Bugzilla - Data

A tool for visualizing Bugzilla data via component, product, assginee, qa_contact, etc. Right now, this tool generates bar charts of Bugzilla bugs. The charts can be sorted according to e.g. bug component, assignee, qa_contact, etc. The charts will automatically be sorted from highest to lowest. This is the image generated from the example query yaml (conf/query.yaml).

Getting started

The basics of using this tool is as simple as,

1. Creating and activating a python 3 virtual environment
2. pip install -e .
3. Running python scripts/make_plot.py

The allowed parameters to make_plot.py are:

usage: make_plot.py [-h] [-q QUERY] [-p PLOT] [-u URL] [--save] [--output]
                  [--login] [--credential_file CREDENTIAL_FILE]

optional arguments:
  -h, --help            show this help message and exit
  -q QUERY, --query QUERY
                        Path to query yaml file (default: conf/query.yaml)
  -p PLOT, --plot PLOT  Plot bar chart for BZs found via <query> sorted
                        according to one of: [component, qa_contact,
                        assigned_to, creator] (default: component)
  -u URL, --url URL     Bugzilla URL (default: bugzilla.redhat.com)
  --save                Save the plot (default: False)
  --output              Output bugzilla data from query to stdout (default:
                        False)
  --login               Login to Bugzilla before making query. Required to use
                        e.g. savedsearch and to get some hidden fields.
                        (default: False)
  --credential_file CREDENTIAL_FILE
                        Path to credential yaml file (default:
                        conf/credentials.yaml)

To effectively use this tool, you must define meaningful queries in conf/query.yaml.

You can define any number of queries within you query file. Each simply must start with - query:.

A simple example query which fetches NEW BZ's for Red Hat Cloudforms is provided, but these can be much more complicated. For example, you can fetch according to specific users who created/reported the BZ via

- query:
   product:
       - Red Hat CloudForms Management Engine
   status:
       - ON_QA
       - NEW
       - POST
       - CLOSED
   include_fields:
       - id
       - summary
       - component
       - description
       - status
       - qa_contact
       - creator
       - assigned_to
       - version
   reporter:
       - <email1>
       - <email2>
       - <email3>

You can also search by qa_contact or assigned_to and then generate plots according to these users. More information about the queries can be found at: https://github.com/python-bugzilla/python-bugzilla, as this is the API that is used for bugzilla queries. Note that the include_fields list is not necessary, and can be commented out if you run into attribute errors in the program call. It can speed up your query if you're dealing with a large number of BZs, since python-bugzilla will only fetch those specific fields.

Another neat feature of python-bugzilla is the ability to login to Bugzilla and use a search saved under your login rather than formulating your own query. To do so, create a credentials file in conf/credentials.yaml, with the following content:

- login_info:
    username: <your_bz_username>
    password: <your_bz_password>

For the query, you can have the following in conf/query.yaml:

- query:
    savedsearch: My Bugs # just an example, can be the name of any search defined in your Bugzilla

Then run the script passing the --login so the script knows that a login is required.

python scripts/make_plot.py --login