ee473 some support for partial scans with explicit context;
wenzelm
parents: 40533
diff changeset
   130
    }
0206466ee473 some support for partial scans with explicit context;
wenzelm
parents: 40533
diff changeset
   131
    (toks.toList, ctxt)
0206466ee473 some support for partial scans with explicit context;
wenzelm
parents: 40533
diff changeset
   132
  }
34166
446a33b874b3 renamed class Outer_Keyword to Outer_Syntax;
wenzelm
parents:
diff changeset
   133
}