You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@calcite.apache.org by Jesus Camacho Rodriguez <jc...@hortonworks.com> on 2015/11/10 09:37:37 UTC
Post 1.5
Master branch is now open again.
Jesús
Re: Post 1.5
Posted by Jesus Camacho Rodriguez <jc...@hortonworks.com>.
OK, I just saw that I do not have to worry anymore about the javadoc. Thanks Julian! :)
On 11/10/15, 11:27 PM, "Jesus Camacho Rodriguez" <jc...@hortonworks.com> wrote:
>Hi Julian,
>
>I guess it was the excitement about releasing :) Honestly, I did not realize I had to wait 24 hours, sorry about that.
>
>* Thanks for the release notes! Thanks also for the instructions in the HOW-TO: they were easy to follow, and I didn't have any major problem.
>The fact that it was our first release as top-level project didn't change anything; however, the recent changes in the ASF infrastructure did. In particular, I had to generate all the artifacts from my own repository, and then push that to the ASF repository. Should I update the HOW-TO? My understanding is that these restrictions in the ASF infrastructure are temporary, so I do not know if it is worth it. Further, using an external repository (github) to create the release does not seem a very good practice...
>* I agree with your point about the vote over the weekend; I proceeded because committers reacted quickly and most active community members had already casted their vote.
>* About the tag, do you mean the tag of the release (it is there)? Or...? I did not get it.
>* I can regenerate the javadoc (stripping out dates). I'll also push the code to origin/branch-release; I did not see that in the instructions, but it makes sense.
>
>--
>Jesús
>
>
>
>
>
>On 11/10/15, 7:42 PM, "Julian Hyde" <jh...@apache.org> wrote:
>
>>Also:
>>* Can you please add a tag? It looks as if maven didn't add one in offline mode.
>>* How were the instructions in HOWTO? This is the first time anyone
>>has used them post-graduation.
>>* Can you push to origin/branch-release? I always make sure that that
>>points to the latest release. I don't know whether I put that in the
>>instructions.
>>
>>Julian
>>
>>
>>On Tue, Nov 10, 2015 at 10:19 AM, Julian Hyde <jh...@apache.org> wrote:
>>> Many thanks, Jesus, for being release manager. You made everything run
>>> really smoothly. The most impressive thing was that I first found out
>>> about the release from a tweet from a senior Gartner analyst[1],
>>> including a link to a news item!
>>>
>>> I can regenerate the javadoc and publish to the site -- shall I do
>>> that? I'd like to strip out dates from the generated javadoc, to
>>> reduce svn churn.
>>>
>>> A couple of minor nits:
>>>
>>> * If the vote is over a weekend, let's make it a 48 hour vote. (I was
>>> impressed that several of our committers heeded my call and voted on a
>>> Sunday evening, but let's respect people's weekends.)
>>> * The release bits haven't propagated to all of the mirrors yet (e.g.
>>> [2] gives 404). The guideline says to wait 24 hours between pushing to
>>> dist/release and announcing the release.
>>>
>>> Julian
>>>
>>> [1] https://twitter.com/merv/status/664127978746437632
>>>
>>> [2] http://mirrors.ibiblio.org/apache/calcite/apache-calcite-1.5.0
>>>
>>>
>>> On Tue, Nov 10, 2015 at 12:37 AM, Jesus Camacho Rodriguez
>>> <jc...@hortonworks.com> wrote:
>>>> Master branch is now open again.
>>>>
>>>> Jesús
>>
Re: Hoisting literals
Posted by Julian Hyde <jh...@apache.org>.
It isn’t, but it would be really useful. I’ve logged https://issues.apache.org/jira/browse/CALCITE-963 <https://issues.apache.org/jira/browse/CALCITE-963> with an outline of how it could be implemented.
Julian
> On Nov 11, 2015, at 9:33 AM, Homer <lo...@gmail.com> wrote:
>
> Hi,
>
> I am wondering if this is something that has already in calcite but I cannot seem to find it.
>
> Imagine that it is common to run a large number of very similar machine generated queries that just change the literals in the sql query
>
> for example (the real queries would be much more complex)
>
> Select * from emp where empno = 1
> Select * from emp where empno = 2
> ...
>
> etc
>
> The plan that is likely being generated for these kind of queries is going to be very much the same each time, so to save some time, I would like to recognize that the literals are all that have changed in a query and use the previously optimized execution plan and just replace the literals?
>
> Is there anything I can leverage in calcite (or is the functionality already there?) for this kind of thing?
>
> thanks,
> Homer
Hoisting literals
Posted by Homer <lo...@gmail.com>.
Hi,
I am wondering if this is something that has already in calcite but I
cannot seem to find it.
Imagine that it is common to run a large number of very similar machine
generated queries that just change the literals in the sql query
for example (the real queries would be much more complex)
Select * from emp where empno = 1
Select * from emp where empno = 2
...
etc
The plan that is likely being generated for these kind of queries is
going to be very much the same each time, so to save some time, I would
like to recognize that the literals are all that have changed in a query
and use the previously optimized execution plan and just replace the
literals?
Is there anything I can leverage in calcite (or is the functionality
already there?) for this kind of thing?
thanks,
Homer
Re: Post 1.5
Posted by Jesus Camacho Rodriguez <jc...@hortonworks.com>.
On 11/11/15, 12:45 AM, "Julian Hyde" <jh...@apache.org> wrote:
>
>> On Nov 10, 2015, at 2:27 PM, Jesus Camacho Rodriguez <jc...@hortonworks.com> wrote:
>>
>> Hi Julian,
>>
>> I guess it was the excitement about releasing :) Honestly, I did not realize I had to wait 24 hours, sorry about that.
>>
>> * Thanks for the release notes! Thanks also for the instructions in the HOW-TO: they were easy to follow, and I didn't have any major problem.
>> The fact that it was our first release as top-level project didn't change anything; however, the recent changes in the ASF infrastructure did. In particular, I had to generate all the artifacts from my own repository, and then push that to the ASF repository. Should I update the HOW-TO? My understanding is that these restrictions in the ASF infrastructure are temporary, so I do not know if it is worth it. Further, using an external repository (github) to create the release does not seem a very good practice…
>
>I think you should at least add a note, to make it easier for the next release manager. We do not know when the temporary measures will end. It’s easy enough to remove the note.
>
>You don’t need to publish the site with the changed HOWTO - it usually gets published every couple of weeks for some reason or other.
Sure, makes sense: I will add a note shortly.
Thanks,
Jesús
>
>> * I agree with your point about the vote over the weekend; I proceeded because committers reacted quickly and most active community members had already casted their vote.
>
>Agreed. Everyone got to cast their vote.
>
>> * About the tag, do you mean the tag of the release (it is there)? Or...? I did not get it.
>
>Sorry, false alarm. For some reason I wasn’t seeing ‘tag: calcite-1.5.0’ when I did
>
> git log --pretty=format:'%h %d %s %cr’
>
>But I see it now.
>
>Julian
>
>
Re: Post 1.5
Posted by Julian Hyde <jh...@apache.org>.
> On Nov 10, 2015, at 2:27 PM, Jesus Camacho Rodriguez <jc...@hortonworks.com> wrote:
>
> Hi Julian,
>
> I guess it was the excitement about releasing :) Honestly, I did not realize I had to wait 24 hours, sorry about that.
>
> * Thanks for the release notes! Thanks also for the instructions in the HOW-TO: they were easy to follow, and I didn't have any major problem.
> The fact that it was our first release as top-level project didn't change anything; however, the recent changes in the ASF infrastructure did. In particular, I had to generate all the artifacts from my own repository, and then push that to the ASF repository. Should I update the HOW-TO? My understanding is that these restrictions in the ASF infrastructure are temporary, so I do not know if it is worth it. Further, using an external repository (github) to create the release does not seem a very good practice…
I think you should at least add a note, to make it easier for the next release manager. We do not know when the temporary measures will end. It’s easy enough to remove the note.
You don’t need to publish the site with the changed HOWTO - it usually gets published every couple of weeks for some reason or other.
> * I agree with your point about the vote over the weekend; I proceeded because committers reacted quickly and most active community members had already casted their vote.
Agreed. Everyone got to cast their vote.
> * About the tag, do you mean the tag of the release (it is there)? Or...? I did not get it.
Sorry, false alarm. For some reason I wasn’t seeing ‘tag: calcite-1.5.0’ when I did
git log --pretty=format:'%h %d %s %cr’
But I see it now.
Julian
Re: Post 1.5
Posted by Jesus Camacho Rodriguez <jc...@hortonworks.com>.
Hi Julian,
I guess it was the excitement about releasing :) Honestly, I did not realize I had to wait 24 hours, sorry about that.
* Thanks for the release notes! Thanks also for the instructions in the HOW-TO: they were easy to follow, and I didn't have any major problem.
The fact that it was our first release as top-level project didn't change anything; however, the recent changes in the ASF infrastructure did. In particular, I had to generate all the artifacts from my own repository, and then push that to the ASF repository. Should I update the HOW-TO? My understanding is that these restrictions in the ASF infrastructure are temporary, so I do not know if it is worth it. Further, using an external repository (github) to create the release does not seem a very good practice...
* I agree with your point about the vote over the weekend; I proceeded because committers reacted quickly and most active community members had already casted their vote.
* About the tag, do you mean the tag of the release (it is there)? Or...? I did not get it.
* I can regenerate the javadoc (stripping out dates). I'll also push the code to origin/branch-release; I did not see that in the instructions, but it makes sense.
--
Jesús
On 11/10/15, 7:42 PM, "Julian Hyde" <jh...@apache.org> wrote:
>Also:
>* Can you please add a tag? It looks as if maven didn't add one in offline mode.
>* How were the instructions in HOWTO? This is the first time anyone
>has used them post-graduation.
>* Can you push to origin/branch-release? I always make sure that that
>points to the latest release. I don't know whether I put that in the
>instructions.
>
>Julian
>
>
>On Tue, Nov 10, 2015 at 10:19 AM, Julian Hyde <jh...@apache.org> wrote:
>> Many thanks, Jesus, for being release manager. You made everything run
>> really smoothly. The most impressive thing was that I first found out
>> about the release from a tweet from a senior Gartner analyst[1],
>> including a link to a news item!
>>
>> I can regenerate the javadoc and publish to the site -- shall I do
>> that? I'd like to strip out dates from the generated javadoc, to
>> reduce svn churn.
>>
>> A couple of minor nits:
>>
>> * If the vote is over a weekend, let's make it a 48 hour vote. (I was
>> impressed that several of our committers heeded my call and voted on a
>> Sunday evening, but let's respect people's weekends.)
>> * The release bits haven't propagated to all of the mirrors yet (e.g.
>> [2] gives 404). The guideline says to wait 24 hours between pushing to
>> dist/release and announcing the release.
>>
>> Julian
>>
>> [1] https://twitter.com/merv/status/664127978746437632
>>
>> [2] http://mirrors.ibiblio.org/apache/calcite/apache-calcite-1.5.0
>>
>>
>> On Tue, Nov 10, 2015 at 12:37 AM, Jesus Camacho Rodriguez
>> <jc...@hortonworks.com> wrote:
>>> Master branch is now open again.
>>>
>>> Jesús
>
Re: Post 1.5
Posted by Julian Hyde <jh...@apache.org>.
Also:
* Can you please add a tag? It looks as if maven didn't add one in offline mode.
* How were the instructions in HOWTO? This is the first time anyone
has used them post-graduation.
* Can you push to origin/branch-release? I always make sure that that
points to the latest release. I don't know whether I put that in the
instructions.
Julian
On Tue, Nov 10, 2015 at 10:19 AM, Julian Hyde <jh...@apache.org> wrote:
> Many thanks, Jesus, for being release manager. You made everything run
> really smoothly. The most impressive thing was that I first found out
> about the release from a tweet from a senior Gartner analyst[1],
> including a link to a news item!
>
> I can regenerate the javadoc and publish to the site -- shall I do
> that? I'd like to strip out dates from the generated javadoc, to
> reduce svn churn.
>
> A couple of minor nits:
>
> * If the vote is over a weekend, let's make it a 48 hour vote. (I was
> impressed that several of our committers heeded my call and voted on a
> Sunday evening, but let's respect people's weekends.)
> * The release bits haven't propagated to all of the mirrors yet (e.g.
> [2] gives 404). The guideline says to wait 24 hours between pushing to
> dist/release and announcing the release.
>
> Julian
>
> [1] https://twitter.com/merv/status/664127978746437632
>
> [2] http://mirrors.ibiblio.org/apache/calcite/apache-calcite-1.5.0
>
>
> On Tue, Nov 10, 2015 at 12:37 AM, Jesus Camacho Rodriguez
> <jc...@hortonworks.com> wrote:
>> Master branch is now open again.
>>
>> Jesús
Re: Post 1.5
Posted by Julian Hyde <jh...@apache.org>.
Many thanks, Jesus, for being release manager. You made everything run
really smoothly. The most impressive thing was that I first found out
about the release from a tweet from a senior Gartner analyst[1],
including a link to a news item!
I can regenerate the javadoc and publish to the site -- shall I do
that? I'd like to strip out dates from the generated javadoc, to
reduce svn churn.
A couple of minor nits:
* If the vote is over a weekend, let's make it a 48 hour vote. (I was
impressed that several of our committers heeded my call and voted on a
Sunday evening, but let's respect people's weekends.)
* The release bits haven't propagated to all of the mirrors yet (e.g.
[2] gives 404). The guideline says to wait 24 hours between pushing to
dist/release and announcing the release.
Julian
[1] https://twitter.com/merv/status/664127978746437632
[2] http://mirrors.ibiblio.org/apache/calcite/apache-calcite-1.5.0
On Tue, Nov 10, 2015 at 12:37 AM, Jesus Camacho Rodriguez
<jc...@hortonworks.com> wrote:
> Master branch is now open again.
>
> Jesús