[Pharo-project] Fwd: [squeak-dev] Units package
we should harvest the Units....latest version
Begin forwarded message:
From: "Marcus Denker" <denker@iam.unibe.ch> Date: June 24, 2008 2:11:45 PM CEDT To: "Balázs Kósi" <rebmekop@gmail.com> Cc: The general-purpose Squeak developers list <squeak-dev@lists.squeakfoundation.org
Subject: Re: [squeak-dev] Units package Reply-To: The general-purpose Squeak developers list <squeak-dev@lists.squeakfoundation.org
Hello,
the bug seems to be fixed in the last version. I harvested the test to make sure (this is in Units-md.16 on SqueakSource)
The versions starting with -14 contain many enhancements (and the bugfix). This was contributed by Pinesoft (http://www.pinesoft.co.uk/).
Maybe the universe version could be updated?
Marcus
-- Marcus Denker -- denker@iam.unibe.ch http://www.iam.unibe.ch/~denker
On Thu, Jun 19, 2008 at 4:56 PM, Balázs Kósi <rebmekop@gmail.com> wrote:
UnitValue >> hash ^value hash bitXor: unit hash
and
testEquality
self deny: (60 units: #seconds) hash = (60 units: #meters) hash. self deny: (60 units: #seconds) = (60 units: #meters). self deny: (60 units: #seconds) hash = (1 units: #minutes) hash. self assert: (60 units: #seconds) = (1 units: #minutes).
Much better :)
Cheers, Balázs
ps: cc-d to Marcus Denker, because he seems to be maintainer of the package.
On 27.06.2008, at 09:04, Stéphane Ducasse wrote:
we should harvest the Units....latest version
Do we want it as part of the image? I am a fan of growing the number (and deepness) of abstractions available as part of what one percives as the "language"... e.g. I am quite sure that the math capabilities of the programming language of the future will be more similar to mathematica then to C. Having "native" support for dealing with Units is, I think, something that a language should have built in. In most situations, numbers are used to model something that has a unit... and today we have the memory and processing power to model this in most cases. The units package allows one to write expressions like: 25 inches as cm and do calculations with these units. But there is always the danger of putting "too much" into a system. All the stuff we are cleaning now in pharo is an example for that. So there is definitely a tension between providing more abstraction vs. simplifying the system to it's bare minimum. For the units, I lean towards including them. Marcus
Begin forwarded message:
From: "Marcus Denker" <denker@iam.unibe.ch> Date: June 24, 2008 2:11:45 PM CEDT To: "Balázs Kósi" <rebmekop@gmail.com> Cc: The general-purpose Squeak developers list <squeak-dev@lists.squeakfoundation.org
Subject: Re: [squeak-dev] Units package Reply-To: The general-purpose Squeak developers list <squeak-dev@lists.squeakfoundation.org
Hello,
the bug seems to be fixed in the last version. I harvested the test to make sure (this is in Units-md.16 on SqueakSource)
The versions starting with -14 contain many enhancements (and the bugfix). This was contributed by Pinesoft (http://www.pinesoft.co.uk/).
Maybe the universe version could be updated?
Marcus
-- Marcus Denker -- denker@iam.unibe.ch http://www.iam.unibe.ch/~denker
On Thu, Jun 19, 2008 at 4:56 PM, Balázs Kósi <rebmekop@gmail.com> wrote:
UnitValue >> hash ^value hash bitXor: unit hash
and
testEquality
self deny: (60 units: #seconds) hash = (60 units: #meters) hash. self deny: (60 units: #seconds) = (60 units: #meters). self deny: (60 units: #seconds) hash = (1 units: #minutes) hash. self assert: (60 units: #seconds) = (1 units: #minutes).
Much better :)
Cheers, Balázs
ps: cc-d to Marcus Denker, because he seems to be maintainer of the package.
_______________________________________________ Pharo-project mailing list Pharo-project@lists.gforge.inria.fr http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project
Marcus Denker wrote:
On 27.06.2008, at 09:04, Stéphane Ducasse wrote:
we should harvest the Units....latest version
Do we want it as part of the image?
Question is whether we view Pharo more as a distribution: small kernel and a (large) set of loadable additions that are know to work (pass tests). Which would be the model I prefer :-) I would even go so far to have minimalistic Date/Time features, but then the ability to load Chronos if I need it. Very simple HTTPClient vs libCurl or a full native implementation etc etc. Michael
On Jun 27, 2008, at 9:35 AM, Michael Rueger wrote:
Marcus Denker wrote:
On 27.06.2008, at 09:04, Stéphane Ducasse wrote:
we should harvest the Units....latest version
Do we want it as part of the image?
Question is whether we view Pharo more as a distribution: small kernel and a (large) set of loadable additions that are know to work (pass tests). Which would be the model I prefer :-)
I would that way too. Even if for regex I would do the inverse to clean some part that deliberately needs a good regex. Now I would like to check sake to see how we can build a process to check out and run tests automatically.
I would even go so far to have minimalistic Date/Time features, but then the ability to load Chronos if I need it. Very simple HTTPClient vs libCurl or a full native implementation etc etc.
Michael
_______________________________________________ Pharo-project mailing list Pharo-project@lists.gforge.inria.fr http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project
participants (4)
-
Hilaire Fernandes -
Marcus Denker -
Michael Rueger -
Stéphane Ducasse