<div dir="ltr"><div>Replies inline.</div><div><br></div><div dir="ltr">On Thu, Apr 11, 2019 at 7:59 PM Sam McCall <<a href="mailto:sammccall@google.com">sammccall@google.com</a>> wrote:<br></div><div class="gmail_quote"><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex"><div dir="auto"><div>Thanks - I think there's quite a lot of background missing here:</div></div></blockquote><div>Indeed, here are clarifications. </div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex"><div dir="auto"><div> what/who is this for,</div></div></blockquote><div>Building the index for clangd without actually running clangd it seems useful, e.g. for cases when you want to share the index (would require the same code paths, but already potentially useful).</div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex"><div dir="auto"><div> why a separate binary from clangd-indexer </div></div></blockquote><div>We discussed this and reusing clangd-indexer could be an option too. There are mainly two reasons:</div><div>- Implementations are very different (should they be?), avoid corresponding complexities in the code.</div><div>- If we decide to have only one index format later, deprecating and removing either of the tools is simple than considerably changing semantics of a single tool.</div><div><br></div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex"><div dir="auto"><div>(or and how do they relate, how much of the BackgroundIndex code structure makes sense with multiple users.</div></div></blockquote><div>The idea is that eventually background-indexer is the default tool to bring the index up-to-date and clangd-indexer can be deprecated (I hope we'll end up with a single on-disk representation eventually). </div><div>Wrt to the code structure, the only thing that we're interested in the context of the tool is the on-disk index format of the index.</div><div>The rest of the code in BackgroundIndex we're actually fighting with to make this work, see other comments about exposing a different API that captures the intention of this use-case (one-shot update of the index) better.</div><div> </div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex"><div dir="auto"><div dir="auto">I assume there's been some offline discussion about this, but i missed it :-)<br></div><div dir="auto"><br><div class="gmail_quote" dir="auto"><div dir="ltr" class="gmail_attr">On Thu, Apr 11, 2019, 18:16 Ilya Biryukov via Phabricator <<a href="mailto:reviews@reviews.llvm.org" target="_blank">reviews@reviews.llvm.org</a>> wrote:<br></div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex">ilya-biryukov added a comment.<br>
<br>
To make it clear, I think the question is not just "which part of functionality is missing in BackgroundIndex", it's rather "which part of BackgroundIndex we **don't** need".<br>
<br>
<br>
Repository:<br>
rG LLVM Github Monorepo<br>
<br>
CHANGES SINCE LAST ACTION<br>
<a href="https://reviews.llvm.org/D59605/new/" rel="noreferrer noreferrer" target="_blank">https://reviews.llvm.org/D59605/new/</a><br>
<br>
<a href="https://reviews.llvm.org/D59605" rel="noreferrer noreferrer" target="_blank">https://reviews.llvm.org/D59605</a><br>
<br>
<br>
<br>
</blockquote></div></div></div>
</blockquote></div><br clear="all"><div><br></div>-- <br><div dir="ltr" class="gmail_signature"><div dir="ltr"><div><div dir="ltr"><div>Regards,</div><div>Ilya Biryukov</div></div></div></div></div></div>