summaryrefslogtreecommitdiff
path: root/apps/drwho/ChangeLog
diff options
context:
space:
mode:
Diffstat (limited to 'apps/drwho/ChangeLog')
-rw-r--r--apps/drwho/ChangeLog250
1 files changed, 0 insertions, 250 deletions
diff --git a/apps/drwho/ChangeLog b/apps/drwho/ChangeLog
deleted file mode 100644
index 114661947ba..00000000000
--- a/apps/drwho/ChangeLog
+++ /dev/null
@@ -1,250 +0,0 @@
-Wed Sep 30 13:00:52 1998 Douglas C. Schmidt <schmidt@tango.cs.wustl.edu>
-
- * Fixed a bunch of warnings related to char * and const char *.
- Thanks to Sandro Doro <doros@aureus.sublink.org> for reporting
- this.
-
-Sat Sep 12 21:21:01 1998 Douglas C. Schmidt <schmidt@tango.cs.wustl.edu>
-
- * Made zillions of changes to "ACE"ify all of the code. Drwho is
- now officially working again!
-
-Sun Sep 6 22:48:52 1998 Douglas C. Schmidt <schmidt@tango.cs.wustl.edu>
-
- * Replaced all Str::*() methods with the new ACE::*() methods,
- which are more "standard".
-
- * Began to work on drwho after a 5 year absence...
-
-Sun Feb 14 16:13:10 1993 Douglas C. Schmidt (schmidt at net1.ics.uci.edu)
-
- * Yow. Once again back at work... This time I fixed things up so
- that the release compiles properly with Solaris 2.1 C++, g++
- 2.3.3, and the Sun C++ compiler on Sun OS 4.1.x. To do this I
- had to change the user-defined memory allocator somewhat, since
- the Sun C++ compiler was using the global NEW operator to
- allocate pools of memory behind my back... Naturally, this
- reaked havoc with the existing version in the server, which
- returns all the allocated memory to the free list once a request
- is satisfied. In addition, g++ had a weird multiple
- inheritance/pure virtual function bug that I fixed by reordering
- certain pure virtual functions. Oh what fun... ;-) Anyhow,
- everything is now back in running order!
-
-Sun Dec 6 16:13:21 1992 Douglas C. Schmidt (schmidt at net1.ics.uci.edu)
-
- * Yow, well, after about 8 months of total neglect I'm finally
- back at work on this program! The current changes I did today
- involved getting drwho to compile with g++ 2.3.2. This required
- a couple of small work arounds in the source code due to bugs,
- but hey, I guess it is better than nothing right?!
-
- Also, I had to fix a couple of places where I didn't correctly
- initialize static variables (the compiler implementations have
- changed since last year obviously). Also, I've changed a couple
- of the default server hosts to reflect the fascist policies of
- our support group ;-) (I've no longer got access to certain
- servers...)
-
- Anyhow, I think drwho is once again running with g++, so now I
- can hand the release over to Ericka... ;-)
-
-Wed May 6 23:12:02 1992 Douglas C. Schmidt (schmidt at net4.ics.uci.edu)
-
- * There is something horribly wrong with g++ 2.1. Therefore, I
- had to make a couple of changes in the pmc-rusers.C and
- pm-client.C files in order to make the blasted thing compile
- when I had pointers to member functions... Make sure to change
- this back when g++ is fixed...
-
-Tue Feb 4 11:23:12 1992 Douglas C. Schmidt (schmidt at net6.ics.uci.edu)
-
- * The blasted program was crashing when there were consecutive
- newlines in the input file. I fixed
- File_Manager::get_login_and_real_name and
- File_Manager::open_friends_file so that they detect and skip
- over these consecutive newlines.
-
-Thu Nov 21 21:55:00 1991 Douglas C. Schmidt (schmidt at net6.ics.uci.edu)
-
- * Make the time that drwho uses to consider a host idle be a
- command-line parameter.
-
-Wed Nov 13 14:24:40 1991 Douglas C. Schmidt (schmidt at bastille.ics.uci.edu)
-
- * Something else that ought to be done:
-
- Add a flag to the -R option that allows the hostname to be
- printed using the internet address and/or the hostname...
-
-Sat Nov 9 13:45:30 1991 Douglas C. Schmidt (schmidt at bastille.ics.uci.edu)
-
- * Added zillions more changes yet again. We now have an rusers
- compatibility mode (-R), to go along with the -a, -w, and
- default (flo) options.
-
- * Changed the -w option so that it only returns the name of the
- host machine where the user is logged in. This way, I can say:
-
- % talk schmidt@`drwho -w schmidt`
- % rsh `drwho -w schmidt` w
-
- etc... ;-)
-
- * Added support for the -L option (print using login name rather
- than real name). Also added support for the -l option (print
- out verbosely, a la ls -l!).
-
-Sun Oct 27 21:32:15 1991 Douglas C. Schmidt (schmidt at bastille.ics.uci.edu)
-
- * Need to complete the -s and -S options to support sorting the
- output by login name and real name, respectively... In order to
- support the '*' indication with this scheme we probably need to
- lists, one for active and one for inactive users!
-
-Tue Oct 22 00:13:21 1991 Douglas C. Schmidt (schmidt at net6.ics.uci.edu)
-
- * Make another zillion changes...
-
-Sun Oct 20 21:35:24 1991 Douglas C. Schmidt (schmidt at net6.ics.uci.edu)
-
- * Added support for the -p option to allow setting the client and
- server port number from the command-line.
-
- * Things done so far:
-
- * Owen also wants an new rflo feature (done)
-
- I also want a version that given a command like:
-
- whereis omalley
-
- would return the login where that login is active if there is
- one (ie. omalley@zola). Then you could have commands like:
-
- talk `whereis omalley`
-
- that would find where I am and try to talk to me there.
-
- * Another neat addition would be: have an option (e.g., `-a') so
- that rflo would return *all* the users logged in and then look
- up their names using the yp passwd stuff! (partially done, but
- not very elegantly yet...). (done)
-
- * Make the port number a command-line option... (done)
-
- * we also need think about how to incorporate inheritance and
- dynamic binding into this thang (probably it can be used for
- the local/remote split, and also perhaps for the
- friends/everyone split (see below)). (done)
-
- * Fix up the options stuff wrt the -F option etc... (done)
-
- * Have I fixed the is_active shit? (done)
-
- * we need a "message abstraction" that abstracts away from the
- details of packets protocols and remote operations protocols.
- (done).
-
- * Note, should make an option so we could read the names of the
- hosts to query from a file... (done)
-
-Fri Oct 18 16:17:39 1991 Douglas C. Schmidt (schmidt at net6.ics.uci.edu)
-
- * I've made countless changes...
-
-Wed Oct 16 17:42:40 1991 Douglas C. Schmidt (schmidt at net6.ics.uci.edu)
-
- * Undid the message manager abstraction and merged it in with the
- Friend_Manager client and server. This really cleans up the
- interface!
-
- * Yow, make zillions of important changes to make the
- client/server split more explicit... Now the Friend_Manager is
- split into client and server, the message manager is split, and
- the communications manager is also split. Each file is much
- smaller and easier to understand!
-
-Mon Oct 14 18:36:55 1991 Douglas C. Schmidt (schmidt at net1.ics.uci.edu)
-
- * Added support for the -h and -? options, that print out a long
- and short usage message, respectively.
-
-Tue Oct 1 09:28:29 1991 Douglas C. Schmidt (schmidt at net4.ics.uci.edu)
-
- * The -a option works a great deal better too... I added a check
- in the File_Manager::open_passwd_file routine to strip off the
- extra subfields in the pw_gecos field, since this info isn't
- really very useful and makes the "real name" field too long!
-
- * Added some extra stuff to the Comm_Manager so that I could
- change the max size of the UDP datagrams that are passed around.
- As it turns out, I don't really need to do this, but it is more
- robust this way...
-
-Thu Sep 26 14:00:45 1991 Douglas C. Schmidt (schmidt at net4.ics.uci.edu)
-
- * Make sun3 and sun4 subdirectories to help the build process!
-
- * Changed the UDP port number from 12346 to 12344 so I wouldn't
- collide with Owen!
-
- * There is a weird bus error problem on the sun 4s... Hum...
-
- * Yow, got everything working again...
-
- * Still to do:
-
- * Think about fixing the -f option to work correctly for
- *relative* filenames...
-
- * Make UDP_BUFFER_SIZE a command-line option...
-
- * Modify server.C to be started by inetd.
-
- * For -a option... if they are a friend, use the name from
- .friend.dta otherwise use the name from the passwd file.
-
- * Don't forget about:
-
- delete Friend_Manager::friend_record;
- delete Friend_Manager::sorted_record;
-
- Need to figure out a good way to deal with this!
-
-Mon Sep 23 16:09:46 1991 Douglas C. Schmidt (schmidt at net4.ics.uci.edu)
-
- * Added a neat feature that now allows the user to specify which
- hosts to examine by specifying an option ('-F') on the
- command-line that reads the contents of that file and inserts it
- into the list of files.
-
-Wed Sep 18 10:46:41 1991 Douglas C. Schmidt (schmidt at bastille.ics.uci.edu)
-
- * We need to make all the interfaces throughout rflo much more
- object-oriented, e.g.:
-
- * Made a host-manager abstraction to handle all the host
- machine related operations. This makes the options stuff
- *much* cleaner!
-
- * Make rflo compile with g++ 1.37.2! Now it compiles with cfront
- 2.0, Saber C++ 1.0.1 and g++-1.39.0 and g++-1.37.2.
-
-Tue Sep 17 19:02:47 1991 Douglas C. Schmidt (schmidt at net4.ics.uci.edu)
-
- * Started merging in the stuff Owen did. I'm trying to maintain a
- consistent programming style... The Owen stuff adds support for
- timeouts and fixes problems with returning a count of the number
- of friends!
-
-Wed Sep 4 10:14:51 1991 Douglas C. Schmidt (schmidt at net4.ics.uci.edu)
-
- * Things to do:
-
- * Add comprehensive daemon support for server.C.
- * Add timeout stuff in case hosts are down!
- * Talk to support about making a standard daemon.
- * Fix the problem with returning the number of friends.
-
-