D13900: Disable class-memaccess compiler warnings where not wanted

[Aaron Puchert]

aaronpuchert added a comment.


  On the surface the compiler is right, this comes very close to undefined behavior. But there is so much crazy stuff happening in the DUChain that it might be fine in the end.
  
  In D13900#287851 <https://phabricator.kde.org/D13900#287851>, @kossebau wrote:
  
  > I have no deep understanding of the actual code, all seem more elaborated low-level memory magic to me, which though has existed for years without (known?) issues, so my motivation to try to rewrite it using proper (modern?) C++ expressions is rather low :)
  
  
  Well, actually I do remember a few crashes in DUChain code. But in the few cases when I didn't have something better to do, I was immediately scared away by the code. I find it really hard to understand.
  
  > Perhaps worth a comment, but then I have no idea what to write besides "Dear clever code reader, please consider to rewrite this code using normal C++ class code" which though is IMHO an implicit with most/any given diagnostic pragmas ;) But feel challenged to be the clever code reader now :P I know it's just me to keep fingers from keyboard here.
  
  Yes, somebody should probably rewrite that. But it's hard to know where to start.

INLINE COMMENTS

> qmljsparser.cpp:51-53
>  // This file is automatically generated from qmljs.g.
>  // Changes should be made to that file, not here. Any change to this file will
>  // be lost!

If that is true, the change should be made somewhere else.

REPOSITORY
  R32 KDevelop

REVISION DETAIL
  https://phabricator.kde.org/D13900

To: kossebau, #kdevelop
Cc: aaronpuchert, kfunk, kdevelop-devel, antismap, iodelay, vbspam, njensen, geetamc, Pilzschaf, akshaydeo, surgenight, arrowd
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.kde.org/pipermail/kdevelop-devel/attachments/20180706/d9218876/attachment.html>