Git Product home page Git Product logo

metrics's People

Contributors

florentk avatar kelson42 avatar popolechien avatar rgaudin avatar

Stargazers

 avatar  avatar  avatar  avatar  avatar

Watchers

 avatar  avatar  avatar  avatar  avatar  avatar

metrics's Issues

Tune "Average time to solve"

  • Title should be "Time to solve"
  • Y axis legend should "Average tickets lifetime (days)"
  • Colors should be like in chart "Backlog size"
  • Same bars as "Backlog size" should be used".

[Cocom] Analysis fail with some repositories

I got this error :

2019-12-31 17:40:08 > 2019-12-31 16:40:05,843 - grimoire_elk.elk - ERROR - Error enriching ocean from cocom (https://github.com/kiwix/tools): 'message'
2019-12-31 17:40:08 > Traceback (most recent call last):
  File "/usr/local/lib/python3.5/dist-packages/grimoire_elk/elk.py", line 672, in enrich_backend
    enrich_count = enrich_items(ocean_backend, enrich_backend)
  File "/usr/local/lib/python3.5/dist-packages/grimoire_elk/elk.py", line 452, in enrich_items
    total = enrich_backend.enrich_items(ocean_backend)
  File "/usr/local/lib/python3.5/dist-packages/grimoire_elk/enriched/cocom.py", line 223, in enrich_items
    rich_items = self.get_rich_items(item)
  File "/usr/local/lib/python3.5/dist-packages/grimoire_elk/enriched/cocom.py", line 183, in get_rich_items
    eitem['message'] = entry['message']
2019-12-31 17:40:08 > KeyError: 'message'

with theses repository :

Implement Dashboard with following KPI

KPIS:

  • Size of the backlog (number of tickets with the status open, differentiation of label "bug" from other)
  • Number of tickets opened and closed
  • Size of of code base in lines with (differentiation per language)
  • Number of lines commited (differenciation per contributor)
  • Number of code contributors/committers (in the last period)
  • Average to ticket time to solve (differenciation bugs from others)
  • Number of ticket participants
  • Number of releases (in the last period)

Remarks:

  • All the stats are over time (period of 1 days or 1 month, ...) in x axis.
  • This KPIs are global (all openzim+kiwix repos together), but each KPI should be computable for one or many selected repos.

25 november

Allow to add new "visualisations" in docker image (needed for new KPIs)
Allow to init Overviews panel on container creation from a config file
Create a new "overview" panel with first KPIs:

  • Size of backlog
  • Number of tickets opened and closed
  • Average to ticket time to solve
  • Number of added line & removed lines

9 december

Integrate Graal to docker image
Add the KPI "Size of code base in lines" to owerview panel

16 december

Add other KPIs :

  • Number of code contributors/committers
  • Number of ticket authors
  • Number defined tags

23 december

Allow to easily use the panel interface to choose differentiation :

  • code language
  • label (bug, enhancement ...)
  • organisation (kiwix/openzim)
  • contributor
  • period

To evolve Docker image to integrating Graal module

https://github.com/chaoss/grimoirelab-graal is a Generic Repository AnALyzer for Grimoirelab. He uses among other things coloc or lizard tools.

The actual docker don't include the Graal integration but a first version of this integration is proposed as a fork here (see docker subdirectory) based on the work of inishchith at Google Summer of code 2019.

Graal integration allow to add a new dashboard to show metric :

  • Total line of code (LOC)
  • Total line of comment
  • Total number of function
  • Code complexity
  • Comments per LOC
  • LOC per function
  • Statistics on file extension

"Size of backlog" is wrong

If I show a chart only over the last two years, the current value is around 500 but if I configure the chart to look over the past 5 years then it show around 1000 tickets. This has to be wrong, it should always display the same amount of ticket with the status open at a certain date.

Use a year as default time scale

Current chart default period seems to be last two years. I think last 12 (a year) should be better in an attempt to reduce the number of of information and focus on the essential.

metrics.kiwix.org down

I have configured HTTPs on Sloppy and restart the container... and now no content is available at all.

Use own Dockerfile-full

To add studies in grimoire-elk, I modified files in grimoire-elk repos, then I have fork this repos. To use this fork in docker file, I must to use the build_grimoirelab modified script and re-build full docker image. We choose to rewrite our own Dockerfile instead a Dockerfile based on grimoirelab/full image.

Tunes "Number of tickets opened and closed"

  • Legend should be "opened" and "closed"
  • Colors should be the same as on Github for an open (green) and closed (red) ticket
  • Y axis legend should be "Number of tickets"
  • I want things displayes in a form of curves
  • A third "ratio" curve in black should show the opened - closed (so can be negative)

Export a chart in a picture

svg or png. That would really help to share charts. Would Kibana allow such an option? We have exports in CSV already.

Average time to solve does not look coherent.

A picture is worth a thousand words, they say:
Capture d’écran 2019-12-09 à 15 06 34

Also, we've been solving a lot more tickets recently (particularly on android) which should mechanically drag that average down. But not create a trough.

What does q.* > count mean

On Metrics.kiwix.org I'm currently looking at the Size of backlog chart and I have no idea what 600 is. But we're between it and 500.

Add Number defined tags in the dashboard

Git backend of GrimoireELK do not set the "tags". We need to modify it to allow to create the new visualization " Number defined tags".

Copy item['data']['defs'] to eitem in grimoire_elk/enriched/git.py:get_rich_item

Tune "Total lines of code"

Tune "Size of backlog" chart

The numbers are more realistic now. Here a re few things I would like to do:

Then we should be good I guess with that chart.

Backlog : Allows filter by label

To support filter by label, we need to enhance backlog studie : compute the average for each repos and for Bug and Enhancement label.

As a label can be defined multiple times, we need to define priority : if Bug and Enhancement are define, considere is Bug.

X axis granularity is not the best

It seems the default duration of any point on the X axis is not optimal. Charts display a lot of points/bars because this point (duration) is really short. This is difficult to read the charts because of this.

It seems to display a day (even if I believe this an aggreate of values over many days). I don't think having more than 20-30 values on the X axis makes a lot of sense. To me depending of the overall period selected, we should display days/weeks/months smartly (but maybe we choose have somehow a chooser)? Or do we have an alternative approach?

Recommend Projects

  • React photo React

    A declarative, efficient, and flexible JavaScript library for building user interfaces.

  • Vue.js photo Vue.js

    🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.

  • Typescript photo Typescript

    TypeScript is a superset of JavaScript that compiles to clean JavaScript output.

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo D3

    Bring data to life with SVG, Canvas and HTML. 📊📈🎉

Recommend Topics

  • javascript

    JavaScript (JS) is a lightweight interpreted programming language with first-class functions.

  • web

    Some thing interesting about web. New door for the world.

  • server

    A server is a program made to process requests and deliver data to clients.

  • Machine learning

    Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

    We are working to build community through open source technology. NB: members must have two-factor auth.

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google ❤️ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.