Re: [Pharo-project] MethodFinder should use pragmas
On Jun 17, 2011, at 10:33 AM, Damien Cassou wrote:
I would like to have your opinion on http://code.google.com/p/pharo/issues/detail?id=4426
The problem is that it looks not nice to have pragmas in all methods as part of the source code.. Imagine we do that for 5 different things (e.g. Slint), than the source is smaller than the annotations and readability suffers... -- Marcus Denker -- http://www.marcusdenker.de INRIA Lille -- Nord Europe. Team RMoD.
(my 2 cents) So maybe it can be cool to implement a way to annotate methods ... I think we should be able to provide more information about methods/classes/packages etc like which is the corresponding class test ? is this method well written ? is this method an extension ? and even comments At some point, it was useful (and maybe the only way to do it) to put such info in source code, but maybe now we should decide not to use strings anymore to attach specific behavior (I mean, is that obvious that an extension is a category starting by * ? or that a test method's selector must begin with 'test' ? ). I know it's easier to say than to do, but I think that a system based on string is out dated today :) So let's start 1.4 and break everything :D Ben On Jun 17, 2011, at 1:40 PM, Marcus Denker wrote:
On Jun 17, 2011, at 10:33 AM, Damien Cassou wrote:
I would like to have your opinion on http://code.google.com/p/pharo/issues/detail?id=4426
The problem is that it looks not nice to have pragmas in all methods as part of the source code.. Imagine we do that for 5 different things (e.g. Slint), than the source is smaller than the annotations and readability suffers...
-- Marcus Denker -- http://www.marcusdenker.de INRIA Lille -- Nord Europe. Team RMoD.
Yes. Couldn't have an hybrid solution where the inclusion would (automatic) be by the the protocol the method is in? -- Cesar Rabak Em 17/06/2011 08:40, Marcus Denker < marcus.denker@inria.fr > escreveu: On Jun 17, 2011, at 10:33 AM, Damien Cassou wrote:
I would like to have your opinion on http://code.google.com/p/pharo/issues/detail?id=4426
The problem is that it looks not nice to have pragmas in all methods as part of the source code.. Imagine we do that for 5 different things (e.g. Slint), than the source is smaller than the annotations and readability suffers... -- Marcus Denker -- http://www.marcusdenker.de INRIA Lille -- Nord Europe. Team RMoD.
I understand your fear but we should try and see. For SmallLint the problem is that there are too many rules and we should not have that many annotation. So metadata should be added to the packages. Stef
On Jun 17, 2011, at 10:33 AM, Damien Cassou wrote:
I would like to have your opinion on http://code.google.com/p/pharo/issues/detail?id=4426
The problem is that it looks not nice to have pragmas in all methods as part of the source code.. Imagine we do that for 5 different things (e.g. Slint), than the source is smaller than the annotations and readability suffers...
-- Marcus Denker -- http://www.marcusdenker.de INRIA Lille -- Nord Europe. Team RMoD.
On Sat, Jun 18, 2011 at 8:51 AM, Stéphane Ducasse <stephane.ducasse@inria.fr> wrote:
I understand your fear but we should try and see.
Ok, I submitted a proposal, please have a look http://code.google.com/p/pharo/issues/detail?id=4426 -- Damien Cassou http://damiencassou.seasidehosting.st "Lambdas are relegated to relative obscurity until Java makes them popular by not having them." James Iry
participants (5)
-
Benjamin -
csrabak＠bol.com.br -
Damien Cassou -
Marcus Denker -
Stéphane Ducasse