We want to provide a simple yet powerful set of tools to allow ARM developers to easily create/manage archives (including package uploads/builds) and generate images.
Organizations need to generate their own images for testing and evaluation purposes, possibly including software that supports unreleased hardware or software with restricted redistribution rights. To generate such images they need to maintain an integrated set of software packages that can be installed on to their devices.
- Ted wants to generate an image for marvel doves, which require non-free software packages that are not included in Ubuntu but instead are packaged in a marvel PPA. Ted must be able to create a new archive (by branching the Ubuntu archive and marvel's PPA) and generate the image using the new archive.
- A partner is developing a custom netbook UI but need to keep it private until it's finalized, so they use the tools to create a slim archive off of the main Ubuntu archive and do any UI changes (or add new packages) on their new archive, which is only accessible by themselves. The slim archive is made possible (instead of a full one) because it should be binary compatible with the main Ubuntu archive. (Not yet clear whether or not the new archive should be hosted on LP, like a private PPA; probably need to allow it to live in LP or outside)
- YAP (Yet Another Partner) is working on optimizing their new (not-yet-released) chip, but for that they need a version of GCC newer than the one on the Ubuntu archive. They want to create a new (private) archive where they'll upload the new GCC version, but upgrading to that new version of GCC is known to break binary compatibility, so it must be possible for them to easily rebuild all packages using the new GCC.
- YAP also has a separate team working on the UI for a device which will use their new chip, so they want to have yet another archive, based on the one containing the new GCC, where they'll make their UI changes without affecting other users of the archive containing the new GCC. The tools should allow them to do that as well.
- During the development of YAP's latest device, it should be possible for them to easily see the changes done to the upstream archive since both archives diverged. They should also be able to review those changes and pull the ones they want into their archive.
- YAP has finished development of a new device and want to upstream the changes they've done when developing. They need to first see what are those changes, decide which ones should go upstream and submit them. Similarly, they should be able to easily pull some/all changes from the upstream archive.
Some of the features described here will depend on DerivedArchiveRebuild
- Archives can be refered to by their URLs. (I think this is how we're going to tell the tools the archives they'll operate on)
- There are two kinds of new archive that can be created: a 'full' archive which starts out containing every package in its parent archive at the version at the time of creation (the debian/ubuntu relationship is a bit like this) or an archive that starts out empty and expects that a user will have another archive to satisfy most package requests (the relationship between the primary archive and a PPA is like this)
By providing a high-level interface with sensible defaults, these tools will abstract some of the low-level complexity of other utilities/frameworks that require more knowledge to be operated.
Users must be able to run any of the tools on either a Desktop or a Server. Also, the tools must not be tied to Launchpad, although they should probably take advantage of anything provided by Launchpad whenever desirable.
The tools should be able to operate on both local and remote archives. It's possible to make the tools able to write to remote locations (e.g. when creating a new archive), but I'm not sure that's a requirement.
Unlike the archives maintained by Launchpad, the archives created using these tools won't have all the infrastructure (e.g. daemons) responsible for detecting new package uploads and then building/publishing the packages, so the tools themselves will be responsible for doing that whenever they're used to upload a new package.
Other features we'll probably want:
- provide a way to create new archives by branching a subset of ubuntu packages.
- visualize (through a web tool, probably) deltas between archives and their (direct or no?) upstreams/downstreams
- manage these deltas by pulling/pushing changes from/to upstreams
- Must allow either short/adhoc (e.g. upload a new upstream version of gstreamer, rebuild things against it and test that everything still works) or long running experiments (e.g. hardware enablement and the like, where you may take months to get things how you like before dropping it all in to Ubuntu at once when the hardware is announced)
deep hierarchies: common archive -> project archive -> project variant archive etc.
- automatic superseding and merging, to keep the archive in sync with its upstream. Must also be able to stop the automatic superseding/merging at any point (e.g. when freezing for a release).
- Need to be able to freeze an archive so that all uploads are held for approval by the release team.
- a way of enforcing version number rules in a particular archive would be good. this is because we'll have to use fairly odd version numbering in downstream archives, so it'd be nice to have archives enforce their own rules for version numbering.
Most tools will probably have no UI other than their command-line arguments, except for showing deltas between archives, in which case we'll probably want some sort of UI. That UI should also allow to push/pull parts of the delta upstream/downstream.
It's important that we are able to test new features, and demonstrate them to users. Use this section to describe a short plan that anybody can follow that demonstrates the feature is working. This can then be used during testing, and to show off after release. Please add an entry to http://testcases.qa.ubuntu.com/Coverage/NewFeatures for tracking test coverage.
This need not be added or completed until the specification is nearing beta.