About MCGitHubRepository and cache

Previous Topic Next Topic
 
classic Classic list List threaded Threaded
5 messages Options
Reply | Threaded
Open this post in threaded view
|

About MCGitHubRepository and cache

demarey
Hi,

I have some issues related to MCGitHubRepository and its cache functionality.
As documented (good point) in MCGitHubRepository>>#flushCache, the directory acts like a cache since we download the directory from github.

It leads to filesystem errors if you build the image and load dependencies on a host and then you move your image to another host. A typical case for that is the use of a continuous integration server to prepare the image, and then it is downloaded by users and used elsewhere.

If we want to keep the cache feature as default, shouldn't we test if the file reference is still valid before using it. If it is not, we just need to flush the cache.

WDYT?

Regards,
Christophe.


--
You received this message because you are subscribed to the Google Groups "Metacello" group.
To unsubscribe from this group and stop receiving emails from it, send an email to [hidden email].
For more options, visit https://groups.google.com/groups/opt_out.
 
 
Reply | Threaded
Open this post in threaded view
|

Re: About MCGitHubRepository and cache

Johan Brichau-3
Hi Christophe,

Indeed. I've been hit by this too after (inadvertently) removing the github cache directory.

A related problem is that when there are multiple images on the same machine using github, you have problems with the github.zip file: https://code.google.com/p/metacello/issues/detail?id=184
Most probably, the issue also applies to Pharo I think.

@Dale: I shall probably create a pull request for the issue I mention above. This issue was created before I knew where the issues should have been reported...

On 24 Jun 2013, at 13:28, ChristopheD <[hidden email]> wrote:

> Hi,
>
> I have some issues related to MCGitHubRepository and its cache functionality.
> As documented (good point) in MCGitHubRepository>>#flushCache, the directory acts like a cache since we download the directory from github.
>
> It leads to filesystem errors if you build the image and load dependencies on a host and then you move your image to another host. A typical case for that is the use of a continuous integration server to prepare the image, and then it is downloaded by users and used elsewhere.
>
> If we want to keep the cache feature as default, shouldn't we test if the file reference is still valid before using it. If it is not, we just need to flush the cache.
>
> WDYT?
>
> Regards,
> Christophe.
>
>
>
> --
> You received this message because you are subscribed to the Google Groups "Metacello" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to [hidden email].
> For more options, visit https://groups.google.com/groups/opt_out.
>  
>  

--
You received this message because you are subscribed to the Google Groups "Metacello" group.
To unsubscribe from this group and stop receiving emails from it, send an email to [hidden email].
For more options, visit https://groups.google.com/groups/opt_out.


Reply | Threaded
Open this post in threaded view
|

Re: About MCGitHubRepository and cache

Dale Henrichs-3
Johan and Christophe,

No need for a pull request ... the code in the google bug report is good enough.

Thanks,

Dale

----- Original Message -----
| From: "Johan Brichau" <[hidden email]>
| To: [hidden email]
| Sent: Monday, June 24, 2013 5:01:20 AM
| Subject: Re: [Metacello] About MCGitHubRepository and cache
|
| Hi Christophe,
|
| Indeed. I've been hit by this too after (inadvertently) removing the
| github cache directory.
|
| A related problem is that when there are multiple images on the same
| machine using github, you have problems with the github.zip file:
| https://code.google.com/p/metacello/issues/detail?id=184
| Most probably, the issue also applies to Pharo I think.
|
| @Dale: I shall probably create a pull request for the issue I mention
| above. This issue was created before I knew where the issues should
| have been reported...
|
| On 24 Jun 2013, at 13:28, ChristopheD <[hidden email]>
| wrote:
|
| > Hi,
| >
| > I have some issues related to MCGitHubRepository and its cache
| > functionality.
| > As documented (good point) in MCGitHubRepository>>#flushCache, the
| > directory acts like a cache since we download the directory from
| > github.
| >
| > It leads to filesystem errors if you build the image and load
| > dependencies on a host and then you move your image to another
| > host. A typical case for that is the use of a continuous
| > integration server to prepare the image, and then it is downloaded
| > by users and used elsewhere.
| >
| > If we want to keep the cache feature as default, shouldn't we test
| > if the file reference is still valid before using it. If it is
| > not, we just need to flush the cache.
| >
| > WDYT?
| >
| > Regards,
| > Christophe.
| >
| >
| >
| > --
| > You received this message because you are subscribed to the Google
| > Groups "Metacello" group.
| > To unsubscribe from this group and stop receiving emails from it,
| > send an email to [hidden email].
| > For more options, visit https://groups.google.com/groups/opt_out.
| >  
| >  
|
| --
| You received this message because you are subscribed to the Google
| Groups "Metacello" group.
| To unsubscribe from this group and stop receiving emails from it,
| send an email to [hidden email].
| For more options, visit https://groups.google.com/groups/opt_out.
|
|
|

--
You received this message because you are subscribed to the Google Groups "Metacello" group.
To unsubscribe from this group and stop receiving emails from it, send an email to [hidden email].
For more options, visit https://groups.google.com/groups/opt_out.


Reply | Threaded
Open this post in threaded view
|

Re: About MCGitHubRepository and cache

Dale Henrichs-3
In reply to this post by demarey
Cristophe,

I think it's a bug:)[1]...

Now that I'm out of the woods regarding the switch from gemstone.com to gemtalksystems.com and the MetacelloPreview appears to be functioning on Pharo-2.0 ... I'm taking this opportunity to knock off a couple of outstanding Metacello bugs ...

I've got Johan's bug fix[2] under test and I'm starting on your issue right now:)

Dale

[1] https://github.com/dalehenrich/metacello-work/issues/175
[2] https://code.google.com/p/metacello/issues/detail?id=184


From: "ChristopheD" <[hidden email]>
To: [hidden email]
Sent: Monday, June 24, 2013 4:28:58 AM
Subject: [Metacello] About MCGitHubRepository and cache

Hi,

I have some issues related to MCGitHubRepository and its cache functionality.
As documented (good point) in MCGitHubRepository>>#flushCache, the directory acts like a cache since we download the directory from github.

It leads to filesystem errors if you build the image and load dependencies on a host and then you move your image to another host. A typical case for that is the use of a continuous integration server to prepare the image, and then it is downloaded by users and used elsewhere.

If we want to keep the cache feature as default, shouldn't we test if the file reference is still valid before using it. If it is not, we just need to flush the cache.

WDYT?

Regards,
Christophe.


--
You received this message because you are subscribed to the Google Groups "Metacello" group.
To unsubscribe from this group and stop receiving emails from it, send an email to [hidden email].
For more options, visit https://groups.google.com/groups/opt_out.
 
 

--
You received this message because you are subscribed to the Google Groups "Metacello" group.
To unsubscribe from this group and stop receiving emails from it, send an email to [hidden email].
For more options, visit https://groups.google.com/groups/opt_out.
 
 
Reply | Threaded
Open this post in threaded view
|

Re: About MCGitHubRepository and cache

Dale Henrichs-3
Johan and Christophe,

Just thought I'd let you know that I've decided to allow folks to explicitly set the github-cache directory[1] (as is done with package-cache for Pharo2.0) so you can have a common github cache directory for all of your images ...

Also have a bugfix for original issue if you move image to a different machine are are using the default..

The fix will be available in Preview 1.0.0-beta.32.8...

Dale

[1] https://github.com/dalehenrich/metacello-work/issues/175#issuecomment-20902186
[2] https://github.com/dalehenrich/metacello-work/issues?direction=desc&milestone=10&page=1&sort=updated&state=open


From: "Dale K. Henrichs" <[hidden email]>
To: [hidden email]
Sent: Monday, July 1, 2013 11:13:58 AM
Subject: Re: [Metacello] About MCGitHubRepository and cache

Cristophe,

I think it's a bug:)[1]...

Now that I'm out of the woods regarding the switch from gemstone.com to gemtalksystems.com and the MetacelloPreview appears to be functioning on Pharo-2.0 ... I'm taking this opportunity to knock off a couple of outstanding Metacello bugs ...

I've got Johan's bug fix[2] under test and I'm starting on your issue right now:)

Dale

[1] https://github.com/dalehenrich/metacello-work/issues/175
[2] https://code.google.com/p/metacello/issues/detail?id=184


From: "ChristopheD" <[hidden email]>
To: [hidden email]
Sent: Monday, June 24, 2013 4:28:58 AM
Subject: [Metacello] About MCGitHubRepository and cache

Hi,

I have some issues related to MCGitHubRepository and its cache functionality.
As documented (good point) in MCGitHubRepository>>#flushCache, the directory acts like a cache since we download the directory from github.

It leads to filesystem errors if you build the image and load dependencies on a host and then you move your image to another host. A typical case for that is the use of a continuous integration server to prepare the image, and then it is downloaded by users and used elsewhere.

If we want to keep the cache feature as default, shouldn't we test if the file reference is still valid before using it. If it is not, we just need to flush the cache.

WDYT?

Regards,
Christophe.


--
You received this message because you are subscribed to the Google Groups "Metacello" group.
To unsubscribe from this group and stop receiving emails from it, send an email to [hidden email].
For more options, visit https://groups.google.com/groups/opt_out.
 
 


--
You received this message because you are subscribed to the Google Groups "Metacello" group.
To unsubscribe from this group and stop receiving emails from it, send an email to [hidden email].
For more options, visit https://groups.google.com/groups/opt_out.