• Immutable Page
• Info
• Attachments
• More Actions: Raw Text Print View Render as Docbook Delete Cache ------------------------ Check Spelling Like Pages Local Site Map ------------------------ Rename Page Delete Page ------------------------ Subscribe User ------------------------ Remove Spam Revert to this revision Package Pages Sync Pages ------------------------ Load Save SlideShow

Purpose

The /distribution/series/+queue page should enable archive admins to accept, override and reject packages in bulk, providing quick access to all the information necessary for the decision.

? Is part of the purpose also for uploaders to check progress? That would affect the design too.

You can also view the input information regarding this page.

Target user audience

• The primary users of this page are technically very adept Launchpad users / admins who will prefer a console script if it gets the job done faster.

Performance restrictions

To enable a larger set of uploads to be included while bulk-reviewing, the resulting binaries for each upload (with any relevant 'old' publishing info) can be loaded into the page *after* the initial page load. There are several options for how this could be done, the most obvious being to include an 'expand all' option both at the top of the page as well as for each source package release grouping. Another option would be to automatically load in the expanded info as soon as the initial page has loaded, but hopefully that's not necessary?

Use-cases

(Note: these use-cases are with the assumption that we'll be able to group by source package release and order by the latest (max) upload date of the grouped uploads)

Bulk-accepting a bunch of debian imports

Colin opens '/ubuntu/karmic/+queue' and sees that there are currently over fifty source package releases with their related binary uploads displayed in the New queue as well as a number of source uploads. Noticing that nearly all of the packages are debian imports, Colin tabs to 'select all' (the first form input) and hits <SPACE> to select all the binary uploads associated with the listed source package releases. He then tabs down through the list deselecting all the binaries for the two packages that are not debian imports - simply by unchecking the checkbox next to each source package release. He then tabs to the Accept button and hits <ENTER>.

Overriding one or more uploads

While bulk-reviewing the uploads for 50 source package releases, one particular source package release near the bottom of the list catches Martin's attention - he clicks (keys?) 'Expand all' for that source package release which loads in the resulting binaries for all the binary uploads, as well as any previous old publishings of the same binary from other packages. Martin notices that one of the resulting binaries has been published in main in the past via a different source package, and so updates the destination of the new binary to main.

Excluding one binary from the review

Rob opens '/ubuntu/karmic/+queue' and sees that there are quite a few packages requiring closer inspection. He tabs to 'Expand all' which immediately begins expanding the binary uploads for each source package release with the resulting binaries and existing overrides information. Rob works his way down the list, checking the box next to the source package release for some packages (which automatically selects all the binaries) and manually selecting certain binaries for others. When he reaches the bottom of the list, he tabs to the Accept button and hits <ENTER>.

Questions

• I wonder if it would be handy to reject certain binaries inline - so it disappears immediately from the queue? I'm assuming that currently, if someone selects all, then deselects one or two, accepts all the select uploads, they would then need to go manually navigate the list again to select and reject the uploads that they previously deselected.
• Is the workflow substantially different for reviewing source uploads vs binary uploads? Or is it easy to bulk-review a mixed list of source/binary uploads.
• Have there ever been concurrency issues with doing bulk reviewing of the queue (multiple people reviewing the same uploads)? Or perhaps the latest review would just over-write?