X-Git-Url: http://dxcluster.net/gitweb/gitweb.cgi?a=blobdiff_plain;f=Changes;h=fe4156cda47a89616dedfc5c8b3d595e9b15f126;hb=b326a39e5cf429a4c8fe52026f937f7b011cbca2;hp=21ab3c66dcd39eb99640d6a9d7ea3bd5a8eb723e;hpb=6e210063f1ef4e5c36ad04ba029b59711cc1eb04;p=spider.git diff --git a/Changes b/Changes index 21ab3c66..fe4156cd 100644 --- a/Changes +++ b/Changes @@ -1,5 +1,73 @@ +27Jun01======================================================================= +1. try adding an ARC mail hack +25Jun01======================================================================= +1. added automatic 'not here' message +21Jun01======================================================================= +1. fix (some) problems with 'init' +2. fix stat/msg a bit more +3. small change to PC28 generation to make sure that the subject always +contains at least a blank +16Jun01======================================================================= +1. fix talk +2. fix stat/msg +14Jun01======================================================================= +1. changed debug api to (potentially) use less cpu time +2. changed the spider.cgi in Arnold's borrowed java client to Ian's perl +script +3. Added input route filtering +4. put back the set/isolate logic into routing which is now overridable with +filters whcih means a default filter is no longer generated. If you want the +OLD behaviour, do rm /spider/filter/route/*.pl*, restart and try again. +5. Add http_proxy to Internet.pm and allow sh/qrz to use an HTTP Proxy +13Jun01======================================================================= +1. fix init cnd rinit cmds +2. add missing clear/route cmd +10Jun01======================================================================= +1. fix set/here and unset/here +2. fix indirect routing to users two hops away +09Jun01======================================================================= +1. more fixes +2. add set/debug filter so that you can see why your filters aren't working +(for sysops only) +3. add ephemera deduping for all those PC41,24,50 etc etc broadcasts +4. Fix a problem with the connect text being sent on outgoing connections +(this confuses db0fhf amongst other things) +5. catch deep recursion bugs before you run out of memory. +6. Make sure that PC16|17|19|21 update as well as add and delete +7. local announces by the sysop will always be displayed +08Jun01======================================================================= +1. first cut with new routing code. Created NEW_ROUTE branch +2. added acc/route and rej/route commands +3. added stat/route_node and stat/route_user commands to look at the routing +table entities +07Jun01======================================================================= +1. move userconfig.pl and nodeconfig.pl to stat/ +2. fix problem with locally connected users not being removed from the +routing tables +3. Fix the build calc to take into account subdirectories +06Jun01======================================================================= +1. add stat/route_node and stat/route_user commands +2. add sh/userconfig command (to show backward links to nodes) +3. add sh/nodeconfig command +05Jun01======================================================================= +1. add set/bbs command +2. more work on Routing code. +3. status/msg on its own will print the status of the msg system. +4. add sh/newconfig command +03Jun01======================================================================= +1. Fix the problem with ExtMsg and unresolvable IP addresses, hopefully +properly this time. +15May01======================================================================= +1. set/lockout now prevents any outgoing connection taking place. +2. Started the new routing stuff which will run in parallel for a while. +3. changed the msg timeout arrangements so that it might not get stuck so +easily. +4. add $DXMsg::bulltopriv global to make msgs to known callsigns private 14May01======================================================================= 1. fix problem with re-reading in db definitions for remote databases. +2. try to prevent situations where two can (semi) successfully login, probably +under load, and get a 'tried to connect whilst already connected' abort. +3. change the max number of days to search for spots to 100 11May01======================================================================= 1. allow stat/msg on priv 1 07May01=======================================================================