<div><div dir="auto">Oh, yeah, that should probably</div></div><div dir="auto">be clamped.</div><div dir="auto"><br></div><div dir="auto">I didn’t have a specific use case in mind, but in general I find that tools that emit “richer” information can be used for more things, hence returning the count, and not some fixed value.</div><div><br><div class="gmail_quote"><div dir="ltr" class="gmail_attr">On Sat, Jun 13, 2020 at 12:58 AM Nathan James via Phabricator <<a href="mailto:reviews@reviews.llvm.org">reviews@reviews.llvm.org</a>> wrote:<br></div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">njames93 added inline comments.<br>
Herald added a subscriber: arphaman.<br>
<br>
<br>
================<br>
Comment at: clang-tidy/tool/ClangTidyMain.cpp:362<br>
+ << Plural << "\n";<br>
+ return WErrorCount;<br>
+ }<br>
----------------<br>
Was there any specific reason for returning the error count instead of returning 1. It results in undefined behaviour on POSIX shells if a value is returned outside the range of 0<=n<=255. See <a href="https://bugs.llvm.org/show_bug.cgi?id=46305" rel="noreferrer" target="_blank">https://bugs.llvm.org/show_bug.cgi?id=46305</a><br>
<br>
<br>
CHANGES SINCE LAST ACTION<br>
<a href="https://reviews.llvm.org/D15528/new/" rel="noreferrer" target="_blank">https://reviews.llvm.org/D15528/new/</a><br>
<br>
<a href="https://reviews.llvm.org/D15528" rel="noreferrer" target="_blank">https://reviews.llvm.org/D15528</a><br>
<br>
<br>
<br>
</blockquote></div></div>