Go: sync ExternalFlow.qll #16558
Go: sync ExternalFlow.qll #16558
Conversation
owen-mc
force-pushed
the
go/sync-external-flow
branch
2 times, most recently
from
d9dfb93 to
9c2667b
Compare
| private predicate canonicalPackageHasASubpackage(string package, string subpkg) { | ||
| canonicalPackage(package) and | ||
| (subpkg = package or packageLink(package, subpkg)) | ||
| } | ||
|
|
||
| /** | ||
| * Holds if MaD framework coverage of `package` is `n` api endpoints of the | ||
| * kind `(kind, part)`, and `pkgs` is the number of subpackages of `package` | ||
| * which have MaD framework coverage (including `package` itself). | ||
| * kind `(kind, part)`. |
There was a problem hiding this comment.
Why does this stop documenting pkgs, which still exists?
There was a problem hiding this comment.
To match the java and C# versions of this file, so they're easier to keep in sync or make into a shared library. You're right that it's better though. Maybe I'll keep it and make a tiny PR to add it to the others.
| ) and | ||
| not exists(Provenance provenance | | ||
| neutralElement(this, "summary", provenance) and | ||
| provenance.isManual() |
There was a problem hiding this comment.
Actually it is correct. The idea is that generated summary models are overridden by manual neutral models. This allows us to fixed generated models in a way which will persist when they are regenerated.
There was a problem hiding this comment.
Let me know if you think there should be a comment briefly explaining this (and feel free to give a suggestion).
|
Looks plausible, except as noted |
This is needed for other languages, but in Go you cannot have two functions which are distinguished only by their parameter types.
owen-mc
force-pushed
the
go/sync-external-flow
branch
from
97807e6 to
cbbdd01
Compare
|
I've addressed the review comments and rebased to check that this is still in a mergeable state, since it was quite old. There has been one change to the java copy of |
Update go's copy of
ExternalFlow.qllto bring it in line with java and csharp's.Argument[-1]. Now it isArgument[receiver]. (Other languages use different terminology, such asArgument[this].)