Difference between revisions of "Equinox/p2/Galileo/Features"
|Line 14:||Line 14:|
; '''Promote the usage of an installer'''
; '''Promote the usage of an installer'''
: In order to promote bundle pooling, we need to promote the usage of an installer and make it more robust than it is today. Priority:
: In order to promote bundle pooling, we need to promote the usage of an installer and make it more robust than it is today. Priority: .
; '''Tighter desktop integration'''
; '''Tighter desktop integration'''
Revision as of 15:43, 10 September 2008
This page highlights the features and function intended to be worked on for p2 during the Galileo release. The general themes for this release are robustness, ease of use, more direct value-add to our users, and be a better platform. Readers interested in additional details can consult the 3.5 bug list. See also the p2 Milestone Plan
Note that this document evolved from the 3.5 contributions page as it was on September 5 2008.
- Improve overall usability
- Review the overall usability of the current UI. See also p2 UI plan. Priority: 1.
- Improve the error messages provided in case of inability to install. Priority: 1. Bug 200380
- Allow for selection of the optional pieces. Priority: 2.
- Closer integration with the VM
- Provide a way to change the VM being used by Eclipse through the UI and thus trigger the appropriate provisioning actions. Priority: 1.
- Promote the usage of an installer
- In order to promote bundle pooling, we need to promote the usage of an installer and make it more robust than it is today. Priority: 2.
- Tighter desktop integration
- Provide ways to better integrate an Eclipse based application with the desktop (e.g. desktop shortcuts, registry keys...). Priority: 2.
- Better shared install
- Improve the management of shared installs to allow for the end user to not lose the extensions installed. Priority: 2.
- Browser-based installation
- Installation of new function can be triggered by users clicking on a link in a browser. Priority: 3. Bug 246628
- Ease the installation of extension and bridges plug-ins
- p2 should make it easier for bridge plug-ins to be found (e.g. The Mylyn/CDT integration). Priority: 2.
- p2 should make it easier for extension plug-ins to be found (e.g. Webserver adaptors for WTP). Priority: 2.
- Ease the management of complementary items
- Provide a way to change language in the UI (and always automatically install the language pack). Priority: 2.
- Provide a way to install doc automatically when available. Priority: 2.
- Provide a way to install source for the installed bundles. Priority: 2.
- Installation duplication
- Provide a way to create a new install from the existing one. Priority: 2.
- Recovery application
- Provide a recovery application allowing the user to revert to a previous well-known state in case where the installation has been left in a broken state as a result of the last installation. Priority: 3.
- Profile interchange
- Profiles can be manipulated and exchanged between users. This allows previous setups to be stored and recreated and for users to exchange profiles. Priority: 3
- Download integrity through MD5/SHA1 and signature verification
- The integrity of downloaded artifacts can be verified using MD5/SHA1 hashing algorithms and/or signature verification. Priority: 1
- Robustness / responsiveness / user friendliness
- p2 needs to be robust when downloading from misbehaving servers and still provide an overall good responsiveness. Also it needs to provide more precise messages wrt download failures (e.g. distinguish between missing file, unknown host, no connectivity, etc.). Priority: 1.
- In order to ease some configuration changes (e.g. VM, locale, etc.), p2 should allow for some artifacts to be pre-downloaded.
- Improve Adaptive downloads and mirror selection
- p2 currently has a simple adaptive download scheme. This could be improved by taking into account aspect of the download such as artifact type available, artifact processing time, download speed, etc. Priority: 2
- Restart from partial downloads
- In presence of very large artifacts, p2 needs to be able to restart a download where it left off. Priority: 2.
- Download time estimation
- Estimation of the download time as the download progresses. Priority: 2
- Media support
- p2 supports and properly manages the interaction with repositories stored on removable and volume-oriented media such as CDs, DVDs. Priority: 2.
Update Manager replacement
- Replace update manager
- In addition of installing and configuring plug-ins, update manager constructs were being used by other components to drive certain aspects of the UI, the branding, primary plug-in, etc. p2 needs to provide a replacement for these constructs. Priority: 1. Bug 196318
- Mechanism to execute arbitrary code
- UM provided with install handlers a way to execute arbitrary code. p2 replaced this with the concept of touchpoints however lost some flexibility there. We need to reinstate this flexibility whether it is by automatic provisioning of touchpoints or by having an escape mechanism. Priority: 1.
- p2 will graduate some of its provisional API into final APIs. Priority: 1.
- Review touchpoint contribution model
- The touchpoint contribution model needs to be reworked to facilitate the addition of touchpoints. Priority: 1. [ https://bugs.eclipse.org/bugs/show_bug.cgi?id=203323 Bug 203323]
- Reaction to configuration changes
- When some configuration aspects of p2 change, we need to properly react to those. Priority: 1.
- Dynamic provisioning of missing touchpoints
- p2 should allow for touchpoints to be dynamically downloaded before the installation starts. Priority: 2.
- Making the agent fully dynamic
- The p2 code base needs to be able to react to parts of its code coming and going dynamically. Priority: 2.
- UI building blocks and programmatic configurability
- Reusable building blocks of UI needs to be provided to facilitate the creation of other kinds of user interface. Priority: 2.
- Provide configuration option of various p2 default dialogs to limit the coding necessary when using p2 controlled environments. Priority: 2.
- Improve robustness of installation
- Provide infrastructure to help touchpoint author create action that will leave the system in a good state in case of installation failure. Priority: 1.
- Improve the eclipse touchpoint
- Add actions to the touchpoint to handle things like preferences, OSGi framework extensions, etc. Priority: 1.
- Dependency model improvement
- Add metadata construct to allow for a separation of the line-ups information from grouping information. Priority: 2.
- Add metadata constructs allowing the expression of negation and choice. Priority: 3.
- Look at making the model of capabilities / requirements more general. Priority: 3.
- Linkability in metadata
- Explore the addition of more links / "bookmarks" into an IU. Priority: 2.
- Garbage collection
- Need a way to ensure that a bundle pool is only used by one agent to avoid problems with GC. Priority: 1.
- Metadata garbage collector. Priority: 2. [ https://bugs.eclipse.org/bugs/show_bug.cgi?id=218564 Bug 218564]
- GC the download cache. Priority: 1.
- GC the profile registry. Priority: 2.
- Sequenced provisioning
- Users and developers can mandate that various update and install operations must be executed prior to attempting subsequent operations. Priority: 2
- Multiple processes modifying the same profile registry / profile
- Lock the repo when writing from multiple profiles. Priority: 1.
- Multiple process modifying the profile registry and the GC. Priority: 1.
- Profile initialization
- Provide the ability for profiles to be pre-populated with various capabilities and variables (e.g. VM, OS, etc.). Priority: 2.
- Improve overall traceability of the install
- Provide better trace of what is going on during the install (which server files have been downloaded, where we unzipped files, etc.). Priority: 1.
- Improve test suite
- We need to have a rich automated test suite as well as detailed smoke tests. Also we should encourage the eclipse team to use p2 for each I-build. Priority: 1.
- Repository enhancements
- Improve the model with which repositories are associated with each others. Priority: 1.
- Ability to do composite repositories. Priority: 1.
- Provide the ability for repositories announce themselves and also be discovered by p2. Priority: 3.
Tooling / plug-in developer
- Repository browsers and editors
- p2 tooling includes browsers and editors for the artifact and metadata repositories. Users can view, add and remove elements from local and remote p2 repositories. Priority: 2.
- Repository tooling
- Artifact and metadata repositories can be duplicated in whole or in part using a set of tools included in p2. Priority: 1.
- Provide IU / artifact correlation tool to check if all the artifacts of an IU are available in an artifact repository, to remove all the artifacts related to an IU, etc. Priority: 2.
- Metadata Authoring
- p2 tooling will offer the ability to author installable units. Priority:2.