| author | haftmann |
| Tue, 29 Mar 2022 15:44:27 +0200 | |
| changeset 75364 | 366f85a10407 |
| parent 74887 | 56247fdb8bbb |
| child 75384 | 20093a63d03b |
| permissions | -rw-r--r-- |
|
36956
21be4832c362
renamed class Outer_Lex to Token and Token_Kind to Token.Kind;
wenzelm
parents:
34311
diff
changeset
|
1 |
/* Title: Pure/Isar/token.scala |
|
34139
d1ded303fe0e
Outer lexical syntax for Isabelle/Isar -- Scala version.
wenzelm
parents:
diff
changeset
|
2 |
Author: Makarius |
|
d1ded303fe0e
Outer lexical syntax for Isabelle/Isar -- Scala version.
wenzelm
parents:
diff
changeset
|
3 |
|
|
36956
21be4832c362
renamed class Outer_Lex to Token and Token_Kind to Token.Kind;
wenzelm
parents:
34311
diff
changeset
|
4 |
Outer token syntax for Isabelle/Isar. |
|
34139
d1ded303fe0e
Outer lexical syntax for Isabelle/Isar -- Scala version.
wenzelm
parents:
diff
changeset
|
5 |
*/ |
|
d1ded303fe0e
Outer lexical syntax for Isabelle/Isar -- Scala version.
wenzelm
parents:
diff
changeset
|
6 |
|
|
d1ded303fe0e
Outer lexical syntax for Isabelle/Isar -- Scala version.
wenzelm
parents:
diff
changeset
|
7 |
package isabelle |
|
d1ded303fe0e
Outer lexical syntax for Isabelle/Isar -- Scala version.
wenzelm
parents:
diff
changeset
|
8 |
|
|
d1ded303fe0e
Outer lexical syntax for Isabelle/Isar -- Scala version.
wenzelm
parents:
diff
changeset
|
9 |
|
| 59083 | 10 |
import scala.collection.mutable |
11 |
import scala.util.parsing.input |
|
12 |
||
13 |
||
|
36956
21be4832c362
renamed class Outer_Lex to Token and Token_Kind to Token.Kind;
wenzelm
parents:
34311
diff
changeset
|
14 |
object Token |
|
34139
d1ded303fe0e
Outer lexical syntax for Isabelle/Isar -- Scala version.
wenzelm
parents:
diff
changeset
|
15 |
{
|
|
34157
0a0a19153626
explicit representation of Token_Kind -- cannot really depend on runtime types due to erasure;
wenzelm
parents:
34143
diff
changeset
|
16 |
/* tokens */ |
|
34139
d1ded303fe0e
Outer lexical syntax for Isabelle/Isar -- Scala version.
wenzelm
parents:
diff
changeset
|
17 |
|
|
36956
21be4832c362
renamed class Outer_Lex to Token and Token_Kind to Token.Kind;
wenzelm
parents:
34311
diff
changeset
|
18 |
object Kind extends Enumeration |
|
34139
d1ded303fe0e
Outer lexical syntax for Isabelle/Isar -- Scala version.
wenzelm
parents:
diff
changeset
|
19 |
{
|
| 59081 | 20 |
/*immediate source*/ |
|
34157
0a0a19153626
explicit representation of Token_Kind -- cannot really depend on runtime types due to erasure;
wenzelm
parents:
34143
diff
changeset
|
21 |
val COMMAND = Value("command")
|
|
0a0a19153626
explicit representation of Token_Kind -- cannot really depend on runtime types due to erasure;
wenzelm
parents:
34143
diff
changeset
|
22 |
val KEYWORD = Value("keyword")
|
|
0a0a19153626
explicit representation of Token_Kind -- cannot really depend on runtime types due to erasure;
wenzelm
parents:
34143
diff
changeset
|
23 |
val IDENT = Value("identifier")
|
|
0a0a19153626
explicit representation of Token_Kind -- cannot really depend on runtime types due to erasure;
wenzelm
parents:
34143
diff
changeset
|
24 |
val LONG_IDENT = Value("long identifier")
|
|
0a0a19153626
explicit representation of Token_Kind -- cannot really depend on runtime types due to erasure;
wenzelm
parents:
34143
diff
changeset
|
25 |
val SYM_IDENT = Value("symbolic identifier")
|
|
0a0a19153626
explicit representation of Token_Kind -- cannot really depend on runtime types due to erasure;
wenzelm
parents:
34143
diff
changeset
|
26 |
val VAR = Value("schematic variable")
|
|
0a0a19153626
explicit representation of Token_Kind -- cannot really depend on runtime types due to erasure;
wenzelm
parents:
34143
diff
changeset
|
27 |
val TYPE_IDENT = Value("type variable")
|
|
0a0a19153626
explicit representation of Token_Kind -- cannot really depend on runtime types due to erasure;
wenzelm
parents:
34143
diff
changeset
|
28 |
val TYPE_VAR = Value("schematic type variable")
|
|
40290
47f572aff50a
support for floating-point tokens in outer syntax (coinciding with inner syntax version);
wenzelm
parents:
38367
diff
changeset
|
29 |
val NAT = Value("natural number")
|
|
47f572aff50a
support for floating-point tokens in outer syntax (coinciding with inner syntax version);
wenzelm
parents:
38367
diff
changeset
|
30 |
val FLOAT = Value("floating-point number")
|
| 59081 | 31 |
val SPACE = Value("white space")
|
32 |
/*delimited content*/ |
|
|
34157
0a0a19153626
explicit representation of Token_Kind -- cannot really depend on runtime types due to erasure;
wenzelm
parents:
34143
diff
changeset
|
33 |
val STRING = Value("string")
|
|
0a0a19153626
explicit representation of Token_Kind -- cannot really depend on runtime types due to erasure;
wenzelm
parents:
34143
diff
changeset
|
34 |
val ALT_STRING = Value("back-quoted string")
|
| 55512 | 35 |
val CARTOUCHE = Value("text cartouche")
|
|
74373
6e4093927dbb
outer syntax: support for control-cartouche tokens;
wenzelm
parents:
72744
diff
changeset
|
36 |
val CONTROL = Value("control cartouche")
|
|
67439
78759a7bd874
more uniform support for formal comments in outer syntax, notably \<^cancel> and \<^latex>;
wenzelm
parents:
67432
diff
changeset
|
37 |
val INFORMAL_COMMENT = Value("informal comment")
|
|
78759a7bd874
more uniform support for formal comments in outer syntax, notably \<^cancel> and \<^latex>;
wenzelm
parents:
67432
diff
changeset
|
38 |
val FORMAL_COMMENT = Value("formal comment")
|
| 59081 | 39 |
/*special content*/ |
|
48754
c2c1e5944536
clarified undefined, unparsed, unfinished command spans;
wenzelm
parents:
48718
diff
changeset
|
40 |
val ERROR = Value("bad input")
|
|
34157
0a0a19153626
explicit representation of Token_Kind -- cannot really depend on runtime types due to erasure;
wenzelm
parents:
34143
diff
changeset
|
41 |
val UNPARSED = Value("unparsed input")
|
|
34139
d1ded303fe0e
Outer lexical syntax for Isabelle/Isar -- Scala version.
wenzelm
parents:
diff
changeset
|
42 |
} |
|
d1ded303fe0e
Outer lexical syntax for Isabelle/Isar -- Scala version.
wenzelm
parents:
diff
changeset
|
43 |
|
|
34157
0a0a19153626
explicit representation of Token_Kind -- cannot really depend on runtime types due to erasure;
wenzelm
parents:
34143
diff
changeset
|
44 |
|
| 55494 | 45 |
/* parsers */ |
46 |
||
47 |
object Parsers extends Parsers |
|
48 |
||
|
67439
78759a7bd874
more uniform support for formal comments in outer syntax, notably \<^cancel> and \<^latex>;
wenzelm
parents:
67432
diff
changeset
|
49 |
trait Parsers extends Scan.Parsers with Comment.Parsers |
| 55494 | 50 |
{
|
51 |
private def delimited_token: Parser[Token] = |
|
52 |
{
|
|
53 |
val string = quoted("\"") ^^ (x => Token(Token.Kind.STRING, x))
|
|
54 |
val alt_string = quoted("`") ^^ (x => Token(Token.Kind.ALT_STRING, x))
|
|
|
67439
78759a7bd874
more uniform support for formal comments in outer syntax, notably \<^cancel> and \<^latex>;
wenzelm
parents:
67432
diff
changeset
|
55 |
val cmt = comment ^^ (x => Token(Token.Kind.INFORMAL_COMMENT, x)) |
|
78759a7bd874
more uniform support for formal comments in outer syntax, notably \<^cancel> and \<^latex>;
wenzelm
parents:
67432
diff
changeset
|
56 |
val formal_cmt = comment_cartouche ^^ (x => Token(Token.Kind.FORMAL_COMMENT, x)) |
|
74373
6e4093927dbb
outer syntax: support for control-cartouche tokens;
wenzelm
parents:
72744
diff
changeset
|
57 |
val cart = cartouche ^^ (x => Token(Token.Kind.CARTOUCHE, x)) |
|
6e4093927dbb
outer syntax: support for control-cartouche tokens;
wenzelm
parents:
72744
diff
changeset
|
58 |
val ctrl = control_cartouche ^^ (x => Token(Token.Kind.CONTROL, x)) |
| 55494 | 59 |
|
| 74887 | 60 |
string | (alt_string | (cmt | (formal_cmt | (cart | ctrl)))) |
| 55494 | 61 |
} |
62 |
||
| 58900 | 63 |
private def other_token(keywords: Keyword.Keywords): Parser[Token] = |
| 55494 | 64 |
{
|
65 |
val letdigs1 = many1(Symbol.is_letdig) |
|
| 62103 | 66 |
val sub = one(s => s == Symbol.sub_decoded || s == Symbol.sub) |
| 55494 | 67 |
val id = |
68 |
one(Symbol.is_letter) ~ |
|
69 |
(rep(letdigs1 | (sub ~ letdigs1 ^^ { case x ~ y => x + y })) ^^ (_.mkString)) ^^
|
|
70 |
{ case x ~ y => x + y }
|
|
71 |
||
72 |
val nat = many1(Symbol.is_digit) |
|
73 |
val natdot = nat ~ "." ~ nat ^^ { case x ~ y ~ z => x + y + z }
|
|
74 |
val id_nat = id ~ opt("." ~ nat) ^^ { case x ~ Some(y ~ z) => x + y + z case x ~ None => x }
|
|
75 |
||
76 |
val ident = id ~ rep("." ~> id) ^^
|
|
77 |
{ case x ~ Nil => Token(Token.Kind.IDENT, x)
|
|
78 |
case x ~ ys => Token(Token.Kind.LONG_IDENT, (x :: ys).mkString(".")) }
|
|
79 |
||
80 |
val var_ = "?" ~ id_nat ^^ { case x ~ y => Token(Token.Kind.VAR, x + y) }
|
|
81 |
val type_ident = "'" ~ id ^^ { case x ~ y => Token(Token.Kind.TYPE_IDENT, x + y) }
|
|
82 |
val type_var = "?'" ~ id_nat ^^ { case x ~ y => Token(Token.Kind.TYPE_VAR, x + y) }
|
|
83 |
val nat_ = nat ^^ (x => Token(Token.Kind.NAT, x)) |
|
84 |
val float = |
|
85 |
("-" ~ natdot ^^ { case x ~ y => x + y } | natdot) ^^ (x => Token(Token.Kind.FLOAT, x))
|
|
86 |
||
87 |
val sym_ident = |
|
88 |
(many1(Symbol.is_symbolic_char) | one(sym => Symbol.is_symbolic(sym))) ^^ |
|
89 |
(x => Token(Token.Kind.SYM_IDENT, x)) |
|
90 |
||
| 58899 | 91 |
val keyword = |
| 58900 | 92 |
literal(keywords.minor) ^^ (x => Token(Token.Kind.KEYWORD, x)) ||| |
93 |
literal(keywords.major) ^^ (x => Token(Token.Kind.COMMAND, x)) |
|
| 55494 | 94 |
|
95 |
val space = many1(Symbol.is_blank) ^^ (x => Token(Token.Kind.SPACE, x)) |
|
96 |
||
97 |
val recover_delimited = |
|
98 |
(recover_quoted("\"") |
|
|
99 |
(recover_quoted("`") |
|
|
| 74887 | 100 |
(recover_cartouche | recover_comment))) ^^ (x => Token(Token.Kind.ERROR, x)) |
| 55494 | 101 |
|
102 |
val bad = one(_ => true) ^^ (x => Token(Token.Kind.ERROR, x)) |
|
103 |
||
104 |
space | (recover_delimited | |
|
105 |
(((ident | (var_ | (type_ident | (type_var | (float | (nat_ | sym_ident)))))) ||| |
|
| 58899 | 106 |
keyword) | bad)) |
| 55494 | 107 |
} |
108 |
||
| 58900 | 109 |
def token(keywords: Keyword.Keywords): Parser[Token] = |
| 67446 | 110 |
delimited_token | other_token(keywords) |
| 55494 | 111 |
|
| 58900 | 112 |
def token_line(keywords: Keyword.Keywords, ctxt: Scan.Line_Context) |
|
55510
1585a65aad64
tuned signature -- emphasize line-oriented aspect;
wenzelm
parents:
55505
diff
changeset
|
113 |
: Parser[(Token, Scan.Line_Context)] = |
| 55494 | 114 |
{
|
115 |
val string = |
|
|
55510
1585a65aad64
tuned signature -- emphasize line-oriented aspect;
wenzelm
parents:
55505
diff
changeset
|
116 |
quoted_line("\"", ctxt) ^^ { case (x, c) => (Token(Token.Kind.STRING, x), c) }
|
| 55494 | 117 |
val alt_string = |
|
55510
1585a65aad64
tuned signature -- emphasize line-oriented aspect;
wenzelm
parents:
55505
diff
changeset
|
118 |
quoted_line("`", ctxt) ^^ { case (x, c) => (Token(Token.Kind.ALT_STRING, x), c) }
|
|
1585a65aad64
tuned signature -- emphasize line-oriented aspect;
wenzelm
parents:
55505
diff
changeset
|
119 |
val cart = cartouche_line(ctxt) ^^ { case (x, c) => (Token(Token.Kind.CARTOUCHE, x), c) }
|
|
67439
78759a7bd874
more uniform support for formal comments in outer syntax, notably \<^cancel> and \<^latex>;
wenzelm
parents:
67432
diff
changeset
|
120 |
val cmt = comment_line(ctxt) ^^ { case (x, c) => (Token(Token.Kind.INFORMAL_COMMENT, x), c) }
|
|
78759a7bd874
more uniform support for formal comments in outer syntax, notably \<^cancel> and \<^latex>;
wenzelm
parents:
67432
diff
changeset
|
121 |
val formal_cmt = |
|
78759a7bd874
more uniform support for formal comments in outer syntax, notably \<^cancel> and \<^latex>;
wenzelm
parents:
67432
diff
changeset
|
122 |
comment_cartouche_line(ctxt) ^^ { case (x, c) => (Token(Token.Kind.FORMAL_COMMENT, x), c) }
|
| 58900 | 123 |
val other = other_token(keywords) ^^ { case x => (x, Scan.Finished) }
|
| 55494 | 124 |
|
| 74887 | 125 |
string | (alt_string | (cart | (cmt | (formal_cmt | other)))) |
| 55494 | 126 |
} |
127 |
} |
|
128 |
||
129 |
||
| 59083 | 130 |
/* explode */ |
131 |
||
132 |
def explode(keywords: Keyword.Keywords, inp: CharSequence): List[Token] = |
|
| 64824 | 133 |
Parsers.parseAll(Parsers.rep(Parsers.token(keywords)), Scan.char_reader(inp)) match {
|
| 59083 | 134 |
case Parsers.Success(tokens, _) => tokens |
135 |
case _ => error("Unexpected failure of tokenizing input:\n" + inp.toString)
|
|
136 |
} |
|
137 |
||
138 |
def explode_line(keywords: Keyword.Keywords, inp: CharSequence, context: Scan.Line_Context) |
|
139 |
: (List[Token], Scan.Line_Context) = |
|
140 |
{
|
|
| 64824 | 141 |
var in: input.Reader[Char] = Scan.char_reader(inp) |
| 59083 | 142 |
val toks = new mutable.ListBuffer[Token] |
143 |
var ctxt = context |
|
144 |
while (!in.atEnd) {
|
|
145 |
Parsers.parse(Parsers.token_line(keywords, ctxt), in) match {
|
|
| 60215 | 146 |
case Parsers.Success((x, c), rest) => toks += x; ctxt = c; in = rest |
| 59083 | 147 |
case Parsers.NoSuccess(_, rest) => |
148 |
error("Unexpected failure of tokenizing input:\n" + rest.source.toString)
|
|
149 |
} |
|
150 |
} |
|
151 |
(toks.toList, ctxt) |
|
152 |
} |
|
153 |
||
| 64671 | 154 |
val newline: Token = explode(Keyword.Keywords.empty, "\n").head |
155 |
||
| 59083 | 156 |
|
| 69603 | 157 |
/* embedded */ |
158 |
||
159 |
def read_embedded(keywords: Keyword.Keywords, inp: CharSequence): Option[Token] = |
|
160 |
explode(keywords, inp) match {
|
|
161 |
case List(tok) if tok.is_embedded => Some(tok) |
|
162 |
case _ => None |
|
163 |
} |
|
164 |
||
165 |
||
| 65523 | 166 |
/* names */ |
167 |
||
168 |
def read_name(keywords: Keyword.Keywords, inp: CharSequence): Option[Token] = |
|
169 |
explode(keywords, inp) match {
|
|
170 |
case List(tok) if tok.is_name => Some(tok) |
|
171 |
case _ => None |
|
172 |
} |
|
173 |
||
174 |
def quote_name(keywords: Keyword.Keywords, name: String): String = |
|
175 |
if (read_name(keywords, name).isDefined) name |
|
176 |
else quote(name.replace("\"", "\\\""))
|
|
177 |
||
178 |
||
| 67132 | 179 |
/* plain antiquotation (0 or 1 args) */ |
180 |
||
181 |
def read_antiq_arg(keywords: Keyword.Keywords, inp: CharSequence): Option[(String, Option[String])] = |
|
182 |
explode(keywords, inp).filter(_.is_proper) match {
|
|
183 |
case List(t) if t.is_name => Some(t.content, None) |
|
184 |
case List(t1, t2) if t1.is_name && t2.is_embedded => Some(t1.content, Some(t2.content)) |
|
185 |
case _ => None |
|
186 |
} |
|
187 |
||
188 |
||
| 59735 | 189 |
/* implode */ |
190 |
||
191 |
def implode(toks: List[Token]): String = |
|
192 |
toks match {
|
|
193 |
case List(tok) => tok.source |
|
| 60215 | 194 |
case _ => toks.map(_.source).mkString |
| 59735 | 195 |
} |
196 |
||
197 |
||
|
34157
0a0a19153626
explicit representation of Token_Kind -- cannot really depend on runtime types due to erasure;
wenzelm
parents:
34143
diff
changeset
|
198 |
/* token reader */ |
|
34139
d1ded303fe0e
Outer lexical syntax for Isabelle/Isar -- Scala version.
wenzelm
parents:
diff
changeset
|
199 |
|
| 56464 | 200 |
object Pos |
201 |
{
|
|
|
59706
bf6ca55aae13
proper command id for inlined errors, which is important for Command.State.accumulate;
wenzelm
parents:
59705
diff
changeset
|
202 |
val none: Pos = new Pos(0, 0, "", "") |
|
bf6ca55aae13
proper command id for inlined errors, which is important for Command.State.accumulate;
wenzelm
parents:
59705
diff
changeset
|
203 |
val start: Pos = new Pos(1, 1, "", "") |
|
bf6ca55aae13
proper command id for inlined errors, which is important for Command.State.accumulate;
wenzelm
parents:
59705
diff
changeset
|
204 |
def file(file: String): Pos = new Pos(1, 1, file, "") |
|
bf6ca55aae13
proper command id for inlined errors, which is important for Command.State.accumulate;
wenzelm
parents:
59705
diff
changeset
|
205 |
def id(id: String): Pos = new Pos(0, 1, "", id) |
|
59715
4f0d0e4ad68d
avoid duplicate header errors, more precise positions;
wenzelm
parents:
59706
diff
changeset
|
206 |
val command: Pos = id(Markup.COMMAND) |
| 56464 | 207 |
} |
208 |
||
|
59671
9715eb8e9408
more precise position information in Isabelle/Scala, with YXML markup as in Isabelle/ML;
wenzelm
parents:
59122
diff
changeset
|
209 |
final class Pos private[Token]( |
| 59696 | 210 |
val line: Int, |
211 |
val offset: Symbol.Offset, |
|
|
59706
bf6ca55aae13
proper command id for inlined errors, which is important for Command.State.accumulate;
wenzelm
parents:
59705
diff
changeset
|
212 |
val file: String, |
|
bf6ca55aae13
proper command id for inlined errors, which is important for Command.State.accumulate;
wenzelm
parents:
59705
diff
changeset
|
213 |
val id: String) |
| 64824 | 214 |
extends input.Position |
|
34139
d1ded303fe0e
Outer lexical syntax for Isabelle/Isar -- Scala version.
wenzelm
parents:
diff
changeset
|
215 |
{
|
|
34157
0a0a19153626
explicit representation of Token_Kind -- cannot really depend on runtime types due to erasure;
wenzelm
parents:
34143
diff
changeset
|
216 |
def column = 0 |
|
0a0a19153626
explicit representation of Token_Kind -- cannot really depend on runtime types due to erasure;
wenzelm
parents:
34143
diff
changeset
|
217 |
def lineContents = "" |
|
0a0a19153626
explicit representation of Token_Kind -- cannot really depend on runtime types due to erasure;
wenzelm
parents:
34143
diff
changeset
|
218 |
|
| 67895 | 219 |
def advance(token: Token): Pos = advance(token.source) |
220 |
def advance(source: String): Pos = |
|
|
34157
0a0a19153626
explicit representation of Token_Kind -- cannot really depend on runtime types due to erasure;
wenzelm
parents:
34143
diff
changeset
|
221 |
{
|
|
59671
9715eb8e9408
more precise position information in Isabelle/Scala, with YXML markup as in Isabelle/ML;
wenzelm
parents:
59122
diff
changeset
|
222 |
var line1 = line |
|
9715eb8e9408
more precise position information in Isabelle/Scala, with YXML markup as in Isabelle/ML;
wenzelm
parents:
59122
diff
changeset
|
223 |
var offset1 = offset |
| 67895 | 224 |
for (s <- Symbol.iterator(source)) {
|
|
59671
9715eb8e9408
more precise position information in Isabelle/Scala, with YXML markup as in Isabelle/ML;
wenzelm
parents:
59122
diff
changeset
|
225 |
if (line1 > 0 && Symbol.is_newline(s)) line1 += 1 |
|
9715eb8e9408
more precise position information in Isabelle/Scala, with YXML markup as in Isabelle/ML;
wenzelm
parents:
59122
diff
changeset
|
226 |
if (offset1 > 0) offset1 += 1 |
|
9715eb8e9408
more precise position information in Isabelle/Scala, with YXML markup as in Isabelle/ML;
wenzelm
parents:
59122
diff
changeset
|
227 |
} |
|
9715eb8e9408
more precise position information in Isabelle/Scala, with YXML markup as in Isabelle/ML;
wenzelm
parents:
59122
diff
changeset
|
228 |
if (line1 == line && offset1 == offset) this |
|
59706
bf6ca55aae13
proper command id for inlined errors, which is important for Command.State.accumulate;
wenzelm
parents:
59705
diff
changeset
|
229 |
else new Pos(line1, offset1, file, id) |
|
34157
0a0a19153626
explicit representation of Token_Kind -- cannot really depend on runtime types due to erasure;
wenzelm
parents:
34143
diff
changeset
|
230 |
} |
| 56464 | 231 |
|
| 59695 | 232 |
private def position(end_offset: Symbol.Offset): Position.T = |
|
59671
9715eb8e9408
more precise position information in Isabelle/Scala, with YXML markup as in Isabelle/ML;
wenzelm
parents:
59122
diff
changeset
|
233 |
(if (line > 0) Position.Line(line) else Nil) ::: |
|
9715eb8e9408
more precise position information in Isabelle/Scala, with YXML markup as in Isabelle/ML;
wenzelm
parents:
59122
diff
changeset
|
234 |
(if (offset > 0) Position.Offset(offset) else Nil) ::: |
|
9715eb8e9408
more precise position information in Isabelle/Scala, with YXML markup as in Isabelle/ML;
wenzelm
parents:
59122
diff
changeset
|
235 |
(if (end_offset > 0) Position.End_Offset(end_offset) else Nil) ::: |
|
59706
bf6ca55aae13
proper command id for inlined errors, which is important for Command.State.accumulate;
wenzelm
parents:
59705
diff
changeset
|
236 |
(if (file != "") Position.File(file) else Nil) ::: |
|
bf6ca55aae13
proper command id for inlined errors, which is important for Command.State.accumulate;
wenzelm
parents:
59705
diff
changeset
|
237 |
(if (id != "") Position.Id_String(id) else Nil) |
|
59671
9715eb8e9408
more precise position information in Isabelle/Scala, with YXML markup as in Isabelle/ML;
wenzelm
parents:
59122
diff
changeset
|
238 |
|
|
9715eb8e9408
more precise position information in Isabelle/Scala, with YXML markup as in Isabelle/ML;
wenzelm
parents:
59122
diff
changeset
|
239 |
def position(): Position.T = position(0) |
|
9715eb8e9408
more precise position information in Isabelle/Scala, with YXML markup as in Isabelle/ML;
wenzelm
parents:
59122
diff
changeset
|
240 |
def position(token: Token): Position.T = position(advance(token).offset) |
| 67895 | 241 |
def position(source: String): Position.T = position(advance(source).offset) |
|
59671
9715eb8e9408
more precise position information in Isabelle/Scala, with YXML markup as in Isabelle/ML;
wenzelm
parents:
59122
diff
changeset
|
242 |
|
| 64728 | 243 |
override def toString: String = Position.here(position(), delimited = false) |
|
34139
d1ded303fe0e
Outer lexical syntax for Isabelle/Isar -- Scala version.
wenzelm
parents:
diff
changeset
|
244 |
} |
|
d1ded303fe0e
Outer lexical syntax for Isabelle/Isar -- Scala version.
wenzelm
parents:
diff
changeset
|
245 |
|
| 64824 | 246 |
abstract class Reader extends input.Reader[Token] |
|
34157
0a0a19153626
explicit representation of Token_Kind -- cannot really depend on runtime types due to erasure;
wenzelm
parents:
34143
diff
changeset
|
247 |
|
| 56464 | 248 |
private class Token_Reader(tokens: List[Token], val pos: Pos) extends Reader |
|
34139
d1ded303fe0e
Outer lexical syntax for Isabelle/Isar -- Scala version.
wenzelm
parents:
diff
changeset
|
249 |
{
|
| 71601 | 250 |
def first: Token = tokens.head |
251 |
def rest: Token_Reader = new Token_Reader(tokens.tail, pos.advance(first)) |
|
252 |
def atEnd: Boolean = tokens.isEmpty |
|
|
34139
d1ded303fe0e
Outer lexical syntax for Isabelle/Isar -- Scala version.
wenzelm
parents:
diff
changeset
|
253 |
} |
|
d1ded303fe0e
Outer lexical syntax for Isabelle/Isar -- Scala version.
wenzelm
parents:
diff
changeset
|
254 |
|
| 59705 | 255 |
def reader(tokens: List[Token], start: Token.Pos): Reader = |
256 |
new Token_Reader(tokens, start) |
|
|
34139
d1ded303fe0e
Outer lexical syntax for Isabelle/Isar -- Scala version.
wenzelm
parents:
diff
changeset
|
257 |
} |
|
d1ded303fe0e
Outer lexical syntax for Isabelle/Isar -- Scala version.
wenzelm
parents:
diff
changeset
|
258 |
|
|
36956
21be4832c362
renamed class Outer_Lex to Token and Token_Kind to Token.Kind;
wenzelm
parents:
34311
diff
changeset
|
259 |
|
| 60215 | 260 |
sealed case class Token(kind: Token.Kind.Value, source: String) |
|
36956
21be4832c362
renamed class Outer_Lex to Token and Token_Kind to Token.Kind;
wenzelm
parents:
34311
diff
changeset
|
261 |
{
|
|
21be4832c362
renamed class Outer_Lex to Token and Token_Kind to Token.Kind;
wenzelm
parents:
34311
diff
changeset
|
262 |
def is_command: Boolean = kind == Token.Kind.COMMAND |
| 63446 | 263 |
def is_command(name: String): Boolean = kind == Token.Kind.COMMAND && source == name |
| 48718 | 264 |
def is_keyword: Boolean = kind == Token.Kind.KEYWORD |
| 63446 | 265 |
def is_keyword(name: String): Boolean = kind == Token.Kind.KEYWORD && source == name |
| 63450 | 266 |
def is_keyword(name: Char): Boolean = |
267 |
kind == Token.Kind.KEYWORD && source.length == 1 && source(0) == name |
|
| 55505 | 268 |
def is_delimiter: Boolean = is_keyword && !Symbol.is_ascii_identifier(source) |
|
48365
d88aefda01c4
basic support for stand-alone options with external string representation;
wenzelm
parents:
48349
diff
changeset
|
269 |
def is_ident: Boolean = kind == Token.Kind.IDENT |
|
48605
e777363440d6
allow negative int values as well, according to real = int | float;
wenzelm
parents:
48599
diff
changeset
|
270 |
def is_sym_ident: Boolean = kind == Token.Kind.SYM_IDENT |
| 46943 | 271 |
def is_string: Boolean = kind == Token.Kind.STRING |
|
48349
a78e5d399599
support Session.Queue with ordering and dependencies;
wenzelm
parents:
48335
diff
changeset
|
272 |
def is_nat: Boolean = kind == Token.Kind.NAT |
|
48365
d88aefda01c4
basic support for stand-alone options with external string representation;
wenzelm
parents:
48349
diff
changeset
|
273 |
def is_float: Boolean = kind == Token.Kind.FLOAT |
|
36956
21be4832c362
renamed class Outer_Lex to Token and Token_Kind to Token.Kind;
wenzelm
parents:
34311
diff
changeset
|
274 |
def is_name: Boolean = |
|
21be4832c362
renamed class Outer_Lex to Token and Token_Kind to Token.Kind;
wenzelm
parents:
34311
diff
changeset
|
275 |
kind == Token.Kind.IDENT || |
| 62969 | 276 |
kind == Token.Kind.LONG_IDENT || |
|
36956
21be4832c362
renamed class Outer_Lex to Token and Token_Kind to Token.Kind;
wenzelm
parents:
34311
diff
changeset
|
277 |
kind == Token.Kind.SYM_IDENT || |
|
21be4832c362
renamed class Outer_Lex to Token and Token_Kind to Token.Kind;
wenzelm
parents:
34311
diff
changeset
|
278 |
kind == Token.Kind.STRING || |
|
21be4832c362
renamed class Outer_Lex to Token and Token_Kind to Token.Kind;
wenzelm
parents:
34311
diff
changeset
|
279 |
kind == Token.Kind.NAT |
|
64471
c40c2975fb02
more uniform path syntax, as in ML (see 5a7c919a4ada);
wenzelm
parents:
63477
diff
changeset
|
280 |
def is_embedded: Boolean = is_name || |
|
c40c2975fb02
more uniform path syntax, as in ML (see 5a7c919a4ada);
wenzelm
parents:
63477
diff
changeset
|
281 |
kind == Token.Kind.CARTOUCHE || |
|
c40c2975fb02
more uniform path syntax, as in ML (see 5a7c919a4ada);
wenzelm
parents:
63477
diff
changeset
|
282 |
kind == Token.Kind.VAR || |
|
c40c2975fb02
more uniform path syntax, as in ML (see 5a7c919a4ada);
wenzelm
parents:
63477
diff
changeset
|
283 |
kind == Token.Kind.TYPE_IDENT || |
|
c40c2975fb02
more uniform path syntax, as in ML (see 5a7c919a4ada);
wenzelm
parents:
63477
diff
changeset
|
284 |
kind == Token.Kind.TYPE_VAR |
|
36956
21be4832c362
renamed class Outer_Lex to Token and Token_Kind to Token.Kind;
wenzelm
parents:
34311
diff
changeset
|
285 |
def is_space: Boolean = kind == Token.Kind.SPACE |
| 67441 | 286 |
def is_informal_comment: Boolean = kind == Token.Kind.INFORMAL_COMMENT |
287 |
def is_formal_comment: Boolean = kind == Token.Kind.FORMAL_COMMENT |
|
|
69891
def3ec9cdb7e
document markers are formal comments, and may thus occur anywhere in the command-span;
wenzelm
parents:
69603
diff
changeset
|
288 |
def is_marker: Boolean = |
|
def3ec9cdb7e
document markers are formal comments, and may thus occur anywhere in the command-span;
wenzelm
parents:
69603
diff
changeset
|
289 |
kind == Token.Kind.FORMAL_COMMENT && |
|
def3ec9cdb7e
document markers are formal comments, and may thus occur anywhere in the command-span;
wenzelm
parents:
69603
diff
changeset
|
290 |
(source.startsWith(Symbol.marker) || source.startsWith(Symbol.marker_decoded)) |
| 67441 | 291 |
def is_comment: Boolean = is_informal_comment || is_formal_comment |
|
68729
3a02b424d5fb
clarified ignored span / core range: include formal comments, e.g. relevant for error messages from antiquotations;
wenzelm
parents:
67895
diff
changeset
|
292 |
def is_ignored: Boolean = is_space || is_informal_comment |
| 48599 | 293 |
def is_proper: Boolean = !is_space && !is_comment |
|
48754
c2c1e5944536
clarified undefined, unparsed, unfinished command spans;
wenzelm
parents:
48718
diff
changeset
|
294 |
def is_error: Boolean = kind == Token.Kind.ERROR |
|
47012
0e246130486b
clarified command span classification: strict Command.is_command, permissive Command.name;
wenzelm
parents:
46943
diff
changeset
|
295 |
def is_unparsed: Boolean = kind == Token.Kind.UNPARSED |
|
36956
21be4832c362
renamed class Outer_Lex to Token and Token_Kind to Token.Kind;
wenzelm
parents:
34311
diff
changeset
|
296 |
|
|
48754
c2c1e5944536
clarified undefined, unparsed, unfinished command spans;
wenzelm
parents:
48718
diff
changeset
|
297 |
def is_unfinished: Boolean = is_error && |
|
c2c1e5944536
clarified undefined, unparsed, unfinished command spans;
wenzelm
parents:
48718
diff
changeset
|
298 |
(source.startsWith("\"") ||
|
|
c2c1e5944536
clarified undefined, unparsed, unfinished command spans;
wenzelm
parents:
48718
diff
changeset
|
299 |
source.startsWith("`") ||
|
|
57021
6a8fd2ac6756
explicit treatment of unfinished cartouches, which is important for Thy_Syntax.consolidate_spans;
wenzelm
parents:
56998
diff
changeset
|
300 |
source.startsWith("(*") ||
|
|
6a8fd2ac6756
explicit treatment of unfinished cartouches, which is important for Thy_Syntax.consolidate_spans;
wenzelm
parents:
56998
diff
changeset
|
301 |
source.startsWith(Symbol.open) || |
|
6a8fd2ac6756
explicit treatment of unfinished cartouches, which is important for Thy_Syntax.consolidate_spans;
wenzelm
parents:
56998
diff
changeset
|
302 |
source.startsWith(Symbol.open_decoded)) |
|
48754
c2c1e5944536
clarified undefined, unparsed, unfinished command spans;
wenzelm
parents:
48718
diff
changeset
|
303 |
|
| 71601 | 304 |
def is_open_bracket: Boolean = is_keyword && Word.open_brackets.exists(is_keyword) |
305 |
def is_close_bracket: Boolean = is_keyword && Word.close_brackets.exists(is_keyword) |
|
| 63450 | 306 |
|
| 63446 | 307 |
def is_begin: Boolean = is_keyword("begin")
|
308 |
def is_end: Boolean = is_command("end")
|
|
|
63477
f5c81436b930
clarified indentation: 'begin' is treated like a separate command without indent;
wenzelm
parents:
63450
diff
changeset
|
309 |
def is_begin_or_command: Boolean = is_begin || is_command |
| 43611 | 310 |
|
| 72669 | 311 |
def symbol_length: Symbol.Offset = Symbol.iterator(source).length |
312 |
||
|
36956
21be4832c362
renamed class Outer_Lex to Token and Token_Kind to Token.Kind;
wenzelm
parents:
34311
diff
changeset
|
313 |
def content: String = |
|
55492
28d4db6c6e79
tuned signature -- separate Lexicon from Parsers (in accordance to ML version);
wenzelm
parents:
55137
diff
changeset
|
314 |
if (kind == Token.Kind.STRING) Scan.Parsers.quoted_content("\"", source)
|
|
28d4db6c6e79
tuned signature -- separate Lexicon from Parsers (in accordance to ML version);
wenzelm
parents:
55137
diff
changeset
|
315 |
else if (kind == Token.Kind.ALT_STRING) Scan.Parsers.quoted_content("`", source)
|
|
28d4db6c6e79
tuned signature -- separate Lexicon from Parsers (in accordance to ML version);
wenzelm
parents:
55137
diff
changeset
|
316 |
else if (kind == Token.Kind.CARTOUCHE) Scan.Parsers.cartouche_content(source) |
|
67439
78759a7bd874
more uniform support for formal comments in outer syntax, notably \<^cancel> and \<^latex>;
wenzelm
parents:
67432
diff
changeset
|
317 |
else if (kind == Token.Kind.INFORMAL_COMMENT) Scan.Parsers.comment_content(source) |
|
78759a7bd874
more uniform support for formal comments in outer syntax, notably \<^cancel> and \<^latex>;
wenzelm
parents:
67432
diff
changeset
|
318 |
else if (kind == Token.Kind.FORMAL_COMMENT) Comment.content(source) |
|
36956
21be4832c362
renamed class Outer_Lex to Token and Token_Kind to Token.Kind;
wenzelm
parents:
34311
diff
changeset
|
319 |
else source |
| 66914 | 320 |
|
|
66915
f4259adc928a
disallow blanks, relevant for session_name / theory_name e.g. in build log files;
wenzelm
parents:
66914
diff
changeset
|
321 |
def is_system_name: Boolean = |
|
f4259adc928a
disallow blanks, relevant for session_name / theory_name e.g. in build log files;
wenzelm
parents:
66914
diff
changeset
|
322 |
{
|
|
f4259adc928a
disallow blanks, relevant for session_name / theory_name e.g. in build log files;
wenzelm
parents:
66914
diff
changeset
|
323 |
val s = content |
|
69551
adb52af5ba55
exclude file name components that are special on Windows;
wenzelm
parents:
68730
diff
changeset
|
324 |
is_name && Path.is_wellformed(s) && |
| 71601 | 325 |
!s.exists(Symbol.is_ascii_blank) && |
|
69551
adb52af5ba55
exclude file name components that are special on Windows;
wenzelm
parents:
68730
diff
changeset
|
326 |
!Path.is_reserved(s) |
|
66915
f4259adc928a
disallow blanks, relevant for session_name / theory_name e.g. in build log files;
wenzelm
parents:
66914
diff
changeset
|
327 |
} |
|
36956
21be4832c362
renamed class Outer_Lex to Token and Token_Kind to Token.Kind;
wenzelm
parents:
34311
diff
changeset
|
328 |
} |