On 02/24/2016 08:02 AM, Hermet Park wrote: > Well... where mentioned it problem? > Now we say it's problem when edje_edit have a probloem. > Of course, that's our problem and our responsibility. but not users.
Yes, it's our problem. But edje_edit very dependent from edc implementation. And if edc, currently edje_cc parser, have problem, edje_edit have this problem too. The edje_edit can't work with NULL part name (edje too), because he manage parts by name, and if group have the unnamed part we can't take access, to this part (edje_util.c +5744). This issue was hidden early, because edje program can't use part with NULL name. So user can create unnamed part, and edje ignore it, while this part not used in edje program. By the way attribute 'relative' can't use unnamed parts too, or rather any attribute can't use unnamed parts. > > Breakage is very critical. > and this kind of rude change brings people hate and leave efl. > > -----Original Message----- > From: "Viacheslav Reutskiy"<[email protected]> > To: "Enlightenment developer list"<[email protected]>; > Cc: > Sent: 2016-02-24 (수) 14:31:40 > Subject: Re: [E-devel] [EGIT] [core/efl] master 01/01: edje_cc: abort compile > when unnamed part exists > > > On 02/24/2016 07:22 AM, Hermet Park wrote: > >> It shouldn't abort right moment. >> People will hate efl if they found breakage. >> Edje_cc warn it first then we could make it abort later. > Please see the thread 'edje: about unnamed parts' in ML. > Raster says: > <quote> > #1. it's an error. i'm surprised people do it. it's also probably good to > catch > </quote> > And I fully agree with him. It's error. > >> >> -----Original Message----- >> From: "David Seikel"<[email protected]> >> To: <[email protected]>; >> Cc: >> Sent: 2016-02-24 (수) 14:11:28 >> Subject: Re: [E-devel] [EGIT] [core/efl] master 01/01: edje_cc: abort >> compile when unnamed part exists >> >> On Wed, 24 Feb 2016 13:53:48 +0900 (KST) Hermet Park >> <[email protected]> wrote: >> >>> I think this changes edc grammar. >>> only one part may not need a part name. >>> and App edc will be broken after this patch. >>> >>> Do you really like it? >> Well, Raster liked it, and he reviewed it - >> >> On Tue, 23 Feb 2016 08:47:31 +0900 Carsten Haitzler (The Rasterman) >> <[email protected]> wrote: >> >>> On Mon, 22 Feb 2016 11:06:40 +0200 Viacheslav Reutskiy >>> <[email protected]> said: >>> >>>> Hello everyone, >>>> >>>> I have a trouble and have the some solutions. So I want to discuss >>>> it. >>>> >>>> Sometimes the group have parts without names. This is not critical >>>> for edje, but critical for Edje_Edit API, because it use the part >>>> name for access. And if group have more then one unnamed part, it >>>> made group not editable, because edje_edit return the first unnamed >>>> part, only first. >>> actually i would call an unnamed part an error. >>> >>>> This problem very painful for editors, and for all who use >>>> Edje_Edit API's. >>>> >>>> So, have the next solutions: >>>> * stop parse edc if find the unnamed part, and print error, >>>> * set to unnamed part predefined name on parse, like >>>> 'part_0xFFFFFFFFFFFFFFFF'. >>> #1. it's an error. i'm surprised people do it. it's also probably >>> good to catch #it and tell them "don't do this". :) >> So yes, this would be an API break, but it was considered broken >> already anyway. >> >>> -----Original Message----- >>> From: "Viacheslav Reutskiy"<[email protected]> >>> To: "Enlightenment developer >>> list"<[email protected]>; Cc: >>> Sent: 2016-02-23 (화) 18:13:30 >>> Subject: Re: [E-devel] [EGIT] [core/efl] master 01/01: edje_cc: abort >>> compile when unnamed part exists >>> Crap, I made a mistake. I made 'arc land' instead 'arc >>> close-revision'. How to fix it? Revert and push correctly? >>> >>> >>> On 02/23/2016 11:06 AM, Vyacheslav Reutskiy wrote: >>>> rimmed pushed a commit to branch master. >>>> >>>> http://git.enlightenment.org/core/efl.git/commit/?id=81f8e8683d90435c6321e643d40047de643d5a58 >>>> >>>> commit 81f8e8683d90435c6321e643d40047de643d5a58 >>>> Author: Vyacheslav Reutskiy <[email protected]> >>>> Date: Tue Feb 23 11:06:55 2016 +0200 >>>> >>>> edje_cc: abort compile when unnamed part exists >>>> >>>> Summary: abort compile when unnamed part exists. >>>> >>>> Reviewers: raster, reutskiy.v.v >>>> >>>> Reviewed By: reutskiy.v.v >>>> >>>> Subscribers: cedric, jpeg >>>> >>>> Differential Revision: https://phab.enlightenment.org/D3723 >> -- Viacheslav Reutskiy (rimmed) ------------------------------------------------------------------------------ Site24x7 APM Insight: Get Deep Visibility into Application Performance APM + Mobile APM + RUM: Monitor 3 App instances at just $35/Month Monitor end-to-end web transactions and take corrective actions now Troubleshoot faster and improve end-user experience. Signup Now! http://pubads.g.doubleclick.net/gampad/clk?id=272487151&iu=/4140 _______________________________________________ enlightenment-devel mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/enlightenment-devel
