author | wenzelm |
Wed, 14 Oct 2015 15:10:32 +0200 | |
changeset 61439 | 2bf52eec4e8a |
parent 61416 | b9a3324e4e62 |
child 61458 | 987533262fc2 |
permissions | -rw-r--r-- |
57344 | 1 |
(*:wrap=hard:maxLineLen=78:*) |
2 |
||
30124 | 3 |
theory Syntax |
4 |
imports Base |
|
5 |
begin |
|
6 |
||
58618 | 7 |
chapter \<open>Concrete syntax and type-checking\<close> |
30124 | 8 |
|
58618 | 9 |
text \<open>Pure @{text "\<lambda>"}-calculus as introduced in \chref{ch:logic} is |
45258 | 10 |
an adequate foundation for logical languages --- in the tradition of |
11 |
\emph{higher-order abstract syntax} --- but end-users require |
|
12 |
additional means for reading and printing of terms and types. This |
|
13 |
important add-on outside the logical core is called \emph{inner |
|
14 |
syntax} in Isabelle jargon, as opposed to the \emph{outer syntax} of |
|
58555 | 15 |
the theory and proof language @{cite "isabelle-isar-ref"}. |
45258 | 16 |
|
58555 | 17 |
For example, according to @{cite church40} quantifiers are represented as |
57496 | 18 |
higher-order constants @{text "All :: ('a \<Rightarrow> bool) \<Rightarrow> bool"} such that @{text |
19 |
"All (\<lambda>x::'a. B x)"} faithfully represents the idea that is displayed in |
|
20 |
Isabelle as @{text "\<forall>x::'a. B x"} via @{keyword "binder"} notation. |
|
58555 | 21 |
Moreover, type-inference in the style of Hindley-Milner @{cite hindleymilner} |
57496 | 22 |
(and extensions) enables users to write @{text "\<forall>x. B x"} concisely, when |
23 |
the type @{text "'a"} is already clear from the |
|
24 |
context.\footnote{Type-inference taken to the extreme can easily confuse |
|
25 |
users. Beginners often stumble over unexpectedly general types inferred by |
|
26 |
the system.} |
|
45258 | 27 |
|
61416 | 28 |
\<^medskip> |
29 |
The main inner syntax operations are \emph{read} for |
|
45258 | 30 |
parsing together with type-checking, and \emph{pretty} for formatted |
31 |
output. See also \secref{sec:read-print}. |
|
32 |
||
33 |
Furthermore, the input and output syntax layers are sub-divided into |
|
34 |
separate phases for \emph{concrete syntax} versus \emph{abstract |
|
35 |
syntax}, see also \secref{sec:parse-unparse} and |
|
36 |
\secref{sec:term-check}, respectively. This results in the |
|
37 |
following decomposition of the main operations: |
|
38 |
||
39 |
\begin{itemize} |
|
40 |
||
61416 | 41 |
\<^item> @{text "read = parse; check"} |
45258 | 42 |
|
61416 | 43 |
\<^item> @{text "pretty = uncheck; unparse"} |
45258 | 44 |
|
45 |
\end{itemize} |
|
46 |
||
57496 | 47 |
For example, some specification package might thus intercept syntax |
48 |
processing at a well-defined stage after @{text "parse"}, to a augment the |
|
49 |
resulting pre-term before full type-reconstruction is performed by @{text |
|
50 |
"check"}. Note that the formal status of bound variables, versus free |
|
51 |
variables, versus constants must not be changed between these phases. |
|
57345 | 52 |
|
61416 | 53 |
\<^medskip> |
54 |
In general, @{text check} and @{text uncheck} operate |
|
57345 | 55 |
simultaneously on a list of terms. This is particular important for |
56 |
type-checking, to reconstruct types for several terms of the same context |
|
57 |
and scope. In contrast, @{text parse} and @{text unparse} operate separately |
|
57496 | 58 |
on single terms. |
57345 | 59 |
|
60 |
There are analogous operations to read and print types, with the same |
|
61 |
sub-division into phases. |
|
58618 | 62 |
\<close> |
45258 | 63 |
|
30272 | 64 |
|
58618 | 65 |
section \<open>Reading and pretty printing \label{sec:read-print}\<close> |
34924 | 66 |
|
58618 | 67 |
text \<open> |
57496 | 68 |
Read and print operations are roughly dual to each other, such that for the |
69 |
user @{text "s' = pretty (read s)"} looks similar to the original source |
|
70 |
text @{text "s"}, but the details depend on many side-conditions. There are |
|
71 |
also explicit options to control the removal of type information in the |
|
72 |
output. The default configuration routinely looses information, so @{text |
|
73 |
"t' = read (pretty t)"} might fail, or produce a differently typed term, or |
|
74 |
a completely different term in the face of syntactic overloading. |
|
58618 | 75 |
\<close> |
34924 | 76 |
|
58618 | 77 |
text %mlref \<open> |
39876
1ff9bce085bd
preliminary material on "Concrete syntax and type-checking";
wenzelm
parents:
39865
diff
changeset
|
78 |
\begin{mldecls} |
57345 | 79 |
@{index_ML Syntax.read_typs: "Proof.context -> string list -> typ list"} \\ |
80 |
@{index_ML Syntax.read_terms: "Proof.context -> string list -> term list"} \\ |
|
81 |
@{index_ML Syntax.read_props: "Proof.context -> string list -> term list"} \\[0.5ex] |
|
39876
1ff9bce085bd
preliminary material on "Concrete syntax and type-checking";
wenzelm
parents:
39865
diff
changeset
|
82 |
@{index_ML Syntax.read_typ: "Proof.context -> string -> typ"} \\ |
1ff9bce085bd
preliminary material on "Concrete syntax and type-checking";
wenzelm
parents:
39865
diff
changeset
|
83 |
@{index_ML Syntax.read_term: "Proof.context -> string -> term"} \\ |
57345 | 84 |
@{index_ML Syntax.read_prop: "Proof.context -> string -> term"} \\[0.5ex] |
39876
1ff9bce085bd
preliminary material on "Concrete syntax and type-checking";
wenzelm
parents:
39865
diff
changeset
|
85 |
@{index_ML Syntax.pretty_typ: "Proof.context -> typ -> Pretty.T"} \\ |
1ff9bce085bd
preliminary material on "Concrete syntax and type-checking";
wenzelm
parents:
39865
diff
changeset
|
86 |
@{index_ML Syntax.pretty_term: "Proof.context -> term -> Pretty.T"} \\ |
57345 | 87 |
@{index_ML Syntax.string_of_typ: "Proof.context -> typ -> string"} \\ |
88 |
@{index_ML Syntax.string_of_term: "Proof.context -> term -> string"} \\ |
|
39876
1ff9bce085bd
preliminary material on "Concrete syntax and type-checking";
wenzelm
parents:
39865
diff
changeset
|
89 |
\end{mldecls} |
1ff9bce085bd
preliminary material on "Concrete syntax and type-checking";
wenzelm
parents:
39865
diff
changeset
|
90 |
|
57345 | 91 |
\begin{description} |
92 |
||
61439 | 93 |
\<^descr> @{ML Syntax.read_typs}~@{text "ctxt strs"} parses and checks a |
57345 | 94 |
simultaneous list of source strings as types of the logic. |
95 |
||
61439 | 96 |
\<^descr> @{ML Syntax.read_terms}~@{text "ctxt strs"} parses and checks a |
57345 | 97 |
simultaneous list of source strings as terms of the logic. |
57346 | 98 |
Type-reconstruction puts all parsed terms into the same scope: types of |
99 |
free variables ultimately need to coincide. |
|
57345 | 100 |
|
101 |
If particular type-constraints are required for some of the arguments, the |
|
57346 | 102 |
read operations needs to be split into its parse and check phases. Then it |
57496 | 103 |
is possible to use @{ML Type.constraint} on the intermediate pre-terms |
57846 | 104 |
(\secref{sec:term-check}). |
57345 | 105 |
|
61439 | 106 |
\<^descr> @{ML Syntax.read_props}~@{text "ctxt strs"} parses and checks a |
57346 | 107 |
simultaneous list of source strings as terms of the logic, with an implicit |
108 |
type-constraint for each argument to enforce type @{typ prop}; this also |
|
57496 | 109 |
affects the inner syntax for parsing. The remaining type-reconstruction |
110 |
works as for @{ML Syntax.read_terms}. |
|
57345 | 111 |
|
61439 | 112 |
\<^descr> @{ML Syntax.read_typ}, @{ML Syntax.read_term}, @{ML Syntax.read_prop} |
57496 | 113 |
are like the simultaneous versions, but operate on a single argument only. |
114 |
This convenient shorthand is adequate in situations where a single item in |
|
115 |
its own scope is processed. Do not use @{ML "map o Syntax.read_term"} where |
|
116 |
@{ML Syntax.read_terms} is actually intended! |
|
57345 | 117 |
|
61439 | 118 |
\<^descr> @{ML Syntax.pretty_typ}~@{text "ctxt T"} and @{ML |
57345 | 119 |
Syntax.pretty_term}~@{text "ctxt t"} uncheck and pretty-print the given type |
120 |
or term, respectively. Although the uncheck phase acts on a simultaneous |
|
57496 | 121 |
list as well, this is rarely used in practice, so only the singleton case is |
122 |
provided as combined pretty operation. There is no distinction of term vs.\ |
|
123 |
proposition. |
|
57345 | 124 |
|
61439 | 125 |
\<^descr> @{ML Syntax.string_of_typ} and @{ML Syntax.string_of_term} are |
57345 | 126 |
convenient compositions of @{ML Syntax.pretty_typ} and @{ML |
127 |
Syntax.pretty_term} with @{ML Pretty.string_of} for output. The result may |
|
128 |
be concatenated with other strings, as long as there is no further |
|
129 |
formatting and line-breaking involved. |
|
130 |
||
131 |
\end{description} |
|
132 |
||
57346 | 133 |
@{ML Syntax.read_term}, @{ML Syntax.read_prop}, and @{ML |
134 |
Syntax.string_of_term} are the most important operations in practice. |
|
57345 | 135 |
|
61416 | 136 |
\<^medskip> |
137 |
Note that the string values that are passed in and out are |
|
57346 | 138 |
annotated by the system, to carry further markup that is relevant for the |
58555 | 139 |
Prover IDE @{cite "isabelle-jedit"}. User code should neither compose its |
140 |
own input strings, nor try to analyze the output strings. Conceptually this |
|
141 |
is an abstract datatype, encoded as concrete string for historical reasons. |
|
57345 | 142 |
|
143 |
The standard way to provide the required position markup for input works via |
|
144 |
the outer syntax parser wrapper @{ML Parse.inner_syntax}, which is already |
|
145 |
part of @{ML Parse.typ}, @{ML Parse.term}, @{ML Parse.prop}. So a string |
|
146 |
obtained from one of the latter may be directly passed to the corresponding |
|
57346 | 147 |
read operation: this yields PIDE markup of the input and precise positions |
148 |
for warning and error messages. |
|
58618 | 149 |
\<close> |
39876
1ff9bce085bd
preliminary material on "Concrete syntax and type-checking";
wenzelm
parents:
39865
diff
changeset
|
150 |
|
1ff9bce085bd
preliminary material on "Concrete syntax and type-checking";
wenzelm
parents:
39865
diff
changeset
|
151 |
|
58618 | 152 |
section \<open>Parsing and unparsing \label{sec:parse-unparse}\<close> |
39876
1ff9bce085bd
preliminary material on "Concrete syntax and type-checking";
wenzelm
parents:
39865
diff
changeset
|
153 |
|
58618 | 154 |
text \<open> |
57496 | 155 |
Parsing and unparsing converts between actual source text and a certain |
156 |
\emph{pre-term} format, where all bindings and scopes are already resolved |
|
157 |
faithfully. Thus the names of free variables or constants are determined in |
|
158 |
the sense of the logical context, but type information might be still |
|
159 |
missing. Pre-terms support an explicit language of \emph{type constraints} |
|
160 |
that may be augmented by user code to guide the later \emph{check} phase. |
|
45258 | 161 |
|
57496 | 162 |
Actual parsing is based on traditional lexical analysis and Earley parsing |
163 |
for arbitrary context-free grammars. The user can specify the grammar |
|
164 |
declaratively via mixfix annotations. Moreover, there are \emph{syntax |
|
165 |
translations} that can be augmented by the user, either declaratively via |
|
166 |
@{command translations} or programmatically via @{command |
|
58555 | 167 |
parse_translation}, @{command print_translation} @{cite |
168 |
"isabelle-isar-ref"}. The final scope-resolution is performed by the system, |
|
169 |
according to name spaces for types, term variables and constants determined |
|
170 |
by the context. |
|
58618 | 171 |
\<close> |
39876
1ff9bce085bd
preliminary material on "Concrete syntax and type-checking";
wenzelm
parents:
39865
diff
changeset
|
172 |
|
58618 | 173 |
text %mlref \<open> |
39876
1ff9bce085bd
preliminary material on "Concrete syntax and type-checking";
wenzelm
parents:
39865
diff
changeset
|
174 |
\begin{mldecls} |
1ff9bce085bd
preliminary material on "Concrete syntax and type-checking";
wenzelm
parents:
39865
diff
changeset
|
175 |
@{index_ML Syntax.parse_typ: "Proof.context -> string -> typ"} \\ |
1ff9bce085bd
preliminary material on "Concrete syntax and type-checking";
wenzelm
parents:
39865
diff
changeset
|
176 |
@{index_ML Syntax.parse_term: "Proof.context -> string -> term"} \\ |
57496 | 177 |
@{index_ML Syntax.parse_prop: "Proof.context -> string -> term"} \\[0.5ex] |
39876
1ff9bce085bd
preliminary material on "Concrete syntax and type-checking";
wenzelm
parents:
39865
diff
changeset
|
178 |
@{index_ML Syntax.unparse_typ: "Proof.context -> typ -> Pretty.T"} \\ |
1ff9bce085bd
preliminary material on "Concrete syntax and type-checking";
wenzelm
parents:
39865
diff
changeset
|
179 |
@{index_ML Syntax.unparse_term: "Proof.context -> term -> Pretty.T"} \\ |
1ff9bce085bd
preliminary material on "Concrete syntax and type-checking";
wenzelm
parents:
39865
diff
changeset
|
180 |
\end{mldecls} |
1ff9bce085bd
preliminary material on "Concrete syntax and type-checking";
wenzelm
parents:
39865
diff
changeset
|
181 |
|
57346 | 182 |
\begin{description} |
183 |
||
61439 | 184 |
\<^descr> @{ML Syntax.parse_typ}~@{text "ctxt str"} parses a source string as |
57346 | 185 |
pre-type that is ready to be used with subsequent check operations. |
186 |
||
61439 | 187 |
\<^descr> @{ML Syntax.parse_term}~@{text "ctxt str"} parses a source string as |
57346 | 188 |
pre-term that is ready to be used with subsequent check operations. |
189 |
||
61439 | 190 |
\<^descr> @{ML Syntax.parse_prop}~@{text "ctxt str"} parses a source string as |
57346 | 191 |
pre-term that is ready to be used with subsequent check operations. The |
192 |
inner syntax category is @{typ prop} and a suitable type-constraint is |
|
57496 | 193 |
included to ensure that this information is observed in subsequent type |
194 |
reconstruction. |
|
57346 | 195 |
|
61439 | 196 |
\<^descr> @{ML Syntax.unparse_typ}~@{text "ctxt T"} unparses a type after |
57346 | 197 |
uncheck operations, to turn it into a pretty tree. |
198 |
||
61439 | 199 |
\<^descr> @{ML Syntax.unparse_term}~@{text "ctxt T"} unparses a term after |
57346 | 200 |
uncheck operations, to turn it into a pretty tree. There is no distinction |
201 |
for propositions here. |
|
202 |
||
203 |
\end{description} |
|
204 |
||
57496 | 205 |
These operations always operate on a single item; use the combinator @{ML |
206 |
map} to apply them to a list. |
|
58618 | 207 |
\<close> |
39876
1ff9bce085bd
preliminary material on "Concrete syntax and type-checking";
wenzelm
parents:
39865
diff
changeset
|
208 |
|
39852 | 209 |
|
58618 | 210 |
section \<open>Checking and unchecking \label{sec:term-check}\<close> |
34924 | 211 |
|
58618 | 212 |
text \<open>These operations define the transition from pre-terms and |
45258 | 213 |
fully-annotated terms in the sense of the logical core |
214 |
(\chref{ch:logic}). |
|
215 |
||
216 |
The \emph{check} phase is meant to subsume a variety of mechanisms |
|
217 |
in the manner of ``type-inference'' or ``type-reconstruction'' or |
|
218 |
``type-improvement'', not just type-checking in the narrow sense. |
|
219 |
The \emph{uncheck} phase is roughly dual, it prunes type-information |
|
220 |
before pretty printing. |
|
221 |
||
222 |
A typical add-on for the check/uncheck syntax layer is the @{command |
|
58555 | 223 |
abbreviation} mechanism @{cite "isabelle-isar-ref"}. Here the user specifies |
57496 | 224 |
syntactic definitions that are managed by the system as polymorphic @{text |
225 |
"let"} bindings. These are expanded during the @{text "check"} phase, and |
|
226 |
contracted during the @{text "uncheck"} phase, without affecting the |
|
227 |
type-assignment of the given terms. |
|
45258 | 228 |
|
61416 | 229 |
\<^medskip> |
230 |
The precise meaning of type checking depends on the context --- |
|
57496 | 231 |
additional check/uncheck modules might be defined in user space. |
45258 | 232 |
|
233 |
For example, the @{command class} command defines a context where |
|
234 |
@{text "check"} treats certain type instances of overloaded |
|
235 |
constants according to the ``dictionary construction'' of its |
|
236 |
logical foundation. This involves ``type improvement'' |
|
237 |
(specialization of slightly too general types) and replacement by |
|
58555 | 238 |
certain locale parameters. See also @{cite "Haftmann-Wenzel:2009"}. |
58618 | 239 |
\<close> |
39852 | 240 |
|
58618 | 241 |
text %mlref \<open> |
39876
1ff9bce085bd
preliminary material on "Concrete syntax and type-checking";
wenzelm
parents:
39865
diff
changeset
|
242 |
\begin{mldecls} |
1ff9bce085bd
preliminary material on "Concrete syntax and type-checking";
wenzelm
parents:
39865
diff
changeset
|
243 |
@{index_ML Syntax.check_typs: "Proof.context -> typ list -> typ list"} \\ |
1ff9bce085bd
preliminary material on "Concrete syntax and type-checking";
wenzelm
parents:
39865
diff
changeset
|
244 |
@{index_ML Syntax.check_terms: "Proof.context -> term list -> term list"} \\ |
57496 | 245 |
@{index_ML Syntax.check_props: "Proof.context -> term list -> term list"} \\[0.5ex] |
39876
1ff9bce085bd
preliminary material on "Concrete syntax and type-checking";
wenzelm
parents:
39865
diff
changeset
|
246 |
@{index_ML Syntax.uncheck_typs: "Proof.context -> typ list -> typ list"} \\ |
1ff9bce085bd
preliminary material on "Concrete syntax and type-checking";
wenzelm
parents:
39865
diff
changeset
|
247 |
@{index_ML Syntax.uncheck_terms: "Proof.context -> term list -> term list"} \\ |
1ff9bce085bd
preliminary material on "Concrete syntax and type-checking";
wenzelm
parents:
39865
diff
changeset
|
248 |
\end{mldecls} |
1ff9bce085bd
preliminary material on "Concrete syntax and type-checking";
wenzelm
parents:
39865
diff
changeset
|
249 |
|
57346 | 250 |
\begin{description} |
251 |
||
61439 | 252 |
\<^descr> @{ML Syntax.check_typs}~@{text "ctxt Ts"} checks a simultaneous list |
57346 | 253 |
of pre-types as types of the logic. Typically, this involves normalization |
254 |
of type synonyms. |
|
255 |
||
61439 | 256 |
\<^descr> @{ML Syntax.check_terms}~@{text "ctxt ts"} checks a simultaneous list |
57346 | 257 |
of pre-terms as terms of the logic. Typically, this involves type-inference |
258 |
and normalization term abbreviations. The types within the given terms are |
|
259 |
treated in the same way as for @{ML Syntax.check_typs}. |
|
260 |
||
261 |
Applications sometimes need to check several types and terms together. The |
|
262 |
standard approach uses @{ML Logic.mk_type} to embed the language of types |
|
263 |
into that of terms; all arguments are appended into one list of terms that |
|
57496 | 264 |
is checked; afterwards the type arguments are recovered with @{ML |
57346 | 265 |
Logic.dest_type}. |
266 |
||
61439 | 267 |
\<^descr> @{ML Syntax.check_props}~@{text "ctxt ts"} checks a simultaneous list |
57346 | 268 |
of pre-terms as terms of the logic, such that all terms are constrained by |
269 |
type @{typ prop}. The remaining check operation works as @{ML |
|
270 |
Syntax.check_terms} above. |
|
271 |
||
61439 | 272 |
\<^descr> @{ML Syntax.uncheck_typs}~@{text "ctxt Ts"} unchecks a simultaneous |
57346 | 273 |
list of types of the logic, in preparation of pretty printing. |
274 |
||
61439 | 275 |
\<^descr> @{ML Syntax.uncheck_terms}~@{text "ctxt ts"} unchecks a simultaneous |
57346 | 276 |
list of terms of the logic, in preparation of pretty printing. There is no |
277 |
distinction for propositions here. |
|
278 |
||
279 |
\end{description} |
|
280 |
||
57496 | 281 |
These operations always operate simultaneously on a list; use the combinator |
282 |
@{ML singleton} to apply them to a single item. |
|
58618 | 283 |
\<close> |
39876
1ff9bce085bd
preliminary material on "Concrete syntax and type-checking";
wenzelm
parents:
39865
diff
changeset
|
284 |
|
30124 | 285 |
end |