[
Date Prev][
Date Next][
Thread Prev][
Thread Next][
Date Index][
Thread Index]
[
List Home]
Re: [eclipse-dev] Planning Meeting Notes -May 18, 2005
|
I would like to pre-emptively clarify my previous cryptic note.
The 'Access restrictions' discussion will center around implementation
details, the usage of the runtime's x-friends and x-internal directives,
and a potential list of TODOs for some teams.
The reference to changing 'the face of plug-in development forever' was a
hyperbole.
It refers to the fact that plug-in developers will now be able to enjoy a
superior classpath/content assist experience never seen before. (another
hyperbole, but very true :-)
I will post another note to the mailing list a little later on how things
will work at an implementation level.
I apologize for any confusion my original note may have caused. It was
just a reflection of my excitement over the new technology and should have
been taken as just that.
At this point, I would like to remind everybody of the Eclipse position on
what is accessible and what is not.
It was posted by Jeff McAffer to eclipse-dev a while back. here is a copy
and paste:
>There is still some confusion around the Export-Package list in the
MANIFEST.MF. Allow me to clarify the situation for all Eclipse SDK
plugins.
* ALL Eclipse SDK plugins must list ALL of their packages in their
MANIFEST.MF files.
* API or not. If you have a package on your plugin's classpath, it must
be listed.
* New? List it! Old? List it! Borrowed from a friend? List it!
Exceptions to this model should be taken up with the PMC.
Notice that this puts us in exactly the same situation as we have always
been in. Everyone has the ability to see everything. As we have seen in
the past couple days, the community's desire/need for the freedom to access
non-API is overwhelming. In fact, this forms a critical part of the
Eclipse ecosystem allowing experimentation and investigation.
This approach is NOT at odds with, and is fundamentally separate from, our
ongoing work to get control over the API and help developers stay on "the
right path". That effort is supported by the addition of the component.xml
files, the use of "x-internal" and "x-friends" annotations on export
entries and the PDE/JDT work to more accurately model the runtime classpath
at development time.
All teams please review your Export-Package lists and ensure that all
packages are listed correctly.
Wassim
Melhem/Toronto/IB
M@IBMCA To
Sent by: "General development mailing list
eclipse-dev-bounc of the Eclipse project."
es@xxxxxxxxxxx <eclipse-dev@xxxxxxxxxxx>
cc
05/17/2005 09:58 Subject
PM Re: [eclipse-dev] Planning Meeting
Notes -May 18, 2005
Please respond to
"General
development
mailing list of
the Eclipse
project."
PDE would like to propose a last-minute discussion topic entitled:
"Access Restrictions: the story going forward"
I don't want to overhype this, but it's safe to say that the outcome of
this discussion will change the face of plug-in development forever.
Wassim.
Mike
Wilson/Ottawa/IBM
@IBMCA To
Sent by: eclipse-dev@xxxxxxxxxxx
eclipse-dev-bounc cc
es@xxxxxxxxxxx
Subject
[eclipse-dev] Planning Meeting
05/17/2005 03:28 Notes -May 18, 2005
PM
Please respond to
"General
development
mailing list of
the Eclipse
project."
-------------------
Discussion Topics
-------------------
----------
Outreach
----------
--------
Status
--------
Debug/Ant:
- M7 Test Pass
- M7 Fix Pass
- bug prioritization for RC1
JDT UI:
- shipping M7
- bug triage
- bug fixing
- performance
- released refactoring heap performance test case +
ExtractInterfaceHeapAcceptanceTest
- optimized usage of type hierarchies during rename method and
change signature refactorings
Platform/JDT Text:
- shipped 3.1 M7
- performance:
- investigated and fixed errors in tests (also backported)
- removed 'Open Untitled Text File' from main File menu to File > New
- added access restriction options from JDT Core to code assist and
code assist preference page
- compound edit detection:
- (Java)MoveLines and HippieCompleteAction wrap multiple consecutive
invocations into one Undo operation
- start to work on adding default line delimiter support
- bug fixing
- bug inbox tracking
- 1 public holiday
JDT Core:
- shipping M7
- bug fixing
- combined work with PDE to address issues leading to full build on
startup; if you still see unexpected such full builds again, please
get back to us.
- added timestamp to batch compiler generated XML log
- user libraries can now retain classpath attributes and access rules
- generic method search can now find overrides with different signatures
(still overriding as per type parameter substitution)
Install/Update:
- investigating an issue that is timing-sensitive (JUnit tests fail
randomly). It is caused by our fix that eliminated two GETs that
seem [... missing content? ed.]
- M7 testing
- defect fixes
User Assistance:
- Intro:
- testing various NL issues
- M7 testing
- Defect fixes
- Help:
- M7 testing
- bug fixes
- PDE:
- dusting up ["dusting off"? ed.] 3.1 PDE test plan
- bug prioritization
- M7 testing
- bug fixes
Haircuts:
- Konrad got a haircut
- Janek got a haircut
Rel. Eng.:
- 3.1M7
- performance test rerun on 3.1M6 and 3.1M7
- testing and moving to 3.1M7 in builds
- added Linux GTK PPC configuration to builds
- testing changes in source builds to build doc with eclipse specific
Ant tasks at build time
- working on support in source builds to build Linux GTK IA64 port of
eclipse
- NL fragment tool fixing
Platform Core:
- releasing M7
- RC1 preparation
- buddy class loading is released
- working on the OSGi package rename
- bug triage
- bug fixing
_______________________________________________
eclipse-dev mailing list
eclipse-dev@xxxxxxxxxxx
To change your delivery options, retrieve your password, or unsubscribe
from this list, visit
https://dev.eclipse.org/mailman/listinfo/eclipse-dev
_______________________________________________
eclipse-dev mailing list
eclipse-dev@xxxxxxxxxxx
To change your delivery options, retrieve your password, or unsubscribe
from this list, visit
https://dev.eclipse.org/mailman/listinfo/eclipse-dev