Notice: This Wiki is now read only and edits are no longer possible. Please see: https://gitlab.eclipse.org/eclipsefdn/helpdesk/-/wikis/Wiki-shutdown-plan for the plan.
Difference between revisions of "COSMOS API Cleanup"
(→COSMOS API Cleanup - Housecleaning) |
|||
Line 43: | Line 43: | ||
<td> | <td> | ||
org.eclipse.cosmos.dc.birt.helper | org.eclipse.cosmos.dc.birt.helper | ||
+ | <b>This should reside under 'data-reporting' instead of 'data-collection' | ||
</td> | </td> | ||
<td><ul><li><font color="orange">org.eclipse.cosmos.dc.birt.helper</font></li> | <td><ul><li><font color="orange">org.eclipse.cosmos.dc.birt.helper</font></li> | ||
Line 525: | Line 526: | ||
<td> | <td> | ||
<ul> | <ul> | ||
− | <li><font color="red">org.eclipse.cosmos.example.mdr.web</font></li> | + | <li><font color="red">org.eclipse.cosmos.example.mdr.web</font> |
+ | <br/> | ||
+ | <b>This should reside in the 'Example' module</b> | ||
+ | </li> | ||
</ul> | </ul> | ||
</td> | </td> | ||
Line 532: | Line 536: | ||
</tr> | </tr> | ||
</table> | </table> | ||
− | + | ||
− | + | <h2>Data Visualization</h2> | |
<table border="1" width="100%"> | <table border="1" width="100%"> | ||
<tr> | <tr> |
Revision as of 12:03, 14 April 2008
Contents
COSMOS API Cleanup - Housecleaning
This wiki page was constructed in response to https://bugs.eclipse.org/bugs/show_bug.cgi?id=218558. In preparation for releasing COSMOS 1.0, committers are required to ensure API guidelines are met. In COSMOS there are three types of APIs:
- Internal - Indicated by having a package name that contains the keyword 'internal' (e.g. org.eclipse.cosmos.dc.cmdbf.services.internal)
- Provisional - Indicated by having the annotation '@provisional' included in the class comment
- Public - Any API that does not fall under (1) or (2)
COSMOS 1.0 does not plan to release any public APIs. All classes are expected to fall under (1) or (2). Provisional classes MUST be thoroughly commented. A recent (April 4th, 2008) manual code scan revealed the following plug-ins to be in violation of the guideline. Package/classes expected to be provisional are indicated by orange and expected internal APIs are indicated by red. The final verdict of what should be internal or provisional is made by the component owner.
<Ruth>
- Ali, can we look into Bill's bugzilla for this as well? He talks about deprecating any bundles and code that are no longer used, so we should identify those and delete them rather than comment them. 220776
- Question: Looks like COSMOS does not use experimental APIs in COSMOS. What are the COSMOS guidelines for using provisional vs experimental? Eclipse Quality Guidelines
</Ruth>
Data Collection
Violating Plug-ins | Packages/Classes | Assignee | Complete |
---|---|---|---|
org.eclipse.cosmos.common |
|
Hubert Leung | No |
org.eclipse.cosmos.dc.birt.helper This should reside under 'data-reporting' instead of 'data-collection' </td> |
</td> | Sheldon Lee-Loy</td> | <b>No |
org.eclipse.cosmos.dc.broker |
|
Hubert Leung | No |
org.eclipse.cosmos.dc.broker.client |
|
Hubert Leung | No |
org.eclipse.cosmos.dc.common |
|
Joel Hawkins | No |
org.eclipse.cosmos.dc.dataManager |
|
Hubert Leung | No |
org.eclipse.cosmos.dc.dataManager.client |
|
Hubert Leung | No |
org.eclipse.cosmos.dc.dataManager.common |
|
Hubert Leung | No |
org.eclipse.cosmos.dc.local.domain |
|
Joel Hawkins | No |
org.eclipse.cosmos.dc.local.registry |
|
Joel Hawkins | No |
org.eclipse.cosmos.dc.logging.datamgr (Is this necessary?) |
|
Hubert Leung | No |
org.eclipse.cosmos.dc.logging.datamgr.client |
|
Hubert Leung | No |
org.eclipse.cosmos.dc.logging.datamgr.common |
|
Hubert Leung | No |
org.eclipse.cosmos.dc.logging.datamgr.web |
|
Hubert Leung | No |
org.eclipse.cosmos.dc.management.domain |
|
Hubert Leung | No |
org.eclipse.cosmos.dc.management.domain.client |
|
Hubert Leung | No |
org.eclipse.cosmos.dc.mdr |
|
Hubert Leung | No |
org.eclipse.cosmos.dc.mdr.client |
|
Hubert Leung | No |
org.eclipse.cosmos.dc.mdr.common |
|
Hubert Leung | No |
org.eclipse.cosmos.dc.mgmt.common (Is this necessary?) |
|
Joel Hawkins | No |
org.eclipse.cosmos.dc.mgmt.wsdm (Is this necessary?) |
|
Joel Hawkins | No |
org.eclipse.cosmos.dc.remote.wsdm (Is this necessary?) |
|
Joel Hawkins | No |
org.eclipse.cosmos.dc.sample.components |
|
Joel Hawkins | No |
org.eclipse.cosmos.dc.sample.configurations (Is this necessary?) |
|
Joel Hawkins | No |
org.eclipse.cosmos.dc.sample.jpa (Is this necessary?) |
|
Joel Hawkins | No |
org.eclipse.cosmos.dc.spec |
|
Joel Hawkins | No |
org.eclipse.cosmos.dc.stat.datamgr (Is this necessary?) |
|
Hubert Leung | No |
org.eclipse.cosmos.dc.stat.datamgr.client |
|
Hubert Leung | No |
org.eclipse.cosmos.dc.stat.datamgr.common |
|
Hubert Leung | No |
org.eclipse.cosmos.dc.stat.datamgr.web |
|
Hubert Leung | No |
org.eclipse.cosmos.dc.tptp.agentcontroller |
|
Hubert Leung | No |
org.eclipse.cosmos.dc.tptp.gla.components |
|
Hubert Leung | No |
org.eclipse.cosmos.example.mdr.web |
|
Hubert Leung | No |
Data Visualization
Violating Plug-ins | Packages/Classes | Assignee | Complete |
---|---|---|---|
org.eclipse.cosmos.dr.drs.service.handler |
|
Sheldon Lee-Loy | No |
org.eclipse.cosmos.dr.ps.common |
|
Sheldon Lee-Loy | No |
Management Enablement
Violating Plug-ins | Packages/Classes | Assignee | Complete |
---|---|---|---|
org.eclipse.cosmos.me.dmtoolkit.common |
|
David Whiteman | No |
org.eclipse.cosmos.me.dmtoolkit.j2ee |
|
David Whiteman | No |
org.eclipse.cosmos.me.management.common |
|
Joel Hawkins | No |
org.eclipse.cosmos.me.management.wsdm |
|
Joel Hawkins | No |
Resource Modeling
Violating Plug-ins | Packages/Classes | Assignee | Complete |
---|---|---|---|
org.eclipse.cosmos.rm.sml.mdr (Is this necessary?) |
|
Hubert Leung | No |
org.eclipse.cosmos.rm.sml.mdr.web |
|
Hubert Leung | No |
Open Questions/Concerns
Harm Sluiman:
It seems the package provisional idea is being skipped and that is the most common form of provisional in Eclipse. Note TPTP introduced the concept of javadoc provisional only as a transient state where api was stable but behavior was not and testing or documentation was not complete. If you only have one provisional state, I strongly suggest you use the package variant.
Will be discussed in the next architecture call