instantiation of formal parameters with only one symbol (first try to see how this would work)

[mcodescu]

Not intended to merge, but pull request shows diff better.

[tillmo]

I have tried this extensively with 29 DOL files that BKB has provided. It mostly works, with two exceptions. The first one is: for compound ids, it is not possible to omit the symbol kind. Consider e.g.

logic OWL

spec S[Class: X] =
 Class: D SubClassOf: X
end

spec SC = S[C[X]]
end

Then C[X] is always recognized as an instantiation of a generic spec, not as a compound id.

[mcodescu]

I made your example work (not an elegant solution, but it goes through). If other of the test files involving compound identifiers do not work, please let me know.

[tillmo]

Good, now more things go through!
However, the following does not work (probably due to the nesting):

logic OWL
ontology A [ObjectProperty: p] [Class: X] = {}
ontology B [ObjectProperty: p] = {}
ontology C [ObjectProperty: p] [Class: D] = B[A[p][Y[D]]]

leads to

*** Error:
file does not exist: D.env
no input found for: D

[tillmo]

Another problem: Let A.dol contain

logic OWL
ontology A = Class: C 

and B.dol contain

from A get A
logic OWL
ontology P[A] = {}
ontology B[A] = P[A]

Then at the left-hand side of = , A correctly is interpreted as an ontology, but a the right-hand side, it is interpreted as a new class.

[tillmo]

Another point: compound IDs with multiple components do not work. Consider

logic OWL
ontology O [Class: A] = {}
ontology P = O[C[D,E]]

I get

*** Error:
file:///home/till/temp/cc.dol:3.19:
unexpected ','

[mcodescu]

The last issue requires a modification of the parser for structured specifications.

[tillmo]

I see. Luckily, the last issue is the least important one.

[mcodescu]

I modified a bit the test for the second issue, to make it more visible:

from A get A
logic OWL
ontology P[A] = {Class: D SubClassOf: C}
ontology B[A] = P[A]

ontology Test = B[AClass]

This goes through now.

[mcodescu]

The first two issues seem to be fixed now, please test and let me know if there are problems.

[tillmo]

many thanks, goes through as expected!
However, there is a problem in Algebra_II:

file:///home/till/CASL/Hets-lib/Basic/Algebra_II.casl:216.35-216.41:
*** Error:
try to instantiate a formal parameter with more than one symbol with a single-symbol spec:Group

[tillmo]

once #1899 has been closed, this one can go