Translation-Based Tools for Z

Model-checking and refinement checking

You are here: Z2SAL Home / Download / Known Bugs /
Department of Computer Science

Known Bugs

The following is a list of known bugs and other issues, which are being dealt with as time allows.

User Interface Issues

Here we will list any requests to modify the user interface of the tool.

    [empty]

Input Parsing Issues

Here we will list examples of inputs that we should parse, but currently cannot.

  • The public beta-version of the tool does not yet translate Z sequences (because we haven't yet tested all the sequence operations). However, in the alpha-version of the tool, we have successfully translated sequence operations. Users may submit a Z2SAL translation request if they wish to see what an example with sequences looks like.

Output Generation Issues

Here we will list examples of incorrectly-generated SAL output, which may cause the SAL tools to object.

    [empty]

The following are points of style, or a wish-list of things to do, rather than semantic errors:

  • In the regular "Translate Z into SAL" for model-checking purposes, it is redundant to specify, in the ELSE clause, that certain posterior primed variables retain their unprimed values. If we don't mention any primed variables after the --> arrow, then the tools ensure that they keep the same values as before, anyway.
  • When translating the Kurbel and Marlowe refinement, the generated SAL context tickets_app_and_corr.sal contains a stylistic quirk, which is not an error. The free type MTicket has a different automatic label, label__1, from the label__0 that was generated in the other context tickets_init.sal. As it happens, references to the label throughout each context are consistent, but it is just odd that the free type is defined slightly differently in each of the two contexts.
  • In the refinement checking XXX_init.sal examples, it would be nice in future if the retrieve relation could be generated as a DEFINITION, in a similar way to the XXX_app_corr.sal examples. This would make it much easier and formulaic to write the initialisation theorem, which could also be generated automatically.

Library Encoding Issues

Here we will list any known issues with the SAL encoding of the Z mathematical toolkit.

    [empty]
Regent Court, 211 Portobello, Sheffield S1 4DP, United Kingdom