You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@cayenne.apache.org by Alex Kolonitsky <ak...@objectstyle.com> on 2015/02/20 09:28:44 UTC
[VOTE] 4.0.M2
Hi All,
I've prepared 4.0.M2 artifacts for voting:
Maven artifacts: https://repository.apache.org/content/repositories/orgapachecayenne-1003 <https://repository.apache.org/content/repositories/orgapachecayenne-1003>
Assemblies: https://dist.apache.org/repos/dist/dev/cayenne/ <https://dist.apache.org/repos/dist/dev/cayenne/>
Please evaluate and cast your votes.
Regards,
Alex Kolonitsky.
Re: [VOTE] 4.0.M2
Posted by Andrus Adamchik <an...@objectstyle.org>.
Hijacking the vote thread for a moment, I should mention how glad I am that we are finally making this release. There's so much new modern API stuff, new cdbimport, etc., etc. We need to make it available to people who wouldn't normally build Cayenne from source :)
Andrus
> On Feb 20, 2015, at 11:28 AM, Alex Kolonitsky <ak...@objectstyle.com> wrote:
>
> Hi All,
> I've prepared 4.0.M2 artifacts for voting:
>
> Maven artifacts: https://repository.apache.org/content/repositories/orgapachecayenne-1003 <https://repository.apache.org/content/repositories/orgapachecayenne-1003>
> Assemblies: https://dist.apache.org/repos/dist/dev/cayenne/ <https://dist.apache.org/repos/dist/dev/cayenne/>
>
> Please evaluate and cast your votes.
>
> Regards,
> Alex Kolonitsky.
Re: [VOTE] 4.0.M2
Posted by Alex Kolonitsky <ak...@objectstyle.com>.
Thanks Andrus
I've fixed issues and rebuild 4.0.M2 artifacts:
Maven artifacts: https://repository.apache.org/content/repositories/orgapachecayenne-1004 <https://repository.apache.org/content/repositories/orgapachecayenne-1004>
Assemblies: https://dist.apache.org/repos/dist/dev/cayenne/4.0.M2 <https://dist.apache.org/repos/dist/dev/cayenne/4.0.M2>
Please evaluate and cast your votes.
Regards,
Alex Kolonitsky.
> On Feb 23, 2015, at 8:21 PM, Andrus Adamchik <an...@objectstyle.org> wrote:
>
> I just went through release validation checklist. Signatures and checksums match. Source build and OS X binary build both appear valid. Maven repo works. Now Rat... I ran rat.sh from an unpacked source distro. Noticed a small problem (see the attached report) :
>
> 1. Rat fails on a bunch of IDEA .iml files. I guess we just add .iml to ignores going forward. So this is NOT a show stopper. Though I guess ideally we should exclude IDE-generated files from the source assembly (and yes, I checked 3.2M1, and sure enough it had Eclipse project files that got there from my environment).
>
> 2. Also Rat fails on a few test cases and a test SQL file.
>
> ./cayenne-server/src/test/java/org/apache/cayenne/testdo/unsupported_distinct_types/Customer.java
> ./cayenne-server/src/test/java/org/apache/cayenne/testdo/unsupported_distinct_types/Product.java
> ./cayenne-server/src/test/java/org/apache/cayenne/testdo/unsupported_distinct_types/UnsupportedDistinctTypes.java
> ./plugins/maven-cayenne-plugin/src/test/resources/org/apache/cayenne/tools/dbimport/testOneToOne.sql
>
> These we need to fix. So I am temporarily -1. Let's add those licenses and re-publish the release (and while we are at it, we might fix #1 as well).
>
> Andrus
>
>
> <report.txt>
>> On Feb 20, 2015, at 11:28 AM, Alex Kolonitsky <ak...@objectstyle.com> wrote:
>>
>> Hi All,
>> I've prepared 4.0.M2 artifacts for voting:
>>
>> Maven artifacts: https://repository.apache.org/content/repositories/orgapachecayenne-1003 <https://repository.apache.org/content/repositories/orgapachecayenne-1003>
>> Assemblies: https://dist.apache.org/repos/dist/dev/cayenne/ <https://dist.apache.org/repos/dist/dev/cayenne/>
>>
>> Please evaluate and cast your votes.
>>
>> Regards,
>> Alex Kolonitsky.
>
Re: [VOTE - CANCELED] 4.0.M2
Posted by Andrus Adamchik <an...@objectstyle.org>.
Yeah, forgot about this one.
Also as [VOTE] threads are our only legal record of each release, I figured to avoid future confusion, we mark this one as CANCELED (which I did in the subject), and then start a new thread with different Message-ID once we are ready for another vote.
Andrus
> On Feb 25, 2015, at 2:12 PM, Alex Kolonitsky <ak...@objectstyle.com> wrote:
>
> Hi all
> I've just found that we still have dependency on GPL library in tests (org.fluttercode.datafactory:datafactory)
> We postpone voting until I fix this issue and rebuild all release artifacts.
>
>
>> On Feb 25, 2015, at 2:45 AM, Aristedes Maniatis <ar...@maniatis.org> wrote:
>>
>> On 24/02/2015 4:21am, Andrus Adamchik wrote:
>>> These we need to fix. So I am temporarily -1. Let's add those licenses and re-publish the release (and while we are at it, we might fix #1 as well).
>>
>>
>> I think that we perhaps need to pause and understand what it is we are releasing. I appreciate the considerable development work being put in at the moment, particularly by Alex, Savva and yourself. But a lot of that work is not being discussed on this list, so I'm a little confused about exactly where we got to with the new API and other features for 4.0.
>>
>> 1. Are we looking for a milestone release which will contain a fully finished API?
>>
>> 2. What are we asking users to review and give us feedback on?
>>
>> 3. What is our policy on removing bits of the API between 4.0M2 and 4.0M3 if we decide we don't like something?
>>
>> 4. What is the state of the documentation to show users how to use the new API? I took a look and couldn't find anything other than the javadocs.
>>
>> I am all in favour of releasing much more often, but I think we should be clear about what this release is there to achieve.
>>
>> Ari
>>
>>
>>
>> --
>> -------------------------->
>> Aristedes Maniatis
>> GPG fingerprint CBFB 84B4 738D 4E87 5E5C 5EFA EF6A 7D2E 3E49 102A
>>
>
>
Re: [VOTE] 4.0.M2
Posted by Alex Kolonitsky <ak...@objectstyle.com>.
Hi all
I've just found that we still have dependency on GPL library in tests (org.fluttercode.datafactory:datafactory)
We postpone voting until I fix this issue and rebuild all release artifacts.
> On Feb 25, 2015, at 2:45 AM, Aristedes Maniatis <ar...@maniatis.org> wrote:
>
> On 24/02/2015 4:21am, Andrus Adamchik wrote:
>> These we need to fix. So I am temporarily -1. Let's add those licenses and re-publish the release (and while we are at it, we might fix #1 as well).
>
>
> I think that we perhaps need to pause and understand what it is we are releasing. I appreciate the considerable development work being put in at the moment, particularly by Alex, Savva and yourself. But a lot of that work is not being discussed on this list, so I'm a little confused about exactly where we got to with the new API and other features for 4.0.
>
> 1. Are we looking for a milestone release which will contain a fully finished API?
>
> 2. What are we asking users to review and give us feedback on?
>
> 3. What is our policy on removing bits of the API between 4.0M2 and 4.0M3 if we decide we don't like something?
>
> 4. What is the state of the documentation to show users how to use the new API? I took a look and couldn't find anything other than the javadocs.
>
> I am all in favour of releasing much more often, but I think we should be clear about what this release is there to achieve.
>
> Ari
>
>
>
> --
> -------------------------->
> Aristedes Maniatis
> GPG fingerprint CBFB 84B4 738D 4E87 5E5C 5EFA EF6A 7D2E 3E49 102A
>
[DISCUSSION] 4.0.M2
Posted by Andrus Adamchik <an...@objectstyle.org>.
(renamed the thread to separate from VOTE)
> On Feb 25, 2015, at 2:45 AM, Aristedes Maniatis <ar...@maniatis.org> wrote:
> I am all in favour of releasing much more often,
I'd be totally in favor of more frequent releases. Perhaps we need an informal poll to gauge our PMC ability to sustain the voting pace.
> but I think we should be clear about what this release is there to achieve.
[..]
> But a lot of that work is not being discussed on this list, so I'm a little confused about exactly where we got to with the new API and other features for 4.0.
I don't think any major features were done without a list discussion (or an attempt at discussion). But yeah, a single overview should be useful to everyone. I tried to do just that in the upgrade guide [1]. And here is this list again with dev@ links:
* fluent APIs (SQLSelect, SelectById, ObjectSelect) [2, and probably more]
* cayenne-crypto [3]
* cdbimport [4]
* OSGI [5]
* ServerRuntimeBuilder [6]
> 1. Are we looking for a milestone release which will contain a fully finished API?
Fully usable I'd say, but I don't preclude (and in fact expect) further changes in the API. Also there are big gaps in fluent API as a whole (SQLUpdate that we started to discuss, procedure queries, etc.). Also adding aggregate capabilities (GROUP BY) to ObjectSelect. So there's much more work to do. The existing stuff is very clean and useful already though.
> 2. What are we asking users to review and give us feedback on?
The list above.
> 3. What is our policy on removing bits of the API between 4.0M2 and 4.0M3 if we decide we don't like something?
If it wasn't in 3.1, we can just remove it with a notice in UPGRADE-NOTES.txt.
> 4. What is the state of the documentation to show users how to use the new API? I took a look and couldn't find anything other than the javadocs.
The docs are behind. Partially this is intentional, as things keep evolving. Short overview of all the new stuff is present in the upgrade guide PDF though [1].
Andrus
[1] https://github.com/apache/cayenne/blob/4.0.M2/docs/docbook/upgrade-guide/src/docbkx/new-features.xml
[2] http://markmail.org/message/3vcxctfviunaqn4m
[3] http://markmail.org/message/fzxroljpb56yhcmz
[4] http://markmail.org/message/3v53ouhjjml2ukbq
[5] http://markmail.org/message/r6go6k2l2qshcbjj
[6] http://markmail.org/message/6elcafod46yktujk
Re: [VOTE] 4.0.M2
Posted by Aristedes Maniatis <ar...@maniatis.org>.
On 24/02/2015 4:21am, Andrus Adamchik wrote:
> These we need to fix. So I am temporarily -1. Let's add those licenses and re-publish the release (and while we are at it, we might fix #1 as well).
I think that we perhaps need to pause and understand what it is we are releasing. I appreciate the considerable development work being put in at the moment, particularly by Alex, Savva and yourself. But a lot of that work is not being discussed on this list, so I'm a little confused about exactly where we got to with the new API and other features for 4.0.
1. Are we looking for a milestone release which will contain a fully finished API?
2. What are we asking users to review and give us feedback on?
3. What is our policy on removing bits of the API between 4.0M2 and 4.0M3 if we decide we don't like something?
4. What is the state of the documentation to show users how to use the new API? I took a look and couldn't find anything other than the javadocs.
I am all in favour of releasing much more often, but I think we should be clear about what this release is there to achieve.
Ari
--
-------------------------->
Aristedes Maniatis
GPG fingerprint CBFB 84B4 738D 4E87 5E5C 5EFA EF6A 7D2E 3E49 102A
Re: [VOTE] 4.0.M2
Posted by Andrus Adamchik <an...@objectstyle.org>.
I just went through release validation checklist. Signatures and checksums match. Source build and OS X binary build both appear valid. Maven repo works. Now Rat... I ran rat.sh from an unpacked source distro. Noticed a small problem (see the attached report) :
1. Rat fails on a bunch of IDEA .iml files. I guess we just add .iml to ignores going forward. So this is NOT a show stopper. Though I guess ideally we should exclude IDE-generated files from the source assembly (and yes, I checked 3.2M1, and sure enough it had Eclipse project files that got there from my environment).
2. Also Rat fails on a few test cases and a test SQL file.
./cayenne-server/src/test/java/org/apache/cayenne/testdo/unsupported_distinct_types/Customer.java
./cayenne-server/src/test/java/org/apache/cayenne/testdo/unsupported_distinct_types/Product.java
./cayenne-server/src/test/java/org/apache/cayenne/testdo/unsupported_distinct_types/UnsupportedDistinctTypes.java
./plugins/maven-cayenne-plugin/src/test/resources/org/apache/cayenne/tools/dbimport/testOneToOne.sql
These we need to fix. So I am temporarily -1. Let's add those licenses and re-publish the release (and while we are at it, we might fix #1 as well).
Andrus