1 - too many too big repaints in themer
3 - fix strings & add KUIT
5 - allow theme to specify gui-style & color-scheme.
7 - themer: interpret altfile attribute to image nodes
9 - initialize theming 'mode' properly. need to adapt action menu construction.
10 problem: the modes do not cover certain kdmrc settings. (partial patch)
12 - do something about GreetString, i18n & theming. need some magic for the default.
14 - error labels use fixed color. fix needs support from QPalette.
16 - message after switching to text mode
17 - report critical startup errors to the console
18 - set LANGUAGE in the backend for i18n-capable PAM libs
20 - handle non-linux VTs:
21 on systems without VT_GETSTATE, try activating all consoles in turn to
22 find free ones. wow, this sucks so much.
23 - BSD: 1st: pcvt, /dev/ttyC[0] (OpenBSD), /dev/ttyv[0] (other),
24 also emulated by wscons on /dev/ttyE.
25 2nd: syscons, /dev/ttyv[0], fallback /dev/vga
27 - Solaris, /dev/vt[00]
28 - SVR4, /dev/vc[00] (ESIX), /dev/vt[00] (other)
29 - SCO, /dev/tty[00], query current with CONS_GETINFO, counts 0-based
30 ref: xorg/programs/Xserver/hw/xfree86/os-support/xf86_OSlib.h
33 - possibly parse Xserver log to find failure cause. this is very hacky.
35 - try harder to get rid of processes, see X servers failure cleanup path
37 - make auto-re-login (& EchoPasswd) per-user option; save in .dmrc.
39 - add Xserver option set selection (#56329)
40 - add support for XRandR (#48602)
41 save these options to .dmrc?
43 - per-display sections in .dmrc. read-only, as far as kdm is concerned, as
44 otherwise the GUI would become insanely complex.
46 - make config position independent
47 - warn about conflicting {Static,Reserver}Servers settings. more sanity checks?
49 - merge multiple kdmrcs in the style of kconfig.
50 how to set section priorities? the included file is recessive to *any*
51 matching entry from the master file.
54 - config inherits paths that point directly into the previous installation:
55 LogoPixmap & Theme in kdmrc, Wallpaper & WallpaperList in backgroundrc -
56 to fix the latter, treat backgroundrc as an ini file, not as a text blob
57 - add proper quoting and dequoting to ini parser & writer
58 - determining *uid could use finding obviously valid uids (valid password,
59 maybe home paths) and rounding
61 - kcm: deal with NoPassUsers=*
62 - kcm: detach backgroundrc change status from kdmrc change status.
63 - write a separate configurator application, as kcontrol does not scale well
64 enough to cover all of kdm's options.
66 - write generic conversation plugin
67 - write modern conv plugin. or maybe this should be a parallel vs. serial
68 setting of the classic plugin? (#148110)
69 - change conversation plugin interface:
70 - make interactions that are not part of the main data flow signal/slot based
71 - make Activity and Changed seperate notifications
73 - actually implement the libpam_client support
75 - check if pam works before trying to authenticate
76 - test whether nis, kerberos4 & kerberos5 work
77 - sync BSD_AUTH from xdm, sync osfc2 from kcheckpass
79 - Move clock from greeter dialog to desktop
80 - add more clock types (#18178)
81 - make kdm use plasma?
82 - add icons to action menu. icon theme selection!
84 - make greeter accels use ctrl instead of alt
86 - make auto-completion in login field cope with case-insensitivity
88 - Add XDMCP _client_ to core (for remote login like in dtlogin).
89 Currently this is done by simply restarting the x-server with -query.
90 Can't be done without extending XDMCP, though.
92 - many simultaneously running choosers will torture the target servers
93 independently from each other.
95 - add login restrictions for reserve displays (#59353)
97 - possibly do the authentication for the reserve display on the display it
98 is launched from (relates #59353)
100 - remote-accessible command sockets for remote shutdown, etc.
101 or maybe implement it as an xdmcp extension?
102 - LoginMode=DirectQuery
104 - "XDMCP over FiFo" - or at least a "manage <dpy> [<xauth>]" command
105 - the per-display sockets are in fact nonsense; gdm's approach is better
106 - use dbus instead of command socket?
108 - add bgset to XDM_MANAGED
109 add socket command "background\t{inprogress,aborted,done}"
110 - actually, remove XDM_MANAGED and do everything via the socket.
112 - lilo boot option <default>, i.e., -R with no argument
113 - support lilo -A mode
115 - support suspend/hibernate/standby in the shutdown menu. should this be really
116 treated like a shutdown? (#33839)
118 - add language selection (export as LC_*). (#55379, #63804)
119 - add keymap selection (via xkb) (#51245, #64642)
120 for both, one would preset a list of available options and make one entry
121 the greeter's own setting. explicitly setting it sets it for both the greeter
122 and the session. .dmrc later affects only the session, not the greeter.
124 - handle failsafe internally, take care of focus. see #32973
126 - TryExec for "custom" session type. always show the entry, but disable it
127 if it is unavailable for the selected user.
129 - cursor theming support via Xcursor (#66829)
131 - add screensaver (#41941)
132 - support DPMS (#18597)
134 - add a minimalistic window manager to the greeter (#17716, #51039)
136 - Different logos for each session type (see #74500)
137 - User pictures in logo field
138 - display user's .plan/.project (or .person? .userinfo?) in the greeter?
139 text area/label would suck -> tooltip?
141 - allow disabling full names or login names in userview (#54110)
142 - user list loading in the background (after first few to get a reasonable
145 - faking session parameters (type, language, etc.) of nonexistent users based
146 on statistical analysis of actual users ... severe overkill!?
148 - export password to the startup/session scripts. somehow ... (#35396)
150 - maybe reset CapsLock in the greeter. there is some CapsLock vs. ShiftLock
153 - maybe add kiosk mode: the user and his options are preset and locked in
154 the greeter. i doubt it's usefulness, though.
156 - make builtin xconsole hideable; it should free the device when invisible.
157 possibly auto-hide it on vt switch - see krunner_lock for the x event
160 - ssh-agent/gpg-agent integration (#44177, #65709)
162 - lbxproxy integration (tell ghakko)
164 - make shutdown an in-daemon function, not a pre-exit action.
165 displays need to wake up again if nothing happens, though - add
166 configurable shutdown timeout.
167 - when a shutdown is scheduled, don't remove all login possibilities.
168 instead, display a warning in the greeter. use SIGUSR1 to notify already
169 running greeters about changes.
170 - user notification about scheduled shutdown (and cancelled forced shutdown):
173 - d-bus message. this would be best, particularly because screen savers
174 would need no special handling then.
175 - maybe bomb DefaultSdMode, save in state file instead. compare with ksmserver.
177 - gdm changelog indicates that PAM sometimes
178 - continues despite PAM_CONV_ERR
179 - asks user name twice
180 - gdm avoids the PAM_MESSAGE message box vs. prompt problem by displaying
181 everything in one "error area". all messages are simply appended; an empty
182 message clears the area.
183 - gdm stops cursor blinking on not used (remote) displays after 20 secs to
187 - improve signal handling in the subdaemon, it's incredibly racy (SIGTERM, etc).
188 depends on proper main loop.
189 - greeter sometimes won't shutdown when killing kdm - prolly some race
190 - the process reaping from gClose should be in sync with the main loop.
191 - kill warning on AIX - see bug #13628 (really present?)
192 - xdmcp.c: make networkAddressToName use networkAddressToHostname?
193 - options for running the greeter and the core unprivileged. problem: xauth.
194 - make the openbsd _x11-user hack in auth.c sane
195 - rethink the coupling of the kdm components, particularily the config reader.
197 - keep things basically as-is, make the Xaccess interface even more flexible,
198 add capability flags.
199 - as previous, but don't use #defines, but textual constants. even more
200 flexible, but slower, bigger, no compile-time checking, and the typing
201 system would have to be more core-based. keys in the rc are considered
202 invalid if they were not queried.
203 - completely opposite: no explicit queries, but hard-code everything. that
204 kills the idea of having one backend binary for multiple frontends, but
205 that's a BlueSkyDream anyway.
206 following that path, the config reader could be nuked at all. will have to
207 do this for performance reasons anyway if some some sort of call-back
208 mechanism for option values is added to allow some automation (depending
209 on boot options, for example). possible callbacks are KConfig-like [$e]
210 interpretation and an include statement that generates a config on the fly.
213 - put the kmenu sidebar image on the left of the greeter
214 - enable the clock by default
216 thoughts (not really todo):
217 - PAM sucks. big time.
218 historically, it is completely incapable of operating in event-driven contexts
219 when it comes to non-console authentication schemes. the module just hangs in
220 pam_sm_authenticate() (pam_authenticate() to the outside), waiting for input
222 then came linux-pam 0.58, introducing PAM_BINARY_{MSG,PROMPT} to the
223 conversation function interface. no conversation function could handle the
224 binary prompts generically, of course. so came linux-pam 0.63 with a client
225 library that would add another layer of indirection, so the conversation
226 function could simply call into it and it would do whatever was configured
227 by the admin. and everbody was happy, right? wrong! i've yet to see a single
228 module (except for the demo module in linux-pam, of course) that actually
229 uses this feature. not to mention the non-existing portability (you don't
230 seriously expect TOG to extend the PAM standard within the next decade, do
231 you?). so we're right where we started from.
232 this imposes problems in two use cases:
233 - cancelling authentication alltogether. this happens when the user changes
234 the authentication method or when the greeter exits for some reason. if
235 the process waits in the conversation function, it can simply return
236 PAM_CONV_ABORT. if the module hangs, we're screwed.
237 - suspending authentication. this is needed for shutdowns that need auth.
238 if the module hangs, we're screwed, of course. if we're waiting in the
239 conversation function, we have three options: 1) just abort the auth
240 cycle and start a new one. this is what is done currently. 2) just open
241 a second pam handle and authenticate with it, all from within the "outer"
242 pam_authenticate(). if we're lucky, no involved modules use static variables
243 and things work out. 3) linux-pam 0.65 introduced the following: the
244 conversation function can return PAM_CONV_AGAIN. this in turn makes the
245 module and consequently libpam return PAM_INCOMPLETE, requesting the
246 application to call the resp. libpam function again. in theory this
247 guarantees that authentication with a second pam handle is safe. of course,
248 PAM_INCOMPLETE is just as popular and thus useful as PAM_BINARY_PROMPT.
249 we could just longjmp() out of hanging modules from a signal handler.
250 however, this might lead to resource leaks and even leave us with an unstable
251 libpam. killing the hanging process seems like the most viable solution.
252 however, for this we first need to make the greeter a child of the master
253 daemon. also, the display sub-daemon (which happens to do the main auth.)
254 is responsible for keeping the initial X connection open. killing it would
255 terminate the session according to the XDMCP spec. other issues are probable.
256 - multiple conv. plugins could be used in a row, each serving a pam module.
257 the plugins would have to detect that it's their turn by filtering messages
259 - consider making the menu an actions-only menu again and put an "options >>"
260 button somewhere. relates #63401, #61492
261 - pipe .xsession-errors through the daemon and put a size limit on it.
262 remove old logs in disk-full situation.
265 last sync with XFree86 HEAD: 2004-04-02