User:Mathieu/Event-B IDE:Needed use cases: Difference between revisions

From Event-B
Jump to navigationJump to search
imported>Mathieu
m new page
 
imported>Mathieu
m A way to quickly jump along an event refinement chain by skipping all extended and unmodified events
 
Line 3: Line 3:
* A mean to follow a slice of a model across refinements. For example, given 2 events, it is useful to be able to see them and only them side by side on several refinement levels (textual export with event markers and ''sed'' do the trick)
* A mean to follow a slice of a model across refinements. For example, given 2 events, it is useful to be able to see them and only them side by side on several refinement levels (textual export with event markers and ''sed'' do the trick)
* A mean to ''diff'' two levels of refinements and quickly see what as change from one level to the next one (textual export and classical diff do the trick).
* A mean to ''diff'' two levels of refinements and quickly see what as change from one level to the next one (textual export and classical diff do the trick).
=== Model Navigation ===
* A way to quickly jump along an event refinement chain by skipping all extended and unmodified events.


== Model refactoring ==
== Model refactoring ==

Latest revision as of 10:02, 24 February 2009

Model understanding

  • A mean to quickly see which variables are changed from a refinement to another one (new variables, removed one)
  • A mean to follow a slice of a model across refinements. For example, given 2 events, it is useful to be able to see them and only them side by side on several refinement levels (textual export with event markers and sed do the trick)
  • A mean to diff two levels of refinements and quickly see what as change from one level to the next one (textual export and classical diff do the trick).

Model Navigation

  • A way to quickly jump along an event refinement chain by skipping all extended and unmodified events.

Model refactoring

  • Be able to change a guard (given by a formula) by another one in the whole model


Other difficulties

  • It is somewhat difficult to propagate the changes down after inserting a new refinement between two existing levels (a way to diff contiguous levels may be useful).