The Rel Forum

I have resolved this issue, though I cannot reliably replicate it. The problem was caused by multiple errors. I had put a capital S in place of a lower case one, in an identifier that was case sensitive: the text contained "...." inside the defining quotes: and (probably) the stack had bee...

Is it possible to introduce a little translation into DBrowser? This need only translate \n\t to <LF><TAB> as appropriate. This would mean I could enter: INSERT Type_predicate RELATION { TUPLE { name identifier ("base_data_type"), predicate text ( "Base data type (BDT)s are defined as...

I am not sure if this is the same as the infinite loop bug you are already investigating, but it causes an infinite loop; and the approach of substituting X for VALUE in the type definition is neither relevant nor a solution. Define TYPE non_negative_integer POSSREP { X CHAR CONSTRAINT  ( CAST_AS_IN...

If it is possible to express the following situation in Rel， could you suggest the syntax to achieve it？ These have been defined： TYPE identifier POSSREP { VALUE CHAR CONSTRAINT LENGTH ( VALUE ) <= 36 } ; VAR BDT BASE RELATION { name identifier } KEY { name } ; TYPE base_data_type IS { identifier CO...

Is there any way in the language to specify initialisation or default values for an attribute of a relvar, and if so what is the syntax? I looked at INIT, but this is to cater for multiple POSSREP definitions, and gave ERROR: an INIT is not necessary for <= 1 POSSREP. I was trying to provide a defau...

Attached is all the information that appeared in the reporting window when I repeated the error. This is the info regarding the original error not the reporting one.

Error occurred when populating new database; and when partially populating database - doing all the steps up to, but not including the population of the BDT_predicate, and then doing that step caused the problem. There was no evidence of looping - the log I sent you was the complete log. I cannot sp...

I have just generated the above error on one of my scripts, but the bug reporting mechanism does not send the bug - the progress bar goes the full extent to done, then an error box comes up saying it could not be sent. I attach the following: Error message stack Original script This one is a peculia...

Comparing attributes of different types VAR Type_predicate BASE RELATION { name CHARACTER,  predicate CHARACTER }  KEY { name }; OPERATOR TREAT_AS_CHARACTER(P ID ) RETURNS CHARACTER ; return THE_VALUE(P ) ; END OPERATOR ; CONSTRAINT Type_predicate_opt  (EXTEND Type_predicate : )Name := TREAT_AS_CHA...

Syntax error in: CONSTRAINT MOVIE_FKEY_DIRECTOR MOVIE { DIRECTOR# } RENAME (DIRECTOR# AS PERSON#) <= PERSON { PERSON# }; CONSTRAINT CASTING_FKEY_MOVIE CASTING { MOVIE# } <= MOVIE { MOVIE# }; CONSTRAINT CASTING_FKEY_ACTOR CASTING { ACTOR# } RENAME (ACTOR# AS PERSON#) <= PERSON { PERSON# }; should be ...

Optional attribute VAR Arbitrary BASE RELATION { name CHARACTER } KEY { name }; VAR Arbitrary_predicate BASE RELATION { name CHARACTER, predicate CHARACTER } KEY { name }; CONSTRAINT Arbitrary_predicate_opt Arbitrary_predicate { name } <= Arbitrary { name }; Notes This is one of (probably) many way...

Dropping an OPERATOR

Thanks for the workaround - I do keep bumping into this issue from a variety of directions, and had not come up with a solution for myself.

Were you aware that for the base data types the operator >= is OP_GREATHEROREQUAL rather than OP_GREATEROREQUAL?

In trying to simplify my query, I made a mistake in the statements listed. Sorting it out also sorted out my original query. What I was trying to do was to define a (inherited) type whose values depended on the values in a (lookup) relvar. As this was a bit of a red herring, might I suggest you eith...