I thought I made efreet not fail if dbus wasn't available? There is some code there which is intended to do it atleast.
Sebastian ons. 24. jun. 2015, 04.07 skrev Carsten Haitzler <[email protected]>: > On Tue, 23 Jun 2015 17:57:38 +0100 Daniel Kolesa <[email protected]> > said: > > > On Tue, Jun 23, 2015 at 5:42 PM, Mike Blumenkrantz > > <[email protected]> wrote: > > > On Tue, Jun 23, 2015 at 12:24 PM, Daniel Kolesa <[email protected]> > > > wrote: > > > > > >> On Tue, Jun 23, 2015 at 5:16 PM, Mike Blumenkrantz > > >> <[email protected]> wrote: > > >> > It would be great if, in the future, you could refrain from reverts > like > > >> > this. An ongoing mailing list discussion was taking place, and, > > >> regardless > > >> > of what you feel "should never be done", now the issue is once again > > >> > unsolved. > > >> > > >> In my experience if a bad solution is left in the tree for too long, > > >> it'll remain there forever, thus I choose to revert it now. This is > > >> one of the most awful pieces of code I've seen in the EFL yet and I > > >> couldn't sleep at night if it was left in, so at least this allows me > > >> to have a peace of mind; the issue was never really "solved" in the > > >> first place anyway - touching internal symbols within libraries is a > > >> big no (I'm sure most of other developers here will agree with my > > >> statement) and I would appreciate if you discussed things like this on > > >> the ML before actually introducing them. > > >> > > >> Regards, > > >> > > >> D5 > > >> > > >> > > > I think this is a very misleading reply. Ignoring the obvious (and > quite > > > unnecessary) use of hyperbole, there are plenty of ways to ensure that > code > > > doesn't remain in the tree forever. I'll list a few of them so that > you can > > > use it as a reference in the event that you have this urge to revert > again > > > in the future: > > > > > > * Participate in ongoing mailing list discussions > > > * Open a ticket on the bug tracker > > > * Engage in discussion on IRC > > > > It's funny it's you who's telling me this... > > hey - enough of the snarking. > > mike is right. there is a reason i replied to the mail and didn't just > revert. > i was expecting a "hmm ok - i see" at least, then a revert either by mike, > or > by me or someone else after that. > > reverting imho is ok IF the commit does seriously break something. people > pull > updates from git all the time and if a break is going to leave others in a > seriously broken setup, then indeed revert rapidly to get back to working, > and > resolve the issue after that emergency fix. > > this is not such a case. > > > > Your claim that the issue was not solved is also inaccurate. This is > > > exactly how the symbol is accessed in other parts of the tree, > > > > You sure? http://pastebin.com/UVUdVuS2 > > > > The symbol is only accessed within efreet, which is fine. I'm talking > > about cross-library internal symbol accesses, which IS bad, and it's > > not being done anywhere in the EFL. > > > > > and having > > > this ability is one of the great benefits of the unified source tree > that > > > we have. Considering that Edje already depends on Efreet for function > > > calls, the worst that could happen here is that calling edje_init() at > some > > > point in the future causes a crash--something which would be instantly > > > noticeable since a general efl tree build would then fail. > > > > That doesn't matter. Using internal symbols across libraries like this > > is bad by definition and should not be encouraged. > > > > > > > > If you can provide me with a comprehensive list of "things like this" > which > > > bother you when they are committed, I can try to initiate discussion > in the > > > community for future cases. > > > > That's ridiculous... that this is an ugly hack should be common sense, > > and if you're unsure about how to make it non-hacky, just initiate a > > discussion and we can try to work it out some other way. > > in this case the root cause is not related to this symbol. i agree that > digging > into the internals of efreet from outside of it is bad/ugly and even if it > was > formalized as an api call, it is working around a core issue (dbus session > requirements). > > > >> > On Tue, Jun 23, 2015 at 5:22 AM, Daniel Kolesa < > [email protected]> > > >> wrote: > > >> > > > >> >> q66 pushed a commit to branch master. > > >> >> > > >> >> > > >> >> > > >> > http://git.enlightenment.org/core/efl.git/commit/?id=d9db8888ac51175e350853597b3b317e9774a00a > > >> >> > > >> >> commit d9db8888ac51175e350853597b3b317e9774a00a > > >> >> Author: Daniel Kolesa <[email protected]> > > >> >> Date: Tue Jun 23 10:22:36 2015 +0100 > > >> >> > > >> >> Revert "edje: unset efreet cache update flag to prevent dbus > > >> >> connections" > > >> >> > > >> >> This reverts commit 1edb35fff3fe54ac7eea1ba2c26e509284b4e470. > > >> >> > > >> >> Accessing symbols from other libs like this should never be > done. > > >> The > > >> >> consequences of this are not significant enough to do this; > better > > >> >> solution > > >> >> can be found but definitely not like this. > > >> >> --- > > >> >> src/lib/edje/edje_main.c | 2 -- > > >> >> 1 file changed, 2 deletions(-) > > >> >> > > >> >> diff --git a/src/lib/edje/edje_main.c b/src/lib/edje/edje_main.c > > >> >> index a891ec5..03c46ea 100644 > > >> >> --- a/src/lib/edje/edje_main.c > > >> >> +++ b/src/lib/edje/edje_main.c > > >> >> @@ -1,6 +1,5 @@ > > >> >> #include "edje_private.h" > > >> >> > > >> >> -extern int efreet_cache_update; > > >> >> static Edje_Version _version = { VMAJ, VMIN, VMIC, VREV }; > > >> >> EAPI Edje_Version * edje_version = &_version; > > >> >> > > >> >> @@ -70,7 +69,6 @@ edje_init(void) > > >> >> goto shutdown_eet; > > >> >> } > > >> >> > > >> >> - efreet_cache_update = 0; > > >> >> if (!efreet_init()) > > >> >> { > > >> >> ERR("Efreet init failed"); > > >> >> > > >> >> -- > > >> >> > > >> >> > > >> >> > > >> > > > >> > ------------------------------------------------------------------------------ > > >> > Monitor 25 network devices or servers for free with OpManager! > > >> > OpManager is web-based network management software that monitors > > >> > network devices and physical & virtual servers, alerts via email & > sms > > >> > for fault. Monitor 25 devices for free with no restriction. > Download now > > >> > http://ad.doubleclick.net/ddm/clk/292181274;119417398;o > > >> > _______________________________________________ > > >> > enlightenment-devel mailing list > > >> > [email protected] > > >> > https://lists.sourceforge.net/lists/listinfo/enlightenment-devel > > >> > > >> > > >> > ------------------------------------------------------------------------------ > > >> Monitor 25 network devices or servers for free with OpManager! > > >> OpManager is web-based network management software that monitors > > >> network devices and physical & virtual servers, alerts via email & sms > > >> for fault. Monitor 25 devices for free with no restriction. Download > now > > >> http://ad.doubleclick.net/ddm/clk/292181274;119417398;o > > >> _______________________________________________ > > >> enlightenment-devel mailing list > > >> [email protected] > > >> https://lists.sourceforge.net/lists/listinfo/enlightenment-devel > > >> > > > > ------------------------------------------------------------------------------ > > > Monitor 25 network devices or servers for free with OpManager! > > > OpManager is web-based network management software that monitors > > > network devices and physical & virtual servers, alerts via email & sms > > > for fault. Monitor 25 devices for free with no restriction. Download > now > > > http://ad.doubleclick.net/ddm/clk/292181274;119417398;o > > > _______________________________________________ > > > enlightenment-devel mailing list > > > [email protected] > > > https://lists.sourceforge.net/lists/listinfo/enlightenment-devel > > > > > ------------------------------------------------------------------------------ > > Monitor 25 network devices or servers for free with OpManager! > > OpManager is web-based network management software that monitors > > network devices and physical & virtual servers, alerts via email & sms > > for fault. Monitor 25 devices for free with no restriction. Download now > > http://ad.doubleclick.net/ddm/clk/292181274;119417398;o > > _______________________________________________ > > enlightenment-devel mailing list > > [email protected] > > https://lists.sourceforge.net/lists/listinfo/enlightenment-devel > > > > > -- > ------------- Codito, ergo sum - "I code, therefore I am" -------------- > The Rasterman (Carsten Haitzler) [email protected] > > > > ------------------------------------------------------------------------------ > Monitor 25 network devices or servers for free with OpManager! > OpManager is web-based network management software that monitors > network devices and physical & virtual servers, alerts via email & sms > for fault. Monitor 25 devices for free with no restriction. Download now > http://ad.doubleclick.net/ddm/clk/292181274;119417398;o > _______________________________________________ > enlightenment-devel mailing list > [email protected] > https://lists.sourceforge.net/lists/listinfo/enlightenment-devel > ------------------------------------------------------------------------------ Monitor 25 network devices or servers for free with OpManager! OpManager is web-based network management software that monitors network devices and physical & virtual servers, alerts via email & sms for fault. Monitor 25 devices for free with no restriction. Download now http://ad.doubleclick.net/ddm/clk/292181274;119417398;o _______________________________________________ enlightenment-devel mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/enlightenment-devel
