Let's make cache name not null in ignite 2.0

[Alexey Kuznetsov-2]

Hi, All!

In ignite-1.x cache.name could be null.

And we have to write a lot of code to handle such name, especially in tools
like web console and Visor.

In ignite 2.0 we could change API and make cache.name not null.

What do you think about such change?

It is worth to create such issue in JIRA?


--
Alexey Kuznetsov
GridGain Systems
www.gridgain.com

[Vladimir Ozerov]

+1

I would also restrict nulls for node names, for IGFS names, etc..

On Wed, Jun 1, 2016 at 10:29 AM, Alexey Kuznetsov <[hidden email]>
wrote:

> Hi, All!
>
> In ignite-1.x cache.name could be null.
>
> And we have to write a lot of code to handle such name, especially in tools
> like web console and Visor.
>
> In ignite 2.0 we could change API and make cache.name not null.
>
> What do you think about such change?
>
> It is worth to create such issue in JIRA?
>
>
> --
> Alexey Kuznetsov
> GridGain Systems
> www.gridgain.com
>

[dsetrakyan]

-1

I don’t think this will give us any advantage other than many frustrated
users who will need to change their code. We should definitely discourage
using nulls though.

D.

On Wed, Jun 1, 2016 at 12:37 AM, Vladimir Ozerov <[hidden email]>
wrote:

> +1
>
> I would also restrict nulls for node names, for IGFS names, etc..
>
> On Wed, Jun 1, 2016 at 10:29 AM, Alexey Kuznetsov <[hidden email]
> >
> wrote:
>
> > Hi, All!
> >
> > In ignite-1.x cache.name could be null.
> >
> > And we have to write a lot of code to handle such name, especially in
> tools
> > like web console and Visor.
> >
> > In ignite 2.0 we could change API and make cache.name not null.
> >
> > What do you think about such change?
> >
> > It is worth to create such issue in JIRA?
> >
> >
> > --
> > Alexey Kuznetsov
> > GridGain Systems
> > www.gridgain.com
> >
>

[Alexey Goncharuk]

We can add an explicit warning in 1.x when a cache with the null name is
used and remove it in 2.0.

2016-06-01 15:49 GMT-07:00 Dmitriy Setrakyan <[hidden email]>:

> -1
>
> I don’t think this will give us any advantage other than many frustrated
> users who will need to change their code. We should definitely discourage
> using nulls though.
>
> D.
>
> On Wed, Jun 1, 2016 at 12:37 AM, Vladimir Ozerov <[hidden email]>
> wrote:
>
> > +1
> >
> > I would also restrict nulls for node names, for IGFS names, etc..
> >
> > On Wed, Jun 1, 2016 at 10:29 AM, Alexey Kuznetsov <
> [hidden email]
> > >
> > wrote:
> >
> > > Hi, All!
> > >
> > > In ignite-1.x cache.name could be null.
> > >
> > > And we have to write a lot of code to handle such name, especially in
> > tools
> > > like web console and Visor.
> > >
> > > In ignite 2.0 we could change API and make cache.name not null.
> > >
> > > What do you think about such change?
> > >
> > > It is worth to create such issue in JIRA?
> > >
> > >
> > > --
> > > Alexey Kuznetsov
> > > GridGain Systems
> > > www.gridgain.com
> > >
> >
>

[dsetrakyan]

Warning is OK, but removing null name after having it for almost 2 years is
a bad idea. People are using it and will be forced to change their code for
no good reason.

On Wed, Jun 1, 2016 at 5:20 PM, Alexey Goncharuk <[hidden email]
> wrote:

> We can add an explicit warning in 1.x when a cache with the null name is
> used and remove it in 2.0.
>
> 2016-06-01 15:49 GMT-07:00 Dmitriy Setrakyan <[hidden email]>:
>
> > -1
> >
> > I don’t think this will give us any advantage other than many frustrated
> > users who will need to change their code. We should definitely discourage
> > using nulls though.
> >
> > D.
> >
> > On Wed, Jun 1, 2016 at 12:37 AM, Vladimir Ozerov <[hidden email]>
> > wrote:
> >
> > > +1
> > >
> > > I would also restrict nulls for node names, for IGFS names, etc..
> > >
> > > On Wed, Jun 1, 2016 at 10:29 AM, Alexey Kuznetsov <
> > [hidden email]
> > > >
> > > wrote:
> > >
> > > > Hi, All!
> > > >
> > > > In ignite-1.x cache.name could be null.
> > > >
> > > > And we have to write a lot of code to handle such name, especially in
> > > tools
> > > > like web console and Visor.
> > > >
> > > > In ignite 2.0 we could change API and make cache.name not null.
> > > >
> > > > What do you think about such change?
> > > >
> > > > It is worth to create such issue in JIRA?
> > > >
> > > >
> > > > --
> > > > Alexey Kuznetsov
> > > > GridGain Systems
> > > > www.gridgain.com
> > > >
> > >
> >
>

[Sergi]

+100500 for dropping it. Having null names is the worst idea I can imagine.

People will have configuration incompatibilities anyways, disallowing nulls
is not a big deal here.

Sergi



2016-06-02 3:50 GMT+03:00 Dmitriy Setrakyan <[hidden email]>:

> Warning is OK, but removing null name after having it for almost 2 years is
> a bad idea. People are using it and will be forced to change their code for
> no good reason.
>
> On Wed, Jun 1, 2016 at 5:20 PM, Alexey Goncharuk <
> [hidden email]
> > wrote:
>
> > We can add an explicit warning in 1.x when a cache with the null name is
> > used and remove it in 2.0.
> >
> > 2016-06-01 15:49 GMT-07:00 Dmitriy Setrakyan <[hidden email]>:
> >
> > > -1
> > >
> > > I don’t think this will give us any advantage other than many
> frustrated
> > > users who will need to change their code. We should definitely
> discourage
> > > using nulls though.
> > >
> > > D.
> > >
> > > On Wed, Jun 1, 2016 at 12:37 AM, Vladimir Ozerov <[hidden email]
> >
> > > wrote:
> > >
> > > > +1
> > > >
> > > > I would also restrict nulls for node names, for IGFS names, etc..
> > > >
> > > > On Wed, Jun 1, 2016 at 10:29 AM, Alexey Kuznetsov <
> > > [hidden email]
> > > > >
> > > > wrote:
> > > >
> > > > > Hi, All!
> > > > >
> > > > > In ignite-1.x cache.name could be null.
> > > > >
> > > > > And we have to write a lot of code to handle such name, especially
> in
> > > > tools
> > > > > like web console and Visor.
> > > > >
> > > > > In ignite 2.0 we could change API and make cache.name not null.
> > > > >
> > > > > What do you think about such change?
> > > > >
> > > > > It is worth to create such issue in JIRA?
> > > > >
> > > > >
> > > > > --
> > > > > Alexey Kuznetsov
> > > > > GridGain Systems
> > > > > www.gridgain.com
> > > > >
> > > >
> > >
> >
>

[dsetrakyan]

Anyone here is even considering the compatibility impact it will have on
our users? I am still struggling to find a single reason for how it would
benefit our user base.

On Wed, Jun 1, 2016 at 7:07 PM, Sergi Vladykin <[hidden email]>
wrote:

> +100500 for dropping it. Having null names is the worst idea I can imagine.
>
> People will have configuration incompatibilities anyways, disallowing nulls
> is not a big deal here.
>
> Sergi
>
>
>
> 2016-06-02 3:50 GMT+03:00 Dmitriy Setrakyan <[hidden email]>:
>
> > Warning is OK, but removing null name after having it for almost 2 years
> is
> > a bad idea. People are using it and will be forced to change their code
> for
> > no good reason.
> >
> > On Wed, Jun 1, 2016 at 5:20 PM, Alexey Goncharuk <
> > [hidden email]
> > > wrote:
> >
> > > We can add an explicit warning in 1.x when a cache with the null name
> is
> > > used and remove it in 2.0.
> > >
> > > 2016-06-01 15:49 GMT-07:00 Dmitriy Setrakyan <[hidden email]>:
> > >
> > > > -1
> > > >
> > > > I don’t think this will give us any advantage other than many
> > frustrated
> > > > users who will need to change their code. We should definitely
> > discourage
> > > > using nulls though.
> > > >
> > > > D.
> > > >
> > > > On Wed, Jun 1, 2016 at 12:37 AM, Vladimir Ozerov <
> [hidden email]
> > >
> > > > wrote:
> > > >
> > > > > +1
> > > > >
> > > > > I would also restrict nulls for node names, for IGFS names, etc..
> > > > >
> > > > > On Wed, Jun 1, 2016 at 10:29 AM, Alexey Kuznetsov <
> > > > [hidden email]
> > > > > >
> > > > > wrote:
> > > > >
> > > > > > Hi, All!
> > > > > >
> > > > > > In ignite-1.x cache.name could be null.
> > > > > >
> > > > > > And we have to write a lot of code to handle such name,
> especially
> > in
> > > > > tools
> > > > > > like web console and Visor.
> > > > > >
> > > > > > In ignite 2.0 we could change API and make cache.name not null.
> > > > > >
> > > > > > What do you think about such change?
> > > > > >
> > > > > > It is worth to create such issue in JIRA?
> > > > > >
> > > > > >
> > > > > > --
> > > > > > Alexey Kuznetsov
> > > > > > GridGain Systems
> > > > > > www.gridgain.com
> > > > > >
> > > > >
> > > >
> > >
> >
>

[Raul Kripalani]

​Hey,​

On Thu, Jun 2, 2016 at 5:21 PM, Dmitriy Setrakyan <[hidden email]>
wrote:

> Anyone here is even considering the compatibility impact it will have on
> our users? I am still struggling to find a single reason for how it would
> benefit our user base.
>

​I've been away for some time. Under what circumstances can you create a
cache with null name?

https://ignite.apache.org/releases/mobile/org/apache/ignite/Ignite.html#cache(java.lang.String)
https://ignite.apache.org/releases/mobile/org/apache/ignite/cache/CacheManager.html#getCache(java.lang.String)

They both require a cache name as an argument.

Do you mean that the user could explicitly pass null as an argument, and
that was something that Ignite swallowed without complaining?

Cheers.

---
Raúl Kripalani
linkedin.com/in/raulkripalani | evosent.com
<http://evosent.com/?utm_source=email&utm_medium=email&utm_campaign=apache> |
blog: raul.io | skype: raul.fuse

[Alexey Kuznetsov-2]

Dmitriy,

I'm talking about possible *** IGNITE 2.0 *** that may be not released
several years :)
I think it will contain a LOT of API changes anyway.
Adding one more change should not be a show stopper?
Or I'm missing something?

Raul,

User can create cache via Spring XML and simply not specify property "name".
And Yes, Ignite will take null without any warning.

I think we at least should start to print warning in current Ignite 1.x.


On Thu, Jun 2, 2016 at 11:46 PM, Raul Kripalani <[hidden email]> wrote:

> ​Hey,​
>
> On Thu, Jun 2, 2016 at 5:21 PM, Dmitriy Setrakyan <[hidden email]>
> wrote:
>
> > Anyone here is even considering the compatibility impact it will have on
> > our users? I am still struggling to find a single reason for how it would
> > benefit our user base.
> >
>
> ​I've been away for some time. Under what circumstances can you create a
> cache with null name?
>
>
> https://ignite.apache.org/releases/mobile/org/apache/ignite/Ignite.html#cache(java.lang.String)
>
> https://ignite.apache.org/releases/mobile/org/apache/ignite/cache/CacheManager.html#getCache(java.lang.String)
>
> They both require a cache name as an argument.
>
> Do you mean that the user could explicitly pass null as an argument, and
> that was something that Ignite swallowed without complaining?
>
> Cheers.
>
> ---
> Raúl Kripalani
> linkedin.com/in/raulkripalani | evosent.com
> <http://evosent.com/?utm_source=email&utm_medium=email&utm_campaign=apache>
> |
> blog: raul.io | skype: raul.fuse
>

--
Alexey Kuznetsov
GridGain Systems
www.gridgain.com

[Raul Kripalani-2]

On Thu, Jun 2, 2016 at 5:56 PM, Alexey Kuznetsov <[hidden email]>
wrote:

> User can create cache via Spring XML and simply not specify property
> "name".
> And Yes, Ignite will take null without any warning.
>
> I think we at least should start to print warning in current Ignite 1.x.
>

​Ok. I would +1 the change too.​ Such API changes are exactly what major
releases are for, and allowing "null" as a cache name seems more like a
leftover than an actual functionality that one would desire.

Cheers.

---
Raúl Kripalani
linkedin.com/in/raulkripalani | evosent.com
<http://evosent.com/?utm_source=email&utm_medium=email&utm_campaign=evosent_raul>
| blog: raul.io
<http://raul.io?utm_source=email&utm_medium=email&utm_campaign=evosent_raul> |
[phone] | skype: raul.fuse

[ruchirc]

where can i find ignite-spark.jar ? or its placed in ignite-core.jar in the
newer versions of Ignite.

will appreciate if anyone can help,

Regards,
Ruchir



On Wed, Jun 1, 2016 at 12:29 AM, Alexey Kuznetsov <[hidden email]>
wrote:

> Hi, All!
>
> In ignite-1.x cache.name could be null.
>
> And we have to write a lot of code to handle such name, especially in tools
> like web console and Visor.
>
> In ignite 2.0 we could change API and make cache.name not null.
>
> What do you think about such change?
>
> It is worth to create such issue in JIRA?
>
>
> --
> Alexey Kuznetsov
> GridGain Systems
> www.gridgain.com
>