I did not know about this convention. I find this design to be unnecessarily confusing.

Why is this needed? Why not use strings / symbols interchangeable?

Cheers, Doru

On 14 Apr 2011, at 23:24, Alexandre Bergel wrote:

...

now when I do

ConfigurationOfRPackage project load: #'1.0'

is telling to me that there is no symbolic version and I do not get it.

Instead, you should do: ConfigurationOfRPackage project load: '1.0'

#'1.0' vs '1.0'

Alexandre

-- _,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;: Alexandre Bergel http://www.bergel.eu ^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;.

---

Moose-dev mailing list Moose-dev@iam.unibe.ch https://www.iam.unibe.ch/mailman/listinfo/moose-dev

-- www.tudorgirba.com

"Some battles are better lost than fought."

I mentioned something similar to this a few months ago. The reason is primarily for making legacy configuration work.

Alexandre

On 14 Apr 2011, at 16:28, Tudor Girba wrote:

I did not know about this convention. I find this design to be unnecessarily confusing.

Why is this needed? Why not use strings / symbols interchangeable?

Cheers, Doru

On 14 Apr 2011, at 23:24, Alexandre Bergel wrote:

...

...

now when I do

ConfigurationOfRPackage project load: #'1.0'

is telling to me that there is no symbolic version and I do not get it.

Instead, you should do: ConfigurationOfRPackage project load: '1.0'

#'1.0' vs '1.0'

Alexandre

-- _,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;: Alexandre Bergel http://www.bergel.eu ^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;.

---

Moose-dev mailing list Moose-dev@iam.unibe.ch https://www.iam.unibe.ch/mailman/listinfo/moose-dev

-- www.tudorgirba.com

"Some battles are better lost than fought."

Could you point me to the original mail where you argued for this?

From this thread I cannot tell (I know, I guess I am just dense:) what the problem is nor what the proposed solution would be, so I cannot comment without some more context...

It's quite possible that I missed the point the last time around, so you'll have to start at the beginning of this conversation to bring me into the loop:)

Dale

On 04/14/2011 01:34 PM, Tudor Girba wrote:

I still do not understand why symbolic versions cannot be treated like normal versions (I argued for this before :)). Their name should be unique within the overall versions anyway, so I do not see where the compatibility would break.

Cheers, Doru

On 14 Apr 2011, at 23:32, Alexandre Bergel wrote:

...

I mentioned something similar to this a few months ago. The reason is primarily for making legacy configuration work.

Alexandre

On 14 Apr 2011, at 16:28, Tudor Girba wrote:

...

I did not know about this convention. I find this design to be unnecessarily confusing.

Why is this needed? Why not use strings / symbols interchangeable?

Cheers, Doru

On 14 Apr 2011, at 23:24, Alexandre Bergel wrote:

...

...

now when I do

ConfigurationOfRPackage project load: #'1.0'

is telling to me that there is no symbolic version and I do not get it.

Instead, you should do: ConfigurationOfRPackage project load: '1.0'

#'1.0' vs '1.0'

Alexandre

-- _,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;: Alexandre Bergel http://www.bergel.eu ^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;.

_______________________________________________ Moose-dev mailing list Moose-dev@iam.unibe.ch https://www.iam.unibe.ch/mailman/listinfo/moose-dev

-- www.tudorgirba.com

"Some battles are better lost than fought."

-- _,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;: Alexandre Bergel http://www.bergel.eu ^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;.

-- www.tudorgirba.com

"No matter how many recipes we know, we still value a chef."

On 04/14/2011 02:09 PM, Tudor Girba wrote:

Hi,

On 14 Apr 2011, at 22:49, Dale Henrichs wrote:

...

Could you point me to the original mail where you argued for this?

I would have to look for it, so better stat afresh :)

...

From this thread I cannot tell (I know, I guess I am just dense:) what the problem is nor what the proposed solution would be, so I cannot comment without some more context...

It's quite possible that I missed the point the last time around, so you'll have to start at the beginning of this conversation to bring me into the loop:)

The whole current discussion started from the realization that:

ConfigurationOfRPackage project load: #'1.0' --> looks for a symbolic version ConfigurationOfRPackage project load: '1.0' --> looks for a regular version

I said that this is confusing, and I believe it does not have to be like that if symbolic versions are treated the same as regular versions.

FWIW, I intended for symbolic versions to be in a separate namespace, so that #stable, #bleedingEdge, #development and any other symbolic versions wouldn't inadvertently collide with a literal version of the same name.

Symbolic versions are defined differently and mean something different from literal versions, so I wanted a user to be able to look at a load expression and know that if she saw something that looked like this:

ConfigurationOfRPackage project load: #'1.0'

That she would not be surprised when the `ConfigurationOfRPackage project currentVersion` answered '1.0-beta.28.3' and not #'1.0'.

So when someone gets the error 'Symbolic version not found' when they try to run `ConfigurationOfRPackage project load: #'1.0'`, I say that it is doing exactly what it should be doing ...

With that said, I don't deny that it is confusing, but really not any more confustion than getting the error 'Version not found' when trying to run the following expression `ConfigurationOfRPackage project load: '1,0'`...There's a typo in the expression and it would be nice if the error said, "you've put a #, instead of $. in the version name"...

Finally, I think that the distinction between symbolic versions and literal versions will be much more obvious in the MetacelloBrowser and typing errors won't be the primary issue...

Dale