[clang-tools-extra] Enforce SL.con.3: Add check to replace operator[] with at() [Cont.] (PR #95220)

[Manuel Pietsch via cfe-commits]

Paul =?utf-8?q?Heidekrüger?= <paul.heidekrueger at tum.de>,
Paul =?utf-8?q?Heidekrüger?= <paul.heidekrueger at tum.de>,
Paul =?utf-8?q?Heidekrüger?= <paul.heidekrueger at tum.de>,
Paul =?utf-8?q?Heidekrüger?= <paul.heidekrueger at tum.de>,
Paul =?utf-8?q?Heidekrüger?= <paul.heidekrueger at tum.de>,
Paul =?utf-8?q?Heidekrüger?= <paul.heidekrueger at tum.de>,
Paul =?utf-8?q?Heidekrüger?= <paul.heidekrueger at tum.de>,
Paul =?utf-8?q?Heidekrüger?= <paul.heidekrueger at tum.de>,
Paul =?utf-8?q?Heidekrüger?= <paul.heidekrueger at tum.de>,
Paul =?utf-8?q?Heidekrüger?= <paul.heidekrueger at tum.de>,
Paul =?utf-8?q?Heidekrüger?= <paul.heidekrueger at tum.de>,
Paul =?utf-8?q?Heidekrüger?= <paul.heidekrueger at tum.de>,
Paul =?utf-8?q?Heidekrüger?= <paul.heidekrueger at tum.de>,
Paul =?utf-8?q?Heidekrüger?= <paul.heidekrueger at tum.de>,
Paul =?utf-8?q?Heidekrüger?= <paul.heidekrueger at tum.de>,
Paul =?utf-8?q?Heidekrüger?= <paul.heidekrueger at tum.de>,
Paul =?utf-8?q?Heidekrüger?= <paul.heidekrueger at tum.de>,
Paul =?utf-8?q?Heidekrüger?= <paul.heidekrueger at tum.de>,
Paul =?utf-8?q?Heidekrüger?= <paul.heidekrueger at tum.de>,
Paul =?utf-8?q?Heidekrüger?= <paul.heidekrueger at tum.de>,
Paul =?utf-8?q?Heidekrüger?= <paul.heidekrueger at tum.de>,
Paul =?utf-8?q?Heidekrüger?= <paul.heidekrueger at tum.de>,
Paul =?utf-8?q?Heidekrüger?= <paul.heidekrueger at tum.de>,
Paul =?utf-8?q?Heidekrüger?= <paul.heidekrueger at tum.de>,
Paul =?utf-8?q?Heidekrüger?= <paul.heidekrueger at tum.de>,
Paul =?utf-8?q?Heidekrüger?= <paul.heidekrueger at tum.de>,
Paul =?utf-8?q?Heidekrüger?= <paul.heidekrueger at tum.de>,
Paul =?utf-8?q?Heidekrüger?= <paul.heidekrueger at tum.de>,
Paul =?utf-8?q?Heidekrüger?= <paul.heidekrueger at tum.de>,
Paul =?utf-8?q?Heidekrüger?= <paul.heidekrueger at tum.de>,
Paul =?utf-8?q?Heidekrüger?= <paul.heidekrueger at tum.de>,
Paul =?utf-8?q?Heidekrüger?= <paul.heidekrueger at tum.de>,
Paul =?utf-8?q?Heidekrüger?= <paul.heidekrueger at tum.de>,
Paul =?utf-8?q?Heidekrüger?= <paul.heidekrueger at tum.de>,
Paul =?utf-8?q?Heidekrüger?= <paul.heidekrueger at tum.de>,
Paul =?utf-8?q?Heidekrüger?= <paul.heidekrueger at tum.de>,
Paul =?utf-8?q?Heidekrüger?= <paul.heidekrueger at tum.de>,
Paul =?utf-8?q?Heidekrüger?= <paul.heidekrueger at tum.de>,
Paul =?utf-8?q?Heidekrüger?= <paul.heidekrueger at tum.de>,
Paul =?utf-8?q?Heidekrüger?= <paul.heidekrueger at tum.de>
Message-ID:
In-Reply-To: <llvm.org/llvm/llvm-project/pull/95220 at github.com>


leunam99 wrote:

We now added FIXITs with the different modes as suggested.
Some notes about the implementation:
- In the None / Function mode, we thought that checking if an at method exists does not make sense anymore and warn on every subscript operator use (besides the explicit exclusions)
  Then, it might be unintuitive that changing a setting called FIXMODE affects if a warning is emmitted, so we emit this general warning in any mode if we do not have a concrete alternative.
  Do you agree with this behaviour? Or will this result in too many false positives?
- The Function mode does not yet check if a valid version of the function exists. Is there a simple way to check if a function template (or function) exists that can fit some specific parameters? 
  Should/Can we do something reasonable if the function is not imported?
- We prefixed the option names with 'Subscript' so that if the check later will be extended, the naming does not get confusing.
- For Templates the fix does not work. We will handle this after it is more clear how Templates should be treated.

https://github.com/llvm/llvm-project/pull/95220