Difference between pages "Event-B Indexers" and "The Use of Theories in Code Generation"

From Event-B
(Difference between pages)
Jump to navigationJump to search
imported>Laurent
 
imported>Andy
 
Line 1: Line 1:
==Purpose==
+
== The Theory Plug-in ==
 +
The theory plug-in is used to add mathematical extensions to Rodin. The theories are created, and deployed, and can then be used in any models in the workspace. When dealing with implementation level models, such as in Tasking Event-B, we need to consider how to translate newly added types and operators into code. We have augmented the theory interface with a Translation Rules section. This enables a user to define translation rules that map Event-B formulas to code.
 +
=== Translation Rules===
 +
Figure 1 shows the interface, and some translations rules of the mapping to Ada.
  
Event-B indexers populate the index repository. Currently, indexers are implemented for the following files :
+
<div id="fig:Translation Rules">
* machine (.bum)
+
<br/>
* context (.buc)
+
[[Image:TheoryCGRules.png|center||caption text]]
 +
<center>'''Figure 1''': Translation Rules</center>
 +
<br/>
 +
</div>
  
 
+
The translation rules are templates used for pattern matching. The meta-variables are defined and typed, and used in the rules. Event-B expressions and predicates are defined on the left hand side of the rule, and the code to be output (as text) appears on the right hand side of the matching rule.
==Event-B Occurrence Kinds==
 
 
 
So far, the following occurrence kinds have been defined :
 
* DECLARATION : when a variable, a carrier set, ..., is declared
 
* REFERENCE : when a variable, a carrier set, ..., is referenced (but not modified)
 
* MODIFICATION : when a variable is modified (e.g., occurs in the left-hand side of an event action)
 
* REDECLARATION : when a variable, parameter or event is redeclared (refined).
 
Please notice the distinction between the index repository notion of declaration ({{class|IDeclaration}}) and the Event-B occurrence kind DECLARATION.
 
 
 
==What Is Indexed==
 
 
 
Currently, indexed elements are :
 
* carrier sets
 
* constants
 
* variables
 
* events
 
* parameters
 
 
 
Formulas which cannot be parsed are ignored.
 
 
 
In the descriptions below, the following notation will be used to specify locations :
 
* [element] : internal location
 
* [element/attribute] : attribute location
 
* [element/attribute/begin..end] : attribute substring location
 
 
 
===Context Indexer===
 
 
 
====Dependencies====
 
* extended contexts
 
 
 
====Declarations====
 
* carrier sets
 
* constants
 
 
 
====Occurrences====
 
 
 
* DECLARATION of carrier sets [Root]
 
* DECLARATION of constants [Root]
 
* REFERENCE of elements in axioms [Axiom/Predicate/b..e]
 
* REFERENCE of elements in theorems [Theorem/Predicate/b..e]
 
 
 
Example:
 
 
 
  Ctx
 
  Sets
 
    S
 
  Constants
 
    C
 
  Axioms
 
    axm1 : C ∈ S
 
 
 
We will end up with the following occurrences :
 
 
 
  S :
 
    DECLARATION in [Ctx.Root]
 
    REFERENCE in [axm1/Predicate/4..5]
 
  C :
 
    DECLARATION in [Ctx.Root]
 
    REFERENCE in [axm1/Predicate/0..1]
 
 
 
 
 
====Exports====
 
* imported and declared carrier sets
 
* imported and declared constants
 
 
 
===Machine Indexer===
 
 
 
====Dependencies====
 
* refined machines
 
* seen contexts
 
 
 
====Declarations====
 
* variables
 
* events
 
* parameters
 
 
 
====Occurrences====
 
* DECLARATION of (local) variables [Root]
 
* DECLARATION of (local) events [Root]
 
* DECLARATION of (local) parameters [Event]
 
* REDECLARATION of abstract variables in the local variables that redeclare them [Variable/Identifier]
 
* REDECLARATION of abstract events in the local events that refine them [Event/RefinesEvent/Target]
 
* REDECLARATION of abstract parameters in the local parameters that redeclare them [Parameter/Identifier]
 
* REFERENCE of abstract parameters or variables in witnesses [Witness/Label]
 
* REFERENCE of elements in invariants [Invariant/Predicate/b..e]
 
* REFERENCE of elements in theorems [Theorem/Predicate/b..e]
 
* REFERENCE of elements in variants [Variant/Expression/b..e]
 
* REFERENCE of elements in guards [Guard/Predicate/b..e]
 
* REFERENCE of elements in witnesses [Witness/Predicate/b..e]
 
* REFERENCE of non assigned elements in actions [Action/Assignment/b..e]
 
* MODIFICATION of assigned elements in actions [Action/Assignment/b..e]
 
 
 
Example :
 
 
 
  M1
 
  VARIABLES
 
    var1
 
  INVARIANTS
 
    inv1 : var1 > 0
 
  EVENTS
 
    INITIALISATION
 
      THEN
 
        act1 : var1 := 1
 
 
 
After indexing M1, we will have the following occurrences:
 
 
 
  M1.var1 :
 
    DECLARATION in [M1.Root]
 
    REFERENCE in [M1.inv1/Predicate/0..4]
 
    MODIFICATION in [M1.act1/Assignment/0..4]
 
  M1.INITIALISATION :
 
    DECLARATION in [M1.Root]
 
 
 
Then, if we add another machine
 
 
 
  M2
 
  REFINES
 
    M1
 
  VARIABLES
 
    var1
 
  EVENTS
 
    INITIALISATION
 
      THEN
 
        act1 : var1 := 1
 
 
 
we will add the following occurrences :
 
 
 
  M1.var1 :
 
    REDECLARATION in [M2.var1/Identifier]
 
  M2.var1 :
 
    DECLARATION in [M2.Root]
 
    MODIFICATION in [M2.act1/Assignment/0..4]
 
  M2.INITIALISATION :
 
    DECLARATION in [M2.Root]
 
 
 
====Exports====
 
* imported carrier sets
 
* imported constants
 
* local variables
 
* local events
 
* local parameters
 
 
 
==Propagation==
 
 
 
Propagators are defined for :
 
* events ({{class|EventPropagator}}) to propagate through refines
 
* parameters ({{class|IdentifierPropagator}}) to propagate through redeclaration
 
* variables ({{class|IdentifierPropagator}} as well) to propagate through redeclaration
 
 
 
In the above example with machines M1 and M2, after both files have been indexed, we can query occurrences of M1.var1.
 
 
 
With no propagator, the result would be :
 
 
 
  M1.var1 :
 
    DECLARATION in [M1.Root]
 
    REFERENCE in [M1.inv1/Predicate/0..4]
 
    MODIFICATION in [M1.act1/Assignment/0..4]
 
    REDECLARATION in [M2.var1/Identifier]
 
 
 
Using the {{class|IdentifierPropagator}}, the result becomes :
 
 
 
  M1.var1 (propagated) :
 
    DECLARATION in [M1.Root]
 
    REFERENCE in [M1.inv1/Predicate/0..4]
 
    MODIFICATION in [M1.act1/Assignment/0..4]
 
    REDECLARATION in [M2.var1/Identifier]
 
    DECLARATION in [M2.Root]
 
    MODIFICATION in [M2.act1/Assignment/0..4]
 
 
 
[[Category:Developer documentation]]
 

Revision as of 08:38, 15 May 2012

The Theory Plug-in

The theory plug-in is used to add mathematical extensions to Rodin. The theories are created, and deployed, and can then be used in any models in the workspace. When dealing with implementation level models, such as in Tasking Event-B, we need to consider how to translate newly added types and operators into code. We have augmented the theory interface with a Translation Rules section. This enables a user to define translation rules that map Event-B formulas to code.

Translation Rules

Figure 1 shows the interface, and some translations rules of the mapping to Ada.


caption text
Figure 1: Translation Rules


The translation rules are templates used for pattern matching. The meta-variables are defined and typed, and used in the rules. Event-B expressions and predicates are defined on the left hand side of the rule, and the code to be output (as text) appears on the right hand side of the matching rule.