bountysource / bountysource/core

Bugzilla query searches/imports seem to be strangely getting actual issues

Open
#814 1 comment 0 reactions 0 assignees View on GitHub
Dominant language
Ruby
Stars
643
Forks
199
PR merge metrics
No merged PRs in 30d

Description

In short bug count don't seem to be the same as the queries or the logical redirect of a query.
When I enter a specific product query for bugzilla, this should most likely be added as tracker with status=**all** or suggest to add all component as trackers seperately.

There needs to be more control over what queries resulted in a tracker and to merge and to split trackers and queries, at least for team owned ones.

For example a query for the product clang or any more specific query always goes to the same tracker, but not with the actual 5000+ bugs, but with a strange count of 102 bugs. (http://llvm.org/bugs/describecomponents.cgi?product=clang)

Another example with polly I tried adding the query "http://llvm.org/bugs/buglist.cgi?query_format=specific&component=polly&bug_status=__all__&product=Projects" which results in 97 bugs, but it get's rewritten to the query: http://llvm.org/bugs/buglist.cgi?product=Projects which should actually hold even more bugs, but only 31 are available on the bountysource platform.

So as a simple solution matching trackers to actual queries as
http://llvm.org/bugs/buglist.cgi?product=Projects&component=polly&bug_status=__all__ -> polly
and
http://llvm.org/bugs/buglist.cgi?product=clang&component=frontend&bug_status=__all__ -> clang
could work.

Additionally we should make it possible to merge or split trackers.
Preventing duplication and too detailed queries needs to be considered to.

(Another idea: Let teams verify and own trackers and let these trackers overwrite/block queries. These trackers could start with the most unspecific query and after that be split into multiple queries/trackers later on...)
##

Want to back this issue? **[Post a bounty on it!](https://www.bountysource.com/issues/2681564-bugzilla-query-searches-imports-seem-to-be-strangely-getting-actual-issues?utm_campaign=plugin&utm_content=tracker%2F47&utm_medium=issues&utm_source=github)** We accept bounties via [Bountysource](https://www.bountysource.com/?utm_campaign=plugin&utm_content=tracker%2F47&utm_medium=issues&utm_source=github).

Contributor guide

Open the contributing guide

Research direction

Start by tracing the Bugzilla query import and tracker-matching flow, then reproduce the clang and polly example URLs and compare their returned counts with the platform's trackers. Done should include a defined approach for preserving query filters and a clear, tested policy for merging, splitting, ownership, and duplicate prevention.

Written by the indexing model from the issue text.

Assessment

Tech stack
ruby
Domain
backend
Issue type
Bug
Difficulty
5/5
Estimated time
Over a week
Activity status
Stale
Clarity
Needs clarification
Newbie friendliness
20/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.