Scribd
Upload
10 views22 pages

Exam Preplecture Solution

The document provides information about an exam for a software maintenance and evolution course, including exam format, topics covered, and example questions and answers. Example questions cover definitions of software maintenance, principles of software evolution, refactoring techniques, and metrics for measuring complexity. Additional material provides background on change propagation, architecture erosion, and managing code clones.

Uploaded by

fareehaiftikhar
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
10 views22 pages

Exam Preplecture Solution

The document provides information about an exam for a software maintenance and evolution course, including exam format, topics covered, and example questions and answers. Example questions cover definitions of software maintenance, principles of software evolution, refactoring techniques, and metrics for measuring complexity. Additional material provides background on change propagation, architecture erosion, and managing code clones.

Uploaded by

fareehaiftikhar
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 22

Exam Preparation

Software Maintenance & Evolution


Final Exam Info
• Written Exam
• Same time slot, same room as the lecture
• Covers the entire course
• Be as concise and short as possible, but
no shorter with your answers
• Use non-red permanent ink pencils
• Write down only one single solution
In the following...
• Possible answers are presented
• There are not exhaustive
• There might be other answers or additional
points to consider
• The title of a slide indicates the task number
from the exam, date of the lecture relevant to
answer the questions, and if available slide
numbers
• No warranty!
1a, 22.2 Slides 7, 28
• “the modification of a software product after delivery to
correct faults, to improve performance or other attributes,
or to adapt the product to a modified environment”
• “standard for software life cycle processes depicts
maintenance as one of the primary life cycle processes”

• “as the process of a software product undergoing


modification to code and associated documentation due
to a problem or the need for improvement. The objective
is to modify the existing software product while
preserving its integrity”
1a, 22.2 Slides 29

• Corrective Maintenance
• Preventive Maintenance
• Adaptive Maintenance
• Perfective Maintenance
1a, 22.2 Slide 29
2a, 1.3 & 8.3
• Process of improving the internal structure of the
code
• During this process the external behavior, i.e., the
functionality, of the system does not change
• Part of reverse engineering, e.g., refactor to
understand
• Refactor to test
• Part of reengineering, e.g., remove duplicated code
• Can also occur during daily development work
outside/without a larger reengineering project
1c, 22.02 Paper

• A series of Law’s that describe


principles of software evolution
• Discusses the driving forces behind the
software evolution
• http://www.ifi.uzh.ch/seal/teaching/
courses/SWEvo13/lehman-IEEE-80.pdf
1c, 22.02 Paper
• Law of increasing Entropy: As an evolving
program is continually changed, its complexity,
reflecting deteriorating structure, increases unless
work is done to maintain or reduce it.
• Untangle dependency structure
• Extract Method
• Make function calls simpler
• In general: Refactoring as ‘‘cure’’ from a
deteriorating structure
2a

• Not explicitly mentioned on slides


• Postconditions: Adapt all method calls
with the new method name (if not there
will be compile and build errors)
• Precondition: Code needs to be
compile-able to find all method calls
2b, 22.3
• Does not take into account the control flow
• Does not take into account the dependency
structure
• Depends significantly on the programming
language
• Are 10 methods with 20 LOC as complex as 1
big method with 200 LOC?
• Just a single, plain number: Hard for
developers to deal with them
2b, 22.3

• Better: Metrics based on reference points


or metrics that are independent of
system size (or other individual
characteristics of a system)

• FAN-IN & FAN-OUT


• Cyclomatic Complexity
2c, 1.3
• Changing one module leads to a rat-tail of
changes in other modules
• More defects
• More difficult to understand
• Longer build times
• Changes take longer to implement
• More testing needed
• ...
2d, 22.2 Slide 12 Part II
2e
• Not explicitly mentioned in lecture slides
• Source configuration management tool: Any tool
(suite) that facilitates the management of code
and artifacts: Version control systems, automated
build and test system, continuos integration
• Defect tracker: Database for reporting new
defects, discuss current defects, submit patches, ..
• There are products that integrate both
functionalities
2f, 1.3

• Captures the context of a system


• Can give information about the
rationale of the current implementation
• Can help to identify the important parts
of a system
2g, 22.3
2h 1.3 & 17.5
• Identify experts & code owners: Who are the
active maintainers fixing defects
• Identify the critical parts of a system
• Get overall impression of the state of the
system: How many critical bugs? How long
does it take to fix a bug?
• Software business analyst
• Build defect prediction models to forecast the
location of defects in the next release
3a, 22.3
3e, 3.5 Slides 14-18
Evolution of the number of methods
(NOM) of classes of a system
3e, 3.5 Slides 14-18
4a, 10.5
• In particular see mentioned paper:”Cloning Considered
Harmful” Considered Harmful”
• Pragmatic: If clones do not change, and hence, cause no
additional effort (or defects) do not refactor
• Due to language or framework features: Boilerplate code
• Intentionally created clones: Hardware variation, e.g.,
Linux SCSI Driver -> for each platform code is cloned
and modified
• Inconclusive results from empirical studies: Questioning
at least the general bad reputation of code clones
• There are tools that support clone management

You might also like