SourceForge has been redesigned. Learn more.
Close

User Activity

  • Posted a comment on ticket #32 on The Unicon Project

    On 08/01/2017 08:44 AM, Steve Wampler wrote: On 07/30/2017 09:29 PM, Jafar wrote: I have to double check, but I think we already retained the old sematics with native co-expressions. Or in other words, native co-expression always have had the semanics when it comes to string scanning environment. It is just with the initial implementation of concurency we didn't even have support for native co-expressions. Now that they are bak, the old semantics came back with them. More testing is needed probably...

  • Posted a comment on ticket #32 on The Unicon Project

    On 07/30/2017 09:29 PM, Jafar wrote: I have to double check, but I think we already retained the old sematics with native co-expressions. Or in other words, native co-expression always have had the semanics when it comes to string scanning environment. It is just with the initial implementation of concurency we didn't even have support for native co-expressions. Now that they are bak, the old semantics came back with them. More testing is needed probably to confirm this and also to uncover and fix...

  • Posted a comment on ticket #32 on The Unicon Project

    On 07/30/2017 10:50 AM, Charles Evans wrote: ... It is my understanding that most functional languages have closures that are essentially snapshots. Hopefully we only want snapshots of local variables, which are already provided by the coexpr code, which also already handles nested creates. ... One change to co-expression 'snapshots' that was made a while back was to include &subject in the snapshot. This change has the unfortunate effect that you can no longer write co-expressions (and, hence, closure)...

  • Posted a comment on ticket #32 on The Unicon Project

    On 07/04/2017 02:49 PM, Bruce Rennie wrote: Much as this would be nice, a problem that exists within the Unicon/Icon language that works directly against this is the fact that there are global variables and only one level of local variables. Procedures cannot have procedures defined in themselves - not block structured. I think there are two factors at play here: -- first class closures should have access to the surrounding environment (locals in the defining environment and globals). Faking closures...

  • Posted a comment on discussion Open Discussion on ROBODoc

    I've been using robodoc for a while to produce html docs and just started trying...

View All

Personal Data

Username:
swampler
Joined:
2000-09-08 17:41:05

Projects

  • The Unicon Project A modern descendant of the Icon programming language. Last Updated:

Skills

  • No skills entered.

Personal Tools