<div dir="ltr"><div class="gmail_extra"><div class="gmail_quote">On Thu, Jan 9, 2014 at 3:18 PM, Marshall Clow <span dir="ltr"><<a href="mailto:mclow.lists@gmail.com" target="_blank">mclow.lists@gmail.com</a>></span> wrote:<br>
<blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left-width:1px;border-left-color:rgb(204,204,204);border-left-style:solid;padding-left:1ex"><div style="word-wrap:break-word"><br><div><div class="im">
<div>On Jan 9, 2014, at 2:51 PM, Richard Smith <<a href="mailto:richard@metafoo.co.uk" target="_blank">richard@metafoo.co.uk</a>> wrote:</div><br><blockquote type="cite"><div dir="ltr"><div class="gmail_extra"><div class="gmail_quote">
On Thu, Jan 9, 2014 at 2:39 PM, Marshall Clow <span dir="ltr"><<a href="mailto:mclow.lists@gmail.com" target="_blank">mclow.lists@gmail.com</a>></span> wrote:<br>
<blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left-width:1px;border-left-color:rgb(204,204,204);border-left-style:solid;padding-left:1ex"><div style="word-wrap:break-word"><div>On Jan 9, 2014, at 2:20 PM, Richard Smith <<a href="mailto:richard@metafoo.co.uk" target="_blank">richard@metafoo.co.uk</a>> wrote:<br>
</div><div><br><blockquote type="cite"><div dir="ltr"><div class="gmail_extra"><div class="gmail_quote"><div>On Thu, Jan 9, 2014 at 7:25 AM, Alp Toker <span dir="ltr"><<a href="mailto:alp@nuanti.com" target="_blank">alp@nuanti.com</a>></span> wrote:<br>
<blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left-width:1px;border-left-color:rgb(204,204,204);border-left-style:solid;padding-left:1ex"> </blockquote></div><div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left-width:1px;border-left-color:rgb(204,204,204);border-left-style:solid;padding-left:1ex">
It's not fair on the user or the system header maintainers to bring up unexpected errors with this flag.<br>
</blockquote><div><br></div><div>I don't think this is clear. If the code in the system header is ill-formed, I would expect the system header maintainers to be *grateful* that we flag it as an error and not just as a warning. But I'll wait to hear what they have to say about that.</div>
</div></div></div></div></blockquote><div><br></div><div>A few points, in no particular order:</div><div><br></div></div><div>* If we have ill-formed code in system headers, I would expect it to fail to compile whether the user specifies -<span style="font-family:LucidaGrande">Wsystem-headers </span>or not.</div>
</div></blockquote><div><br></div><div>And yet in some cases we don't, because we deliberately support some flavors of broken system headers. </div></div></div></div></blockquote><blockquote type="cite"><div dir="ltr">
<div class="gmail_extra"><div class="gmail_quote"><div>Given that, would you want -Wsystem-headers to make such code produce an error or only a warning?</div></div></div></div></blockquote><div><br></div></div>Eww.</div><div>
<br></div><div>First - is any of that code in libc++?</div><div>Second, there’s two audiences that I think are affected here, and they have different needs:</div><div>* People trying to use the system headers. These people are, in general, unable/unwilling to change them. Giving them warnings/errors is just noise.</div>
<div>* People who are working on the system headers. They need to see the warnings.</div><div><br></div><div><br></div><div><div class="im"><blockquote type="cite"><div dir="ltr"><div class="gmail_extra"><div class="gmail_quote">
<blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left-width:1px;border-left-color:rgb(204,204,204);border-left-style:solid;padding-left:1ex"><div style="word-wrap:break-word"><div><br></div><div>* If a user has a clean build, and then rebuilds with -<font face="LucidaGrande">Wsystem-headers, I would expect to get warnings - not errors. [ Isn’t that what PR</font>18327 is all about? ]</div>
</div></blockquote><div><br></div><div>Do you expect this because the flag starts with -W? </div></div></div></div></blockquote><div><br></div></div>Yes.</div><div class="im"><div><br><blockquote type="cite"><div dir="ltr">
<div class="gmail_extra"><div class="gmail_quote"><div>(Maybe that's the problem here -- it's *not* a warning flag in the sense that the other -W flags are. But that's not unprecedented -- nor is -Werror.)</div>
</div></div></div></blockquote><div><br></div></div><div><br></div></div><div><div class="im"><blockquote type="cite"><div dir="ltr"><div class="gmail_extra"><div class="gmail_quote"><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left-width:1px;border-left-color:rgb(204,204,204);border-left-style:solid;padding-left:1ex">
<div style="word-wrap:break-word"><div><span style="font-family:LucidaGrande">* There are some “interesting” language features which are only enabled for system headers, and cause warnings if used in user code.</span></div>
<div><font face="LucidaGrande">[ User-defined suffixes that do not start with an underscore, for example. ]</font></div></div></blockquote><div><br></div><div>This is a really great example, thanks. </div></div></div></div>
</blockquote><div><br></div></div><div>Note that putting something like this in your source file gives you a warning;</div><div><br></div><div> std::complex<long double> operator"" ilx(long double __im)</div>
<div> {</div><div> return { 0.0l, __im };</div><div> }</div><div><br></div><div>$ clang11 junk.cpp<br>junk.cpp:5:31: warning: user-defined literal suffixes not starting with '_' are<br> reserved; no literal will invoke this operator [-Wuser-defined-literals]<br>
std::complex<long double> operator"" ilx(long double __im)<br> ^<br>1 warning generated.<br><br></div><div class="im"><br><blockquote type="cite"><div dir="ltr"><div class="gmail_extra">
<div class="gmail_quote"><div>So it seems there are at least three different classes of errors that we might think about producing in system headers:</div>
<div><br></div><div> 1) warnings that the user has turned into errors with -Werror or -Werror=foo</div><div> 2) errors for ill-formed code that we suppress by default in system headers as a workaround for a system header bug</div>
<div> 3) errors for code that is ill-formed outside system headers but valid within system headers (using reserved names, adding names to namespace std, that sort of thing)</div><div><br></div><div>With -Wsystem-headers enabled, I think (1) should be an error, (3) should remain suppressed (not even a warning), and (2) should be either a warning or an error (and your first two bullets don't give me a clear idea of which way these cases should go).</div>
</div></div></div>
</blockquote></div></div><br><div>Agreed for #1 and #3.</div><div>And for #2, I’d like to see a warning; “This is code we wouldn’t allow outside a system header, but we do so here”</div><div><br></div><div>I don’t see the point of making #2 an error. </div>
<div>For most people (who can’t fix it), it’s just a hassle.</div><div>For system developers, they can read warning output just as well as error output.</div></div></blockquote><div><br></div><div><div class="" style="font-family:arial,sans-serif;font-size:medium">
<div class=""><div id=":11h" class="" style="font-size:13px"><div id=":11g" style="overflow:hidden"><div dir="ltr"><div class="gmail_extra"><div class="gmail_quote"><div>OK, so I think this is Alp's original patch plus a change to treat #1 as an error. As you observe, -Wuser-defined-literals is currently not an error outside of a system header, so perhaps we have no diagnostics in case #3 at the moment, but it's something to keep in mind for the future.</div>
</div></div></div></div></div></div></div></div></div></div></div>