[Mlir-commits] [mlir] [mlir] Method to iterate over registered operations for a given dialect class. (PR #112344)

[llvmlistbot at llvm.org]

================
@@ -711,6 +714,21 @@ ArrayRef<RegisteredOperationName> MLIRContext::getRegisteredOperations() {
   return impl->sortedRegisteredOperations;
 }
 
+/// Return information for registered operations by dialect.
+ArrayRef<RegisteredOperationName>
+MLIRContext::getRegisteredOperationsByDialect(StringRef dialectName) {
+  auto lowerBound = std::lower_bound(
+      impl->sortedRegisteredOperations.begin(),
+      impl->sortedRegisteredOperations.end(), std::make_pair(dialectName, ""),
+      [](auto &lhs, auto &rhs) {
+        return lhs.getDialect().getNamespace().compare(
+            rhs.getDialect.getNamespace());
----------------
graphite-app[bot] wrote:

There appears to be a mismatch in the lambda function's comparison. The right-hand side of the comparison should be `rhs.first` to correctly match the pair created in the `lower_bound` call. The current `rhs.getDialect.getNamespace()` is both syntactically incorrect (missing parentheses) and doesn't align with the pair structure. Consider updating the lambda to:

```cpp
[](auto &lhs, auto &rhs) {
    return lhs.getDialect().getNamespace().compare(rhs.first);
}
```

This change will ensure the comparison is performed correctly between the dialect namespaces.

*Spotted by [Graphite Reviewer](https://app.graphite.dev/graphite-reviewer/?org=llvm&ref=ai-review-comment)*<i class='graphite__hidden'><br /><br />Is this helpful? React 👍 or 👎 to let us know.</i>

https://github.com/llvm/llvm-project/pull/112344