<div dir="ltr"><div>Thanks for trying it Artyom --- I intended to do it later today.<br><br>I think it is best to add the Args.erase() in order to really do what is meant. I will commit it later today --- unless someone objects.<br>
<br></div><div>Cheers,<br>--<br></div><div>Arnaud<br></div><div><br></div></div><div class="gmail_extra"><br><br><div class="gmail_quote">On Wed, Apr 30, 2014 at 4:46 PM, Artyom Skrobov <span dir="ltr"><<a href="mailto:Artyom.Skrobov@arm.com" target="_blank">Artyom.Skrobov@arm.com</a>></span> wrote:<br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div class="">>>> In clang/lib/Tooling/CompilationDatabase.cpp:241, this really looks<br>
weird:<br>
>>><br>
>>> // Remove -no-integrated-as; it's not used for syntax checking,<br>
>>> // and it confuses targets which don't support this option.<br>
>>> std::remove_if(Args.begin(), Args.end(),<br>
>>> MatchesAny(std::string("-no-integrated-as")));<br>
>>><br>
>>> The attached patch adds the extra erase step necessary to really remove<br>
>>> the elements --- as the comment states. But it bothers me to see that<br>
the<br>
>>> current code seems to work as it is, hinting that those lines may no<br>
longer<br>
>>> be necessary.<br>
>><br>
>> I don't see how those lines work as is. I think they are only working in<br>
>> the absence of that flag.<br>
><br>
> I guess integrated-as works on Win32 now? Either that, or the test is<br>
> broken too.<br>
<br>
</div>I don't immediately see why those lines work as is, but I can confirm that<br>
removing them causes a failure in Tooling/multi-jobs.cpp; while adding<br>
Args.erase() keeps the test passing.<br>
<br>
<br>
</blockquote></div><br></div>