Hi,
as written in other threads. PetitParser is using binarySelectors that allow more than two chars (allowed by Ansi Smalltalk) and we can not parse that, the other is that Magritte/PetitParser use #classSide and STClassLoader didn't know how to resolve that. For the binary selector I have added a patch that allows to parse it when it is coming from Squeak/Pharo and then we can rewrite it to => or such. For the #classSide I have used Paolo's patch and propose it for inclusion. I have not seen that the the STClassLoader allows to have dialect specific code so I think it is best to include it. For the origin of #classSide, I have clicked on versions in the Pharo browser and it appears to have a date from 2004, I don't know who added it though. so please comment on the ChangeLog and code. _______________________________________________ help-smalltalk mailing list [hidden email] http://lists.gnu.org/mailman/listinfo/help-smalltalk 0002-SqueakParser.st-Handle-binary-selectors-with-more-th.patch (1K) Download Attachment 0003-STClassLoader.st-Handle-classSide-to-be-able-to-pars.patch (1K) Download Attachment |
On 08/26/2010 04:33 PM, Holger Hans Peter Freyther wrote:
> Hi, > > as written in other threads. PetitParser is using binarySelectors that allow > more than two chars (allowed by Ansi Smalltalk) and we can not parse that, the > other is that Magritte/PetitParser use #classSide and STClassLoader didn't > know how to resolve that. > > For the binary selector I have added a patch that allows to parse it when it > is coming from Squeak/Pharo and then we can rewrite it to => or such. > > For the #classSide I have used Paolo's patch and propose it for inclusion. I > have not seen that the the STClassLoader allows to have dialect specific code > so I think it is best to include it. For the origin of #classSide, I have > clicked on versions in the Pharo browser and it appears to have a date from > 2004, I don't know who added it though. > > so please comment on the ChangeLog and code. All looks good, thanks! So now Magritte uses PP? :) Paolo _______________________________________________ help-smalltalk mailing list [hidden email] http://lists.gnu.org/mailman/listinfo/help-smalltalk |
On 08/26/2010 10:36 PM, Paolo Bonzini wrote:
> > So now Magritte uses PP? :) No, but I had issues converting magritte-tests with gst-convert for weeks (well everytime I tried) and it appears to be due this classSide thing. My plan for PetitParser is to parse Media Gatway Control Protocol (MGCP) messages and implement a simple Call Agent... progess is still rather slow as I have to find a good way to develop things. _______________________________________________ help-smalltalk mailing list [hidden email] http://lists.gnu.org/mailman/listinfo/help-smalltalk |
In reply to this post by Holger Freyther
On 08/26/2010 04:33 PM, Holger Hans Peter Freyther wrote:
> Hi, > > as written in other threads. PetitParser is using binarySelectors that allow > more than two chars (allowed by Ansi Smalltalk) and we can not parse that, the > other is that Magritte/PetitParser use #classSide and STClassLoader didn't > know how to resolve that. > > For the binary selector I have added a patch that allows to parse it when it > is coming from Squeak/Pharo and then we can rewrite it to => or such. > > For the #classSide I have used Paolo's patch and propose it for inclusion. I > have not seen that the the STClassLoader allows to have dialect specific code > so I think it is best to include it. For the origin of #classSide, I have > clicked on versions in the Pharo browser and it appears to have a date from > 2004, I don't know who added it though. > > so please comment on the ChangeLog and code. Both applied, thanks! Paolo _______________________________________________ help-smalltalk mailing list [hidden email] http://lists.gnu.org/mailman/listinfo/help-smalltalk |
Free forum by Nabble | Edit this page |