R2spec and version number

[Pierre-Yves Chibon]

Dear all,

I would like your opinion regarding: [Bug 503108] New: R2spec-2.5.2
loses Version information

It basically says that when the package version is 2.3-50 R2spec only
retrieve 2.3 as being the version number.
To me and based on the packages R-abind, R-acepack, R-mAR (and some
other that I didn't list) this is not a bug.

Do you agree with me ?

Thanks,
Best regards,

Pierre

[José Abílio Matos]

On Friday 29 May 2009 06:53:54 Pierre-Yves wrote:
>  Dear all,
>
>  I would like your opinion regarding: [Bug 503108] New: R2spec-2.5.2
>  loses Version information
>
>  It basically says that when the package version is 2.3-50 R2spec only
>  retrieve 2.3 as being the version number.
>  To me and based on the packages R-abind, R-acepack, R-mAR (and some
>  other that I didn't list) this is not a bug.
>
>  Do you agree with me ? 
I agree that we are loosing information in the version and that on the other 
hand the dash is just a fancy replacement for the dot.

So we could agree to replace all the dashes in R package versions by dots and 
be done with it.

Whatever we decide we should apply this uniformly to all R packages that have 
this numbering.

>  Thanks,
>  Best regards,
>
>  Pierre 
-- 
José Abílio

[Pierre-Yves Chibon]

On Fri, 2009-05-29 at 10:53 +0100, José Matos wrote:
>  On Friday 29 May 2009 06:53:54 Pierre-Yves wrote:
>  > Dear all,
>  >
>  > I would like your opinion regarding: [Bug 503108] New: R2spec-2.5.2
>  > loses Version information
>  >
>  > It basically says that when the package version is 2.3-50 R2spec only
>  > retrieve 2.3 as being the version number.
>  > To me and based on the packages R-abind, R-acepack, R-mAR (and some
>  > other that I didn't list) this is not a bug.
>  >
>  > Do you agree with me ?
>  
>  I agree that we are loosing information in the version and that on the other 
>  hand the dash is just a fancy replacement for the dot.
>  
>  So we could agree to replace all the dashes in R package versions by dots and 
>  be done with it.
>  
>  Whatever we decide we should apply this uniformly to all R packages that have 
>  this numbering. 
Thanks for the feedback José :)

Spot, does the guideline say something about this ?
Should they ?
Should we bring the discussion to a broader audience ?

Regards,
Pierre

[Martyn Plummer]

On Wed, 2009-06-03 at 15:09 +0200, Pierre-Yves wrote:
>  On Fri, 2009-05-29 at 10:53 +0100, José Matos wrote:
>  > On Friday 29 May 2009 06:53:54 Pierre-Yves wrote:
>  > > Dear all,
>  > >
>  > > I would like your opinion regarding: [Bug 503108] New: R2spec-2.5.2
>  > > loses Version information
>  > >
>  > > It basically says that when the package version is 2.3-50 R2spec only
>  > > retrieve 2.3 as being the version number.
>  > > To me and based on the packages R-abind, R-acepack, R-mAR (and some
>  > > other that I didn't list) this is not a bug.
>  > >
>  > > Do you agree with me ?
>  > 
>  > I agree that we are loosing information in the version and that on the other 
>  > hand the dash is just a fancy replacement for the dot.
>  > 
>  > So we could agree to replace all the dashes in R package versions by dots and 
>  > be done with it.
>  > 
>  > Whatever we decide we should apply this uniformly to all R packages that have 
>  > this numbering.
>  
>  Thanks for the feedback José :)
>  
>  Spot, does the guideline say something about this ?
>  Should they ?
>  Should we bring the discussion to a broader audience ?
>  
>  Regards,
>  Pierre 
For R, there is no difference between "-" and "." in package version
numbers. Here is what the R Extension manual says:

"The version is a sequence of at least two (and usually three)
non-negative integers separated by single ‘.’ or ‘-’ characters. The
canonical form is as shown in the example [0.5-1 - Martyn], and a
version such as ‘0.01’ or ‘0.01.0’ will be handled as if it were
‘0.1-0’."

So you should just replace dashes with dots.

Martyn

-----------------------------------------------------------------------
This message and its attachments are strictly confidential. If you are
not the intended recipient of this message, please immediately notify 
the sender and delete it. Since its integrity cannot be guaranteed, 
its content cannot involve the sender's responsibility. Any misuse, 
any disclosure or publication of its content, either whole or partial, 
is prohibited, exception made of formally approved use
-----------------------------------------------------------------------

[Tom Callaway]

On 06/03/2009 12:50 PM, Martyn Plummer wrote:

>  For R, there is no difference between "-" and "."
> in package version
>  numbers. Here is what the R Extension manual says:
>  
>  "The version is a sequence of at least two (and usually three)
>  non-negative integers separated by single ‘.’ or ‘-’ characters. The
>  canonical form is as shown in the example [0.5-1 - Martyn], and a
>  version such as ‘0.01’ or ‘0.01.0’ will be handled as if it were
>  ‘0.1-0’."
>  
>  So you should just replace dashes with dots. 
Based on that, I'd say we should be embedding the whole version (with
dot, not dash). Pretty much every R package is going to need to be fixed
for this, include the core R.

~spot

[Pierre-Yves Chibon]

On Wed, 2009-06-03 at 17:07 -0400, Tom "spot" Callaway wrote:
>  On 06/03/2009 12:50 PM, Martyn Plummer wrote:
>  
>  > For R, there is no difference between "-" and "." in package
> version
>  > numbers. Here is what the R Extension manual says:
>  > 
>  > "The version is a sequence of at least two (and usually three)
>  > non-negative integers separated by single ‘.’ or ‘-’ characters. The
>  > canonical form is as shown in the example [0.5-1 - Martyn], and a
>  > version such as ‘0.01’ or ‘0.01.0’ will be handled as if it were
>  > ‘0.1-0’."
>  > 
>  > So you should just replace dashes with dots.
>  
>  Based on that, I'd say we should be embedding the whole version (with
>  dot, not dash). Pretty much every R package is going to need to be fixed
>  for this, include the core R. 
I'd also propose that we put a small note on the packaging guidelines :)

Thanks all,

Pierre