Yea, there may be value in both. If i ever tried to do this I'd probably take the approach of converting all the obvious cases first that should enforce checking and then see what's left. Then we could use llvm:Error and lldb::Status, or something <br><div class="gmail_quote"><div dir="ltr">On Tue, Jan 31, 2017 at 5:39 PM Jim Ingham <<a href="mailto:jingham@apple.com">jingham@apple.com</a>> wrote:<br></div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">Yeah, I have no idea how you'd make sure that the SBError returned to Python in a Python SBValue was checked before it went out of scope, and I'm not sure it's our business to be enforcing that... So we're going to have to do something special for those errors. We also use the pattern where we return a count and an error, but ofttimes you don't care why you got nothing, so you just check the count. The error is informational. Making that obnoxious would also not be a plus. I actually think there's some value to having "programming errors that must be checked" and informational errors. Maybe we need both.<br class="gmail_msg">
<br class="gmail_msg">
Jim<br class="gmail_msg">
<br class="gmail_msg">
> On Jan 31, 2017, at 5:32 PM, Zachary Turner <<a href="mailto:zturner@google.com" class="gmail_msg" target="_blank">zturner@google.com</a>> wrote:<br class="gmail_msg">
><br class="gmail_msg">
> llvm::Error only enforces checking at the point that it goes out of scope. So the example you mention should be supported, as long as you propagate the value all the way up and don't destroy it. There's also ways to explicitly ignore an Error (similar to casting to void to make the compiler not warn about unused variables), so as a last resort we could put code like that in the SB implementation if we had to<br class="gmail_msg">
> On Tue, Jan 31, 2017 at 5:23 PM Jim Ingham <<a href="mailto:jingham@apple.com" class="gmail_msg" target="_blank">jingham@apple.com</a>> wrote:<br class="gmail_msg">
> I think we discussed this before, but we need an informational error object. IIRC, the llvm::Error has to be checked. But for instance if you ask a value object to evaluate itself, but you've moved to a section of code where the variable has no location, then evaluating that value will result in an error. But that isn't an error the value object code needs to do anything about. And it might go all the way up through the SB & Python API's before it's appropriate to check the error.<br class="gmail_msg">
><br class="gmail_msg">
> IIRC, the llvm::Error is one of those "you have to check it or you get smacked by the compiler" dealies. That's appropriate for some of our uses of Error, but not all.<br class="gmail_msg">
><br class="gmail_msg">
> Jim<br class="gmail_msg">
><br class="gmail_msg">
><br class="gmail_msg">
> > On Jan 31, 2017, at 4:01 PM, Zachary Turner via Phabricator via lldb-commits <<a href="mailto:lldb-commits@lists.llvm.org" class="gmail_msg" target="_blank">lldb-commits@lists.llvm.org</a>> wrote:<br class="gmail_msg">
> ><br class="gmail_msg">
> > zturner added a comment.<br class="gmail_msg">
> ><br class="gmail_msg">
> >> Move Error from Core -> Utility<br class="gmail_msg">
> ><br class="gmail_msg">
> > Also, I almost forgot.<br class="gmail_msg">
> ><br class="gmail_msg">
> > Long term: Delete and use `llvm::Error`<br class="gmail_msg">
> ><br class="gmail_msg">
> ><br class="gmail_msg">
> > <a href="https://reviews.llvm.org/D29359" rel="noreferrer" class="gmail_msg" target="_blank">https://reviews.llvm.org/D29359</a><br class="gmail_msg">
> ><br class="gmail_msg">
> ><br class="gmail_msg">
> ><br class="gmail_msg">
> > _______________________________________________<br class="gmail_msg">
> > lldb-commits mailing list<br class="gmail_msg">
> > <a href="mailto:lldb-commits@lists.llvm.org" class="gmail_msg" target="_blank">lldb-commits@lists.llvm.org</a><br class="gmail_msg">
> > <a href="http://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits" rel="noreferrer" class="gmail_msg" target="_blank">http://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits</a><br class="gmail_msg">
><br class="gmail_msg">
<br class="gmail_msg">
</blockquote></div>