Merge with upstream #40
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
* Parse and print `//`-style comments as docs This updates `wit-parser` to account for WebAssembly/component-model#281. Doc comments are now considered to be all forms of comments and are collected in the `Docs` structure. The structure already stripped all leading `/` characters so it works for either `///` or `//`. Additionally docs are now printed with `//` instead of `///`. * Update test expectations
…codealliance#1317) This is useful when doing shared-everything linking and we have a library module that exports `cabi_realloc`, but not the adapter-specific `cabi_import_realloc` and `cabi_export_realloc` functions. For context: this came up while adding Preview 2 support to `wasi-libc` using bindings generated by `wit-bindgen`, which only exports `cabi_realloc`. That works fine if you're statically linking `libc.a` and using `wit-component` the traditional way, but if you're using `libc.so` and shared-everything linking, each .so module is treated as an "adapter", which meant `wit-component` was looking for `cabi_import_realloc` and `cabi_export_realloc`. Signed-off-by: Joel Dice <[email protected]>
…iance#1320) This commit updates the `wasm-tools component wit` subcommand to support wasm inputs which don't necessarily have the `*.wasm` extension (or `*.wat`). Instead the file contents themselves are always inspected. Documentation has additionally been updated in a few places to account for this change.
This commit ensures that if a core wasm binary is fed into the WIT-decoding process it'll fail quickly with an up-front error message that it's not a component.
* Update import name validation in `wasmparser`. This commit updates the import name validation in wasmparser to conform to the current component-model spec. It allows for nested namespaces in interface, locked dep, and unlock dep names. Additionally, locked and unlocked dep names can now contain one or more projections. Interface names can now have more than one projection. A check was added to ensure url and relative-url names don't contain a `<`. Fixes bytecodealliance#1314. * Add a feature gate for the extended import names. This commit adds the `component_model_nested_names` feature which controls whether or not the validator will except nested namespaces and projections in import names. It adds a new test for the extended forms, enabling the feature only for that test. * Fix fuzz.
* Allow renaming imports in `ComponentEncoder`. This commit adds an `import_name_map` option to `ComponentEncoder` to allow renaming imports in the output component. Language toolchains can use this option to substitute `unlocked-dep` and `locked-dep` names in the components they build. * Add `--import-name` option to `wasm-tools component new`. This commit adds a `--import-name` option to support renaming import names via the `wasm-tools component new` command.
…ytecodealliance#1316) * implement from_reader for ComponentInfo for incremental wit decoding * preserve wasmparser lifetimes and use from_reader in ComponentInfo::new
) This commit fixes a bug in `Resolve::merge` which is one of the workhorses of the `wasm-tools component new` subcommand. Previously it was assumed that `Resolve::packages` always listed packages in topological order, but `Resolve::merge` updated the internal state of `Resolve` to violate this invariant. This happened because new packages which might be dependencies of existing packages (through newly discovered interfaces for example) would be appended at the end. After wrestling with a few options to solve this issue I've settled on relaxing the invariants of `Resolve` such that it does not always contain a topologically sorted list of packages. Instead a new method, `topological_packages`, is explicitly used to order the packages within `Resolve`. This method is then used as part of `Resolve::merge` which relies on visiting in a topological order. For testing this it wasn't previously found via fuzzing because fuzzing does not execute `Resolve::merge`. Extra validation about topological-sorted-ness caught the problem in the `merge` tests, however. The fuzzer for WIT has been updated to test merging and assert validity now as well too.
* make import_names optional * Update src/bin/wasm-tools/component.rs * Update src/bin/wasm-tools/component.rs --------- Co-authored-by: Peter Huene <[email protected]>
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
No description provided.