Week 42/2013: Migration
Last week I migrated all the bug reports and feature requests into our own tracker.
We could not find a ready-made solution for migrating SF.net tickets into Redmine, so I ended up writing my own script that parsed the JSON data exported from SF.net and inserted it into the new tracker via its REST API and a bunch of direct database manipulations.
I have since been combing through the database of bugs and feature requests and trying to organize and reprioritize them. There is a wealth of good ideas particularly in the feature requests, which I must say personally I've been neglecting to peruse in recent years. I've largely been focusing on the proposals in the wiki, but now that they are being consolidated into the tracker, it is beneficial to go through all the old ideas and see how compatible they are with current thinking.
While I've been rejecting/closing some ideas outright, many of them link nicely into our current plans for Doomsday 2. Much of the work involves clarifying issue titles, updating tags, and linking issues together with related ones. So don't be surprised if you suddenly receive notifications about changes to 10 year old RFEs.
All in all the new Redmine-based tracker is really showing its power. It is fast and one can easily search issues based on keywords, tags, or filtering. In the long run, migrating away from the SF.net trackers will be very beneficial for daily development work, as we'll hopefully have to rely less on private todo lists, the wiki, and other ad hoc mechanisms to manage the work.
My plan is to continue going through the issue database and then end the brief development hiatus by doing some fixes for an upcoming 1.12.1 patch.
We could not find a ready-made solution for migrating SF.net tickets into Redmine, so I ended up writing my own script that parsed the JSON data exported from SF.net and inserted it into the new tracker via its REST API and a bunch of direct database manipulations.
I have since been combing through the database of bugs and feature requests and trying to organize and reprioritize them. There is a wealth of good ideas particularly in the feature requests, which I must say personally I've been neglecting to peruse in recent years. I've largely been focusing on the proposals in the wiki, but now that they are being consolidated into the tracker, it is beneficial to go through all the old ideas and see how compatible they are with current thinking.
While I've been rejecting/closing some ideas outright, many of them link nicely into our current plans for Doomsday 2. Much of the work involves clarifying issue titles, updating tags, and linking issues together with related ones. So don't be surprised if you suddenly receive notifications about changes to 10 year old RFEs.
All in all the new Redmine-based tracker is really showing its power. It is fast and one can easily search issues based on keywords, tags, or filtering. In the long run, migrating away from the SF.net trackers will be very beneficial for daily development work, as we'll hopefully have to rely less on private todo lists, the wiki, and other ad hoc mechanisms to manage the work.
My plan is to continue going through the issue database and then end the brief development hiatus by doing some fixes for an upcoming 1.12.1 patch.
Comments
Why can't website related issues simply be tagged as such in a single tracker?