How to avoid re-creating a recipe (+my meta-bucket approach)

How to avoid re-creating a recipe (+my meta-bucket approach)

[Tim O'Callaghan]

Hi,

After seeing the lmsensors thread float by I thought it about time I started the discussion on how people manage to avoid re-inventing recipes, and to get the ball rolling, I thought I would describe mine.

As I do not have the any idea what recipes may be kicking about, I devised what I call the meta-bucket approach. It is pretty simple, I have a script that clones every existing OE/Yocto repository I know about into one easily grep-able place[1]. 

The to the main question, how are others managing this?

Best regards,

Tim.
[1] https://github.com/timoc/meta-bucket

Re: How to avoid re-creating a recipe (+my meta-bucket approach)

[Paul Eggleton]

On Wednesday 17 October 2012 08:39:36 Tim O'Callaghan wrote:
> After seeing the lmsensors thread float by I thought it about time I started
> the discussion on how people manage to avoid re-inventing recipes, and to
> get the ball rolling, I thought I would describe mine.
> 
> As I do not have the any idea what recipes may be kicking about, I devised
> what I call the meta-bucket approach. It is pretty simple, I have a script
> that clones every existing OE/Yocto repository I know about into one easily
> grep-able place[1].
> 
> The to the main question, how are others managing this?

I clone the layer repositories most likely to hold new recipes and then use 
some scripts I wrote to process them into a list of recipes that can be easily 
searched using grep. I wouldn't mind sharing those scripts if some people 
would find them useful.

However, what I hope we can do in the 1.4 timeframe would be to implement my 
proposal here:

 https://bugzilla.yoctoproject.org/show_bug.cgi?id=3272

This would provide us with a way to search all published layers down to the 
recipe level from the web.

Cheers,
Paul

-- 

Paul Eggleton
Intel Open Source Technology Centre

Re: How to avoid re-creating a recipe (+my meta-bucket approach)

[Philip Balister]

On 10/17/2012 01:49 AM, Paul Eggleton wrote:
> On Wednesday 17 October 2012 08:39:36 Tim O'Callaghan wrote:
>> After seeing the lmsensors thread float by I thought it about time I started
>> the discussion on how people manage to avoid re-inventing recipes, and to
>> get the ball rolling, I thought I would describe mine.
>>
>> As I do not have the any idea what recipes may be kicking about, I devised
>> what I call the meta-bucket approach. It is pretty simple, I have a script
>> that clones every existing OE/Yocto repository I know about into one easily
>> grep-able place[1].
>>
>> The to the main question, how are others managing this?
>
> I clone the layer repositories most likely to hold new recipes and then use
> some scripts I wrote to process them into a list of recipes that can be easily
> searched using grep. I wouldn't mind sharing those scripts if some people
> would find them useful.
>
> However, what I hope we can do in the 1.4 timeframe would be to implement my
> proposal here:
>
>   https://bugzilla.yoctoproject.org/show_bug.cgi?id=3272
>
> This would provide us with a way to search all published layers down to the
> recipe level from the web.

This is an excellent idea!

Philip

Re: How to avoid re-creating a recipe (+my meta-bucket approach)

[Tom Zanussi]

On Wed, 2012-10-17 at 08:39 +0000, Tim O'Callaghan wrote:
> Hi,
> 
> After seeing the lmsensors thread float by I thought it about time I started the discussion on how people manage to avoid re-inventing recipes, and to get the ball rolling, I thought I would describe mine.
> 
> As I do not have the any idea what recipes may be kicking about, I devised what I call the meta-bucket approach. It is pretty simple, I have a script that clones every existing OE/Yocto repository I know about into one easily grep-able place[1]. 
> 
> The to the main question, how are others managing this?
> 
> Best regards,
> 
> Tim.
> [1] https://github.com/timoc/meta-bucket
> 

I do a similar thing - clone all the repos I can think of, and then have
OpenGrok index them all.  That makes everything available as a fully
searchable cross-referenced set of web pages.  Couldn't live without
it...

http://hub.opensolaris.org/bin/view/Project+opengrok/

Tom

> _______________________________________________
> yocto mailing list
> yocto@yoctoproject.org
> https://lists.yoctoproject.org/listinfo/yocto