From kde-core-devel Fri Jul 06 13:36:12 2012 From: "Maks Orlovich" Date: Fri, 06 Jul 2012 13:36:12 +0000 To: kde-core-devel Subject: Re: Review Request: kjs: Mark Date Infinity as invalid Message-Id: <20120706133612.22878.89140 () vidsolbach ! de> X-MARC-Message: https://marc.info/?l=kde-core-devel&m=134158202403375 MIME-Version: 1 Content-Type: multipart/mixed; boundary="--===============9187201419007293098==" --===============9187201419007293098== Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable ----------------------------------------------------------- This is an automatically generated e-mail. To reply, visit: http://git.reviewboard.kde.org/r/105387/#review15458 ----------------------------------------------------------- kjs/date_object.cpp Same fix here? (And perhaps in other spots?) = Really, this probably needs some cleanup to not dupe this chunk of code= ; the spec defines one in terms of the other, after all. - Maks Orlovich On June 29, 2012, 6:06 p.m., Bernd Buschinski wrote: > = > ----------------------------------------------------------- > This is an automatically generated e-mail. To reply, visit: > http://git.reviewboard.kde.org/r/105387/ > ----------------------------------------------------------- > = > (Updated June 29, 2012, 6:06 p.m.) > = > = > Review request for kdelibs. > = > = > Description > ------- > = > kjs: Mark Date Infinity as invalid > = > we currently accept Infinity as a valid date, but its not, we even calc w= ith it, > which may or may not lead to a result, but always to the wrong result. > = > The correct way it to just report it as invalid, as we do by remembering = it as NaN. > = > = > Diffs > ----- > = > kjs/date_object.cpp 8a1fc2c = > = > Diff: http://git.reviewboard.kde.org/r/105387/diff/ > = > = > Testing > ------- > = > = > Thanks, > = > Bernd Buschinski > = > --===============9187201419007293098== Content-Type: text/html; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable
This is an automatically generated e-mail. To reply, visit: http://git.revie= wboard.kde.org/r/105387/

= =
kjs/date_object.cpp (Diff revision 1) 
JSValue *DateObjectFuncImp::callAsFunction(ExecState* exec, JSObjec=
t*, const List& args)
889 
        || (n >=3D=
 7 && isNaN(args=
[6]->toNumber(exec)))) {
 894 
        || (n >=3D=
 7 && isNaN(args=
[6]->toNumber(exec)))) {
 
Same fix here? (And perhaps in other spots?)

Really, this probably needs some cleanup to not dupe this chunk of code; th=
e spec
defines one in terms of the other, after all.

- Maks


On June 29th, 2012, 6:06 p.m., Bernd Buschinski wrote:

Review request for kdelibs.
By Bernd Buschinski.

Updated June 29, 2012, 6:06 p.m.

Descripti= on 

kjs: Mark Date Infinity as invalid

we currently accept Infinity as a valid date, but its not, we even calc wit=
h it,
which may or may not lead to a result, but always to the wrong result.

The correct way it to just report it as invalid, as we do by remembering it=
 as NaN.

Diffs=

  • kjs/date_object.cpp (8a1fc2c)

View Diff
--===============9187201419007293098==--