/jiralint

JIRA utilities used in Jenkins jobs

Primary LanguagePython

This repo contains utilities that use Jenkins for early warnings/error detection in jiras and other jira tools.

Contents of this repository

  • README: this file

  • render-testcase.py: takes a report.json to render jira query results into junit.xml test reports

  • *.json: reports used by JBoss Tools team for detecting stale/non-triaged/incorrect jiras

  • bzira.py: utility to make bugs returned by bugzilla query show up in jira to use in sprint-planning.

Steps to use it:

  1. Create a <reportname>.json file with the following format:

     [
 {
   "nofixversion": {
"jql": "(project = JBIDE AND fixVersion is EMPTY AND createdDate <= -7d",
       "description": "Issues have no fix version and was created 7 dags ago. Please triage."
   }
 },
 {
   "nocomponent": {
     "jql": "project in (JBIDE,JBDS) AND component is EMPTY AND resolution is EMPTY",
     "description": "Issue has no component - please triage and make sure it is put in the right component or set of components."
   }
 }
 ]

    Each entry has a report name used for identifying its results and then a jql for the query and finally a description which is added to each jira "error" message.

  2. Test it by running:

    ```
$ python render-testcase.py -u user -p secret -s jiraserver -r <reportname>.json
```

    Now for each report a -test.xml file will be created containing "test failures" for each jira matching the result.

  3. Setup a Jenkins build which executes the above script and enable junit reporting and you got automatic notification and tracking to help improve your jira content.

Bugzilla in Jira for sprint planning

bzira.py is an experiment that takes the issues returned by a bugzilla query to make available in jira for prioritization. Main usecase is for JBoss Tools team to see Eclipse bugzillas together with its planned jiras for sprint planning.

The created issues are purely intended for sprint planning - not for mirroring. This makes the setup and mechanics simpler and require much less issues like double comments etc.

Requirements:

As root, run pip to install new or upgrade existing python modules:

# pip install --upgrade bugzilla bugzillatools python-bugzilla jiraclient jira-python jira jiracli jira-cli requests

If that doesn’t work, see complete list of python modules here: https://gist.github.com/nickboldt/a3ba926186fb6b7442f6/d4297f984871ea8dbd09ad15ac197fb72e709053

To run bzira:

python -i bzira.py -u username -p password

For now this script is hardcoded to staging Jira and a single RHT query.

Updating filters

To add a completely new filter requires a manual step since jira for some obscure reason does not support looking up filters by name.

When a new filter is added you have to do the following:

  1. Add it without an 'id' to the .json file

  2. Run the script, ie. python setup_filters.py -u jiralint -p secret -f filters.json,filters-composite.json

  3. Commit the updated .json file which now will contain the newly created filter

Have fun!