Hello all, A late January release date for 5.0.6 seems more and more likely. Here's an overview of what's in store for you: New Developments ---------------- The MacPerl BBEdit extension glue now works quite well. While BB Software has sort of scooped me by doing a version of BBEdit which supports the AppleEvent Object Model on text, my more complicated strategy (of translating every BBEdit extension callback into an AppleEvent exchange) should give comparable performance and makes quite nice looking scripts. As an Example: -------- #!perl use Mac::BBEdit; PresetUndo; $text = Copy; $text =~ s/the/da/mg; Paste $text; SetUndo; --------- does a simple text substitution in an Undo-friendly way. While the Mac::BBEdit module is not object oriented, it also uses more Perl5 features than most Perl code I had written before. Badly needed improvements ------------------------- There are still a few open bugs (the help menu and syntax checking, for instance). Here's what I have fixed so far: - Redirecting STDIN would not work with <> [Jrgen Christoffel]. - After the first script execution, STDOUT would become unbuffered [Jeff Johnson]. - Perl would crash when running two CGI scripts in sequence (actually a bug in the CGI glue) [Steve Goodwin]. - localtime, gmtime, and related routines and libraries exhibited various problems [Brad Cox, Dave Wodelet]. - Fixed a memory leak in sort [Vicki Brown]. - Performance in the background and for CGIs should be significantly improved. - Added a new AppleEvent communication model for CGIs and BBEdit glue, which cuts the number of AppleEvents exchanged by a factor of 2 to 3. Final "Forwarding..." Fallout Follow-Up --------------------------------------- The MacPerl list is now on autopilot again. The test message I had to send to everyone had a quite beneficial effect in that numerous subscribers used the opportunity to inform me of minor address improvements. Those of you who personally complained to the NetCS manager probably have gotten a reply from him by now. He also directed one reply to the list, but I didn't think it had anything substantial to add; he seemed fairly unrepentant and blamed the user. I have pointed out the technical problems with NetCS tactics to him. The "offending" user is off the list until he resubscribes under a NetCS-free address. Going through my log of bounces, I noticed that a somewhat more agressive tactics at dealing with bounces (which currently get appended to a huge file that I scan about once every 3 months) might be beneficial to the smooth operation of the mailing list. I'll see what I can do in that respect. Matthias ----- Matthias Neeracher <neeri@iis.ee.ethz.ch> http://err.ethz.ch/members/neeri.html "They are rich and famous artists, two of the major talents of their time -- but they would have long since been legally put to sleep in any properly organized society" -- Hunter S. Thompson, _Generation of Swine_