Hello, everyone Clinton? Would you be willing to turn up the Discussion feature on the SourceForge project space? I've been managing the GnuCOBOL project space there, and it's a pretty good forum, out of some few thousand threads, and many tens of thousands of posts, we've had to deal with only 3 pieces of spam, and that's with one of the groups (Help getting started) open to anonymous, unregistered submissions.
This is still pending, but the Pygments lexer is coming along as part of the docset I'm working on (long roadmap, 100 pages in and it's still not even a full outline, let alone getting into any nitty-gritties), Syntax highlighted listings make for a better conversation, in my humble opinion, and thread web pages seem a little accessible than just a mailing list. The highlighter won't be visible on the forge until I get the code accepted by team Pocoo for the Pygments layer, and then it'd be another wait until a source release makes it to the Allura team, but it will be in the not too distant future. I know it can split focus, but the project space there already hosts the mailing lists, so access to the archives will always be visible on the top-level menu of the Allura Unicon space, along with a Discussion tab (if you deem it worthy of turning on). Just curious, and a little bit hopeful. I find it more accessible, even with the bad rap the previous owners (Dice) tainted SourceForge with. The new owners (BIZX) seem quite a bit more concerned about reputation then scraping nickels with sleazy tactics. I see it as a fight with good speech to drown out the bad. Next up; how fixed is the 22 (MVS) and 13 (everybody else) column showline tracer field widths? I find -t with 13 character filename space way too short. I've tweaked my own copy or rdebug.r to go out as far as 26, (with a 5 digit line-number field, as I'm thinking that a 10,000 line Unicon program is not that far of a stretch. Large, yes, but with some of the features in Unicon, programming in the large doesn't seem to be a restriction). But, not knowing the ins and outs of all the support utilities yet, I'm wondering what tracer support programs that will break, or if you'd be up for accepting a patch request? Thought I'd ask here before I get too used to the extra wiggle room provided by 26 characters in the filename slot. http://peoplecards.ca/unicon/statements.html#suspend (for example) Next; I have to do a little more testing, but I'm getting a segfault with https access. # # Try https # procedure main() w := open("https://sourceforge.net/projects/open-cobol/";, "m") | stop("Can't") line := read(w) write("Line: ", line) while write(read(w)) end prompt$ unicon https.icn - Parsing https.icn: . /home/btiffin/inst/langs/unicon-svn/bin/icont -c -O https.icn /tmp/uni21056032 Translating: https.icn: main No errors /home/btiffin/inst/langs/unicon-svn/bin/icont https.u -x Linking: Executing: Run-time error 302 File https.icn; Line 6 memory violation Traceback: main() open("https://sourcefo...","m";) from line 6 in https.icn This is rev 4470 on a 64bit Ubuntu node, but I'll do a little more digging before making a bug report. I may well have a borked openssl config, blended with gnutls. And lastly; Thanks again for Unicon, Clint, Jafar, Steve, Robert and crew. Having way too much fun, and the rabbit holes are a code spelunker's delight. Next few days will be getting used to all the neato tools that Robert Parlett has written up. Deep, rabbits, Mmmm. Have good, make well, Brian ------------------------------------------------------------------------------ _______________________________________________ Unicon-group mailing list Unicon-group@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/unicon-group
