Slide 1
Slide 1 text
SUPRIYA SRIVATSA | SOFTWARE ENGINEER | @SUPRIYASRIVATSA
Forgotten Histories
Hidden Insights Lying Within Your Git Repository
Slide 2
Slide 2 text
Codebase should be viewed
not as the ultimate truth
Slide 3
Slide 3 text
Codebase should be viewed
not as the ultimate truth, but
as a story of evolutions.
Slide 4
Slide 4 text
No content
Slide 5
Slide 5 text
No content
Slide 6
Slide 6 text
There is more to code than code.
Slide 7
Slide 7 text
There is more to code than code.
What
Slide 8
Slide 8 text
When
There is more to code than code.
Who
What
Slide 9
Slide 9 text
When
A time dimension
Who
A social/organisational dimension
There is more to code than code.
Slide 10
Slide 10 text
Tech debt is not just technical.
Slide 11
Slide 11 text
Behavioural Code Analysis
Slide 12
Slide 12 text
No content
Slide 13
Slide 13 text
No content
Slide 14
Slide 14 text
No content
Slide 15
Slide 15 text
No content
Slide 16
Slide 16 text
No content
Slide 17
Slide 17 text
BRING OUT YOUR CODEBASE!
Slide 18
Slide 18 text
BRING OUT YOUR CODEBASE!
Slide 19
Slide 19 text
BRING OUT YOUR CODEBASE!
Slide 20
Slide 20 text
http://adamtornhill.com/code/maatdistro.htm
GET CODE MAAT
Slide 21
Slide 21 text
GIT LOGS
$ git log --all --numstat --date=short --pretty=format:'--%h--
%ad--%aN' --no-renames --after=YYYY-MM-DD > log.log
Slide 22
Slide 22 text
CHANGE FREQUENCY
$ java -jar code-maat-1.0-standalone.jar -c git2 -l log.log -a
revisions -r 50 > .csv
git log --format=format: --name-only --after=2019-12-01 | sort | uniq
-c | sort -r | head -50
Slide 23
Slide 23 text
No content
Slide 24
Slide 24 text
Halstead Complexity
Measuring Complexity
Cyclomatic
Complexity
Slide 25
Slide 25 text
Halstead Complexity
Measuring Complexity
Lines of code,
Indentation based
complexity
Cyclomatic
Complexity
Slide 26
Slide 26 text
Most active parts of code are
often most complex parts of
code.
Slide 27
Slide 27 text
- Low cohesion
Most active parts of code are
often most complex parts of
code.
Slide 28
Slide 28 text
- Low cohesion
- Developer congestion
Most active parts of code are
often most complex parts of
code.
Slide 29
Slide 29 text
- Low cohesion
- Developer congestion
REFACTOR HOTSPOT
Most active parts of code are
often most complex parts of
code.
Slide 30
Slide 30 text
HOTSPOT
A complicated piece of code you have to work with often
Slide 31
Slide 31 text
Change Coupling
Files that change together
Slide 32
Slide 32 text
CHANGE COUPLING
- The files change together in at least X commits.
- The files change together in at least Y% of the total
commits to each file.
Slide 33
Slide 33 text
CHANGE COUPLING
$ java -jar code-maat-1.0-standalone.jar -c git2 -l log.log -a
coupling -r 30 > .csv
Slide 34
Slide 34 text
Refactor
Hotspots
Change
Coupling
Slide 35
Slide 35 text
Refactor
Hotspots
Change
Coupling
Code Age
Slide 36
Slide 36 text
Refactor
Hotspots
Change
Coupling
Code Age Knowledge
Maps
Slide 37
Slide 37 text
No content
Slide 38
Slide 38 text
No content
Slide 39
Slide 39 text
No content
Slide 40
Slide 40 text
No content
Slide 41
Slide 41 text
No content
Slide 42
Slide 42 text
No content
Slide 43
Slide 43 text
Software Design X-Rays
Adam Tornhill
Slide 44
Slide 44 text
Supriya Srivatsa
@SupriyaSrivatsa