juergbi opened a new pull request, #2097: URL: https://github.com/apache/buildstream/pull/2097
Duplicate dependencies are normally merged already in `extract_depends_from_node()`. However, that only works for dependencies that are declared with the same filename. If a link element is in a dependency list, merging may still be required but can be done only after the link element has been loaded. The element state updates don't work correctly if there are duplicate dependencies that aren't merged. Merging `Dependency` objects does not adjust the `name` or `path`. This means that if an element has duplicate dependencies, the name from the first dependency declaration is used. One could argue that we should always use the name of the link target, however, this would break cache keys (via `configure_dependencies()`) and I can't think of a real problem with the current approach. Fixes #2002. -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected]
