Sustainable Software

1
Sustainable Software

• Karl Lieberherr
• Northeastern University
• College of Computer Science
• Demeter Group
• Joint work with Mira Mezini, David Lorenz, Linda
  Seiter, Mitchell Wand, Johan Ovlinger, Doug
  Orleans

2
Sustainable Softwarewhat the dictionary says

• To sustain
• to keep up
• to maintain
• to prolong
• to supply (with provisions)
• to support

3
Sustainable Software

• Easy to extend can easily add new code that
  interacts with old code old code can call new
  code. Needs the right hooks that allow
  extensions to be added.
• Easy to evolve can easily replace old code by
  new code to improve system.
• Easy to adapt can easily replace context in
  which code runs with new context.

4
Sustainable Software

• Extending Need the right hooks that allow
  extensions to be added.
• Evolving Need to factor in evolutionary
  replacement.
• Adapting Need to factor in context replacement.

5
Sustainable Software Examples

• Extending Add new alternative (subclass).
• Evolving Replace algorithm by better one.
• Adapting
• Switch from sequential to concurrent platform.
• Switch from single processor to distributed
  multiprocessor implementation.
• Use the same basic business rule with more
  complex business data.

6
From Crista Lopes PhD thesis (NU/Xerox)
Instead of writing this
7
Idea

• How can we make extension, evolution and
  adaptation easier?
• Localize changes!
• But A change might intrinsically influence many
  methods or classes. The code needed for the
  change might be spread over a good fraction of
  the program.
• Solution Allow programmer to refer to many
  points in execution of program in a localized
  manner. Allow additional actions to be executed
  at those points.

8
Two things are important!

• Specify
• additional or modified actions.
• Actions come first!
• when to call those actions.
• Specification may cut across many methods.
  Crosscutting comes second.
• We use separate constructs to specify crosscuts
  and actions on crosscuts.

9
Constructs to use

• UML collaborations (adapted) for expressing
  actions decoupled from crosscuts. New allow
  rewriting of methods.
• Composite adapters to express the crosscutting
  that maps actions to execution points. May
  contain arbitrary Java code to implement required
  interface of collaboration with provided interface

10
Cross-cutting of aspects
better program
ordinary program
Basic classes structure
Concern 1
Slice of functionality
Concern 2
synchronization description
Concern 3
...
11
Example 1 outline

• ReadersWriters pattern
• Have a data structure with read and write methods
  that are used in a multi-threaded environment
• Rules
• at most one writer and no reader
• several readers and no writer

12
Example 1 plan

• Describe synchronization pattern as a UML
  (Unified Modeling Language) collaboration.
• Describe instantiation of pattern using an
  adapter.

13
Example 1 collaboration

• collaboration ReadersWriters
• protected int activeReaders_ 0 ...
• participant DataToProtect
• expect Object read(Object args)
• expect void write(Object args)
• replace Object read(Object args)
• beforeRead() Object r expected(args)
• afterRead() return r
• replace write(Object args)
• beforeWrite() expected(args)
• afterWrite()

14
Example 1 collaboration

• protected boolean allowReader()
• return waitingWriters_ 0
• activeWriters_ 0
• protected boolean allowWriter()
• return activeReaders_0
• activeWriters_ 0
• protected synchronized void beforeRead()
• waitingReaders_
• while (!allowReader())
• try wait() catch (...)
• -- waitingReaders_ activeReaders_
  ...

15
Example 1 adapter

• adapter ReadersWritersUse
• MyHashtable is ReadersWriters.DataToProtect
• with
• read clone(), get(Object),
• contains(Object), elements(), isEmpty(),
• ...
• write clear(), put(Object,Object),
• remove(Object), ...

16
Example 1 discussion

• Simple case one class only. Typical case one
  collaboration affects many classes.
• One collaboration may affect program at many
  different join points. Collaboration localizes
  many changes cutting across one or several
  classes.
• Adapter describes the crosscutting.

17
Example 2 Problem with synchronization code it
is tangled with basic code

• // Buffer with in() and out()
• class BoundedBuffer
• Object array
• int inPtr 0, outPtr 0
• int usedSlots 0
• BoundedBuffer(int capacity)
• array new Objectcapacity

18
Example 2 Tangling
synchronized void in(Object o) while
(usedSlots array.length) try wait()
catch (InterruptedException e)
arrayinPtr o inPtr (inPtr 1 )
array.length usedSlots if (usedSlots1)
notifyall() ...
19
Example 2 Bounded Buffer Pattern

• Describe the bounded buffer pattern ready to be
  instantiated many times
• introduce domain specific language for
  coordination
• exclusion sets, method managers
• Later implementation Translate the bounded
  buffer pattern into a collaboration at Java level

20
Example 2 Coordinator
coordinator BoundedDataStructure selfex put,
take mutex put, take condition emptytrue,
fullfalse expect void put (Object o)
expect Object take() expect int used()
expect int length()
exclusion sets
coordinator variables
21
Example 2 Coordinator continued
method managers with requires clauses and
entry/exit clauses
put requires (!full) on exit
emptyfalse if (used()length())
fulltrue take requires (!empty) on
exit fullfalse if (used()0)
emptytrue
22
How to use a coordinator

• Two steps
• Define a basic Buffer class that allows us to
  implement the required interface of the
  coordinator
• Provide an adapter that gives the details of the
  implementation of the required interface of the
  coordinator with the provided interface of the
  Buffer

23
How to use an aspect
Enhanced classes
aspect
Aspect deployment
classes
Adapter
24
Example 2 Basic Buffer
class Buffer //declarations public void
in(Object o) arrayputPtr o inPtr
(inPtr1)array.length usedSlots
public Object out() Object old
arraytakePtr arraytakePtr null
outPtr (outPtr1)array.length
usedSlots-- return old
25
Example 2 Use the coordinator with basic Buffer

• adapter BoundedDataStructureToBuffer
• Buffer is BoundedDataStructure
• with // adaptation body in Java
• void put (Object o) in(o)
• Object take() return out()
• int used() return usedSlots
• int length() return array.length

26
Example 2 Many uses of coordinator

• The coordinator can be used with many different
  data structures
• different kinds of bounded buffers
• different kinds of bounded stacks

27
Use the coordinator with basic BoundedStack

• adapter BoundedDataStructureToMyStack
• BoundedStack is BoundedDataStructure
• with
• void put (Object o) push(o)
• Object take() return pop()
• int used() return size()
• int length() return limit()

28
How to use an adapter

• lt BoundedDataStructureToBuffer gt b
• turns the ordinary buffer b into a synchronized
  buffer.

29
More control

• If we want direct control how wait and notify are
  used use a generic language for AOP
  collaboration-adapter language (will be part of
  new Demeter language).
• The collaboration-adapter language can also be
  used as intermediate language to implement
  aspects.
• Example follows.

30
Collaboration-adapter language

• We find it to be very expressive for AOP in
  general
• see OOPSLA 98 paper (Mezini/Lieberherr) and
  follow-on technical report

31
Example 2 in collaboration language

• collaboration BoundedDataStructure
• participant D
• expect void put (Object o)
• expect Object take()
• expect int used()
• expect int length()
• protected boolean fullfalse protected
  boolean emptytrue
• replace synchronized void put (Object o)
• while (!full) try wait()
• catch (InterruptedException e)
• expected()
• if (used()1) notifyall()
• emptyfalse
• if (used()length()) fulltrue

32
Collaboration language example continued

• replace synchronized Object take (Object o)
• while (!empty) try wait()
• catch (InterruptedException e)
• Object r expected()
• if (used()length - 1) notifyall()
• fullfalse
• if (used()0) emptytrue return r
• // end participant
• // end collaboration

33
Example 2 Adapter 1 Use the coordinator with
basic Buffer

• adapter BoundedDataStructureToBuffer
• Buffer is BoundedDataStructure.D
• with // adaptation body in Java
• void put (Object o) in(o)
• Object take() return out()
• int used() return usedSlots
• int length() return array.length

34
Example 2 Adapter 2 Use the coordinator with
basic BoundedStack

• adapter BoundedDataStructureToMyStack
• BoundedStack is BoundedDataStructure.D
• with
• void put (Object o) push(o)
• Object take() return pop()
• int used() return size()
• int length() return limit()

35
Adapters

• Implement required interface in terms of provided
  interface.
• The adaptation bodies are written in terms of
  three self variables The environment of the
  participant, the base environment and the adapter
  environment.
• Adapters express the crosscutting.

36
What is an aspect?

• An aspect language is a domain-specific language
  for specifying a collaboration pattern.
• An adapter language is a gluing language to
  instantiate a collaboration pattern and to
  express how the pattern crosscuts the
  application.

37
What is an aspect?

• Adaptation
• one participant may be mapped to a set of
  otherwise not structurally related classes
• two neighboring participants may be mapped to
  classes that are far apart (many intermediate
  classes)
• Adaptation crosscuts method/object structure.

38
Problems without AspectsEffects of Uncontrolled
Tangling
During implementation separate higher-level functi
ons are mixed together
During maintenance/evolution individual
collaborations need to be factored out of
the tangled code
39
Collaborations
minimal assumptions on application structure
Participant Graph
P1
P3
P2

expected interfaces
Behavior Definition
P
P1
add new functionality enhance the expected
provided everything declared public

...
written to the PG similar to an OO
program is written to a concrete class graph

P3
...
40
Effects of scattering
bad, because the unit of reuse is generally not
a class, but a slice of behavior affecting
several classes

• tangled code
• difficult to reason about
• why is this here?
• what does this connect to?
• difficult to change
• difficult to evolve
• code duplication

essentially, lack of modularity
41
The intuition behind aspects/adapters
modification
Woven code
expected
provided
adapters
Concrete application
42
A simple multi-class collaboration

• Solve simple counting problems
• Define Count collaboration and use it twice
• Demonstrates concept of adaptive programming used
  in Demeter.
• Aaptive programming is good to express certain
  kinds of crosscuts in robust way
• Example of a functional aspect

43
Example 3 Count Collaboration
collaboration Count participant
Source expect TraversalGraph getT()
// new TraversalGraph(classGraph, //
new Strategy(from Source to Target)) public
int count () // traversal/visitor weaving
getT().traverse(this, new Visitor() int r
public void before(Target host) r
public void start() r 0 ) //
ClassGraph classGraph new ClassGraph()
44
Example 3 Count Collaboration
participant Target
Use in Bus simulation example Source --
BusRoute Target -- Person
45
Use 1
Count all persons waiting at any bus stop on a
bus route
from BusRoute via BusStop to Person
busStops
BusRoute
BusStopList
buses
0..
BusStop
BusList
waiting
0..
passengers
Bus
PersonList
Person
0..
46
Use 2
count all persons waiting at any bus stop on a
bus route
from BusRoute via BusStop to Person
villages
BusRoute
BusStopList
buses
VillageList
busStops
0..
0..
BusStop
BusList
Village
waiting
0..
passengers
Bus
PersonList
Person
0..
47
Adapter 1

• adapter CountingForBusRoute1
• BusRoute is Counting.Source
• with
• TraversalGraph getT() return
• new TraversalGraph(classGraph1,
• new Strategy(from BusRoute via
  BusStop to Person))
• Person is Counting.Target
• // ClassGraph classGraph new ClassGraph()

48
Adapter 2

• adapter CountingForBusRoute2
• BusRoute is Counting.Source
• with
• TraversalGraph getT() return
• new TraversalGraph(classGraph2,
• new Strategy(from BusRoute via
  BusStop to Person))
• Person is Counting.Target
• // ClassGraph classGraph new ClassGraph()

49
Discussion

• Program (collaboration and adapter) adapts to
  changing class graph
• Collaborations work well both for non-functional
  aspects (like synchronization) as well as
  functional aspects (like counting)

50
Design Choices

• Abstract crosscuts

• Crosscuts in adapters

51
Crosscut

• Crosscut refers to certain times during the
  execution of the program. Examples when certain
  messages are received by certain objects

52
Crosscut

• Set of methods
• all methods with certain signature properties
  attached to a set of classes
• Sequence of methods
• A subgraph of a graph
• traversal through the subgraph
• A path set of a graph

53
Graph-embedding

• XML
• OCL
• Collaboration
• Visitor

54
Graph-embedding

• XML
• explicit navigation
• abbreviated navigation
• OCL
• Collaboration
• Visitor

55
Enterprise Java Beans (EJB) and Aspects

• EJB a Java component technology from SUN/IBM
• Aspects a conceptual tool for the design of
  enterprise Java beans (and other components)

56
Enterprise JavaBeans (EJB)

• Addresses aspectual decomposition.
• An enterprise Bean provider usually does not
  program transactions, concurrency, security,
  distribution and other services into the
  enterprise Beans.
• An enterprise Bean provider relies on an EJB
  container provider for these services.

57
EJB

• Beans
• Containers to manage and adapt the beans.
  Intercept messages sent to beans and can execute
  additional code. Similar to reimplementation of
  expected interface in collaboration.

58
Example EJB persistence

• As an example we consider how persistence is
  handled by EJB containers. The deployment
  descriptor of a bean contains an instance
  variable ContainerManagedFields defining the
  instance variables that need to be read or
  written. This will be used to generate the
  database access code automatically and protects
  the bean from database specific code.

59
Persistence

• collaboration Persistence PerMem p
• participant Source
• expect Target targets
• expect void writeOp()
• // for all targetswriteOp
• participant Target
• expect void writeOp()
• replace void writeOp()
• // write to persistent memory p
• expected()

60
Deployment

• adapter PersistenceConn1
• ClassGraph g // from Company to
• Company is Persistence.Source
• Nodes(g) is Persistence.Target
• with writeOp write
• // must be the same writeOp for both
• // Source and Target

61
Generate deployment descriptor

• Connector contains information about
  ContainerManagedFields
• Connector localizes information it is not spread
  through several classes

62
Demeter explained with XML orhow can XML use
OO/AOP?
Schema (similar to an XML schema) Object
descriptions (e.g., XML documents)
Demeter produces Java classes with basic
capabilities to process descriptions parser,
various kind of visitor classes for printing,
copying, comparing, etc. Java objects
e.g., produced by the parser from object
descriptions.
Behavior (Java with support for traversal/visitor
programming) Synchronization descriptions
(COOL) Remote invocation/data transfer
descriptions (RIDL)
63
Separated concerns in Demeter

• Object structure (.cd)
• Object description (.input)
• Behavior (.beh)
• Navigation (where do you want to go?)
• Synchronization (.cool)
• Remote invocation/data transfer (.ridl)

64
What we want to add to Demeter

• Collaborations (with rewriting)
• Adapters
• to have a general purpose language for AOP that
  supports reuse. Higher-level aspect descriptions
  (such as .cool and .ridl descriptions) will be
  translated to collaborations and adapters.