[PATCH] D52984: [analyzer] Checker reviewer's checklist

Artem Dergachev via Phabricator via cfe-commits cfe-commits at lists.llvm.org
Sat Nov 10 08:13:49 PST 2018


NoQ added a comment.

In https://reviews.llvm.org/D52984#1294233, @aaron.ballman wrote:

> Personally, I think it's detrimental to the community for subprojects to come up with their own coding guidelines. My preference is for the static analyzer to come more in line with the rest of the project (over time, organically) in terms of style, terminology, diagnostic wording, etc. However, if the consensus is that we want a separate coding standard, I think it should be explicitly documented somewhere public and then maintained as part of the project.


I tihnk it's mostly conventions of using Analyzer-specific APIs, eg. avoid `addTransition()` hell - i guess we already have that, or how to register custom immutable maps, or how to implement checker dependencies or inter-checker APIs, or how much do we want to split modeling and checking into separate checkers, stuff like that.



================
Comment at: www/analyzer/checker_dev_manual.html:719
+<ul>
+<li>User facing documentation is important for adoption! Make sure the check list updated
+    at the homepage of the analyzer. Also ensure that the description is good quality in
----------------
aaron.ballman wrote:
> xazax.hun wrote:
> > Szelethus wrote:
> > > xazax.hun wrote:
> > > > Szelethus wrote:
> > > > > Make sure the **checker** list **is** updated
> > > > I think at some point we should decide if we prefer the term check or checker to refer to these things :)  Clang Tidy clearly prefers check.
> > > That is the distinction I'm aware of too: checkers in the Static Analyzer, checks in clang-tidy.
> > My understanding is the following: we want users to use the term `check`, since that is more widespread and used by other (non-clang) tools as well. The term `checker` is something like a historical artifact in the developer community of the static analyzer. But if this is not the case, I am happy to change the terminology. But I do want to have some input from rest of the community too :)
> I grew up with the term "checker", but I feel like "check" may have won the war. I don't have a strong opinion here though.
We have the word "checker" all over the website, in option names, and, most importantly, in the "How to write a //checker// in 24 hours" video. I don't think we have much choice (:


https://reviews.llvm.org/D52984





More information about the cfe-commits mailing list