1st Analysis of Model Transformations Workshop, 2012 Several important tasks performed by model driven development tools — such as well-formedness constraint validation or model transformations — rely on evaluating model queries. If the model changes rapidly or frequently, it is beneficial to provide live queries that automatically propagate these model changes into the query results. To ease the development and debugging of live queries, the development environment should provide a way to evaluate the query results continuously, helping to understand how the created query works. This presentation presents a generic live model query visualizer that displays and updates the query results depending on their source models. It has been implemented for the EMF-IncQuery framework and presented here for validating BPMN models.

1. Developing
and
Visualizing
Live
Model
Queries
Zoltán
Ujhelyi,
Tamás
Szabó
István
Ráth,
Dániel Varró
Ábel
Hegedüs
(demonstrator)
Budapest University of Technology and Economics
Department of Measurement and Informa<on Systems

2. What
is
a
Model
Query?
 ModelQuery(A,B)
Cond(A,B)
o Retrieve
tuples
of
model
elements
A,B
o SaOsfying
the
query
condiOon
Cond
o Enumerate
one
/
all
instances
o With
A,B
as
input
or
output
parameters
2

3. What
is
a
Model
Query?
 ModelQuery(A,B)
AUTOSAR
Cond(A,B)
• Select pairs of SystemSignal
and
o Retrieve
tuples
of
its signal group which are not in
the same IPDU
model
elements
A,B
o SaOsfying
the
query
condiOon
Cond
IMA
• Find instances of cri;cal func;ons
o Enumerate
not running on different
one
/
all
instances
processors of different chassis
o With
A,B
as
input
or
output
BPMN
parameters
• Show sink ac;vi;es with no
outgoing flow
3

4. What
is
a
Model
Query?
 ModelQuery(A,B)
AUTOSAR
Cond(A,B)
• Select pairs of SystemSignal
and
o Retrieve
tuples
of
its signal group which are not in
the same IPDU
model
elements
A,B
o SaOsfying
the
query
condiOon
Cond
IMA
• Find instances of cri;cal func;ons
o Enumerate
not running on different
one
/
all
instances
processors of different chassis
o With
A,B
as
input
or
output
BPMN
parameters
• Show sink ac;vi;es with no
outgoing flow
4

5. Live
Model
Queries
 Live
Query
EvaluaOon:
Incremental
cache
of
matches
o Maintain
a
cheap
cache
(only
memory
overhead)
o NoOfy
about
relevant
changes
o Enable
reacOons
to
complex
structural
events
5

6. Live
Model
Queries
 Live
Query
EvaluaOon:
Incremental
cache
of
matches
o Maintain
a
cheap
cache
(only
memory
overhead)
o NoOfy
about
relevant
changes
o Enable
reacOons
to
complex
structural
events
6

7. Live
Model
Queries
 Live
Query
EvaluaOon:
Incremental
cache
of
matches
o Maintain
a
cheap
cache
(only
memory
overhead)
o NoOfy
about
relevant
changes
o Enable
reacOons
to
complex
structural
events
7

8. Live
Model
Queries
 Live
Query
EvaluaOon:
Live
Model
Queries
for
Model
TransformaOons
Incremental
cache
of
matches
o Maintain
a
cheap
cache
• Find
all
contexts
a
MT
rule
is
applicable
for
(only
memory
overhead)
• Live
MT
rules:
triggered
o NoOfy
about
relevant
changes
automaOcally
upon
change
o Enable
reacOons
to
complex
structural
events
Live
Model
Queries
for
Early
ValidaOon
• Find
all
violaOons
of
a
well-­‐formedness
constraint
• Immediate
re-­‐validaOon
upon
model
change
Live
Model
Queries
for
Traceability
Management
• See
talk
tomorrow
at
11:30
8

9. Example:
ValidaOon
of
BPMN
Models
 BPMN
models
o Describing
business
processes
o Models
control
and
dataflow
• Flowchart-­‐like
notaOon
9

10. Example:
ValidaOon
of
BPMN
Models
 Well-­‐formedness
validaOon
during
ediOng
o Sink
AcOvity:
acOvity
without
outgoing
edge
10

11. Example:
ValidaOon
of
BPMN
Models
 Well-­‐formedness
validaOon
during
ediOng
o Sink
AcOvity:
acOvity
without
outgoing
edge
11

12. EMF-­‐IncQuery Framework

13. EMF-­‐IncQuery:
Architecture
Overview
Application
Application code
Generated pattern Generative tooling Query
matcher specification
API
Validation Derived feature Interpreted
Engine support pattern matcher
INCQUERY Base RETE Core INCQUERY Core
Framework

14. EMF-­‐IncQuery:
Architecture
Overview
Application • Graph pattern-based
query language
Application code • Xtext2-based tooling
• Execution, testing and
debugging for queries
Generated pattern Generative tooling Query
matcher specification
API
Validation Derived feature Interpreted
Engine support pattern matcher
INCQUERY Base RETE Core INCQUERY Core
Framework

15. EMF-­‐IncQuery:
Architecture
Overview
Application • Graph pattern-based
query language
Application code • Xtext2-based tooling
Typesafe query
wrappers (DTOs) • Execution, testing and
for easy integration debugging for queries
Generated pattern Generative tooling Query
matcher specification
API
Validation Derived feature Interpreted
Engine support pattern matcher
INCQUERY Base RETE Core INCQUERY Core
Framework

16. EMF-­‐IncQuery:
Architecture
Overview
Application • Graph pattern-based
query language
Application code • Xtext2-based tooling
Typesafe query
wrappers (DTOs) • Execution, testing and
for easy integration debugging for queries
Generated pattern Generative tooling Query
matcher specification
API
Validation Derived feature Interpreted
Engine support pattern matcher
• Easy-to-use API for
On-the-fly validation Efficient maintenance
query
and problem marker of derived attributes
management
management and references
• Support for
programmed and
INCQUERY Base RETE Core INCQUERY Core dynamic queries
• Fully dynamic
parameterization
Framework

17. EMF-­‐IncQuery:
Architecture
Overview
Application • Graph pattern-based
query language
Application code • Xtext2-based tooling
Typesafe query
wrappers (DTOs) • Execution, testing and
for easy integration debugging for queries
Generated pattern Generative tooling Query
matcher specification
API
Validation Derived feature Interpreted
Engine support pattern matcher
• Easy-to-use API for
On-the-fly validation Efficient maintenance
query
and problem marker of derived attributes
management
management and references
• Support for
programmed and
INCQUERY Base RETE Core INCQUERY Core dynamic queries
• Fully dynamic
High performance Scalable evaluation for parameterization
backward getters and complex queries
Framework
transitive closure (over 2 000 000 EObjects)

18. Simple
Graph
Pajern
in
EMF-­‐IncQuery
pattern sequenceFlowEdge
(Flow:SequenceEdge,
Src:Activity, Dst:Activity)= {
SequenceEdge.source(Flow, Src);
SequenceEdge.target(Flow, Dst);
}
source
Flow:
target
Src
:
Dst:
Sequence
AcOvity
AcOvity
Edge
18

19. And
Some
More
Complex
Examples…
pattern hasOutEdge(A: Activity) {
find sequenceFlowEdge(_Fr, A, _Other);
} or {
find messageFlowEdge(_Fr, A, _Other);
Pajern
composiOon
}
(for
reuse)
pattern sinkActivityNames(Name) {
Activity(A);
Activity.name(A, Name);
neg find hasOutEdge(A);
}
NegaOve
composiOon
(negaOon,
quanOficaOon)
19

20. EMF-­‐INCQUERY
Development
Tools

21. EMF-­‐INCQUERY
Development
Tools
Pajern
Editor
EMF
Instance
Model
Query
Explorer

22. EMF-­‐INCQUERY
Development
Tools
Pajern
Editor
EMF
Instance
Model
Query
Explorer

23. EMF-­‐INCQUERY
Development
Tools
• Works
with
most
EMF
editors
out-­‐of-­‐the-­‐box
• Reveals
matches
as
selecOon
Pajern
Editor
EMF
Instance
Model
Query
results
recalculated
on-­‐the-­‐fly
Query
Explorer

24. Visualiza<on of Live Model Queries
in EMF-­‐IncQuery

25. Requirements
for
Query
VisualizaOon
Genericity
Incrementality
• MulOple
Model
Sources
• Query
Changes
• Model
Editors
• Model
Changes
• Current
SelecOon
Traceability
PresentaOon
• Query
DefiniOons
• Filtering
• Input
Model
• Grouping
25

26. Proposed
Architecture
of
Query
Visualizer
Edited
Model
Edited
Query
Other
Model
Query
Editors
Sources
Connectors
Model
and
Filtered
group
of
NoOficaOons
queries
Query
Query
Repository
Engine
Query
Query
Result
Query
Results
DefiniOon
Deltas
Query
Result
Viewer
26

27. Model
Source
Connectors
for
Genericity
 Different
model
sources
Edited
Edited
Model
Other
Query
Model
o Graphical
(e.g
BPMN,
UML)
Query
Editors
Sources
o Textual
(e.g.
OCL)
Connectors
o Different
implementaOon
Model
and
technologies
Query
NoOficaOons
Model source connectors Engine
 OperaOons
Genericity Incrementality
o Open
model
• Multiple Model Sources • Query
Changes
o Send
noOficaOon
upon
model
• Model
Editors
• Current
Selection
• Model
Changes
change
o Get
current
selecOon
Traceability
• Query
Definitions
Presentation
• Filtering
o Close
editor
• Input
Model • Grouping
27

28. Query-­‐based
Indexing
for
Incrementality
 Update
query
results
Query-­‐based indexing
incrementally
upon
o Model
updates
• Contains
model
elements
accessed
• E.g.
using
the
built-­‐in
editor
by
the
query
• Engine:
handles
this
case
• Populated
by
an
exhausOve
model
internally
traversal
(can
be
slow!)
o Query
updates
• Goal:
Avoid
unnecessary
re-­‐
• E.g.
using
the
query
editor
traversal
• Engine:
rebuilds
internal
model
indexes
Strategy
1:
Generic
model
indexer
Genericity Incrementality
• “Wildcard
mode”:
• Multiple Model Sources • Query
Changes
• Model
Editors • Model
Changes
index
every
model
element
• Current
Selection • Higher
memory
consumpOon
Traceability Presentation Strategy
2:
• Query
Definitions • Filtering Group
iniOalizaOon
• Input
Model • Grouping
• IniOalize
several
pajerns
together
28

29. Traceability
and
NavigaOon
 Maintaining
source
models
Query
Editor
Edited
Edited
Model
Query
Query
Editors
Other
Model
Sources
and
queries
Connectors
 Support
for
navigaOng
to/
highlighOng
Query
Repository
Query
Engine
o Corresponding
query
definiOons
Query
Result
Viewer
• HighlighOng
funcOonality
in
the
query
editor
Genericity Incrementality
o Referenced
model
element(s)
• Multiple Model Sources
• Model
Editors
• Query
Changes
• Model
Changes
• Editor-­‐dependent
• Current
Selection
implementaOon
Traceability Presentation
• Model source connector • Query
Definitions • Filtering
handles
funcOonality
• Input
Model • Grouping
29

30. Traceability
in
Query
Visualizer
Pajern
Editor
EMF
Instance
Model
Query
Explorer

31. Traceability
in
Query
Visualizer
Pajern
Editor
EMF
Instance
Model
Query
Explorer

32. Grouping
and
Filtering
for
PresentaOon
 Define
query
groups
(„related”
queries)
Query
Query
Query
Result
o StaOc
groups
DefiniOons
Results
Deltas
• E.g.
Namespaces
Query
Result
Viewer
o Dynamic
groups
• E.g.
Dependency
based
 Filtering
Genericity Incrementality
o Filter
visible
queries
• Multiple Model Sources • Query
Changes
• Model
Editors • Model
Changes
• Uses
hints
from
developer
• Current
Selection
• Manual
overriding
Traceability Presentation
o Filter
query
results
• Query
Definitions • Filtering
• Input
Model • Grouping
• By
binding
query
parameters
32

33. Result
Viewer
33

34. Result
Viewer
Different
model
Szöveg
sources
34

35. Result
Viewer
Loaded
Queries
35

36. Result
Viewer
Query
Results
36

37. Filtering
Visible
Queries
37

38. Filtering
Visible
Queries
Elements
in
Query
Registry
38

39. Filtering
Visible
Queries
Queries
Loaded
39

40. Query
Result
Filtering
40

41. Query
Result
Filtering
41

42. Conclusions
and
Future
Work
R1.
Genericity
Edited
Model
Edited
Other
Model
R3.
Traceability
Query
Query
Editors
Sources
R3.
Traceability
Connectors
R4.
PresentaOon
Model
and
Filtered
group
of
R2.
Incrementality
NoOficaOons
queries
Query
Query
Repository
Engine
Query
Result
Query
Query
Results
Deltas
DefiniOon
Query
Result
Viewer
R4.
PresentaOon
42

43. Conclusions
and
Future
Work
R1.
Genericity
Edited
Model
Edited
Other
Model
R3.
Traceability
Query
Query
Editors
Sources
R3.
Traceability
Connectors
R4.
PresentaOon
Model
and
Filtered
group
of
R2.
Incrementality
NoOficaOons
queries
Query
Query
Repository
Engine
Query
Result
Query
Query
Results
Deltas
DefiniOon
Query
Result
Viewer
R4.
PresentaOon
Future
Plans
• Analysis-­‐based
grouping
• Graph-­‐based
result
visualizaOon
43