+07Nov99=======================================================================
+1. reduced the amount of stuff compared against for dup announces.
+2. Check the regex for DK0WCY announces produces results before storing it as
+a WWV.
+3. Added some logging for set/priv (un)set/lockout.
+4. Added test long path calc to sh/muf
+5. Only wonder down the msg queue every minute
+6. Put in the initial DB code (at last), you can create and remove local and
+standard remote dbs, you can import AK1A style .FUL ascii databases, you can
+enquire on a local or remote database.
+7. A return ping to a node will clear down all outstanding pings to
+that node (which might cause some confusion if more then one ping is
+outstanding for a node, but then - shit happens).
+04Nov99=======================================================================
+1. Removed ~ from the end of the PC18.
+2. Removed a hangover from duff character checking in cluster.pl
+3. Stop PC frames that are to be routed, being routed back to themselves.
+03Nov99=======================================================================
+1. Simplified command caching so it uses anonymous subroutines, you should
+also get error messages back on the console now when developing.
+2. Allow locally connected AK1A clusters that for some obscure reason don't
+issue PC19s to still appear as connected and allow them to acquire users.
+31Oct99=======================================================================
+1. updated Minimuf.pm and show/muf.pl to the fixed versions sent to me by
+Steve Franke K9AN.
+2. Made set/badnodes work, added show/badnodes. This is an interim fix until
+input spot filtering works.
+24Oct99=======================================================================
+1. fixed announces so that nodes only get one of them (not two).
+2. Altered the select timeouts to prevent apparent excessive CPU usage on
+multiprocessor machines (or at least that is what seems to be the trouble
+here).
22Oct99=======================================================================
-1. allow filter names to be in upper or lower case.
+1. allow filter CALLSIGNS to be in upper or lower case.
2. create the concept of input and output filters, input filters are of the
-form "in_$call.pl" eg: "in_g1tlh.pl", output filters are "g1tlh.pl".
+form "in_$call.pl" eg: "in_g1tlh.pl" (or "in_G1TLH.pl"), output filters are
+"g1tlh.pl" (or "G1TLH.pl").
3. all users can now filter on spots, ann and wwv (assuming the sysop has set
the filters up - still no user creatable filters. Nodes can filter on both
-input and output for all of these.
+input for announces and output for all of these. I would recommend setting
+up a few filters that are likely to be wanted and then symlinking callsigns
+to them as you get requests.
4. added callsign and length of history/position of screen stats to console.
-DO REMEMBER to copy /spider/perl/Console.pm to /spider/local (doing any changes
-on the way as it has extra data items in it now).
+If you want to change the colour of the callsign you will need to copy the
+Console.pm to your /spider/local directory and change the $mycallcolor equate.
21Oct99=======================================================================
1. fixed reply so that reply b or reply nop generate a bulletin with the
correct address.
2. implemented sh/muf command.
-3. reduced privilege of catchup and uncatchup to 5.
+3. reduced privilege of catchup and uncatchup to 5. This so that, suitably
+privileged, nodes that occasionally link can rcmd a "catchup <nodecall> all"
+to the node they are connecting to and thus prevent a large tranche of,
+potentially duplicate, mail. This isn't an issue for DXSpider, but may be
+an issue for any AK1A clusters you are connected to.
20Oct99=======================================================================
1. Translated all the subroutines of minimuf into perl as Minimuf.pm
2. Limited the length of an ANN that is saved for de-duping to $pc12_dup_lth
projects / spider.git / blobdiff