diff options
author | Daniel Baumann <daniel.baumann@progress-linux.org> | 2024-04-17 13:39:29 +0000 |
---|---|---|
committer | Daniel Baumann <daniel.baumann@progress-linux.org> | 2024-04-17 13:39:29 +0000 |
commit | 1bbf3d3cebcb6ff09231ee43296d62d6a93fcfd0 (patch) | |
tree | 6dc13e28a6078185b871acc5b7f9a2df6c5d948a | |
parent | Initial commit. (diff) | |
download | rust-proc-macro2-1bbf3d3cebcb6ff09231ee43296d62d6a93fcfd0.tar.xz rust-proc-macro2-1bbf3d3cebcb6ff09231ee43296d62d6a93fcfd0.zip |
Adding upstream version 1.0.76.upstream/1.0.76
Signed-off-by: Daniel Baumann <daniel.baumann@progress-linux.org>
-rw-r--r-- | .cargo_vcs_info.json | 6 | ||||
-rw-r--r-- | .github/FUNDING.yml | 1 | ||||
-rw-r--r-- | .github/workflows/ci.yml | 180 | ||||
-rw-r--r-- | .gitignore | 3 | ||||
-rw-r--r-- | Cargo.toml | 67 | ||||
-rw-r--r-- | Cargo.toml.orig | 59 | ||||
-rw-r--r-- | LICENSE-APACHE | 176 | ||||
-rw-r--r-- | LICENSE-MIT | 23 | ||||
-rw-r--r-- | README.md | 94 | ||||
-rw-r--r-- | build.rs | 202 | ||||
-rw-r--r-- | build/probe.rs | 21 | ||||
-rw-r--r-- | rust-toolchain.toml | 2 | ||||
-rw-r--r-- | src/detection.rs | 75 | ||||
-rw-r--r-- | src/extra.rs | 84 | ||||
-rw-r--r-- | src/fallback.rs | 1143 | ||||
-rw-r--r-- | src/lib.rs | 1328 | ||||
-rw-r--r-- | src/location.rs | 29 | ||||
-rw-r--r-- | src/marker.rs | 21 | ||||
-rw-r--r-- | src/parse.rs | 996 | ||||
-rw-r--r-- | src/rcvec.rs | 145 | ||||
-rw-r--r-- | src/wrapper.rs | 930 | ||||
-rw-r--r-- | tests/comments.rs | 105 | ||||
-rw-r--r-- | tests/features.rs | 8 | ||||
-rw-r--r-- | tests/marker.rs | 99 | ||||
-rw-r--r-- | tests/test.rs | 759 | ||||
-rw-r--r-- | tests/test_fmt.rs | 28 | ||||
-rw-r--r-- | tests/test_size.rs | 42 |
27 files changed, 6626 insertions, 0 deletions
diff --git a/.cargo_vcs_info.json b/.cargo_vcs_info.json new file mode 100644 index 0000000..0448b81 --- /dev/null +++ b/.cargo_vcs_info.json @@ -0,0 +1,6 @@ +{ + "git": { + "sha1": "ea778eb80729a01094b033c45633122052d96fd4" + }, + "path_in_vcs": "" +}
\ No newline at end of file diff --git a/.github/FUNDING.yml b/.github/FUNDING.yml new file mode 100644 index 0000000..7507077 --- /dev/null +++ b/.github/FUNDING.yml @@ -0,0 +1 @@ +github: dtolnay diff --git a/.github/workflows/ci.yml b/.github/workflows/ci.yml new file mode 100644 index 0000000..c51ca9c --- /dev/null +++ b/.github/workflows/ci.yml @@ -0,0 +1,180 @@ +name: CI + +on: + push: + pull_request: + workflow_dispatch: + schedule: [cron: "40 1 * * *"] + +permissions: + contents: read + +env: + RUSTFLAGS: -Dwarnings + +jobs: + pre_ci: + uses: dtolnay/.github/.github/workflows/pre_ci.yml@master + + test: + name: Rust ${{matrix.rust}} + needs: pre_ci + if: needs.pre_ci.outputs.continue + runs-on: ubuntu-latest + strategy: + fail-fast: false + matrix: + rust: [1.56.0, stable, beta] + timeout-minutes: 45 + steps: + - uses: actions/checkout@v4 + - uses: dtolnay/rust-toolchain@master + with: + toolchain: ${{matrix.rust}} + components: rust-src + - run: cargo test + - run: cargo test --no-default-features + - run: cargo test --features span-locations + - name: RUSTFLAGS='--cfg procmacro2_semver_exempt' cargo test + run: cargo test + env: + RUSTFLAGS: --cfg procmacro2_semver_exempt ${{env.RUSTFLAGS}} + - name: RUSTFLAGS='--cfg procmacro2_semver_exempt' cargo test --no-default-features + run: cargo test --no-default-features + env: + RUSTFLAGS: --cfg procmacro2_semver_exempt ${{env.RUSTFLAGS}} + + nightly: + name: Rust nightly + needs: pre_ci + if: needs.pre_ci.outputs.continue + runs-on: ubuntu-latest + timeout-minutes: 45 + steps: + - uses: actions/checkout@v4 + - uses: dtolnay/rust-toolchain@nightly + with: + components: rust-src + - name: Enable type layout randomization + run: echo RUSTFLAGS=${RUSTFLAGS}\ -Zrandomize-layout >> $GITHUB_ENV + - run: cargo check + env: + RUSTFLAGS: --cfg procmacro2_nightly_testing ${{env.RUSTFLAGS}} + - run: cargo test + - run: cargo test --no-default-features + - run: cargo test --no-default-features --test features -- --ignored make_sure_no_proc_macro # run the ignored test to make sure the `proc-macro` feature is disabled + - run: cargo test --features span-locations + - run: cargo test --manifest-path tests/ui/Cargo.toml + - name: RUSTFLAGS='--cfg procmacro2_semver_exempt' cargo test + run: cargo test + env: + RUSTFLAGS: --cfg procmacro2_semver_exempt ${{env.RUSTFLAGS}} + - name: RUSTFLAGS='--cfg procmacro2_semver_exempt' cargo test --no-default-features + run: cargo test --no-default-features + env: + RUSTFLAGS: --cfg procmacro2_semver_exempt ${{env.RUSTFLAGS}} + - name: RUSTFLAGS='-Z allow-features=' cargo test + run: cargo test + env: + RUSTFLAGS: -Z allow-features= --cfg procmacro2_backtrace ${{env.RUSTFLAGS}} + + minimal: + name: Minimal versions + needs: pre_ci + if: needs.pre_ci.outputs.continue + runs-on: ubuntu-latest + timeout-minutes: 45 + steps: + - uses: actions/checkout@v4 + - uses: dtolnay/rust-toolchain@nightly + - run: cargo generate-lockfile -Z minimal-versions + - run: cargo check --locked + + webassembly: + name: WebAssembly + needs: pre_ci + if: needs.pre_ci.outputs.continue + runs-on: ubuntu-latest + timeout-minutes: 45 + steps: + - uses: actions/checkout@v4 + - uses: dtolnay/rust-toolchain@nightly + with: + target: wasm32-unknown-unknown + components: rust-src + - run: cargo test --target wasm32-unknown-unknown --no-run + + fuzz: + name: Fuzz + needs: pre_ci + if: needs.pre_ci.outputs.continue + runs-on: ubuntu-latest + timeout-minutes: 45 + steps: + - uses: actions/checkout@v4 + - uses: dtolnay/rust-toolchain@nightly + with: + components: rust-src + - uses: dtolnay/install@cargo-fuzz + - run: cargo fuzz check + - run: cargo check --no-default-features --features afl + working-directory: fuzz + - uses: dtolnay/install@honggfuzz + - run: sudo apt-get update # https://github.com/actions/runner-images/issues/8953 + - run: sudo apt-get install binutils-dev libunwind-dev + - run: cargo hfuzz build --no-default-features --features honggfuzz + working-directory: fuzz + + doc: + name: Documentation + needs: pre_ci + if: needs.pre_ci.outputs.continue + runs-on: ubuntu-latest + timeout-minutes: 45 + env: + RUSTDOCFLAGS: -Dwarnings + steps: + - uses: actions/checkout@v4 + - uses: dtolnay/rust-toolchain@nightly + with: + components: rust-src + - uses: dtolnay/install@cargo-docs-rs + - run: cargo docs-rs + + clippy: + name: Clippy + runs-on: ubuntu-latest + if: github.event_name != 'pull_request' + timeout-minutes: 45 + steps: + - uses: actions/checkout@v4 + - uses: dtolnay/rust-toolchain@nightly + with: + components: clippy, rust-src + - run: cargo clippy --tests -- -Dclippy::all -Dclippy::pedantic + - run: cargo clippy --tests --all-features -- -Dclippy::all -Dclippy::pedantic + + miri: + name: Miri + needs: pre_ci + if: needs.pre_ci.outputs.continue + runs-on: ubuntu-latest + timeout-minutes: 45 + steps: + - uses: actions/checkout@v4 + - uses: dtolnay/rust-toolchain@miri + - run: cargo miri setup + - run: cargo miri test + env: + MIRIFLAGS: -Zmiri-strict-provenance + + outdated: + name: Outdated + runs-on: ubuntu-latest + if: github.event_name != 'pull_request' + timeout-minutes: 45 + steps: + - uses: actions/checkout@v4 + - uses: dtolnay/install@cargo-outdated + - run: cargo outdated --workspace --exit-code 1 + - run: cargo outdated --manifest-path fuzz/Cargo.toml --exit-code 1 diff --git a/.gitignore b/.gitignore new file mode 100644 index 0000000..6936990 --- /dev/null +++ b/.gitignore @@ -0,0 +1,3 @@ +/target +**/*.rs.bk +Cargo.lock diff --git a/Cargo.toml b/Cargo.toml new file mode 100644 index 0000000..280bbdf --- /dev/null +++ b/Cargo.toml @@ -0,0 +1,67 @@ +# THIS FILE IS AUTOMATICALLY GENERATED BY CARGO +# +# When uploading crates to the registry Cargo will automatically +# "normalize" Cargo.toml files for maximal compatibility +# with all versions of Cargo and also rewrite `path` dependencies +# to registry (e.g., crates.io) dependencies. +# +# If you are reading this file be aware that the original Cargo.toml +# will likely look very different (and much more reasonable). +# See Cargo.toml.orig for the original contents. + +[package] +edition = "2021" +rust-version = "1.56" +name = "proc-macro2" +version = "1.0.76" +authors = [ + "David Tolnay <dtolnay@gmail.com>", + "Alex Crichton <alex@alexcrichton.com>", +] +autobenches = false +description = "A substitute implementation of the compiler's `proc_macro` API to decouple token-based libraries from the procedural macro use case." +documentation = "https://docs.rs/proc-macro2" +readme = "README.md" +keywords = [ + "macros", + "syn", +] +categories = ["development-tools::procedural-macro-helpers"] +license = "MIT OR Apache-2.0" +repository = "https://github.com/dtolnay/proc-macro2" + +[package.metadata.docs.rs] +rustc-args = [ + "--cfg", + "procmacro2_semver_exempt", +] +rustdoc-args = [ + "--cfg", + "procmacro2_semver_exempt", + "--cfg", + "doc_cfg", + "--generate-link-to-definition", +] +targets = ["x86_64-unknown-linux-gnu"] + +[package.metadata.playground] +features = ["span-locations"] + +[lib] +doc-scrape-examples = false + +[dependencies.unicode-ident] +version = "1.0" + +[dev-dependencies.quote] +version = "1.0" +default_features = false + +[dev-dependencies.rustversion] +version = "1" + +[features] +default = ["proc-macro"] +nightly = [] +proc-macro = [] +span-locations = [] diff --git a/Cargo.toml.orig b/Cargo.toml.orig new file mode 100644 index 0000000..6f7c786 --- /dev/null +++ b/Cargo.toml.orig @@ -0,0 +1,59 @@ +[package] +name = "proc-macro2" +version = "1.0.76" +authors = ["David Tolnay <dtolnay@gmail.com>", "Alex Crichton <alex@alexcrichton.com>"] +autobenches = false +categories = ["development-tools::procedural-macro-helpers"] +description = "A substitute implementation of the compiler's `proc_macro` API to decouple token-based libraries from the procedural macro use case." +documentation = "https://docs.rs/proc-macro2" +edition = "2021" +keywords = ["macros", "syn"] +license = "MIT OR Apache-2.0" +repository = "https://github.com/dtolnay/proc-macro2" +rust-version = "1.56" + +[package.metadata.docs.rs] +rustc-args = ["--cfg", "procmacro2_semver_exempt"] +rustdoc-args = ["--cfg", "procmacro2_semver_exempt", "--cfg", "doc_cfg", "--generate-link-to-definition"] +targets = ["x86_64-unknown-linux-gnu"] + +[package.metadata.playground] +features = ["span-locations"] + +[dependencies] +unicode-ident = "1.0" + +[dev-dependencies] +quote = { version = "1.0", default_features = false } +rustversion = "1" + +[features] +proc-macro = [] +default = ["proc-macro"] + +# Expose methods Span::start and Span::end which give the line/column location +# of a token. +span-locations = [] + +# This feature no longer means anything. +nightly = [] + +[lib] +doc-scrape-examples = false + +[workspace] +members = ["benches/bench-libproc-macro", "tests/ui"] + +[patch.crates-io] +# Our doc tests depend on quote which depends on proc-macro2. Without this line, +# the proc-macro2 dependency of quote would be the released version of +# proc-macro2. Quote would implement its traits for types from that proc-macro2, +# meaning impls would be missing when tested against types from the local +# proc-macro2. +# +# GitHub Actions builds that are in progress at the time that you publish may +# spuriously fail. This is because they'll be building a local proc-macro2 which +# carries the second-most-recent version number, pulling in quote which resolves +# to a dependency on the just-published most recent version number. Thus the +# patch will fail to apply because the version numbers are different. +proc-macro2 = { path = "." } diff --git a/LICENSE-APACHE b/LICENSE-APACHE new file mode 100644 index 0000000..1b5ec8b --- /dev/null +++ b/LICENSE-APACHE @@ -0,0 +1,176 @@ + Apache License + Version 2.0, January 2004 + http://www.apache.org/licenses/ + +TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION + +1. Definitions. + + "License" shall mean the terms and conditions for use, reproduction, + and distribution as defined by Sections 1 through 9 of this document. + + "Licensor" shall mean the copyright owner or entity authorized by + the copyright owner that is granting the License. + + "Legal Entity" shall mean the union of the acting entity and all + other entities that control, are controlled by, or are under common + control with that entity. For the purposes of this definition, + "control" means (i) the power, direct or indirect, to cause the + direction or management of such entity, whether by contract or + otherwise, or (ii) ownership of fifty percent (50%) or more of the + outstanding shares, or (iii) beneficial ownership of such entity. + + "You" (or "Your") shall mean an individual or Legal Entity + exercising permissions granted by this License. + + "Source" form shall mean the preferred form for making modifications, + including but not limited to software source code, documentation + source, and configuration files. + + "Object" form shall mean any form resulting from mechanical + transformation or translation of a Source form, including but + not limited to compiled object code, generated documentation, + and conversions to other media types. + + "Work" shall mean the work of authorship, whether in Source or + Object form, made available under the License, as indicated by a + copyright notice that is included in or attached to the work + (an example is provided in the Appendix below). + + "Derivative Works" shall mean any work, whether in Source or Object + form, that is based on (or derived from) the Work and for which the + editorial revisions, annotations, elaborations, or other modifications + represent, as a whole, an original work of authorship. For the purposes + of this License, Derivative Works shall not include works that remain + separable from, or merely link (or bind by name) to the interfaces of, + the Work and Derivative Works thereof. + + "Contribution" shall mean any work of authorship, including + the original version of the Work and any modifications or additions + to that Work or Derivative Works thereof, that is intentionally + submitted to Licensor for inclusion in the Work by the copyright owner + or by an individual or Legal Entity authorized to submit on behalf of + the copyright owner. For the purposes of this definition, "submitted" + means any form of electronic, verbal, or written communication sent + to the Licensor or its representatives, including but not limited to + communication on electronic mailing lists, source code control systems, + and issue tracking systems that are managed by, or on behalf of, the + Licensor for the purpose of discussing and improving the Work, but + excluding communication that is conspicuously marked or otherwise + designated in writing by the copyright owner as "Not a Contribution." + + "Contributor" shall mean Licensor and any individual or Legal Entity + on behalf of whom a Contribution has been received by Licensor and + subsequently incorporated within the Work. + +2. Grant of Copyright License. Subject to the terms and conditions of + this License, each Contributor hereby grants to You a perpetual, + worldwide, non-exclusive, no-charge, royalty-free, irrevocable + copyright license to reproduce, prepare Derivative Works of, + publicly display, publicly perform, sublicense, and distribute the + Work and such Derivative Works in Source or Object form. + +3. Grant of Patent License. Subject to the terms and conditions of + this License, each Contributor hereby grants to You a perpetual, + worldwide, non-exclusive, no-charge, royalty-free, irrevocable + (except as stated in this section) patent license to make, have made, + use, offer to sell, sell, import, and otherwise transfer the Work, + where such license applies only to those patent claims licensable + by such Contributor that are necessarily infringed by their + Contribution(s) alone or by combination of their Contribution(s) + with the Work to which such Contribution(s) was submitted. If You + institute patent litigation against any entity (including a + cross-claim or counterclaim in a lawsuit) alleging that the Work + or a Contribution incorporated within the Work constitutes direct + or contributory patent infringement, then any patent licenses + granted to You under this License for that Work shall terminate + as of the date such litigation is filed. + +4. Redistribution. You may reproduce and distribute copies of the + Work or Derivative Works thereof in any medium, with or without + modifications, and in Source or Object form, provided that You + meet the following conditions: + + (a) You must give any other recipients of the Work or + Derivative Works a copy of this License; and + + (b) You must cause any modified files to carry prominent notices + stating that You changed the files; and + + (c) You must retain, in the Source form of any Derivative Works + that You distribute, all copyright, patent, trademark, and + attribution notices from the Source form of the Work, + excluding those notices that do not pertain to any part of + the Derivative Works; and + + (d) If the Work includes a "NOTICE" text file as part of its + distribution, then any Derivative Works that You distribute must + include a readable copy of the attribution notices contained + within such NOTICE file, excluding those notices that do not + pertain to any part of the Derivative Works, in at least one + of the following places: within a NOTICE text file distributed + as part of the Derivative Works; within the Source form or + documentation, if provided along with the Derivative Works; or, + within a display generated by the Derivative Works, if and + wherever such third-party notices normally appear. The contents + of the NOTICE file are for informational purposes only and + do not modify the License. You may add Your own attribution + notices within Derivative Works that You distribute, alongside + or as an addendum to the NOTICE text from the Work, provided + that such additional attribution notices cannot be construed + as modifying the License. + + You may add Your own copyright statement to Your modifications and + may provide additional or different license terms and conditions + for use, reproduction, or distribution of Your modifications, or + for any such Derivative Works as a whole, provided Your use, + reproduction, and distribution of the Work otherwise complies with + the conditions stated in this License. + +5. Submission of Contributions. Unless You explicitly state otherwise, + any Contribution intentionally submitted for inclusion in the Work + by You to the Licensor shall be under the terms and conditions of + this License, without any additional terms or conditions. + Notwithstanding the above, nothing herein shall supersede or modify + the terms of any separate license agreement you may have executed + with Licensor regarding such Contributions. + +6. Trademarks. This License does not grant permission to use the trade + names, trademarks, service marks, or product names of the Licensor, + except as required for reasonable and customary use in describing the + origin of the Work and reproducing the content of the NOTICE file. + +7. Disclaimer of Warranty. Unless required by applicable law or + agreed to in writing, Licensor provides the Work (and each + Contributor provides its Contributions) on an "AS IS" BASIS, + WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or + implied, including, without limitation, any warranties or conditions + of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A + PARTICULAR PURPOSE. You are solely responsible for determining the + appropriateness of using or redistributing the Work and assume any + risks associated with Your exercise of permissions under this License. + +8. Limitation of Liability. In no event and under no legal theory, + whether in tort (including negligence), contract, or otherwise, + unless required by applicable law (such as deliberate and grossly + negligent acts) or agreed to in writing, shall any Contributor be + liable to You for damages, including any direct, indirect, special, + incidental, or consequential damages of any character arising as a + result of this License or out of the use or inability to use the + Work (including but not limited to damages for loss of goodwill, + work stoppage, computer failure or malfunction, or any and all + other commercial damages or losses), even if such Contributor + has been advised of the possibility of such damages. + +9. Accepting Warranty or Additional Liability. While redistributing + the Work or Derivative Works thereof, You may choose to offer, + and charge a fee for, acceptance of support, warranty, indemnity, + or other liability obligations and/or rights consistent with this + License. However, in accepting such obligations, You may act only + on Your own behalf and on Your sole responsibility, not on behalf + of any other Contributor, and only if You agree to indemnify, + defend, and hold each Contributor harmless for any liability + incurred by, or claims asserted against, such Contributor by reason + of your accepting any such warranty or additional liability. + +END OF TERMS AND CONDITIONS diff --git a/LICENSE-MIT b/LICENSE-MIT new file mode 100644 index 0000000..31aa793 --- /dev/null +++ b/LICENSE-MIT @@ -0,0 +1,23 @@ +Permission is hereby granted, free of charge, to any +person obtaining a copy of this software and associated +documentation files (the "Software"), to deal in the +Software without restriction, including without +limitation the rights to use, copy, modify, merge, +publish, distribute, sublicense, and/or sell copies of +the Software, and to permit persons to whom the Software +is furnished to do so, subject to the following +conditions: + +The above copyright notice and this permission notice +shall be included in all copies or substantial portions +of the Software. + +THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF +ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED +TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A +PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT +SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY +CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION +OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR +IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER +DEALINGS IN THE SOFTWARE. diff --git a/README.md b/README.md new file mode 100644 index 0000000..3a29ce8 --- /dev/null +++ b/README.md @@ -0,0 +1,94 @@ +# proc-macro2 + +[<img alt="github" src="https://img.shields.io/badge/github-dtolnay/proc--macro2-8da0cb?style=for-the-badge&labelColor=555555&logo=github" height="20">](https://github.com/dtolnay/proc-macro2) +[<img alt="crates.io" src="https://img.shields.io/crates/v/proc-macro2.svg?style=for-the-badge&color=fc8d62&logo=rust" height="20">](https://crates.io/crates/proc-macro2) +[<img alt="docs.rs" src="https://img.shields.io/badge/docs.rs-proc--macro2-66c2a5?style=for-the-badge&labelColor=555555&logo=docs.rs" height="20">](https://docs.rs/proc-macro2) +[<img alt="build status" src="https://img.shields.io/github/actions/workflow/status/dtolnay/proc-macro2/ci.yml?branch=master&style=for-the-badge" height="20">](https://github.com/dtolnay/proc-macro2/actions?query=branch%3Amaster) + +A wrapper around the procedural macro API of the compiler's `proc_macro` crate. +This library serves two purposes: + +- **Bring proc-macro-like functionality to other contexts like build.rs and + main.rs.** Types from `proc_macro` are entirely specific to procedural macros + and cannot ever exist in code outside of a procedural macro. Meanwhile + `proc_macro2` types may exist anywhere including non-macro code. By developing + foundational libraries like [syn] and [quote] against `proc_macro2` rather + than `proc_macro`, the procedural macro ecosystem becomes easily applicable to + many other use cases and we avoid reimplementing non-macro equivalents of + those libraries. + +- **Make procedural macros unit testable.** As a consequence of being specific + to procedural macros, nothing that uses `proc_macro` can be executed from a + unit test. In order for helper libraries or components of a macro to be + testable in isolation, they must be implemented using `proc_macro2`. + +[syn]: https://github.com/dtolnay/syn +[quote]: https://github.com/dtolnay/quote + +## Usage + +```toml +[dependencies] +proc-macro2 = "1.0" +``` + +The skeleton of a typical procedural macro typically looks like this: + +```rust +extern crate proc_macro; + +#[proc_macro_derive(MyDerive)] +pub fn my_derive(input: proc_macro::TokenStream) -> proc_macro::TokenStream { + let input = proc_macro2::TokenStream::from(input); + + let output: proc_macro2::TokenStream = { + /* transform input */ + }; + + proc_macro::TokenStream::from(output) +} +``` + +If parsing with [Syn], you'll use [`parse_macro_input!`] instead to propagate +parse errors correctly back to the compiler when parsing fails. + +[`parse_macro_input!`]: https://docs.rs/syn/2.0/syn/macro.parse_macro_input.html + +## Unstable features + +The default feature set of proc-macro2 tracks the most recent stable compiler +API. Functionality in `proc_macro` that is not yet stable is not exposed by +proc-macro2 by default. + +To opt into the additional APIs available in the most recent nightly compiler, +the `procmacro2_semver_exempt` config flag must be passed to rustc. We will +polyfill those nightly-only APIs back to Rust 1.56.0. As these are unstable APIs +that track the nightly compiler, minor versions of proc-macro2 may make breaking +changes to them at any time. + +``` +RUSTFLAGS='--cfg procmacro2_semver_exempt' cargo build +``` + +Note that this must not only be done for your crate, but for any crate that +depends on your crate. This infectious nature is intentional, as it serves as a +reminder that you are outside of the normal semver guarantees. + +Semver exempt methods are marked as such in the proc-macro2 documentation. + +<br> + +#### License + +<sup> +Licensed under either of <a href="LICENSE-APACHE">Apache License, Version +2.0</a> or <a href="LICENSE-MIT">MIT license</a> at your option. +</sup> + +<br> + +<sub> +Unless you explicitly state otherwise, any contribution intentionally submitted +for inclusion in this crate by you, as defined in the Apache-2.0 license, shall +be dual licensed as above, without any additional terms or conditions. +</sub> diff --git a/build.rs b/build.rs new file mode 100644 index 0000000..3347f87 --- /dev/null +++ b/build.rs @@ -0,0 +1,202 @@ +// rustc-cfg emitted by the build script: +// +// "wrap_proc_macro" +// Wrap types from libproc_macro rather than polyfilling the whole API. +// Enabled on rustc 1.29+ as long as procmacro2_semver_exempt is not set, +// because we can't emulate the unstable API without emulating everything +// else. Also enabled unconditionally on nightly, in which case the +// procmacro2_semver_exempt surface area is implemented by using the +// nightly-only proc_macro API. +// +// "hygiene" +// Enable Span::mixed_site() and non-dummy behavior of Span::resolved_at +// and Span::located_at. Enabled on Rust 1.45+. +// +// "proc_macro_span" +// Enable non-dummy behavior of Span::start and Span::end methods which +// requires an unstable compiler feature. Enabled when building with +// nightly, unless `-Z allow-feature` in RUSTFLAGS disallows unstable +// features. +// +// "super_unstable" +// Implement the semver exempt API in terms of the nightly-only proc_macro +// API. Enabled when using procmacro2_semver_exempt on a nightly compiler. +// +// "span_locations" +// Provide methods Span::start and Span::end which give the line/column +// location of a token. Enabled by procmacro2_semver_exempt or the +// "span-locations" Cargo cfg. This is behind a cfg because tracking +// location inside spans is a performance hit. +// +// "is_available" +// Use proc_macro::is_available() to detect if the proc macro API is +// available or needs to be polyfilled instead of trying to use the proc +// macro API and catching a panic if it isn't available. Enabled on Rust +// 1.57+. + +use std::env; +use std::ffi::OsString; +use std::path::Path; +use std::process::{self, Command, Stdio}; +use std::str; +use std::u32; + +fn main() { + let rustc = rustc_minor_version().unwrap_or(u32::MAX); + + let docs_rs = env::var_os("DOCS_RS").is_some(); + let semver_exempt = cfg!(procmacro2_semver_exempt) || docs_rs; + if semver_exempt { + // https://github.com/dtolnay/proc-macro2/issues/147 + println!("cargo:rustc-cfg=procmacro2_semver_exempt"); + } + + if semver_exempt || cfg!(feature = "span-locations") { + println!("cargo:rustc-cfg=span_locations"); + } + + if rustc < 57 { + println!("cargo:rustc-cfg=no_is_available"); + } + + if rustc < 66 { + println!("cargo:rustc-cfg=no_source_text"); + } + + if !cfg!(feature = "proc-macro") { + println!("cargo:rerun-if-changed=build.rs"); + return; + } + + println!("cargo:rerun-if-changed=build/probe.rs"); + + let proc_macro_span; + let consider_rustc_bootstrap; + if compile_probe(false) { + // This is a nightly or dev compiler, so it supports unstable features + // regardless of RUSTC_BOOTSTRAP. No need to rerun build script if + // RUSTC_BOOTSTRAP is changed. + proc_macro_span = true; + consider_rustc_bootstrap = false; + } else if let Some(rustc_bootstrap) = env::var_os("RUSTC_BOOTSTRAP") { + if compile_probe(true) { + // This is a stable or beta compiler for which the user has set + // RUSTC_BOOTSTRAP to turn on unstable features. Rerun build script + // if they change it. + proc_macro_span = true; + consider_rustc_bootstrap = true; + } else if rustc_bootstrap == "1" { + // This compiler does not support the proc macro Span API in the + // form that proc-macro2 expects. No need to pay attention to + // RUSTC_BOOTSTRAP. + proc_macro_span = false; + consider_rustc_bootstrap = false; + } else { + // This is a stable or beta compiler for which RUSTC_BOOTSTRAP is + // set to restrict the use of unstable features by this crate. + proc_macro_span = false; + consider_rustc_bootstrap = true; + } + } else { + // Without RUSTC_BOOTSTRAP, this compiler does not support the proc + // macro Span API in the form that proc-macro2 expects, but try again if + // the user turns on unstable features. + proc_macro_span = false; + consider_rustc_bootstrap = true; + } + + if proc_macro_span || !semver_exempt { + println!("cargo:rustc-cfg=wrap_proc_macro"); + } + + if proc_macro_span { + println!("cargo:rustc-cfg=proc_macro_span"); + } + + if semver_exempt && proc_macro_span { + println!("cargo:rustc-cfg=super_unstable"); + } + + if consider_rustc_bootstrap { + println!("cargo:rerun-if-env-changed=RUSTC_BOOTSTRAP"); + } +} + +fn compile_probe(rustc_bootstrap: bool) -> bool { + if env::var_os("RUSTC_STAGE").is_some() { + // We are running inside rustc bootstrap. This is a highly non-standard + // environment with issues such as: + // + // https://github.com/rust-lang/cargo/issues/11138 + // https://github.com/rust-lang/rust/issues/114839 + // + // Let's just not use nightly features here. + return false; + } + + let rustc = cargo_env_var("RUSTC"); + let out_dir = cargo_env_var("OUT_DIR"); + let probefile = Path::new("build").join("probe.rs"); + + // Make sure to pick up Cargo rustc configuration. + let mut cmd = if let Some(wrapper) = env::var_os("RUSTC_WRAPPER") { + let mut cmd = Command::new(wrapper); + // The wrapper's first argument is supposed to be the path to rustc. + cmd.arg(rustc); + cmd + } else { + Command::new(rustc) + }; + + if !rustc_bootstrap { + cmd.env_remove("RUSTC_BOOTSTRAP"); + } + + cmd.stderr(Stdio::null()) + .arg("--edition=2021") + .arg("--crate-name=proc_macro2") + .arg("--crate-type=lib") + .arg("--emit=dep-info,metadata") + .arg("--out-dir") + .arg(out_dir) + .arg(probefile); + + if let Some(target) = env::var_os("TARGET") { + cmd.arg("--target").arg(target); + } + + // If Cargo wants to set RUSTFLAGS, use that. + if let Ok(rustflags) = env::var("CARGO_ENCODED_RUSTFLAGS") { + if !rustflags.is_empty() { + for arg in rustflags.split('\x1f') { + cmd.arg(arg); + } + } + } + + match cmd.status() { + Ok(status) => status.success(), + Err(_) => false, + } +} + +fn rustc_minor_version() -> Option<u32> { + let rustc = cargo_env_var("RUSTC"); + let output = Command::new(rustc).arg("--version").output().ok()?; + let version = str::from_utf8(&output.stdout).ok()?; + let mut pieces = version.split('.'); + if pieces.next() != Some("rustc 1") { + return None; + } + pieces.next()?.parse().ok() +} + +fn cargo_env_var(key: &str) -> OsString { + env::var_os(key).unwrap_or_else(|| { + eprintln!( + "Environment variable ${} is not set during execution of build script", + key, + ); + process::exit(1); + }) +} diff --git a/build/probe.rs b/build/probe.rs new file mode 100644 index 0000000..5afa13a --- /dev/null +++ b/build/probe.rs @@ -0,0 +1,21 @@ +// This code exercises the surface area that we expect of Span's unstable API. +// If the current toolchain is able to compile it, then proc-macro2 is able to +// offer these APIs too. + +#![feature(proc_macro_span)] + +extern crate proc_macro; + +use core::ops::RangeBounds; +use proc_macro::{Literal, Span}; + +pub fn join(this: &Span, other: Span) -> Option<Span> { + this.join(other) +} + +pub fn subspan<R: RangeBounds<usize>>(this: &Literal, range: R) -> Option<Span> { + this.subspan(range) +} + +// Include in sccache cache key. +const _: Option<&str> = option_env!("RUSTC_BOOTSTRAP"); diff --git a/rust-toolchain.toml b/rust-toolchain.toml new file mode 100644 index 0000000..20fe888 --- /dev/null +++ b/rust-toolchain.toml @@ -0,0 +1,2 @@ +[toolchain] +components = ["rust-src"] diff --git a/src/detection.rs b/src/detection.rs new file mode 100644 index 0000000..beba7b2 --- /dev/null +++ b/src/detection.rs @@ -0,0 +1,75 @@ +use core::sync::atomic::{AtomicUsize, Ordering}; +use std::sync::Once; + +static WORKS: AtomicUsize = AtomicUsize::new(0); +static INIT: Once = Once::new(); + +pub(crate) fn inside_proc_macro() -> bool { + match WORKS.load(Ordering::Relaxed) { + 1 => return false, + 2 => return true, + _ => {} + } + + INIT.call_once(initialize); + inside_proc_macro() +} + +pub(crate) fn force_fallback() { + WORKS.store(1, Ordering::Relaxed); +} + +pub(crate) fn unforce_fallback() { + initialize(); +} + +#[cfg(not(no_is_available))] +fn initialize() { + let available = proc_macro::is_available(); + WORKS.store(available as usize + 1, Ordering::Relaxed); +} + +// Swap in a null panic hook to avoid printing "thread panicked" to stderr, +// then use catch_unwind to determine whether the compiler's proc_macro is +// working. When proc-macro2 is used from outside of a procedural macro all +// of the proc_macro crate's APIs currently panic. +// +// The Once is to prevent the possibility of this ordering: +// +// thread 1 calls take_hook, gets the user's original hook +// thread 1 calls set_hook with the null hook +// thread 2 calls take_hook, thinks null hook is the original hook +// thread 2 calls set_hook with the null hook +// thread 1 calls set_hook with the actual original hook +// thread 2 calls set_hook with what it thinks is the original hook +// +// in which the user's hook has been lost. +// +// There is still a race condition where a panic in a different thread can +// happen during the interval that the user's original panic hook is +// unregistered such that their hook is incorrectly not called. This is +// sufficiently unlikely and less bad than printing panic messages to stderr +// on correct use of this crate. Maybe there is a libstd feature request +// here. For now, if a user needs to guarantee that this failure mode does +// not occur, they need to call e.g. `proc_macro2::Span::call_site()` from +// the main thread before launching any other threads. +#[cfg(no_is_available)] +fn initialize() { + use std::panic::{self, PanicInfo}; + + type PanicHook = dyn Fn(&PanicInfo) + Sync + Send + 'static; + + let null_hook: Box<PanicHook> = Box::new(|_panic_info| { /* ignore */ }); + let sanity_check = &*null_hook as *const PanicHook; + let original_hook = panic::take_hook(); + panic::set_hook(null_hook); + + let works = panic::catch_unwind(proc_macro::Span::call_site).is_ok(); + WORKS.store(works as usize + 1, Ordering::Relaxed); + + let hopefully_null_hook = panic::take_hook(); + panic::set_hook(original_hook); + if sanity_check != &*hopefully_null_hook { + panic!("observed race condition in proc_macro2::inside_proc_macro"); + } +} diff --git a/src/extra.rs b/src/extra.rs new file mode 100644 index 0000000..4a69d46 --- /dev/null +++ b/src/extra.rs @@ -0,0 +1,84 @@ +//! Items which do not have a correspondence to any API in the proc_macro crate, +//! but are necessary to include in proc-macro2. + +use crate::fallback; +use crate::imp; +use crate::marker::Marker; +use crate::Span; +use core::fmt::{self, Debug}; + +/// An object that holds a [`Group`]'s `span_open()` and `span_close()` together +/// (in a more compact representation than holding those 2 spans individually. +/// +/// [`Group`]: crate::Group +#[derive(Copy, Clone)] +pub struct DelimSpan { + inner: DelimSpanEnum, + _marker: Marker, +} + +#[derive(Copy, Clone)] +enum DelimSpanEnum { + #[cfg(wrap_proc_macro)] + Compiler { + join: proc_macro::Span, + open: proc_macro::Span, + close: proc_macro::Span, + }, + Fallback(fallback::Span), +} + +impl DelimSpan { + pub(crate) fn new(group: &imp::Group) -> Self { + #[cfg(wrap_proc_macro)] + let inner = match group { + imp::Group::Compiler(group) => DelimSpanEnum::Compiler { + join: group.span(), + open: group.span_open(), + close: group.span_close(), + }, + imp::Group::Fallback(group) => DelimSpanEnum::Fallback(group.span()), + }; + + #[cfg(not(wrap_proc_macro))] + let inner = DelimSpanEnum::Fallback(group.span()); + + DelimSpan { + inner, + _marker: Marker, + } + } + + /// Returns a span covering the entire delimited group. + pub fn join(&self) -> Span { + match &self.inner { + #[cfg(wrap_proc_macro)] + DelimSpanEnum::Compiler { join, .. } => Span::_new(imp::Span::Compiler(*join)), + DelimSpanEnum::Fallback(span) => Span::_new_fallback(*span), + } + } + + /// Returns a span for the opening punctuation of the group only. + pub fn open(&self) -> Span { + match &self.inner { + #[cfg(wrap_proc_macro)] + DelimSpanEnum::Compiler { open, .. } => Span::_new(imp::Span::Compiler(*open)), + DelimSpanEnum::Fallback(span) => Span::_new_fallback(span.first_byte()), + } + } + + /// Returns a span for the closing punctuation of the group only. + pub fn close(&self) -> Span { + match &self.inner { + #[cfg(wrap_proc_macro)] + DelimSpanEnum::Compiler { close, .. } => Span::_new(imp::Span::Compiler(*close)), + DelimSpanEnum::Fallback(span) => Span::_new_fallback(span.last_byte()), + } + } +} + +impl Debug for DelimSpan { + fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result { + Debug::fmt(&self.join(), f) + } +} diff --git a/src/fallback.rs b/src/fallback.rs new file mode 100644 index 0000000..7b40427 --- /dev/null +++ b/src/fallback.rs @@ -0,0 +1,1143 @@ +#[cfg(span_locations)] +use crate::location::LineColumn; +use crate::parse::{self, Cursor}; +use crate::rcvec::{RcVec, RcVecBuilder, RcVecIntoIter, RcVecMut}; +use crate::{Delimiter, Spacing, TokenTree}; +#[cfg(all(span_locations, not(fuzzing)))] +use alloc::collections::BTreeMap; +#[cfg(all(span_locations, not(fuzzing)))] +use core::cell::RefCell; +#[cfg(span_locations)] +use core::cmp; +use core::fmt::{self, Debug, Display, Write}; +use core::mem::ManuallyDrop; +use core::ops::RangeBounds; +use core::ptr; +use core::str::FromStr; +use std::path::PathBuf; + +/// Force use of proc-macro2's fallback implementation of the API for now, even +/// if the compiler's implementation is available. +pub fn force() { + #[cfg(wrap_proc_macro)] + crate::detection::force_fallback(); +} + +/// Resume using the compiler's implementation of the proc macro API if it is +/// available. +pub fn unforce() { + #[cfg(wrap_proc_macro)] + crate::detection::unforce_fallback(); +} + +#[derive(Clone)] +pub(crate) struct TokenStream { + inner: RcVec<TokenTree>, +} + +#[derive(Debug)] +pub(crate) struct LexError { + pub(crate) span: Span, +} + +impl LexError { + pub(crate) fn span(&self) -> Span { + self.span + } + + pub(crate) fn call_site() -> Self { + LexError { + span: Span::call_site(), + } + } +} + +impl TokenStream { + pub fn new() -> Self { + TokenStream { + inner: RcVecBuilder::new().build(), + } + } + + pub fn is_empty(&self) -> bool { + self.inner.len() == 0 + } + + fn take_inner(self) -> RcVecBuilder<TokenTree> { + let nodrop = ManuallyDrop::new(self); + unsafe { ptr::read(&nodrop.inner) }.make_owned() + } +} + +fn push_token_from_proc_macro(mut vec: RcVecMut<TokenTree>, token: TokenTree) { + // https://github.com/dtolnay/proc-macro2/issues/235 + match token { + TokenTree::Literal(crate::Literal { + #[cfg(wrap_proc_macro)] + inner: crate::imp::Literal::Fallback(literal), + #[cfg(not(wrap_proc_macro))] + inner: literal, + .. + }) if literal.repr.starts_with('-') => { + push_negative_literal(vec, literal); + } + _ => vec.push(token), + } + + #[cold] + fn push_negative_literal(mut vec: RcVecMut<TokenTree>, mut literal: Literal) { + literal.repr.remove(0); + let mut punct = crate::Punct::new('-', Spacing::Alone); + punct.set_span(crate::Span::_new_fallback(literal.span)); + vec.push(TokenTree::Punct(punct)); + vec.push(TokenTree::Literal(crate::Literal::_new_fallback(literal))); + } +} + +// Nonrecursive to prevent stack overflow. +impl Drop for TokenStream { + fn drop(&mut self) { + let mut inner = match self.inner.get_mut() { + Some(inner) => inner, + None => return, + }; + while let Some(token) = inner.pop() { + let group = match token { + TokenTree::Group(group) => group.inner, + _ => continue, + }; + #[cfg(wrap_proc_macro)] + let group = match group { + crate::imp::Group::Fallback(group) => group, + crate::imp::Group::Compiler(_) => continue, + }; + inner.extend(group.stream.take_inner()); + } + } +} + +pub(crate) struct TokenStreamBuilder { + inner: RcVecBuilder<TokenTree>, +} + +impl TokenStreamBuilder { + pub fn new() -> Self { + TokenStreamBuilder { + inner: RcVecBuilder::new(), + } + } + + pub fn with_capacity(cap: usize) -> Self { + TokenStreamBuilder { + inner: RcVecBuilder::with_capacity(cap), + } + } + + pub fn push_token_from_parser(&mut self, tt: TokenTree) { + self.inner.push(tt); + } + + pub fn build(self) -> TokenStream { + TokenStream { + inner: self.inner.build(), + } + } +} + +#[cfg(span_locations)] +fn get_cursor(src: &str) -> Cursor { + #[cfg(fuzzing)] + return Cursor { rest: src, off: 1 }; + + // Create a dummy file & add it to the source map + #[cfg(not(fuzzing))] + SOURCE_MAP.with(|cm| { + let mut cm = cm.borrow_mut(); + let span = cm.add_file(src); + Cursor { + rest: src, + off: span.lo, + } + }) +} + +#[cfg(not(span_locations))] +fn get_cursor(src: &str) -> Cursor { + Cursor { rest: src } +} + +impl FromStr for TokenStream { + type Err = LexError; + + fn from_str(src: &str) -> Result<TokenStream, LexError> { + // Create a dummy file & add it to the source map + let mut cursor = get_cursor(src); + + // Strip a byte order mark if present + const BYTE_ORDER_MARK: &str = "\u{feff}"; + if cursor.starts_with(BYTE_ORDER_MARK) { + cursor = cursor.advance(BYTE_ORDER_MARK.len()); + } + + parse::token_stream(cursor) + } +} + +impl Display for LexError { + fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result { + f.write_str("cannot parse string into token stream") + } +} + +impl Display for TokenStream { + fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result { + let mut joint = false; + for (i, tt) in self.inner.iter().enumerate() { + if i != 0 && !joint { + write!(f, " ")?; + } + joint = false; + match tt { + TokenTree::Group(tt) => Display::fmt(tt, f), + TokenTree::Ident(tt) => Display::fmt(tt, f), + TokenTree::Punct(tt) => { + joint = tt.spacing() == Spacing::Joint; + Display::fmt(tt, f) + } + TokenTree::Literal(tt) => Display::fmt(tt, f), + }?; + } + + Ok(()) + } +} + +impl Debug for TokenStream { + fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result { + f.write_str("TokenStream ")?; + f.debug_list().entries(self.clone()).finish() + } +} + +#[cfg(feature = "proc-macro")] +impl From<proc_macro::TokenStream> for TokenStream { + fn from(inner: proc_macro::TokenStream) -> Self { + inner + .to_string() + .parse() + .expect("compiler token stream parse failed") + } +} + +#[cfg(feature = "proc-macro")] +impl From<TokenStream> for proc_macro::TokenStream { + fn from(inner: TokenStream) -> Self { + inner + .to_string() + .parse() + .expect("failed to parse to compiler tokens") + } +} + +impl From<TokenTree> for TokenStream { + fn from(tree: TokenTree) -> Self { + let mut stream = RcVecBuilder::new(); + push_token_from_proc_macro(stream.as_mut(), tree); + TokenStream { + inner: stream.build(), + } + } +} + +impl FromIterator<TokenTree> for TokenStream { + fn from_iter<I: IntoIterator<Item = TokenTree>>(tokens: I) -> Self { + let mut stream = TokenStream::new(); + stream.extend(tokens); + stream + } +} + +impl FromIterator<TokenStream> for TokenStream { + fn from_iter<I: IntoIterator<Item = TokenStream>>(streams: I) -> Self { + let mut v = RcVecBuilder::new(); + + for stream in streams { + v.extend(stream.take_inner()); + } + + TokenStream { inner: v.build() } + } +} + +impl Extend<TokenTree> for TokenStream { + fn extend<I: IntoIterator<Item = TokenTree>>(&mut self, tokens: I) { + let mut vec = self.inner.make_mut(); + tokens + .into_iter() + .for_each(|token| push_token_from_proc_macro(vec.as_mut(), token)); + } +} + +impl Extend<TokenStream> for TokenStream { + fn extend<I: IntoIterator<Item = TokenStream>>(&mut self, streams: I) { + self.inner.make_mut().extend(streams.into_iter().flatten()); + } +} + +pub(crate) type TokenTreeIter = RcVecIntoIter<TokenTree>; + +impl IntoIterator for TokenStream { + type Item = TokenTree; + type IntoIter = TokenTreeIter; + + fn into_iter(self) -> TokenTreeIter { + self.take_inner().into_iter() + } +} + +#[derive(Clone, PartialEq, Eq)] +pub(crate) struct SourceFile { + path: PathBuf, +} + +impl SourceFile { + /// Get the path to this source file as a string. + pub fn path(&self) -> PathBuf { + self.path.clone() + } + + pub fn is_real(&self) -> bool { + false + } +} + +impl Debug for SourceFile { + fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result { + f.debug_struct("SourceFile") + .field("path", &self.path()) + .field("is_real", &self.is_real()) + .finish() + } +} + +#[cfg(all(span_locations, not(fuzzing)))] +thread_local! { + static SOURCE_MAP: RefCell<SourceMap> = RefCell::new(SourceMap { + // Start with a single dummy file which all call_site() and def_site() + // spans reference. + files: vec![FileInfo { + source_text: String::new(), + span: Span { lo: 0, hi: 0 }, + lines: vec![0], + char_index_to_byte_offset: BTreeMap::new(), + }], + }); +} + +#[cfg(all(span_locations, not(fuzzing)))] +struct FileInfo { + source_text: String, + span: Span, + lines: Vec<usize>, + char_index_to_byte_offset: BTreeMap<usize, usize>, +} + +#[cfg(all(span_locations, not(fuzzing)))] +impl FileInfo { + fn offset_line_column(&self, offset: usize) -> LineColumn { + assert!(self.span_within(Span { + lo: offset as u32, + hi: offset as u32, + })); + let offset = offset - self.span.lo as usize; + match self.lines.binary_search(&offset) { + Ok(found) => LineColumn { + line: found + 1, + column: 0, + }, + Err(idx) => LineColumn { + line: idx, + column: offset - self.lines[idx - 1], + }, + } + } + + fn span_within(&self, span: Span) -> bool { + span.lo >= self.span.lo && span.hi <= self.span.hi + } + + fn source_text(&mut self, span: Span) -> String { + let lo_char = (span.lo - self.span.lo) as usize; + + // Look up offset of the largest already-computed char index that is + // less than or equal to the current requested one. We resume counting + // chars from that point. + let (&last_char_index, &last_byte_offset) = self + .char_index_to_byte_offset + .range(..=lo_char) + .next_back() + .unwrap_or((&0, &0)); + + let lo_byte = if last_char_index == lo_char { + last_byte_offset + } else { + let total_byte_offset = match self.source_text[last_byte_offset..] + .char_indices() + .nth(lo_char - last_char_index) + { + Some((additional_offset, _ch)) => last_byte_offset + additional_offset, + None => self.source_text.len(), + }; + self.char_index_to_byte_offset + .insert(lo_char, total_byte_offset); + total_byte_offset + }; + + let trunc_lo = &self.source_text[lo_byte..]; + let char_len = (span.hi - span.lo) as usize; + let source_text = match trunc_lo.char_indices().nth(char_len) { + Some((offset, _ch)) => &trunc_lo[..offset], + None => trunc_lo, + }; + source_text.to_owned() + } +} + +/// Computes the offsets of each line in the given source string +/// and the total number of characters +#[cfg(all(span_locations, not(fuzzing)))] +fn lines_offsets(s: &str) -> (usize, Vec<usize>) { + let mut lines = vec![0]; + let mut total = 0; + + for ch in s.chars() { + total += 1; + if ch == '\n' { + lines.push(total); + } + } + + (total, lines) +} + +#[cfg(all(span_locations, not(fuzzing)))] +struct SourceMap { + files: Vec<FileInfo>, +} + +#[cfg(all(span_locations, not(fuzzing)))] +impl SourceMap { + fn next_start_pos(&self) -> u32 { + // Add 1 so there's always space between files. + // + // We'll always have at least 1 file, as we initialize our files list + // with a dummy file. + self.files.last().unwrap().span.hi + 1 + } + + fn add_file(&mut self, src: &str) -> Span { + let (len, lines) = lines_offsets(src); + let lo = self.next_start_pos(); + let span = Span { + lo, + hi: lo + (len as u32), + }; + + self.files.push(FileInfo { + source_text: src.to_owned(), + span, + lines, + // Populated lazily by source_text(). + char_index_to_byte_offset: BTreeMap::new(), + }); + + span + } + + #[cfg(procmacro2_semver_exempt)] + fn filepath(&self, span: Span) -> PathBuf { + for (i, file) in self.files.iter().enumerate() { + if file.span_within(span) { + return PathBuf::from(if i == 0 { + "<unspecified>".to_owned() + } else { + format!("<parsed string {}>", i) + }); + } + } + unreachable!("Invalid span with no related FileInfo!"); + } + + fn fileinfo(&self, span: Span) -> &FileInfo { + for file in &self.files { + if file.span_within(span) { + return file; + } + } + unreachable!("Invalid span with no related FileInfo!"); + } + + fn fileinfo_mut(&mut self, span: Span) -> &mut FileInfo { + for file in &mut self.files { + if file.span_within(span) { + return file; + } + } + unreachable!("Invalid span with no related FileInfo!"); + } +} + +#[derive(Clone, Copy, PartialEq, Eq)] +pub(crate) struct Span { + #[cfg(span_locations)] + pub(crate) lo: u32, + #[cfg(span_locations)] + pub(crate) hi: u32, +} + +impl Span { + #[cfg(not(span_locations))] + pub fn call_site() -> Self { + Span {} + } + + #[cfg(span_locations)] + pub fn call_site() -> Self { + Span { lo: 0, hi: 0 } + } + + pub fn mixed_site() -> Self { + Span::call_site() + } + + #[cfg(procmacro2_semver_exempt)] + pub fn def_site() -> Self { + Span::call_site() + } + + pub fn resolved_at(&self, _other: Span) -> Span { + // Stable spans consist only of line/column information, so + // `resolved_at` and `located_at` only select which span the + // caller wants line/column information from. + *self + } + + pub fn located_at(&self, other: Span) -> Span { + other + } + + #[cfg(procmacro2_semver_exempt)] + pub fn source_file(&self) -> SourceFile { + #[cfg(fuzzing)] + return SourceFile { + path: PathBuf::from("<unspecified>"), + }; + + #[cfg(not(fuzzing))] + SOURCE_MAP.with(|cm| { + let cm = cm.borrow(); + let path = cm.filepath(*self); + SourceFile { path } + }) + } + + #[cfg(span_locations)] + pub fn start(&self) -> LineColumn { + #[cfg(fuzzing)] + return LineColumn { line: 0, column: 0 }; + + #[cfg(not(fuzzing))] + SOURCE_MAP.with(|cm| { + let cm = cm.borrow(); + let fi = cm.fileinfo(*self); + fi.offset_line_column(self.lo as usize) + }) + } + + #[cfg(span_locations)] + pub fn end(&self) -> LineColumn { + #[cfg(fuzzing)] + return LineColumn { line: 0, column: 0 }; + + #[cfg(not(fuzzing))] + SOURCE_MAP.with(|cm| { + let cm = cm.borrow(); + let fi = cm.fileinfo(*self); + fi.offset_line_column(self.hi as usize) + }) + } + + #[cfg(not(span_locations))] + pub fn join(&self, _other: Span) -> Option<Span> { + Some(Span {}) + } + + #[cfg(span_locations)] + pub fn join(&self, other: Span) -> Option<Span> { + #[cfg(fuzzing)] + return { + let _ = other; + None + }; + + #[cfg(not(fuzzing))] + SOURCE_MAP.with(|cm| { + let cm = cm.borrow(); + // If `other` is not within the same FileInfo as us, return None. + if !cm.fileinfo(*self).span_within(other) { + return None; + } + Some(Span { + lo: cmp::min(self.lo, other.lo), + hi: cmp::max(self.hi, other.hi), + }) + }) + } + + #[cfg(not(span_locations))] + pub fn source_text(&self) -> Option<String> { + None + } + + #[cfg(span_locations)] + pub fn source_text(&self) -> Option<String> { + #[cfg(fuzzing)] + return None; + + #[cfg(not(fuzzing))] + { + if self.is_call_site() { + None + } else { + Some(SOURCE_MAP.with(|cm| cm.borrow_mut().fileinfo_mut(*self).source_text(*self))) + } + } + } + + #[cfg(not(span_locations))] + pub(crate) fn first_byte(self) -> Self { + self + } + + #[cfg(span_locations)] + pub(crate) fn first_byte(self) -> Self { + Span { + lo: self.lo, + hi: cmp::min(self.lo.saturating_add(1), self.hi), + } + } + + #[cfg(not(span_locations))] + pub(crate) fn last_byte(self) -> Self { + self + } + + #[cfg(span_locations)] + pub(crate) fn last_byte(self) -> Self { + Span { + lo: cmp::max(self.hi.saturating_sub(1), self.lo), + hi: self.hi, + } + } + + #[cfg(span_locations)] + fn is_call_site(&self) -> bool { + self.lo == 0 && self.hi == 0 + } +} + +impl Debug for Span { + fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result { + #[cfg(span_locations)] + return write!(f, "bytes({}..{})", self.lo, self.hi); + + #[cfg(not(span_locations))] + write!(f, "Span") + } +} + +pub(crate) fn debug_span_field_if_nontrivial(debug: &mut fmt::DebugStruct, span: Span) { + #[cfg(span_locations)] + { + if span.is_call_site() { + return; + } + } + + if cfg!(span_locations) { + debug.field("span", &span); + } +} + +#[derive(Clone)] +pub(crate) struct Group { + delimiter: Delimiter, + stream: TokenStream, + span: Span, +} + +impl Group { + pub fn new(delimiter: Delimiter, stream: TokenStream) -> Self { + Group { + delimiter, + stream, + span: Span::call_site(), + } + } + + pub fn delimiter(&self) -> Delimiter { + self.delimiter + } + + pub fn stream(&self) -> TokenStream { + self.stream.clone() + } + + pub fn span(&self) -> Span { + self.span + } + + pub fn span_open(&self) -> Span { + self.span.first_byte() + } + + pub fn span_close(&self) -> Span { + self.span.last_byte() + } + + pub fn set_span(&mut self, span: Span) { + self.span = span; + } +} + +impl Display for Group { + // We attempt to match libproc_macro's formatting. + // Empty parens: () + // Nonempty parens: (...) + // Empty brackets: [] + // Nonempty brackets: [...] + // Empty braces: { } + // Nonempty braces: { ... } + fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result { + let (open, close) = match self.delimiter { + Delimiter::Parenthesis => ("(", ")"), + Delimiter::Brace => ("{ ", "}"), + Delimiter::Bracket => ("[", "]"), + Delimiter::None => ("", ""), + }; + + f.write_str(open)?; + Display::fmt(&self.stream, f)?; + if self.delimiter == Delimiter::Brace && !self.stream.inner.is_empty() { + f.write_str(" ")?; + } + f.write_str(close)?; + + Ok(()) + } +} + +impl Debug for Group { + fn fmt(&self, fmt: &mut fmt::Formatter) -> fmt::Result { + let mut debug = fmt.debug_struct("Group"); + debug.field("delimiter", &self.delimiter); + debug.field("stream", &self.stream); + debug_span_field_if_nontrivial(&mut debug, self.span); + debug.finish() + } +} + +#[derive(Clone)] +pub(crate) struct Ident { + sym: String, + span: Span, + raw: bool, +} + +impl Ident { + #[track_caller] + pub fn new_checked(string: &str, span: Span) -> Self { + validate_ident(string); + Ident::new_unchecked(string, span) + } + + pub fn new_unchecked(string: &str, span: Span) -> Self { + Ident { + sym: string.to_owned(), + span, + raw: false, + } + } + + #[track_caller] + pub fn new_raw_checked(string: &str, span: Span) -> Self { + validate_ident_raw(string); + Ident::new_raw_unchecked(string, span) + } + + pub fn new_raw_unchecked(string: &str, span: Span) -> Self { + Ident { + sym: string.to_owned(), + span, + raw: true, + } + } + + pub fn span(&self) -> Span { + self.span + } + + pub fn set_span(&mut self, span: Span) { + self.span = span; + } +} + +pub(crate) fn is_ident_start(c: char) -> bool { + c == '_' || unicode_ident::is_xid_start(c) +} + +pub(crate) fn is_ident_continue(c: char) -> bool { + unicode_ident::is_xid_continue(c) +} + +#[track_caller] +fn validate_ident(string: &str) { + if string.is_empty() { + panic!("Ident is not allowed to be empty; use Option<Ident>"); + } + + if string.bytes().all(|digit| b'0' <= digit && digit <= b'9') { + panic!("Ident cannot be a number; use Literal instead"); + } + + fn ident_ok(string: &str) -> bool { + let mut chars = string.chars(); + let first = chars.next().unwrap(); + if !is_ident_start(first) { + return false; + } + for ch in chars { + if !is_ident_continue(ch) { + return false; + } + } + true + } + + if !ident_ok(string) { + panic!("{:?} is not a valid Ident", string); + } +} + +#[track_caller] +fn validate_ident_raw(string: &str) { + validate_ident(string); + + match string { + "_" | "super" | "self" | "Self" | "crate" => { + panic!("`r#{}` cannot be a raw identifier", string); + } + _ => {} + } +} + +impl PartialEq for Ident { + fn eq(&self, other: &Ident) -> bool { + self.sym == other.sym && self.raw == other.raw + } +} + +impl<T> PartialEq<T> for Ident +where + T: ?Sized + AsRef<str>, +{ + fn eq(&self, other: &T) -> bool { + let other = other.as_ref(); + if self.raw { + other.starts_with("r#") && self.sym == other[2..] + } else { + self.sym == other + } + } +} + +impl Display for Ident { + fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result { + if self.raw { + f.write_str("r#")?; + } + Display::fmt(&self.sym, f) + } +} + +#[allow(clippy::missing_fields_in_debug)] +impl Debug for Ident { + // Ident(proc_macro), Ident(r#union) + #[cfg(not(span_locations))] + fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result { + let mut debug = f.debug_tuple("Ident"); + debug.field(&format_args!("{}", self)); + debug.finish() + } + + // Ident { + // sym: proc_macro, + // span: bytes(128..138) + // } + #[cfg(span_locations)] + fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result { + let mut debug = f.debug_struct("Ident"); + debug.field("sym", &format_args!("{}", self)); + debug_span_field_if_nontrivial(&mut debug, self.span); + debug.finish() + } +} + +#[derive(Clone)] +pub(crate) struct Literal { + repr: String, + span: Span, +} + +macro_rules! suffixed_numbers { + ($($name:ident => $kind:ident,)*) => ($( + pub fn $name(n: $kind) -> Literal { + Literal::_new(format!(concat!("{}", stringify!($kind)), n)) + } + )*) +} + +macro_rules! unsuffixed_numbers { + ($($name:ident => $kind:ident,)*) => ($( + pub fn $name(n: $kind) -> Literal { + Literal::_new(n.to_string()) + } + )*) +} + +impl Literal { + pub(crate) fn _new(repr: String) -> Self { + Literal { + repr, + span: Span::call_site(), + } + } + + pub(crate) unsafe fn from_str_unchecked(repr: &str) -> Self { + Literal::_new(repr.to_owned()) + } + + suffixed_numbers! { + u8_suffixed => u8, + u16_suffixed => u16, + u32_suffixed => u32, + u64_suffixed => u64, + u128_suffixed => u128, + usize_suffixed => usize, + i8_suffixed => i8, + i16_suffixed => i16, + i32_suffixed => i32, + i64_suffixed => i64, + i128_suffixed => i128, + isize_suffixed => isize, + + f32_suffixed => f32, + f64_suffixed => f64, + } + + unsuffixed_numbers! { + u8_unsuffixed => u8, + u16_unsuffixed => u16, + u32_unsuffixed => u32, + u64_unsuffixed => u64, + u128_unsuffixed => u128, + usize_unsuffixed => usize, + i8_unsuffixed => i8, + i16_unsuffixed => i16, + i32_unsuffixed => i32, + i64_unsuffixed => i64, + i128_unsuffixed => i128, + isize_unsuffixed => isize, + } + + pub fn f32_unsuffixed(f: f32) -> Literal { + let mut s = f.to_string(); + if !s.contains('.') { + s.push_str(".0"); + } + Literal::_new(s) + } + + pub fn f64_unsuffixed(f: f64) -> Literal { + let mut s = f.to_string(); + if !s.contains('.') { + s.push_str(".0"); + } + Literal::_new(s) + } + + pub fn string(t: &str) -> Literal { + let mut repr = String::with_capacity(t.len() + 2); + repr.push('"'); + let mut chars = t.chars(); + while let Some(ch) = chars.next() { + if ch == '\0' { + repr.push_str( + if chars + .as_str() + .starts_with(|next| '0' <= next && next <= '7') + { + // circumvent clippy::octal_escapes lint + "\\x00" + } else { + "\\0" + }, + ); + } else if ch == '\'' { + // escape_debug turns this into "\'" which is unnecessary. + repr.push(ch); + } else { + repr.extend(ch.escape_debug()); + } + } + repr.push('"'); + Literal::_new(repr) + } + + pub fn character(t: char) -> Literal { + let mut repr = String::new(); + repr.push('\''); + if t == '"' { + // escape_debug turns this into '\"' which is unnecessary. + repr.push(t); + } else { + repr.extend(t.escape_debug()); + } + repr.push('\''); + Literal::_new(repr) + } + + pub fn byte_string(bytes: &[u8]) -> Literal { + let mut escaped = "b\"".to_string(); + let mut bytes = bytes.iter(); + while let Some(&b) = bytes.next() { + #[allow(clippy::match_overlapping_arm)] + match b { + b'\0' => escaped.push_str(match bytes.as_slice().first() { + // circumvent clippy::octal_escapes lint + Some(b'0'..=b'7') => r"\x00", + _ => r"\0", + }), + b'\t' => escaped.push_str(r"\t"), + b'\n' => escaped.push_str(r"\n"), + b'\r' => escaped.push_str(r"\r"), + b'"' => escaped.push_str("\\\""), + b'\\' => escaped.push_str("\\\\"), + b'\x20'..=b'\x7E' => escaped.push(b as char), + _ => { + let _ = write!(escaped, "\\x{:02X}", b); + } + } + } + escaped.push('"'); + Literal::_new(escaped) + } + + pub fn span(&self) -> Span { + self.span + } + + pub fn set_span(&mut self, span: Span) { + self.span = span; + } + + pub fn subspan<R: RangeBounds<usize>>(&self, range: R) -> Option<Span> { + #[cfg(not(span_locations))] + { + let _ = range; + None + } + + #[cfg(span_locations)] + { + use core::ops::Bound; + + let lo = match range.start_bound() { + Bound::Included(start) => { + let start = u32::try_from(*start).ok()?; + self.span.lo.checked_add(start)? + } + Bound::Excluded(start) => { + let start = u32::try_from(*start).ok()?; + self.span.lo.checked_add(start)?.checked_add(1)? + } + Bound::Unbounded => self.span.lo, + }; + let hi = match range.end_bound() { + Bound::Included(end) => { + let end = u32::try_from(*end).ok()?; + self.span.lo.checked_add(end)?.checked_add(1)? + } + Bound::Excluded(end) => { + let end = u32::try_from(*end).ok()?; + self.span.lo.checked_add(end)? + } + Bound::Unbounded => self.span.hi, + }; + if lo <= hi && hi <= self.span.hi { + Some(Span { lo, hi }) + } else { + None + } + } + } +} + +impl FromStr for Literal { + type Err = LexError; + + fn from_str(repr: &str) -> Result<Self, Self::Err> { + let mut cursor = get_cursor(repr); + #[cfg(span_locations)] + let lo = cursor.off; + + let negative = cursor.starts_with_char('-'); + if negative { + cursor = cursor.advance(1); + if !cursor.starts_with_fn(|ch| ch.is_ascii_digit()) { + return Err(LexError::call_site()); + } + } + + if let Ok((rest, mut literal)) = parse::literal(cursor) { + if rest.is_empty() { + if negative { + literal.repr.insert(0, '-'); + } + literal.span = Span { + #[cfg(span_locations)] + lo, + #[cfg(span_locations)] + hi: rest.off, + }; + return Ok(literal); + } + } + Err(LexError::call_site()) + } +} + +impl Display for Literal { + fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result { + Display::fmt(&self.repr, f) + } +} + +impl Debug for Literal { + fn fmt(&self, fmt: &mut fmt::Formatter) -> fmt::Result { + let mut debug = fmt.debug_struct("Literal"); + debug.field("lit", &format_args!("{}", self.repr)); + debug_span_field_if_nontrivial(&mut debug, self.span); + debug.finish() + } +} diff --git a/src/lib.rs b/src/lib.rs new file mode 100644 index 0000000..7e8f543 --- /dev/null +++ b/src/lib.rs @@ -0,0 +1,1328 @@ +//! [![github]](https://github.com/dtolnay/proc-macro2) [![crates-io]](https://crates.io/crates/proc-macro2) [![docs-rs]](crate) +//! +//! [github]: https://img.shields.io/badge/github-8da0cb?style=for-the-badge&labelColor=555555&logo=github +//! [crates-io]: https://img.shields.io/badge/crates.io-fc8d62?style=for-the-badge&labelColor=555555&logo=rust +//! [docs-rs]: https://img.shields.io/badge/docs.rs-66c2a5?style=for-the-badge&labelColor=555555&logo=docs.rs +//! +//! <br> +//! +//! A wrapper around the procedural macro API of the compiler's [`proc_macro`] +//! crate. This library serves two purposes: +//! +//! [`proc_macro`]: https://doc.rust-lang.org/proc_macro/ +//! +//! - **Bring proc-macro-like functionality to other contexts like build.rs and +//! main.rs.** Types from `proc_macro` are entirely specific to procedural +//! macros and cannot ever exist in code outside of a procedural macro. +//! Meanwhile `proc_macro2` types may exist anywhere including non-macro code. +//! By developing foundational libraries like [syn] and [quote] against +//! `proc_macro2` rather than `proc_macro`, the procedural macro ecosystem +//! becomes easily applicable to many other use cases and we avoid +//! reimplementing non-macro equivalents of those libraries. +//! +//! - **Make procedural macros unit testable.** As a consequence of being +//! specific to procedural macros, nothing that uses `proc_macro` can be +//! executed from a unit test. In order for helper libraries or components of +//! a macro to be testable in isolation, they must be implemented using +//! `proc_macro2`. +//! +//! [syn]: https://github.com/dtolnay/syn +//! [quote]: https://github.com/dtolnay/quote +//! +//! # Usage +//! +//! The skeleton of a typical procedural macro typically looks like this: +//! +//! ``` +//! extern crate proc_macro; +//! +//! # const IGNORE: &str = stringify! { +//! #[proc_macro_derive(MyDerive)] +//! # }; +//! # #[cfg(wrap_proc_macro)] +//! pub fn my_derive(input: proc_macro::TokenStream) -> proc_macro::TokenStream { +//! let input = proc_macro2::TokenStream::from(input); +//! +//! let output: proc_macro2::TokenStream = { +//! /* transform input */ +//! # input +//! }; +//! +//! proc_macro::TokenStream::from(output) +//! } +//! ``` +//! +//! If parsing with [Syn], you'll use [`parse_macro_input!`] instead to +//! propagate parse errors correctly back to the compiler when parsing fails. +//! +//! [`parse_macro_input!`]: https://docs.rs/syn/2.0/syn/macro.parse_macro_input.html +//! +//! # Unstable features +//! +//! The default feature set of proc-macro2 tracks the most recent stable +//! compiler API. Functionality in `proc_macro` that is not yet stable is not +//! exposed by proc-macro2 by default. +//! +//! To opt into the additional APIs available in the most recent nightly +//! compiler, the `procmacro2_semver_exempt` config flag must be passed to +//! rustc. We will polyfill those nightly-only APIs back to Rust 1.56.0. As +//! these are unstable APIs that track the nightly compiler, minor versions of +//! proc-macro2 may make breaking changes to them at any time. +//! +//! ```sh +//! RUSTFLAGS='--cfg procmacro2_semver_exempt' cargo build +//! ``` +//! +//! Note that this must not only be done for your crate, but for any crate that +//! depends on your crate. This infectious nature is intentional, as it serves +//! as a reminder that you are outside of the normal semver guarantees. +//! +//! Semver exempt methods are marked as such in the proc-macro2 documentation. +//! +//! # Thread-Safety +//! +//! Most types in this crate are `!Sync` because the underlying compiler +//! types make use of thread-local memory, meaning they cannot be accessed from +//! a different thread. + +// Proc-macro2 types in rustdoc of other crates get linked to here. +#![doc(html_root_url = "https://docs.rs/proc-macro2/1.0.76")] +#![cfg_attr(any(proc_macro_span, super_unstable), feature(proc_macro_span))] +#![cfg_attr(super_unstable, feature(proc_macro_def_site))] +#![cfg_attr(doc_cfg, feature(doc_cfg))] +#![deny(unsafe_op_in_unsafe_fn)] +#![allow( + clippy::cast_lossless, + clippy::cast_possible_truncation, + clippy::checked_conversions, + clippy::doc_markdown, + clippy::items_after_statements, + clippy::iter_without_into_iter, + clippy::let_underscore_untyped, + clippy::manual_assert, + clippy::manual_range_contains, + clippy::missing_safety_doc, + clippy::must_use_candidate, + clippy::needless_doctest_main, + clippy::new_without_default, + clippy::return_self_not_must_use, + clippy::shadow_unrelated, + clippy::trivially_copy_pass_by_ref, + clippy::unnecessary_wraps, + clippy::unused_self, + clippy::used_underscore_binding, + clippy::vec_init_then_push +)] + +#[cfg(all(procmacro2_semver_exempt, wrap_proc_macro, not(super_unstable)))] +compile_error! {"\ + Something is not right. If you've tried to turn on \ + procmacro2_semver_exempt, you need to ensure that it \ + is turned on for the compilation of the proc-macro2 \ + build script as well. +"} + +#[cfg(all( + procmacro2_nightly_testing, + feature = "proc-macro", + not(proc_macro_span) +))] +compile_error! {"\ + Build script probe failed to compile. +"} + +extern crate alloc; + +#[cfg(feature = "proc-macro")] +extern crate proc_macro; + +mod marker; +mod parse; +mod rcvec; + +#[cfg(wrap_proc_macro)] +mod detection; + +// Public for proc_macro2::fallback::force() and unforce(), but those are quite +// a niche use case so we omit it from rustdoc. +#[doc(hidden)] +pub mod fallback; + +pub mod extra; + +#[cfg(not(wrap_proc_macro))] +use crate::fallback as imp; +#[path = "wrapper.rs"] +#[cfg(wrap_proc_macro)] +mod imp; + +#[cfg(span_locations)] +mod location; + +use crate::extra::DelimSpan; +use crate::marker::Marker; +use core::cmp::Ordering; +use core::fmt::{self, Debug, Display}; +use core::hash::{Hash, Hasher}; +use core::ops::RangeBounds; +use core::str::FromStr; +use std::error::Error; +#[cfg(procmacro2_semver_exempt)] +use std::path::PathBuf; + +#[cfg(span_locations)] +#[cfg_attr(doc_cfg, doc(cfg(feature = "span-locations")))] +pub use crate::location::LineColumn; + +/// An abstract stream of tokens, or more concretely a sequence of token trees. +/// +/// This type provides interfaces for iterating over token trees and for +/// collecting token trees into one stream. +/// +/// Token stream is both the input and output of `#[proc_macro]`, +/// `#[proc_macro_attribute]` and `#[proc_macro_derive]` definitions. +#[derive(Clone)] +pub struct TokenStream { + inner: imp::TokenStream, + _marker: Marker, +} + +/// Error returned from `TokenStream::from_str`. +pub struct LexError { + inner: imp::LexError, + _marker: Marker, +} + +impl TokenStream { + fn _new(inner: imp::TokenStream) -> Self { + TokenStream { + inner, + _marker: Marker, + } + } + + fn _new_fallback(inner: fallback::TokenStream) -> Self { + TokenStream { + inner: inner.into(), + _marker: Marker, + } + } + + /// Returns an empty `TokenStream` containing no token trees. + pub fn new() -> Self { + TokenStream::_new(imp::TokenStream::new()) + } + + /// Checks if this `TokenStream` is empty. + pub fn is_empty(&self) -> bool { + self.inner.is_empty() + } +} + +/// `TokenStream::default()` returns an empty stream, +/// i.e. this is equivalent with `TokenStream::new()`. +impl Default for TokenStream { + fn default() -> Self { + TokenStream::new() + } +} + +/// Attempts to break the string into tokens and parse those tokens into a token +/// stream. +/// +/// May fail for a number of reasons, for example, if the string contains +/// unbalanced delimiters or characters not existing in the language. +/// +/// NOTE: Some errors may cause panics instead of returning `LexError`. We +/// reserve the right to change these errors into `LexError`s later. +impl FromStr for TokenStream { + type Err = LexError; + + fn from_str(src: &str) -> Result<TokenStream, LexError> { + let e = src.parse().map_err(|e| LexError { + inner: e, + _marker: Marker, + })?; + Ok(TokenStream::_new(e)) + } +} + +#[cfg(feature = "proc-macro")] +#[cfg_attr(doc_cfg, doc(cfg(feature = "proc-macro")))] +impl From<proc_macro::TokenStream> for TokenStream { + fn from(inner: proc_macro::TokenStream) -> Self { + TokenStream::_new(inner.into()) + } +} + +#[cfg(feature = "proc-macro")] +#[cfg_attr(doc_cfg, doc(cfg(feature = "proc-macro")))] +impl From<TokenStream> for proc_macro::TokenStream { + fn from(inner: TokenStream) -> Self { + inner.inner.into() + } +} + +impl From<TokenTree> for TokenStream { + fn from(token: TokenTree) -> Self { + TokenStream::_new(imp::TokenStream::from(token)) + } +} + +impl Extend<TokenTree> for TokenStream { + fn extend<I: IntoIterator<Item = TokenTree>>(&mut self, streams: I) { + self.inner.extend(streams); + } +} + +impl Extend<TokenStream> for TokenStream { + fn extend<I: IntoIterator<Item = TokenStream>>(&mut self, streams: I) { + self.inner + .extend(streams.into_iter().map(|stream| stream.inner)); + } +} + +/// Collects a number of token trees into a single stream. +impl FromIterator<TokenTree> for TokenStream { + fn from_iter<I: IntoIterator<Item = TokenTree>>(streams: I) -> Self { + TokenStream::_new(streams.into_iter().collect()) + } +} +impl FromIterator<TokenStream> for TokenStream { + fn from_iter<I: IntoIterator<Item = TokenStream>>(streams: I) -> Self { + TokenStream::_new(streams.into_iter().map(|i| i.inner).collect()) + } +} + +/// Prints the token stream as a string that is supposed to be losslessly +/// convertible back into the same token stream (modulo spans), except for +/// possibly `TokenTree::Group`s with `Delimiter::None` delimiters and negative +/// numeric literals. +impl Display for TokenStream { + fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result { + Display::fmt(&self.inner, f) + } +} + +/// Prints token in a form convenient for debugging. +impl Debug for TokenStream { + fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result { + Debug::fmt(&self.inner, f) + } +} + +impl LexError { + pub fn span(&self) -> Span { + Span::_new(self.inner.span()) + } +} + +impl Debug for LexError { + fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result { + Debug::fmt(&self.inner, f) + } +} + +impl Display for LexError { + fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result { + Display::fmt(&self.inner, f) + } +} + +impl Error for LexError {} + +/// The source file of a given `Span`. +/// +/// This type is semver exempt and not exposed by default. +#[cfg(all(procmacro2_semver_exempt, any(not(wrap_proc_macro), super_unstable)))] +#[cfg_attr(doc_cfg, doc(cfg(procmacro2_semver_exempt)))] +#[derive(Clone, PartialEq, Eq)] +pub struct SourceFile { + inner: imp::SourceFile, + _marker: Marker, +} + +#[cfg(all(procmacro2_semver_exempt, any(not(wrap_proc_macro), super_unstable)))] +impl SourceFile { + fn _new(inner: imp::SourceFile) -> Self { + SourceFile { + inner, + _marker: Marker, + } + } + + /// Get the path to this source file. + /// + /// ### Note + /// + /// If the code span associated with this `SourceFile` was generated by an + /// external macro, this may not be an actual path on the filesystem. Use + /// [`is_real`] to check. + /// + /// Also note that even if `is_real` returns `true`, if + /// `--remap-path-prefix` was passed on the command line, the path as given + /// may not actually be valid. + /// + /// [`is_real`]: #method.is_real + pub fn path(&self) -> PathBuf { + self.inner.path() + } + + /// Returns `true` if this source file is a real source file, and not + /// generated by an external macro's expansion. + pub fn is_real(&self) -> bool { + self.inner.is_real() + } +} + +#[cfg(all(procmacro2_semver_exempt, any(not(wrap_proc_macro), super_unstable)))] +impl Debug for SourceFile { + fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result { + Debug::fmt(&self.inner, f) + } +} + +/// A region of source code, along with macro expansion information. +#[derive(Copy, Clone)] +pub struct Span { + inner: imp::Span, + _marker: Marker, +} + +impl Span { + fn _new(inner: imp::Span) -> Self { + Span { + inner, + _marker: Marker, + } + } + + fn _new_fallback(inner: fallback::Span) -> Self { + Span { + inner: inner.into(), + _marker: Marker, + } + } + + /// The span of the invocation of the current procedural macro. + /// + /// Identifiers created with this span will be resolved as if they were + /// written directly at the macro call location (call-site hygiene) and + /// other code at the macro call site will be able to refer to them as well. + pub fn call_site() -> Self { + Span::_new(imp::Span::call_site()) + } + + /// The span located at the invocation of the procedural macro, but with + /// local variables, labels, and `$crate` resolved at the definition site + /// of the macro. This is the same hygiene behavior as `macro_rules`. + pub fn mixed_site() -> Self { + Span::_new(imp::Span::mixed_site()) + } + + /// A span that resolves at the macro definition site. + /// + /// This method is semver exempt and not exposed by default. + #[cfg(procmacro2_semver_exempt)] + #[cfg_attr(doc_cfg, doc(cfg(procmacro2_semver_exempt)))] + pub fn def_site() -> Self { + Span::_new(imp::Span::def_site()) + } + + /// Creates a new span with the same line/column information as `self` but + /// that resolves symbols as though it were at `other`. + pub fn resolved_at(&self, other: Span) -> Span { + Span::_new(self.inner.resolved_at(other.inner)) + } + + /// Creates a new span with the same name resolution behavior as `self` but + /// with the line/column information of `other`. + pub fn located_at(&self, other: Span) -> Span { + Span::_new(self.inner.located_at(other.inner)) + } + + /// Convert `proc_macro2::Span` to `proc_macro::Span`. + /// + /// This method is available when building with a nightly compiler, or when + /// building with rustc 1.29+ *without* semver exempt features. + /// + /// # Panics + /// + /// Panics if called from outside of a procedural macro. Unlike + /// `proc_macro2::Span`, the `proc_macro::Span` type can only exist within + /// the context of a procedural macro invocation. + #[cfg(wrap_proc_macro)] + pub fn unwrap(self) -> proc_macro::Span { + self.inner.unwrap() + } + + // Soft deprecated. Please use Span::unwrap. + #[cfg(wrap_proc_macro)] + #[doc(hidden)] + pub fn unstable(self) -> proc_macro::Span { + self.unwrap() + } + + /// The original source file into which this span points. + /// + /// This method is semver exempt and not exposed by default. + #[cfg(all(procmacro2_semver_exempt, any(not(wrap_proc_macro), super_unstable)))] + #[cfg_attr(doc_cfg, doc(cfg(procmacro2_semver_exempt)))] + pub fn source_file(&self) -> SourceFile { + SourceFile::_new(self.inner.source_file()) + } + + /// Get the starting line/column in the source file for this span. + /// + /// This method requires the `"span-locations"` feature to be enabled. + /// + /// When executing in a procedural macro context, the returned line/column + /// are only meaningful if compiled with a nightly toolchain. The stable + /// toolchain does not have this information available. When executing + /// outside of a procedural macro, such as main.rs or build.rs, the + /// line/column are always meaningful regardless of toolchain. + #[cfg(span_locations)] + #[cfg_attr(doc_cfg, doc(cfg(feature = "span-locations")))] + pub fn start(&self) -> LineColumn { + self.inner.start() + } + + /// Get the ending line/column in the source file for this span. + /// + /// This method requires the `"span-locations"` feature to be enabled. + /// + /// When executing in a procedural macro context, the returned line/column + /// are only meaningful if compiled with a nightly toolchain. The stable + /// toolchain does not have this information available. When executing + /// outside of a procedural macro, such as main.rs or build.rs, the + /// line/column are always meaningful regardless of toolchain. + #[cfg(span_locations)] + #[cfg_attr(doc_cfg, doc(cfg(feature = "span-locations")))] + pub fn end(&self) -> LineColumn { + self.inner.end() + } + + /// Create a new span encompassing `self` and `other`. + /// + /// Returns `None` if `self` and `other` are from different files. + /// + /// Warning: the underlying [`proc_macro::Span::join`] method is + /// nightly-only. When called from within a procedural macro not using a + /// nightly compiler, this method will always return `None`. + /// + /// [`proc_macro::Span::join`]: https://doc.rust-lang.org/proc_macro/struct.Span.html#method.join + pub fn join(&self, other: Span) -> Option<Span> { + self.inner.join(other.inner).map(Span::_new) + } + + /// Compares two spans to see if they're equal. + /// + /// This method is semver exempt and not exposed by default. + #[cfg(procmacro2_semver_exempt)] + #[cfg_attr(doc_cfg, doc(cfg(procmacro2_semver_exempt)))] + pub fn eq(&self, other: &Span) -> bool { + self.inner.eq(&other.inner) + } + + /// Returns the source text behind a span. This preserves the original + /// source code, including spaces and comments. It only returns a result if + /// the span corresponds to real source code. + /// + /// Note: The observable result of a macro should only rely on the tokens + /// and not on this source text. The result of this function is a best + /// effort to be used for diagnostics only. + pub fn source_text(&self) -> Option<String> { + self.inner.source_text() + } +} + +/// Prints a span in a form convenient for debugging. +impl Debug for Span { + fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result { + Debug::fmt(&self.inner, f) + } +} + +/// A single token or a delimited sequence of token trees (e.g. `[1, (), ..]`). +#[derive(Clone)] +pub enum TokenTree { + /// A token stream surrounded by bracket delimiters. + Group(Group), + /// An identifier. + Ident(Ident), + /// A single punctuation character (`+`, `,`, `$`, etc.). + Punct(Punct), + /// A literal character (`'a'`), string (`"hello"`), number (`2.3`), etc. + Literal(Literal), +} + +impl TokenTree { + /// Returns the span of this tree, delegating to the `span` method of + /// the contained token or a delimited stream. + pub fn span(&self) -> Span { + match self { + TokenTree::Group(t) => t.span(), + TokenTree::Ident(t) => t.span(), + TokenTree::Punct(t) => t.span(), + TokenTree::Literal(t) => t.span(), + } + } + + /// Configures the span for *only this token*. + /// + /// Note that if this token is a `Group` then this method will not configure + /// the span of each of the internal tokens, this will simply delegate to + /// the `set_span` method of each variant. + pub fn set_span(&mut self, span: Span) { + match self { + TokenTree::Group(t) => t.set_span(span), + TokenTree::Ident(t) => t.set_span(span), + TokenTree::Punct(t) => t.set_span(span), + TokenTree::Literal(t) => t.set_span(span), + } + } +} + +impl From<Group> for TokenTree { + fn from(g: Group) -> Self { + TokenTree::Group(g) + } +} + +impl From<Ident> for TokenTree { + fn from(g: Ident) -> Self { + TokenTree::Ident(g) + } +} + +impl From<Punct> for TokenTree { + fn from(g: Punct) -> Self { + TokenTree::Punct(g) + } +} + +impl From<Literal> for TokenTree { + fn from(g: Literal) -> Self { + TokenTree::Literal(g) + } +} + +/// Prints the token tree as a string that is supposed to be losslessly +/// convertible back into the same token tree (modulo spans), except for +/// possibly `TokenTree::Group`s with `Delimiter::None` delimiters and negative +/// numeric literals. +impl Display for TokenTree { + fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result { + match self { + TokenTree::Group(t) => Display::fmt(t, f), + TokenTree::Ident(t) => Display::fmt(t, f), + TokenTree::Punct(t) => Display::fmt(t, f), + TokenTree::Literal(t) => Display::fmt(t, f), + } + } +} + +/// Prints token tree in a form convenient for debugging. +impl Debug for TokenTree { + fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result { + // Each of these has the name in the struct type in the derived debug, + // so don't bother with an extra layer of indirection + match self { + TokenTree::Group(t) => Debug::fmt(t, f), + TokenTree::Ident(t) => { + let mut debug = f.debug_struct("Ident"); + debug.field("sym", &format_args!("{}", t)); + imp::debug_span_field_if_nontrivial(&mut debug, t.span().inner); + debug.finish() + } + TokenTree::Punct(t) => Debug::fmt(t, f), + TokenTree::Literal(t) => Debug::fmt(t, f), + } + } +} + +/// A delimited token stream. +/// +/// A `Group` internally contains a `TokenStream` which is surrounded by +/// `Delimiter`s. +#[derive(Clone)] +pub struct Group { + inner: imp::Group, +} + +/// Describes how a sequence of token trees is delimited. +#[derive(Copy, Clone, Debug, Eq, PartialEq)] +pub enum Delimiter { + /// `( ... )` + Parenthesis, + /// `{ ... }` + Brace, + /// `[ ... ]` + Bracket, + /// `Ø ... Ø` + /// + /// An implicit delimiter, that may, for example, appear around tokens + /// coming from a "macro variable" `$var`. It is important to preserve + /// operator priorities in cases like `$var * 3` where `$var` is `1 + 2`. + /// Implicit delimiters may not survive roundtrip of a token stream through + /// a string. + None, +} + +impl Group { + fn _new(inner: imp::Group) -> Self { + Group { inner } + } + + fn _new_fallback(inner: fallback::Group) -> Self { + Group { + inner: inner.into(), + } + } + + /// Creates a new `Group` with the given delimiter and token stream. + /// + /// This constructor will set the span for this group to + /// `Span::call_site()`. To change the span you can use the `set_span` + /// method below. + pub fn new(delimiter: Delimiter, stream: TokenStream) -> Self { + Group { + inner: imp::Group::new(delimiter, stream.inner), + } + } + + /// Returns the punctuation used as the delimiter for this group: a set of + /// parentheses, square brackets, or curly braces. + pub fn delimiter(&self) -> Delimiter { + self.inner.delimiter() + } + + /// Returns the `TokenStream` of tokens that are delimited in this `Group`. + /// + /// Note that the returned token stream does not include the delimiter + /// returned above. + pub fn stream(&self) -> TokenStream { + TokenStream::_new(self.inner.stream()) + } + + /// Returns the span for the delimiters of this token stream, spanning the + /// entire `Group`. + /// + /// ```text + /// pub fn span(&self) -> Span { + /// ^^^^^^^ + /// ``` + pub fn span(&self) -> Span { + Span::_new(self.inner.span()) + } + + /// Returns the span pointing to the opening delimiter of this group. + /// + /// ```text + /// pub fn span_open(&self) -> Span { + /// ^ + /// ``` + pub fn span_open(&self) -> Span { + Span::_new(self.inner.span_open()) + } + + /// Returns the span pointing to the closing delimiter of this group. + /// + /// ```text + /// pub fn span_close(&self) -> Span { + /// ^ + /// ``` + pub fn span_close(&self) -> Span { + Span::_new(self.inner.span_close()) + } + + /// Returns an object that holds this group's `span_open()` and + /// `span_close()` together (in a more compact representation than holding + /// those 2 spans individually). + pub fn delim_span(&self) -> DelimSpan { + DelimSpan::new(&self.inner) + } + + /// Configures the span for this `Group`'s delimiters, but not its internal + /// tokens. + /// + /// This method will **not** set the span of all the internal tokens spanned + /// by this group, but rather it will only set the span of the delimiter + /// tokens at the level of the `Group`. + pub fn set_span(&mut self, span: Span) { + self.inner.set_span(span.inner); + } +} + +/// Prints the group as a string that should be losslessly convertible back +/// into the same group (modulo spans), except for possibly `TokenTree::Group`s +/// with `Delimiter::None` delimiters. +impl Display for Group { + fn fmt(&self, formatter: &mut fmt::Formatter) -> fmt::Result { + Display::fmt(&self.inner, formatter) + } +} + +impl Debug for Group { + fn fmt(&self, formatter: &mut fmt::Formatter) -> fmt::Result { + Debug::fmt(&self.inner, formatter) + } +} + +/// A `Punct` is a single punctuation character like `+`, `-` or `#`. +/// +/// Multicharacter operators like `+=` are represented as two instances of +/// `Punct` with different forms of `Spacing` returned. +#[derive(Clone)] +pub struct Punct { + ch: char, + spacing: Spacing, + span: Span, +} + +/// Whether a `Punct` is followed immediately by another `Punct` or followed by +/// another token or whitespace. +#[derive(Copy, Clone, Debug, Eq, PartialEq)] +pub enum Spacing { + /// E.g. `+` is `Alone` in `+ =`, `+ident` or `+()`. + Alone, + /// E.g. `+` is `Joint` in `+=` or `'` is `Joint` in `'#`. + /// + /// Additionally, single quote `'` can join with identifiers to form + /// lifetimes `'ident`. + Joint, +} + +impl Punct { + /// Creates a new `Punct` from the given character and spacing. + /// + /// The `ch` argument must be a valid punctuation character permitted by the + /// language, otherwise the function will panic. + /// + /// The returned `Punct` will have the default span of `Span::call_site()` + /// which can be further configured with the `set_span` method below. + pub fn new(ch: char, spacing: Spacing) -> Self { + Punct { + ch, + spacing, + span: Span::call_site(), + } + } + + /// Returns the value of this punctuation character as `char`. + pub fn as_char(&self) -> char { + self.ch + } + + /// Returns the spacing of this punctuation character, indicating whether + /// it's immediately followed by another `Punct` in the token stream, so + /// they can potentially be combined into a multicharacter operator + /// (`Joint`), or it's followed by some other token or whitespace (`Alone`) + /// so the operator has certainly ended. + pub fn spacing(&self) -> Spacing { + self.spacing + } + + /// Returns the span for this punctuation character. + pub fn span(&self) -> Span { + self.span + } + + /// Configure the span for this punctuation character. + pub fn set_span(&mut self, span: Span) { + self.span = span; + } +} + +/// Prints the punctuation character as a string that should be losslessly +/// convertible back into the same character. +impl Display for Punct { + fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result { + Display::fmt(&self.ch, f) + } +} + +impl Debug for Punct { + fn fmt(&self, fmt: &mut fmt::Formatter) -> fmt::Result { + let mut debug = fmt.debug_struct("Punct"); + debug.field("char", &self.ch); + debug.field("spacing", &self.spacing); + imp::debug_span_field_if_nontrivial(&mut debug, self.span.inner); + debug.finish() + } +} + +/// A word of Rust code, which may be a keyword or legal variable name. +/// +/// An identifier consists of at least one Unicode code point, the first of +/// which has the XID_Start property and the rest of which have the XID_Continue +/// property. +/// +/// - The empty string is not an identifier. Use `Option<Ident>`. +/// - A lifetime is not an identifier. Use `syn::Lifetime` instead. +/// +/// An identifier constructed with `Ident::new` is permitted to be a Rust +/// keyword, though parsing one through its [`Parse`] implementation rejects +/// Rust keywords. Use `input.call(Ident::parse_any)` when parsing to match the +/// behaviour of `Ident::new`. +/// +/// [`Parse`]: https://docs.rs/syn/2.0/syn/parse/trait.Parse.html +/// +/// # Examples +/// +/// A new ident can be created from a string using the `Ident::new` function. +/// A span must be provided explicitly which governs the name resolution +/// behavior of the resulting identifier. +/// +/// ``` +/// use proc_macro2::{Ident, Span}; +/// +/// fn main() { +/// let call_ident = Ident::new("calligraphy", Span::call_site()); +/// +/// println!("{}", call_ident); +/// } +/// ``` +/// +/// An ident can be interpolated into a token stream using the `quote!` macro. +/// +/// ``` +/// use proc_macro2::{Ident, Span}; +/// use quote::quote; +/// +/// fn main() { +/// let ident = Ident::new("demo", Span::call_site()); +/// +/// // Create a variable binding whose name is this ident. +/// let expanded = quote! { let #ident = 10; }; +/// +/// // Create a variable binding with a slightly different name. +/// let temp_ident = Ident::new(&format!("new_{}", ident), Span::call_site()); +/// let expanded = quote! { let #temp_ident = 10; }; +/// } +/// ``` +/// +/// A string representation of the ident is available through the `to_string()` +/// method. +/// +/// ``` +/// # use proc_macro2::{Ident, Span}; +/// # +/// # let ident = Ident::new("another_identifier", Span::call_site()); +/// # +/// // Examine the ident as a string. +/// let ident_string = ident.to_string(); +/// if ident_string.len() > 60 { +/// println!("Very long identifier: {}", ident_string) +/// } +/// ``` +#[derive(Clone)] +pub struct Ident { + inner: imp::Ident, + _marker: Marker, +} + +impl Ident { + fn _new(inner: imp::Ident) -> Self { + Ident { + inner, + _marker: Marker, + } + } + + /// Creates a new `Ident` with the given `string` as well as the specified + /// `span`. + /// + /// The `string` argument must be a valid identifier permitted by the + /// language, otherwise the function will panic. + /// + /// Note that `span`, currently in rustc, configures the hygiene information + /// for this identifier. + /// + /// As of this time `Span::call_site()` explicitly opts-in to "call-site" + /// hygiene meaning that identifiers created with this span will be resolved + /// as if they were written directly at the location of the macro call, and + /// other code at the macro call site will be able to refer to them as well. + /// + /// Later spans like `Span::def_site()` will allow to opt-in to + /// "definition-site" hygiene meaning that identifiers created with this + /// span will be resolved at the location of the macro definition and other + /// code at the macro call site will not be able to refer to them. + /// + /// Due to the current importance of hygiene this constructor, unlike other + /// tokens, requires a `Span` to be specified at construction. + /// + /// # Panics + /// + /// Panics if the input string is neither a keyword nor a legal variable + /// name. If you are not sure whether the string contains an identifier and + /// need to handle an error case, use + /// <a href="https://docs.rs/syn/2.0/syn/fn.parse_str.html"><code + /// style="padding-right:0;">syn::parse_str</code></a><code + /// style="padding-left:0;">::<Ident></code> + /// rather than `Ident::new`. + #[track_caller] + pub fn new(string: &str, span: Span) -> Self { + Ident::_new(imp::Ident::new_checked(string, span.inner)) + } + + /// Same as `Ident::new`, but creates a raw identifier (`r#ident`). The + /// `string` argument must be a valid identifier permitted by the language + /// (including keywords, e.g. `fn`). Keywords which are usable in path + /// segments (e.g. `self`, `super`) are not supported, and will cause a + /// panic. + #[track_caller] + pub fn new_raw(string: &str, span: Span) -> Self { + Ident::_new(imp::Ident::new_raw_checked(string, span.inner)) + } + + /// Returns the span of this `Ident`. + pub fn span(&self) -> Span { + Span::_new(self.inner.span()) + } + + /// Configures the span of this `Ident`, possibly changing its hygiene + /// context. + pub fn set_span(&mut self, span: Span) { + self.inner.set_span(span.inner); + } +} + +impl PartialEq for Ident { + fn eq(&self, other: &Ident) -> bool { + self.inner == other.inner + } +} + +impl<T> PartialEq<T> for Ident +where + T: ?Sized + AsRef<str>, +{ + fn eq(&self, other: &T) -> bool { + self.inner == other + } +} + +impl Eq for Ident {} + +impl PartialOrd for Ident { + fn partial_cmp(&self, other: &Ident) -> Option<Ordering> { + Some(self.cmp(other)) + } +} + +impl Ord for Ident { + fn cmp(&self, other: &Ident) -> Ordering { + self.to_string().cmp(&other.to_string()) + } +} + +impl Hash for Ident { + fn hash<H: Hasher>(&self, hasher: &mut H) { + self.to_string().hash(hasher); + } +} + +/// Prints the identifier as a string that should be losslessly convertible back +/// into the same identifier. +impl Display for Ident { + fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result { + Display::fmt(&self.inner, f) + } +} + +impl Debug for Ident { + fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result { + Debug::fmt(&self.inner, f) + } +} + +/// A literal string (`"hello"`), byte string (`b"hello"`), character (`'a'`), +/// byte character (`b'a'`), an integer or floating point number with or without +/// a suffix (`1`, `1u8`, `2.3`, `2.3f32`). +/// +/// Boolean literals like `true` and `false` do not belong here, they are +/// `Ident`s. +#[derive(Clone)] +pub struct Literal { + inner: imp::Literal, + _marker: Marker, +} + +macro_rules! suffixed_int_literals { + ($($name:ident => $kind:ident,)*) => ($( + /// Creates a new suffixed integer literal with the specified value. + /// + /// This function will create an integer like `1u32` where the integer + /// value specified is the first part of the token and the integral is + /// also suffixed at the end. Literals created from negative numbers may + /// not survive roundtrips through `TokenStream` or strings and may be + /// broken into two tokens (`-` and positive literal). + /// + /// Literals created through this method have the `Span::call_site()` + /// span by default, which can be configured with the `set_span` method + /// below. + pub fn $name(n: $kind) -> Literal { + Literal::_new(imp::Literal::$name(n)) + } + )*) +} + +macro_rules! unsuffixed_int_literals { + ($($name:ident => $kind:ident,)*) => ($( + /// Creates a new unsuffixed integer literal with the specified value. + /// + /// This function will create an integer like `1` where the integer + /// value specified is the first part of the token. No suffix is + /// specified on this token, meaning that invocations like + /// `Literal::i8_unsuffixed(1)` are equivalent to + /// `Literal::u32_unsuffixed(1)`. Literals created from negative numbers + /// may not survive roundtrips through `TokenStream` or strings and may + /// be broken into two tokens (`-` and positive literal). + /// + /// Literals created through this method have the `Span::call_site()` + /// span by default, which can be configured with the `set_span` method + /// below. + pub fn $name(n: $kind) -> Literal { + Literal::_new(imp::Literal::$name(n)) + } + )*) +} + +impl Literal { + fn _new(inner: imp::Literal) -> Self { + Literal { + inner, + _marker: Marker, + } + } + + fn _new_fallback(inner: fallback::Literal) -> Self { + Literal { + inner: inner.into(), + _marker: Marker, + } + } + + suffixed_int_literals! { + u8_suffixed => u8, + u16_suffixed => u16, + u32_suffixed => u32, + u64_suffixed => u64, + u128_suffixed => u128, + usize_suffixed => usize, + i8_suffixed => i8, + i16_suffixed => i16, + i32_suffixed => i32, + i64_suffixed => i64, + i128_suffixed => i128, + isize_suffixed => isize, + } + + unsuffixed_int_literals! { + u8_unsuffixed => u8, + u16_unsuffixed => u16, + u32_unsuffixed => u32, + u64_unsuffixed => u64, + u128_unsuffixed => u128, + usize_unsuffixed => usize, + i8_unsuffixed => i8, + i16_unsuffixed => i16, + i32_unsuffixed => i32, + i64_unsuffixed => i64, + i128_unsuffixed => i128, + isize_unsuffixed => isize, + } + + /// Creates a new unsuffixed floating-point literal. + /// + /// This constructor is similar to those like `Literal::i8_unsuffixed` where + /// the float's value is emitted directly into the token but no suffix is + /// used, so it may be inferred to be a `f64` later in the compiler. + /// Literals created from negative numbers may not survive round-trips + /// through `TokenStream` or strings and may be broken into two tokens (`-` + /// and positive literal). + /// + /// # Panics + /// + /// This function requires that the specified float is finite, for example + /// if it is infinity or NaN this function will panic. + pub fn f64_unsuffixed(f: f64) -> Literal { + assert!(f.is_finite()); + Literal::_new(imp::Literal::f64_unsuffixed(f)) + } + + /// Creates a new suffixed floating-point literal. + /// + /// This constructor will create a literal like `1.0f64` where the value + /// specified is the preceding part of the token and `f64` is the suffix of + /// the token. This token will always be inferred to be an `f64` in the + /// compiler. Literals created from negative numbers may not survive + /// round-trips through `TokenStream` or strings and may be broken into two + /// tokens (`-` and positive literal). + /// + /// # Panics + /// + /// This function requires that the specified float is finite, for example + /// if it is infinity or NaN this function will panic. + pub fn f64_suffixed(f: f64) -> Literal { + assert!(f.is_finite()); + Literal::_new(imp::Literal::f64_suffixed(f)) + } + + /// Creates a new unsuffixed floating-point literal. + /// + /// This constructor is similar to those like `Literal::i8_unsuffixed` where + /// the float's value is emitted directly into the token but no suffix is + /// used, so it may be inferred to be a `f64` later in the compiler. + /// Literals created from negative numbers may not survive round-trips + /// through `TokenStream` or strings and may be broken into two tokens (`-` + /// and positive literal). + /// + /// # Panics + /// + /// This function requires that the specified float is finite, for example + /// if it is infinity or NaN this function will panic. + pub fn f32_unsuffixed(f: f32) -> Literal { + assert!(f.is_finite()); + Literal::_new(imp::Literal::f32_unsuffixed(f)) + } + + /// Creates a new suffixed floating-point literal. + /// + /// This constructor will create a literal like `1.0f32` where the value + /// specified is the preceding part of the token and `f32` is the suffix of + /// the token. This token will always be inferred to be an `f32` in the + /// compiler. Literals created from negative numbers may not survive + /// round-trips through `TokenStream` or strings and may be broken into two + /// tokens (`-` and positive literal). + /// + /// # Panics + /// + /// This function requires that the specified float is finite, for example + /// if it is infinity or NaN this function will panic. + pub fn f32_suffixed(f: f32) -> Literal { + assert!(f.is_finite()); + Literal::_new(imp::Literal::f32_suffixed(f)) + } + + /// String literal. + pub fn string(string: &str) -> Literal { + Literal::_new(imp::Literal::string(string)) + } + + /// Character literal. + pub fn character(ch: char) -> Literal { + Literal::_new(imp::Literal::character(ch)) + } + + /// Byte string literal. + pub fn byte_string(s: &[u8]) -> Literal { + Literal::_new(imp::Literal::byte_string(s)) + } + + /// Returns the span encompassing this literal. + pub fn span(&self) -> Span { + Span::_new(self.inner.span()) + } + + /// Configures the span associated for this literal. + pub fn set_span(&mut self, span: Span) { + self.inner.set_span(span.inner); + } + + /// Returns a `Span` that is a subset of `self.span()` containing only + /// the source bytes in range `range`. Returns `None` if the would-be + /// trimmed span is outside the bounds of `self`. + /// + /// Warning: the underlying [`proc_macro::Literal::subspan`] method is + /// nightly-only. When called from within a procedural macro not using a + /// nightly compiler, this method will always return `None`. + /// + /// [`proc_macro::Literal::subspan`]: https://doc.rust-lang.org/proc_macro/struct.Literal.html#method.subspan + pub fn subspan<R: RangeBounds<usize>>(&self, range: R) -> Option<Span> { + self.inner.subspan(range).map(Span::_new) + } + + // Intended for the `quote!` macro to use when constructing a proc-macro2 + // token out of a macro_rules $:literal token, which is already known to be + // a valid literal. This avoids reparsing/validating the literal's string + // representation. This is not public API other than for quote. + #[doc(hidden)] + pub unsafe fn from_str_unchecked(repr: &str) -> Self { + Literal::_new(unsafe { imp::Literal::from_str_unchecked(repr) }) + } +} + +impl FromStr for Literal { + type Err = LexError; + + fn from_str(repr: &str) -> Result<Self, LexError> { + repr.parse().map(Literal::_new).map_err(|inner| LexError { + inner, + _marker: Marker, + }) + } +} + +impl Debug for Literal { + fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result { + Debug::fmt(&self.inner, f) + } +} + +impl Display for Literal { + fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result { + Display::fmt(&self.inner, f) + } +} + +/// Public implementation details for the `TokenStream` type, such as iterators. +pub mod token_stream { + use crate::marker::Marker; + use crate::{imp, TokenTree}; + use core::fmt::{self, Debug}; + + pub use crate::TokenStream; + + /// An iterator over `TokenStream`'s `TokenTree`s. + /// + /// The iteration is "shallow", e.g. the iterator doesn't recurse into + /// delimited groups, and returns whole groups as token trees. + #[derive(Clone)] + pub struct IntoIter { + inner: imp::TokenTreeIter, + _marker: Marker, + } + + impl Iterator for IntoIter { + type Item = TokenTree; + + fn next(&mut self) -> Option<TokenTree> { + self.inner.next() + } + + fn size_hint(&self) -> (usize, Option<usize>) { + self.inner.size_hint() + } + } + + impl Debug for IntoIter { + fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result { + f.write_str("TokenStream ")?; + f.debug_list().entries(self.clone()).finish() + } + } + + impl IntoIterator for TokenStream { + type Item = TokenTree; + type IntoIter = IntoIter; + + fn into_iter(self) -> IntoIter { + IntoIter { + inner: self.inner.into_iter(), + _marker: Marker, + } + } + } +} diff --git a/src/location.rs b/src/location.rs new file mode 100644 index 0000000..463026c --- /dev/null +++ b/src/location.rs @@ -0,0 +1,29 @@ +use core::cmp::Ordering; + +/// A line-column pair representing the start or end of a `Span`. +/// +/// This type is semver exempt and not exposed by default. +#[cfg_attr(doc_cfg, doc(cfg(feature = "span-locations")))] +#[derive(Copy, Clone, Debug, PartialEq, Eq, Hash)] +pub struct LineColumn { + /// The 1-indexed line in the source file on which the span starts or ends + /// (inclusive). + pub line: usize, + /// The 0-indexed column (in UTF-8 characters) in the source file on which + /// the span starts or ends (inclusive). + pub column: usize, +} + +impl Ord for LineColumn { + fn cmp(&self, other: &Self) -> Ordering { + self.line + .cmp(&other.line) + .then(self.column.cmp(&other.column)) + } +} + +impl PartialOrd for LineColumn { + fn partial_cmp(&self, other: &Self) -> Option<Ordering> { + Some(self.cmp(other)) + } +} diff --git a/src/marker.rs b/src/marker.rs new file mode 100644 index 0000000..e8874bd --- /dev/null +++ b/src/marker.rs @@ -0,0 +1,21 @@ +use alloc::rc::Rc; +use core::marker::PhantomData; +use core::panic::{RefUnwindSafe, UnwindSafe}; + +// Zero sized marker with the correct set of autotrait impls we want all proc +// macro types to have. +pub(crate) type Marker = PhantomData<ProcMacroAutoTraits>; + +pub(crate) use self::value::*; + +mod value { + pub(crate) use core::marker::PhantomData as Marker; +} + +pub(crate) struct ProcMacroAutoTraits( + #[allow(dead_code)] // https://github.com/rust-lang/rust/issues/119645 + Rc<()>, +); + +impl UnwindSafe for ProcMacroAutoTraits {} +impl RefUnwindSafe for ProcMacroAutoTraits {} diff --git a/src/parse.rs b/src/parse.rs new file mode 100644 index 0000000..07239bc --- /dev/null +++ b/src/parse.rs @@ -0,0 +1,996 @@ +use crate::fallback::{ + self, is_ident_continue, is_ident_start, Group, LexError, Literal, Span, TokenStream, + TokenStreamBuilder, +}; +use crate::{Delimiter, Punct, Spacing, TokenTree}; +use core::char; +use core::str::{Bytes, CharIndices, Chars}; + +#[derive(Copy, Clone, Eq, PartialEq)] +pub(crate) struct Cursor<'a> { + pub rest: &'a str, + #[cfg(span_locations)] + pub off: u32, +} + +impl<'a> Cursor<'a> { + pub fn advance(&self, bytes: usize) -> Cursor<'a> { + let (_front, rest) = self.rest.split_at(bytes); + Cursor { + rest, + #[cfg(span_locations)] + off: self.off + _front.chars().count() as u32, + } + } + + pub fn starts_with(&self, s: &str) -> bool { + self.rest.starts_with(s) + } + + pub fn starts_with_char(&self, ch: char) -> bool { + self.rest.starts_with(ch) + } + + pub fn starts_with_fn<Pattern>(&self, f: Pattern) -> bool + where + Pattern: FnMut(char) -> bool, + { + self.rest.starts_with(f) + } + + pub fn is_empty(&self) -> bool { + self.rest.is_empty() + } + + fn len(&self) -> usize { + self.rest.len() + } + + fn as_bytes(&self) -> &'a [u8] { + self.rest.as_bytes() + } + + fn bytes(&self) -> Bytes<'a> { + self.rest.bytes() + } + + fn chars(&self) -> Chars<'a> { + self.rest.chars() + } + + fn char_indices(&self) -> CharIndices<'a> { + self.rest.char_indices() + } + + fn parse(&self, tag: &str) -> Result<Cursor<'a>, Reject> { + if self.starts_with(tag) { + Ok(self.advance(tag.len())) + } else { + Err(Reject) + } + } +} + +pub(crate) struct Reject; +type PResult<'a, O> = Result<(Cursor<'a>, O), Reject>; + +fn skip_whitespace(input: Cursor) -> Cursor { + let mut s = input; + + while !s.is_empty() { + let byte = s.as_bytes()[0]; + if byte == b'/' { + if s.starts_with("//") + && (!s.starts_with("///") || s.starts_with("////")) + && !s.starts_with("//!") + { + let (cursor, _) = take_until_newline_or_eof(s); + s = cursor; + continue; + } else if s.starts_with("/**/") { + s = s.advance(4); + continue; + } else if s.starts_with("/*") + && (!s.starts_with("/**") || s.starts_with("/***")) + && !s.starts_with("/*!") + { + match block_comment(s) { + Ok((rest, _)) => { + s = rest; + continue; + } + Err(Reject) => return s, + } + } + } + match byte { + b' ' | 0x09..=0x0d => { + s = s.advance(1); + continue; + } + b if b.is_ascii() => {} + _ => { + let ch = s.chars().next().unwrap(); + if is_whitespace(ch) { + s = s.advance(ch.len_utf8()); + continue; + } + } + } + return s; + } + s +} + +fn block_comment(input: Cursor) -> PResult<&str> { + if !input.starts_with("/*") { + return Err(Reject); + } + + let mut depth = 0usize; + let bytes = input.as_bytes(); + let mut i = 0usize; + let upper = bytes.len() - 1; + + while i < upper { + if bytes[i] == b'/' && bytes[i + 1] == b'*' { + depth += 1; + i += 1; // eat '*' + } else if bytes[i] == b'*' && bytes[i + 1] == b'/' { + depth -= 1; + if depth == 0 { + return Ok((input.advance(i + 2), &input.rest[..i + 2])); + } + i += 1; // eat '/' + } + i += 1; + } + + Err(Reject) +} + +fn is_whitespace(ch: char) -> bool { + // Rust treats left-to-right mark and right-to-left mark as whitespace + ch.is_whitespace() || ch == '\u{200e}' || ch == '\u{200f}' +} + +fn word_break(input: Cursor) -> Result<Cursor, Reject> { + match input.chars().next() { + Some(ch) if is_ident_continue(ch) => Err(Reject), + Some(_) | None => Ok(input), + } +} + +// Rustc's representation of a macro expansion error in expression position or +// type position. +const ERROR: &str = "(/*ERROR*/)"; + +pub(crate) fn token_stream(mut input: Cursor) -> Result<TokenStream, LexError> { + let mut trees = TokenStreamBuilder::new(); + let mut stack = Vec::new(); + + loop { + input = skip_whitespace(input); + + if let Ok((rest, ())) = doc_comment(input, &mut trees) { + input = rest; + continue; + } + + #[cfg(span_locations)] + let lo = input.off; + + let first = match input.bytes().next() { + Some(first) => first, + None => match stack.last() { + None => return Ok(trees.build()), + #[cfg(span_locations)] + Some((lo, _frame)) => { + return Err(LexError { + span: Span { lo: *lo, hi: *lo }, + }) + } + #[cfg(not(span_locations))] + Some(_frame) => return Err(LexError { span: Span {} }), + }, + }; + + if let Some(open_delimiter) = match first { + b'(' if !input.starts_with(ERROR) => Some(Delimiter::Parenthesis), + b'[' => Some(Delimiter::Bracket), + b'{' => Some(Delimiter::Brace), + _ => None, + } { + input = input.advance(1); + let frame = (open_delimiter, trees); + #[cfg(span_locations)] + let frame = (lo, frame); + stack.push(frame); + trees = TokenStreamBuilder::new(); + } else if let Some(close_delimiter) = match first { + b')' => Some(Delimiter::Parenthesis), + b']' => Some(Delimiter::Bracket), + b'}' => Some(Delimiter::Brace), + _ => None, + } { + let frame = match stack.pop() { + Some(frame) => frame, + None => return Err(lex_error(input)), + }; + #[cfg(span_locations)] + let (lo, frame) = frame; + let (open_delimiter, outer) = frame; + if open_delimiter != close_delimiter { + return Err(lex_error(input)); + } + input = input.advance(1); + let mut g = Group::new(open_delimiter, trees.build()); + g.set_span(Span { + #[cfg(span_locations)] + lo, + #[cfg(span_locations)] + hi: input.off, + }); + trees = outer; + trees.push_token_from_parser(TokenTree::Group(crate::Group::_new_fallback(g))); + } else { + let (rest, mut tt) = match leaf_token(input) { + Ok((rest, tt)) => (rest, tt), + Err(Reject) => return Err(lex_error(input)), + }; + tt.set_span(crate::Span::_new_fallback(Span { + #[cfg(span_locations)] + lo, + #[cfg(span_locations)] + hi: rest.off, + })); + trees.push_token_from_parser(tt); + input = rest; + } + } +} + +fn lex_error(cursor: Cursor) -> LexError { + #[cfg(not(span_locations))] + let _ = cursor; + LexError { + span: Span { + #[cfg(span_locations)] + lo: cursor.off, + #[cfg(span_locations)] + hi: cursor.off, + }, + } +} + +fn leaf_token(input: Cursor) -> PResult<TokenTree> { + if let Ok((input, l)) = literal(input) { + // must be parsed before ident + Ok((input, TokenTree::Literal(crate::Literal::_new_fallback(l)))) + } else if let Ok((input, p)) = punct(input) { + Ok((input, TokenTree::Punct(p))) + } else if let Ok((input, i)) = ident(input) { + Ok((input, TokenTree::Ident(i))) + } else if input.starts_with(ERROR) { + let rest = input.advance(ERROR.len()); + let repr = crate::Literal::_new_fallback(Literal::_new(ERROR.to_owned())); + Ok((rest, TokenTree::Literal(repr))) + } else { + Err(Reject) + } +} + +fn ident(input: Cursor) -> PResult<crate::Ident> { + if [ + "r\"", "r#\"", "r##", "b\"", "b\'", "br\"", "br#", "c\"", "cr\"", "cr#", + ] + .iter() + .any(|prefix| input.starts_with(prefix)) + { + Err(Reject) + } else { + ident_any(input) + } +} + +fn ident_any(input: Cursor) -> PResult<crate::Ident> { + let raw = input.starts_with("r#"); + let rest = input.advance((raw as usize) << 1); + + let (rest, sym) = ident_not_raw(rest)?; + + if !raw { + let ident = crate::Ident::_new(crate::imp::Ident::new_unchecked( + sym, + fallback::Span::call_site(), + )); + return Ok((rest, ident)); + } + + match sym { + "_" | "super" | "self" | "Self" | "crate" => return Err(Reject), + _ => {} + } + + let ident = crate::Ident::_new(crate::imp::Ident::new_raw_unchecked( + sym, + fallback::Span::call_site(), + )); + Ok((rest, ident)) +} + +fn ident_not_raw(input: Cursor) -> PResult<&str> { + let mut chars = input.char_indices(); + + match chars.next() { + Some((_, ch)) if is_ident_start(ch) => {} + _ => return Err(Reject), + } + + let mut end = input.len(); + for (i, ch) in chars { + if !is_ident_continue(ch) { + end = i; + break; + } + } + + Ok((input.advance(end), &input.rest[..end])) +} + +pub(crate) fn literal(input: Cursor) -> PResult<Literal> { + let rest = literal_nocapture(input)?; + let end = input.len() - rest.len(); + Ok((rest, Literal::_new(input.rest[..end].to_string()))) +} + +fn literal_nocapture(input: Cursor) -> Result<Cursor, Reject> { + if let Ok(ok) = string(input) { + Ok(ok) + } else if let Ok(ok) = byte_string(input) { + Ok(ok) + } else if let Ok(ok) = c_string(input) { + Ok(ok) + } else if let Ok(ok) = byte(input) { + Ok(ok) + } else if let Ok(ok) = character(input) { + Ok(ok) + } else if let Ok(ok) = float(input) { + Ok(ok) + } else if let Ok(ok) = int(input) { + Ok(ok) + } else { + Err(Reject) + } +} + +fn literal_suffix(input: Cursor) -> Cursor { + match ident_not_raw(input) { + Ok((input, _)) => input, + Err(Reject) => input, + } +} + +fn string(input: Cursor) -> Result<Cursor, Reject> { + if let Ok(input) = input.parse("\"") { + cooked_string(input) + } else if let Ok(input) = input.parse("r") { + raw_string(input) + } else { + Err(Reject) + } +} + +fn cooked_string(mut input: Cursor) -> Result<Cursor, Reject> { + let mut chars = input.char_indices(); + + while let Some((i, ch)) = chars.next() { + match ch { + '"' => { + let input = input.advance(i + 1); + return Ok(literal_suffix(input)); + } + '\r' => match chars.next() { + Some((_, '\n')) => {} + _ => break, + }, + '\\' => match chars.next() { + Some((_, 'x')) => { + backslash_x_char(&mut chars)?; + } + Some((_, 'n' | 'r' | 't' | '\\' | '\'' | '"' | '0')) => {} + Some((_, 'u')) => { + backslash_u(&mut chars)?; + } + Some((newline, ch @ ('\n' | '\r'))) => { + input = input.advance(newline + 1); + trailing_backslash(&mut input, ch as u8)?; + chars = input.char_indices(); + } + _ => break, + }, + _ch => {} + } + } + Err(Reject) +} + +fn raw_string(input: Cursor) -> Result<Cursor, Reject> { + let (input, delimiter) = delimiter_of_raw_string(input)?; + let mut bytes = input.bytes().enumerate(); + while let Some((i, byte)) = bytes.next() { + match byte { + b'"' if input.rest[i + 1..].starts_with(delimiter) => { + let rest = input.advance(i + 1 + delimiter.len()); + return Ok(literal_suffix(rest)); + } + b'\r' => match bytes.next() { + Some((_, b'\n')) => {} + _ => break, + }, + _ => {} + } + } + Err(Reject) +} + +fn byte_string(input: Cursor) -> Result<Cursor, Reject> { + if let Ok(input) = input.parse("b\"") { + cooked_byte_string(input) + } else if let Ok(input) = input.parse("br") { + raw_byte_string(input) + } else { + Err(Reject) + } +} + +fn cooked_byte_string(mut input: Cursor) -> Result<Cursor, Reject> { + let mut bytes = input.bytes().enumerate(); + while let Some((offset, b)) = bytes.next() { + match b { + b'"' => { + let input = input.advance(offset + 1); + return Ok(literal_suffix(input)); + } + b'\r' => match bytes.next() { + Some((_, b'\n')) => {} + _ => break, + }, + b'\\' => match bytes.next() { + Some((_, b'x')) => { + backslash_x_byte(&mut bytes)?; + } + Some((_, b'n' | b'r' | b't' | b'\\' | b'0' | b'\'' | b'"')) => {} + Some((newline, b @ (b'\n' | b'\r'))) => { + input = input.advance(newline + 1); + trailing_backslash(&mut input, b)?; + bytes = input.bytes().enumerate(); + } + _ => break, + }, + b if b.is_ascii() => {} + _ => break, + } + } + Err(Reject) +} + +fn delimiter_of_raw_string(input: Cursor) -> PResult<&str> { + for (i, byte) in input.bytes().enumerate() { + match byte { + b'"' => { + if i > 255 { + // https://github.com/rust-lang/rust/pull/95251 + return Err(Reject); + } + return Ok((input.advance(i + 1), &input.rest[..i])); + } + b'#' => {} + _ => break, + } + } + Err(Reject) +} + +fn raw_byte_string(input: Cursor) -> Result<Cursor, Reject> { + let (input, delimiter) = delimiter_of_raw_string(input)?; + let mut bytes = input.bytes().enumerate(); + while let Some((i, byte)) = bytes.next() { + match byte { + b'"' if input.rest[i + 1..].starts_with(delimiter) => { + let rest = input.advance(i + 1 + delimiter.len()); + return Ok(literal_suffix(rest)); + } + b'\r' => match bytes.next() { + Some((_, b'\n')) => {} + _ => break, + }, + other => { + if !other.is_ascii() { + break; + } + } + } + } + Err(Reject) +} + +fn c_string(input: Cursor) -> Result<Cursor, Reject> { + if let Ok(input) = input.parse("c\"") { + cooked_c_string(input) + } else if let Ok(input) = input.parse("cr") { + raw_c_string(input) + } else { + Err(Reject) + } +} + +fn raw_c_string(input: Cursor) -> Result<Cursor, Reject> { + let (input, delimiter) = delimiter_of_raw_string(input)?; + let mut bytes = input.bytes().enumerate(); + while let Some((i, byte)) = bytes.next() { + match byte { + b'"' if input.rest[i + 1..].starts_with(delimiter) => { + let rest = input.advance(i + 1 + delimiter.len()); + return Ok(literal_suffix(rest)); + } + b'\r' => match bytes.next() { + Some((_, b'\n')) => {} + _ => break, + }, + b'\0' => break, + _ => {} + } + } + Err(Reject) +} + +fn cooked_c_string(mut input: Cursor) -> Result<Cursor, Reject> { + let mut chars = input.char_indices(); + + while let Some((i, ch)) = chars.next() { + match ch { + '"' => { + let input = input.advance(i + 1); + return Ok(literal_suffix(input)); + } + '\r' => match chars.next() { + Some((_, '\n')) => {} + _ => break, + }, + '\\' => match chars.next() { + Some((_, 'x')) => { + backslash_x_nonzero(&mut chars)?; + } + Some((_, 'n' | 'r' | 't' | '\\' | '\'' | '"')) => {} + Some((_, 'u')) => { + if backslash_u(&mut chars)? == '\0' { + break; + } + } + Some((newline, ch @ ('\n' | '\r'))) => { + input = input.advance(newline + 1); + trailing_backslash(&mut input, ch as u8)?; + chars = input.char_indices(); + } + _ => break, + }, + '\0' => break, + _ch => {} + } + } + Err(Reject) +} + +fn byte(input: Cursor) -> Result<Cursor, Reject> { + let input = input.parse("b'")?; + let mut bytes = input.bytes().enumerate(); + let ok = match bytes.next().map(|(_, b)| b) { + Some(b'\\') => match bytes.next().map(|(_, b)| b) { + Some(b'x') => backslash_x_byte(&mut bytes).is_ok(), + Some(b'n' | b'r' | b't' | b'\\' | b'0' | b'\'' | b'"') => true, + _ => false, + }, + b => b.is_some(), + }; + if !ok { + return Err(Reject); + } + let (offset, _) = bytes.next().ok_or(Reject)?; + if !input.chars().as_str().is_char_boundary(offset) { + return Err(Reject); + } + let input = input.advance(offset).parse("'")?; + Ok(literal_suffix(input)) +} + +fn character(input: Cursor) -> Result<Cursor, Reject> { + let input = input.parse("'")?; + let mut chars = input.char_indices(); + let ok = match chars.next().map(|(_, ch)| ch) { + Some('\\') => match chars.next().map(|(_, ch)| ch) { + Some('x') => backslash_x_char(&mut chars).is_ok(), + Some('u') => backslash_u(&mut chars).is_ok(), + Some('n' | 'r' | 't' | '\\' | '0' | '\'' | '"') => true, + _ => false, + }, + ch => ch.is_some(), + }; + if !ok { + return Err(Reject); + } + let (idx, _) = chars.next().ok_or(Reject)?; + let input = input.advance(idx).parse("'")?; + Ok(literal_suffix(input)) +} + +macro_rules! next_ch { + ($chars:ident @ $pat:pat) => { + match $chars.next() { + Some((_, ch)) => match ch { + $pat => ch, + _ => return Err(Reject), + }, + None => return Err(Reject), + } + }; +} + +fn backslash_x_char<I>(chars: &mut I) -> Result<(), Reject> +where + I: Iterator<Item = (usize, char)>, +{ + next_ch!(chars @ '0'..='7'); + next_ch!(chars @ '0'..='9' | 'a'..='f' | 'A'..='F'); + Ok(()) +} + +fn backslash_x_byte<I>(chars: &mut I) -> Result<(), Reject> +where + I: Iterator<Item = (usize, u8)>, +{ + next_ch!(chars @ b'0'..=b'9' | b'a'..=b'f' | b'A'..=b'F'); + next_ch!(chars @ b'0'..=b'9' | b'a'..=b'f' | b'A'..=b'F'); + Ok(()) +} + +fn backslash_x_nonzero<I>(chars: &mut I) -> Result<(), Reject> +where + I: Iterator<Item = (usize, char)>, +{ + let first = next_ch!(chars @ '0'..='9' | 'a'..='f' | 'A'..='F'); + let second = next_ch!(chars @ '0'..='9' | 'a'..='f' | 'A'..='F'); + if first == '0' && second == '0' { + Err(Reject) + } else { + Ok(()) + } +} + +fn backslash_u<I>(chars: &mut I) -> Result<char, Reject> +where + I: Iterator<Item = (usize, char)>, +{ + next_ch!(chars @ '{'); + let mut value = 0; + let mut len = 0; + for (_, ch) in chars { + let digit = match ch { + '0'..='9' => ch as u8 - b'0', + 'a'..='f' => 10 + ch as u8 - b'a', + 'A'..='F' => 10 + ch as u8 - b'A', + '_' if len > 0 => continue, + '}' if len > 0 => return char::from_u32(value).ok_or(Reject), + _ => break, + }; + if len == 6 { + break; + } + value *= 0x10; + value += u32::from(digit); + len += 1; + } + Err(Reject) +} + +fn trailing_backslash(input: &mut Cursor, mut last: u8) -> Result<(), Reject> { + let mut whitespace = input.bytes().enumerate(); + loop { + if last == b'\r' && whitespace.next().map_or(true, |(_, b)| b != b'\n') { + return Err(Reject); + } + match whitespace.next() { + Some((_, b @ (b' ' | b'\t' | b'\n' | b'\r'))) => { + last = b; + } + Some((offset, _)) => { + *input = input.advance(offset); + return Ok(()); + } + None => return Err(Reject), + } + } +} + +fn float(input: Cursor) -> Result<Cursor, Reject> { + let mut rest = float_digits(input)?; + if let Some(ch) = rest.chars().next() { + if is_ident_start(ch) { + rest = ident_not_raw(rest)?.0; + } + } + word_break(rest) +} + +fn float_digits(input: Cursor) -> Result<Cursor, Reject> { + let mut chars = input.chars().peekable(); + match chars.next() { + Some(ch) if '0' <= ch && ch <= '9' => {} + _ => return Err(Reject), + } + + let mut len = 1; + let mut has_dot = false; + let mut has_exp = false; + while let Some(&ch) = chars.peek() { + match ch { + '0'..='9' | '_' => { + chars.next(); + len += 1; + } + '.' => { + if has_dot { + break; + } + chars.next(); + if chars + .peek() + .map_or(false, |&ch| ch == '.' || is_ident_start(ch)) + { + return Err(Reject); + } + len += 1; + has_dot = true; + } + 'e' | 'E' => { + chars.next(); + len += 1; + has_exp = true; + break; + } + _ => break, + } + } + + if !(has_dot || has_exp) { + return Err(Reject); + } + + if has_exp { + let token_before_exp = if has_dot { + Ok(input.advance(len - 1)) + } else { + Err(Reject) + }; + let mut has_sign = false; + let mut has_exp_value = false; + while let Some(&ch) = chars.peek() { + match ch { + '+' | '-' => { + if has_exp_value { + break; + } + if has_sign { + return token_before_exp; + } + chars.next(); + len += 1; + has_sign = true; + } + '0'..='9' => { + chars.next(); + len += 1; + has_exp_value = true; + } + '_' => { + chars.next(); + len += 1; + } + _ => break, + } + } + if !has_exp_value { + return token_before_exp; + } + } + + Ok(input.advance(len)) +} + +fn int(input: Cursor) -> Result<Cursor, Reject> { + let mut rest = digits(input)?; + if let Some(ch) = rest.chars().next() { + if is_ident_start(ch) { + rest = ident_not_raw(rest)?.0; + } + } + word_break(rest) +} + +fn digits(mut input: Cursor) -> Result<Cursor, Reject> { + let base = if input.starts_with("0x") { + input = input.advance(2); + 16 + } else if input.starts_with("0o") { + input = input.advance(2); + 8 + } else if input.starts_with("0b") { + input = input.advance(2); + 2 + } else { + 10 + }; + + let mut len = 0; + let mut empty = true; + for b in input.bytes() { + match b { + b'0'..=b'9' => { + let digit = (b - b'0') as u64; + if digit >= base { + return Err(Reject); + } + } + b'a'..=b'f' => { + let digit = 10 + (b - b'a') as u64; + if digit >= base { + break; + } + } + b'A'..=b'F' => { + let digit = 10 + (b - b'A') as u64; + if digit >= base { + break; + } + } + b'_' => { + if empty && base == 10 { + return Err(Reject); + } + len += 1; + continue; + } + _ => break, + }; + len += 1; + empty = false; + } + if empty { + Err(Reject) + } else { + Ok(input.advance(len)) + } +} + +fn punct(input: Cursor) -> PResult<Punct> { + let (rest, ch) = punct_char(input)?; + if ch == '\'' { + if ident_any(rest)?.0.starts_with_char('\'') { + Err(Reject) + } else { + Ok((rest, Punct::new('\'', Spacing::Joint))) + } + } else { + let kind = match punct_char(rest) { + Ok(_) => Spacing::Joint, + Err(Reject) => Spacing::Alone, + }; + Ok((rest, Punct::new(ch, kind))) + } +} + +fn punct_char(input: Cursor) -> PResult<char> { + if input.starts_with("//") || input.starts_with("/*") { + // Do not accept `/` of a comment as a punct. + return Err(Reject); + } + + let mut chars = input.chars(); + let first = match chars.next() { + Some(ch) => ch, + None => { + return Err(Reject); + } + }; + let recognized = "~!@#$%^&*-=+|;:,<.>/?'"; + if recognized.contains(first) { + Ok((input.advance(first.len_utf8()), first)) + } else { + Err(Reject) + } +} + +fn doc_comment<'a>(input: Cursor<'a>, trees: &mut TokenStreamBuilder) -> PResult<'a, ()> { + #[cfg(span_locations)] + let lo = input.off; + let (rest, (comment, inner)) = doc_comment_contents(input)?; + let fallback_span = Span { + #[cfg(span_locations)] + lo, + #[cfg(span_locations)] + hi: rest.off, + }; + let span = crate::Span::_new_fallback(fallback_span); + + let mut scan_for_bare_cr = comment; + while let Some(cr) = scan_for_bare_cr.find('\r') { + let rest = &scan_for_bare_cr[cr + 1..]; + if !rest.starts_with('\n') { + return Err(Reject); + } + scan_for_bare_cr = rest; + } + + let mut pound = Punct::new('#', Spacing::Alone); + pound.set_span(span); + trees.push_token_from_parser(TokenTree::Punct(pound)); + + if inner { + let mut bang = Punct::new('!', Spacing::Alone); + bang.set_span(span); + trees.push_token_from_parser(TokenTree::Punct(bang)); + } + + let doc_ident = crate::Ident::_new(crate::imp::Ident::new_unchecked("doc", fallback_span)); + let mut equal = Punct::new('=', Spacing::Alone); + equal.set_span(span); + let mut literal = crate::Literal::string(comment); + literal.set_span(span); + let mut bracketed = TokenStreamBuilder::with_capacity(3); + bracketed.push_token_from_parser(TokenTree::Ident(doc_ident)); + bracketed.push_token_from_parser(TokenTree::Punct(equal)); + bracketed.push_token_from_parser(TokenTree::Literal(literal)); + let group = Group::new(Delimiter::Bracket, bracketed.build()); + let mut group = crate::Group::_new_fallback(group); + group.set_span(span); + trees.push_token_from_parser(TokenTree::Group(group)); + + Ok((rest, ())) +} + +fn doc_comment_contents(input: Cursor) -> PResult<(&str, bool)> { + if input.starts_with("//!") { + let input = input.advance(3); + let (input, s) = take_until_newline_or_eof(input); + Ok((input, (s, true))) + } else if input.starts_with("/*!") { + let (input, s) = block_comment(input)?; + Ok((input, (&s[3..s.len() - 2], true))) + } else if input.starts_with("///") { + let input = input.advance(3); + if input.starts_with_char('/') { + return Err(Reject); + } + let (input, s) = take_until_newline_or_eof(input); + Ok((input, (s, false))) + } else if input.starts_with("/**") && !input.rest[3..].starts_with('*') { + let (input, s) = block_comment(input)?; + Ok((input, (&s[3..s.len() - 2], false))) + } else { + Err(Reject) + } +} + +fn take_until_newline_or_eof(input: Cursor) -> (Cursor, &str) { + let chars = input.char_indices(); + + for (i, ch) in chars { + if ch == '\n' { + return (input.advance(i), &input.rest[..i]); + } else if ch == '\r' && input.rest[i + 1..].starts_with('\n') { + return (input.advance(i + 1), &input.rest[..i]); + } + } + + (input.advance(input.len()), input.rest) +} diff --git a/src/rcvec.rs b/src/rcvec.rs new file mode 100644 index 0000000..37955af --- /dev/null +++ b/src/rcvec.rs @@ -0,0 +1,145 @@ +use alloc::rc::Rc; +use alloc::vec; +use core::mem; +use core::panic::RefUnwindSafe; +use core::slice; + +pub(crate) struct RcVec<T> { + inner: Rc<Vec<T>>, +} + +pub(crate) struct RcVecBuilder<T> { + inner: Vec<T>, +} + +pub(crate) struct RcVecMut<'a, T> { + inner: &'a mut Vec<T>, +} + +#[derive(Clone)] +pub(crate) struct RcVecIntoIter<T> { + inner: vec::IntoIter<T>, +} + +impl<T> RcVec<T> { + pub fn is_empty(&self) -> bool { + self.inner.is_empty() + } + + pub fn len(&self) -> usize { + self.inner.len() + } + + pub fn iter(&self) -> slice::Iter<T> { + self.inner.iter() + } + + pub fn make_mut(&mut self) -> RcVecMut<T> + where + T: Clone, + { + RcVecMut { + inner: Rc::make_mut(&mut self.inner), + } + } + + pub fn get_mut(&mut self) -> Option<RcVecMut<T>> { + let inner = Rc::get_mut(&mut self.inner)?; + Some(RcVecMut { inner }) + } + + pub fn make_owned(mut self) -> RcVecBuilder<T> + where + T: Clone, + { + let vec = if let Some(owned) = Rc::get_mut(&mut self.inner) { + mem::take(owned) + } else { + Vec::clone(&self.inner) + }; + RcVecBuilder { inner: vec } + } +} + +impl<T> RcVecBuilder<T> { + pub fn new() -> Self { + RcVecBuilder { inner: Vec::new() } + } + + pub fn with_capacity(cap: usize) -> Self { + RcVecBuilder { + inner: Vec::with_capacity(cap), + } + } + + pub fn push(&mut self, element: T) { + self.inner.push(element); + } + + pub fn extend(&mut self, iter: impl IntoIterator<Item = T>) { + self.inner.extend(iter); + } + + pub fn as_mut(&mut self) -> RcVecMut<T> { + RcVecMut { + inner: &mut self.inner, + } + } + + pub fn build(self) -> RcVec<T> { + RcVec { + inner: Rc::new(self.inner), + } + } +} + +impl<'a, T> RcVecMut<'a, T> { + pub fn push(&mut self, element: T) { + self.inner.push(element); + } + + pub fn extend(&mut self, iter: impl IntoIterator<Item = T>) { + self.inner.extend(iter); + } + + pub fn pop(&mut self) -> Option<T> { + self.inner.pop() + } + + pub fn as_mut(&mut self) -> RcVecMut<T> { + RcVecMut { inner: self.inner } + } +} + +impl<T> Clone for RcVec<T> { + fn clone(&self) -> Self { + RcVec { + inner: Rc::clone(&self.inner), + } + } +} + +impl<T> IntoIterator for RcVecBuilder<T> { + type Item = T; + type IntoIter = RcVecIntoIter<T>; + + fn into_iter(self) -> Self::IntoIter { + RcVecIntoIter { + inner: self.inner.into_iter(), + } + } +} + +impl<T> Iterator for RcVecIntoIter<T> { + type Item = T; + + fn next(&mut self) -> Option<Self::Item> { + self.inner.next() + } + + fn size_hint(&self) -> (usize, Option<usize>) { + self.inner.size_hint() + } +} + +impl<T> RefUnwindSafe for RcVec<T> where T: RefUnwindSafe {} diff --git a/src/wrapper.rs b/src/wrapper.rs new file mode 100644 index 0000000..f5eb826 --- /dev/null +++ b/src/wrapper.rs @@ -0,0 +1,930 @@ +use crate::detection::inside_proc_macro; +#[cfg(span_locations)] +use crate::location::LineColumn; +use crate::{fallback, Delimiter, Punct, Spacing, TokenTree}; +use core::fmt::{self, Debug, Display}; +use core::ops::RangeBounds; +use core::str::FromStr; +use std::panic; +#[cfg(super_unstable)] +use std::path::PathBuf; + +#[derive(Clone)] +pub(crate) enum TokenStream { + Compiler(DeferredTokenStream), + Fallback(fallback::TokenStream), +} + +// Work around https://github.com/rust-lang/rust/issues/65080. +// In `impl Extend<TokenTree> for TokenStream` which is used heavily by quote, +// we hold on to the appended tokens and do proc_macro::TokenStream::extend as +// late as possible to batch together consecutive uses of the Extend impl. +#[derive(Clone)] +pub(crate) struct DeferredTokenStream { + stream: proc_macro::TokenStream, + extra: Vec<proc_macro::TokenTree>, +} + +pub(crate) enum LexError { + Compiler(proc_macro::LexError), + Fallback(fallback::LexError), + + // Rustc was supposed to return a LexError, but it panicked instead. + // https://github.com/rust-lang/rust/issues/58736 + CompilerPanic, +} + +#[cold] +fn mismatch(line: u32) -> ! { + #[cfg(procmacro2_backtrace)] + { + let backtrace = std::backtrace::Backtrace::force_capture(); + panic!("compiler/fallback mismatch #{}\n\n{}", line, backtrace) + } + #[cfg(not(procmacro2_backtrace))] + { + panic!("compiler/fallback mismatch #{}", line) + } +} + +impl DeferredTokenStream { + fn new(stream: proc_macro::TokenStream) -> Self { + DeferredTokenStream { + stream, + extra: Vec::new(), + } + } + + fn is_empty(&self) -> bool { + self.stream.is_empty() && self.extra.is_empty() + } + + fn evaluate_now(&mut self) { + // If-check provides a fast short circuit for the common case of `extra` + // being empty, which saves a round trip over the proc macro bridge. + // Improves macro expansion time in winrt by 6% in debug mode. + if !self.extra.is_empty() { + self.stream.extend(self.extra.drain(..)); + } + } + + fn into_token_stream(mut self) -> proc_macro::TokenStream { + self.evaluate_now(); + self.stream + } +} + +impl TokenStream { + pub fn new() -> Self { + if inside_proc_macro() { + TokenStream::Compiler(DeferredTokenStream::new(proc_macro::TokenStream::new())) + } else { + TokenStream::Fallback(fallback::TokenStream::new()) + } + } + + pub fn is_empty(&self) -> bool { + match self { + TokenStream::Compiler(tts) => tts.is_empty(), + TokenStream::Fallback(tts) => tts.is_empty(), + } + } + + fn unwrap_nightly(self) -> proc_macro::TokenStream { + match self { + TokenStream::Compiler(s) => s.into_token_stream(), + TokenStream::Fallback(_) => mismatch(line!()), + } + } + + fn unwrap_stable(self) -> fallback::TokenStream { + match self { + TokenStream::Compiler(_) => mismatch(line!()), + TokenStream::Fallback(s) => s, + } + } +} + +impl FromStr for TokenStream { + type Err = LexError; + + fn from_str(src: &str) -> Result<TokenStream, LexError> { + if inside_proc_macro() { + Ok(TokenStream::Compiler(DeferredTokenStream::new( + proc_macro_parse(src)?, + ))) + } else { + Ok(TokenStream::Fallback(src.parse()?)) + } + } +} + +// Work around https://github.com/rust-lang/rust/issues/58736. +fn proc_macro_parse(src: &str) -> Result<proc_macro::TokenStream, LexError> { + let result = panic::catch_unwind(|| src.parse().map_err(LexError::Compiler)); + result.unwrap_or_else(|_| Err(LexError::CompilerPanic)) +} + +impl Display for TokenStream { + fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result { + match self { + TokenStream::Compiler(tts) => Display::fmt(&tts.clone().into_token_stream(), f), + TokenStream::Fallback(tts) => Display::fmt(tts, f), + } + } +} + +impl From<proc_macro::TokenStream> for TokenStream { + fn from(inner: proc_macro::TokenStream) -> Self { + TokenStream::Compiler(DeferredTokenStream::new(inner)) + } +} + +impl From<TokenStream> for proc_macro::TokenStream { + fn from(inner: TokenStream) -> Self { + match inner { + TokenStream::Compiler(inner) => inner.into_token_stream(), + TokenStream::Fallback(inner) => inner.to_string().parse().unwrap(), + } + } +} + +impl From<fallback::TokenStream> for TokenStream { + fn from(inner: fallback::TokenStream) -> Self { + TokenStream::Fallback(inner) + } +} + +// Assumes inside_proc_macro(). +fn into_compiler_token(token: TokenTree) -> proc_macro::TokenTree { + match token { + TokenTree::Group(tt) => tt.inner.unwrap_nightly().into(), + TokenTree::Punct(tt) => { + let spacing = match tt.spacing() { + Spacing::Joint => proc_macro::Spacing::Joint, + Spacing::Alone => proc_macro::Spacing::Alone, + }; + let mut punct = proc_macro::Punct::new(tt.as_char(), spacing); + punct.set_span(tt.span().inner.unwrap_nightly()); + punct.into() + } + TokenTree::Ident(tt) => tt.inner.unwrap_nightly().into(), + TokenTree::Literal(tt) => tt.inner.unwrap_nightly().into(), + } +} + +impl From<TokenTree> for TokenStream { + fn from(token: TokenTree) -> Self { + if inside_proc_macro() { + TokenStream::Compiler(DeferredTokenStream::new(into_compiler_token(token).into())) + } else { + TokenStream::Fallback(token.into()) + } + } +} + +impl FromIterator<TokenTree> for TokenStream { + fn from_iter<I: IntoIterator<Item = TokenTree>>(trees: I) -> Self { + if inside_proc_macro() { + TokenStream::Compiler(DeferredTokenStream::new( + trees.into_iter().map(into_compiler_token).collect(), + )) + } else { + TokenStream::Fallback(trees.into_iter().collect()) + } + } +} + +impl FromIterator<TokenStream> for TokenStream { + fn from_iter<I: IntoIterator<Item = TokenStream>>(streams: I) -> Self { + let mut streams = streams.into_iter(); + match streams.next() { + Some(TokenStream::Compiler(mut first)) => { + first.evaluate_now(); + first.stream.extend(streams.map(|s| match s { + TokenStream::Compiler(s) => s.into_token_stream(), + TokenStream::Fallback(_) => mismatch(line!()), + })); + TokenStream::Compiler(first) + } + Some(TokenStream::Fallback(mut first)) => { + first.extend(streams.map(|s| match s { + TokenStream::Fallback(s) => s, + TokenStream::Compiler(_) => mismatch(line!()), + })); + TokenStream::Fallback(first) + } + None => TokenStream::new(), + } + } +} + +impl Extend<TokenTree> for TokenStream { + fn extend<I: IntoIterator<Item = TokenTree>>(&mut self, stream: I) { + match self { + TokenStream::Compiler(tts) => { + // Here is the reason for DeferredTokenStream. + for token in stream { + tts.extra.push(into_compiler_token(token)); + } + } + TokenStream::Fallback(tts) => tts.extend(stream), + } + } +} + +impl Extend<TokenStream> for TokenStream { + fn extend<I: IntoIterator<Item = TokenStream>>(&mut self, streams: I) { + match self { + TokenStream::Compiler(tts) => { + tts.evaluate_now(); + tts.stream + .extend(streams.into_iter().map(TokenStream::unwrap_nightly)); + } + TokenStream::Fallback(tts) => { + tts.extend(streams.into_iter().map(TokenStream::unwrap_stable)); + } + } + } +} + +impl Debug for TokenStream { + fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result { + match self { + TokenStream::Compiler(tts) => Debug::fmt(&tts.clone().into_token_stream(), f), + TokenStream::Fallback(tts) => Debug::fmt(tts, f), + } + } +} + +impl LexError { + pub(crate) fn span(&self) -> Span { + match self { + LexError::Compiler(_) | LexError::CompilerPanic => Span::call_site(), + LexError::Fallback(e) => Span::Fallback(e.span()), + } + } +} + +impl From<proc_macro::LexError> for LexError { + fn from(e: proc_macro::LexError) -> Self { + LexError::Compiler(e) + } +} + +impl From<fallback::LexError> for LexError { + fn from(e: fallback::LexError) -> Self { + LexError::Fallback(e) + } +} + +impl Debug for LexError { + fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result { + match self { + LexError::Compiler(e) => Debug::fmt(e, f), + LexError::Fallback(e) => Debug::fmt(e, f), + LexError::CompilerPanic => { + let fallback = fallback::LexError::call_site(); + Debug::fmt(&fallback, f) + } + } + } +} + +impl Display for LexError { + fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result { + match self { + LexError::Compiler(e) => Display::fmt(e, f), + LexError::Fallback(e) => Display::fmt(e, f), + LexError::CompilerPanic => { + let fallback = fallback::LexError::call_site(); + Display::fmt(&fallback, f) + } + } + } +} + +#[derive(Clone)] +pub(crate) enum TokenTreeIter { + Compiler(proc_macro::token_stream::IntoIter), + Fallback(fallback::TokenTreeIter), +} + +impl IntoIterator for TokenStream { + type Item = TokenTree; + type IntoIter = TokenTreeIter; + + fn into_iter(self) -> TokenTreeIter { + match self { + TokenStream::Compiler(tts) => { + TokenTreeIter::Compiler(tts.into_token_stream().into_iter()) + } + TokenStream::Fallback(tts) => TokenTreeIter::Fallback(tts.into_iter()), + } + } +} + +impl Iterator for TokenTreeIter { + type Item = TokenTree; + + fn next(&mut self) -> Option<TokenTree> { + let token = match self { + TokenTreeIter::Compiler(iter) => iter.next()?, + TokenTreeIter::Fallback(iter) => return iter.next(), + }; + Some(match token { + proc_macro::TokenTree::Group(tt) => crate::Group::_new(Group::Compiler(tt)).into(), + proc_macro::TokenTree::Punct(tt) => { + let spacing = match tt.spacing() { + proc_macro::Spacing::Joint => Spacing::Joint, + proc_macro::Spacing::Alone => Spacing::Alone, + }; + let mut o = Punct::new(tt.as_char(), spacing); + o.set_span(crate::Span::_new(Span::Compiler(tt.span()))); + o.into() + } + proc_macro::TokenTree::Ident(s) => crate::Ident::_new(Ident::Compiler(s)).into(), + proc_macro::TokenTree::Literal(l) => crate::Literal::_new(Literal::Compiler(l)).into(), + }) + } + + fn size_hint(&self) -> (usize, Option<usize>) { + match self { + TokenTreeIter::Compiler(tts) => tts.size_hint(), + TokenTreeIter::Fallback(tts) => tts.size_hint(), + } + } +} + +#[derive(Clone, PartialEq, Eq)] +#[cfg(super_unstable)] +pub(crate) enum SourceFile { + Compiler(proc_macro::SourceFile), + Fallback(fallback::SourceFile), +} + +#[cfg(super_unstable)] +impl SourceFile { + fn nightly(sf: proc_macro::SourceFile) -> Self { + SourceFile::Compiler(sf) + } + + /// Get the path to this source file as a string. + pub fn path(&self) -> PathBuf { + match self { + SourceFile::Compiler(a) => a.path(), + SourceFile::Fallback(a) => a.path(), + } + } + + pub fn is_real(&self) -> bool { + match self { + SourceFile::Compiler(a) => a.is_real(), + SourceFile::Fallback(a) => a.is_real(), + } + } +} + +#[cfg(super_unstable)] +impl Debug for SourceFile { + fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result { + match self { + SourceFile::Compiler(a) => Debug::fmt(a, f), + SourceFile::Fallback(a) => Debug::fmt(a, f), + } + } +} + +#[derive(Copy, Clone)] +pub(crate) enum Span { + Compiler(proc_macro::Span), + Fallback(fallback::Span), +} + +impl Span { + pub fn call_site() -> Self { + if inside_proc_macro() { + Span::Compiler(proc_macro::Span::call_site()) + } else { + Span::Fallback(fallback::Span::call_site()) + } + } + + pub fn mixed_site() -> Self { + if inside_proc_macro() { + Span::Compiler(proc_macro::Span::mixed_site()) + } else { + Span::Fallback(fallback::Span::mixed_site()) + } + } + + #[cfg(super_unstable)] + pub fn def_site() -> Self { + if inside_proc_macro() { + Span::Compiler(proc_macro::Span::def_site()) + } else { + Span::Fallback(fallback::Span::def_site()) + } + } + + pub fn resolved_at(&self, other: Span) -> Span { + match (self, other) { + (Span::Compiler(a), Span::Compiler(b)) => Span::Compiler(a.resolved_at(b)), + (Span::Fallback(a), Span::Fallback(b)) => Span::Fallback(a.resolved_at(b)), + (Span::Compiler(_), Span::Fallback(_)) => mismatch(line!()), + (Span::Fallback(_), Span::Compiler(_)) => mismatch(line!()), + } + } + + pub fn located_at(&self, other: Span) -> Span { + match (self, other) { + (Span::Compiler(a), Span::Compiler(b)) => Span::Compiler(a.located_at(b)), + (Span::Fallback(a), Span::Fallback(b)) => Span::Fallback(a.located_at(b)), + (Span::Compiler(_), Span::Fallback(_)) => mismatch(line!()), + (Span::Fallback(_), Span::Compiler(_)) => mismatch(line!()), + } + } + + pub fn unwrap(self) -> proc_macro::Span { + match self { + Span::Compiler(s) => s, + Span::Fallback(_) => panic!("proc_macro::Span is only available in procedural macros"), + } + } + + #[cfg(super_unstable)] + pub fn source_file(&self) -> SourceFile { + match self { + Span::Compiler(s) => SourceFile::nightly(s.source_file()), + Span::Fallback(s) => SourceFile::Fallback(s.source_file()), + } + } + + #[cfg(span_locations)] + pub fn start(&self) -> LineColumn { + match self { + Span::Compiler(_) => LineColumn { line: 0, column: 0 }, + Span::Fallback(s) => s.start(), + } + } + + #[cfg(span_locations)] + pub fn end(&self) -> LineColumn { + match self { + Span::Compiler(_) => LineColumn { line: 0, column: 0 }, + Span::Fallback(s) => s.end(), + } + } + + pub fn join(&self, other: Span) -> Option<Span> { + let ret = match (self, other) { + #[cfg(proc_macro_span)] + (Span::Compiler(a), Span::Compiler(b)) => Span::Compiler(a.join(b)?), + (Span::Fallback(a), Span::Fallback(b)) => Span::Fallback(a.join(b)?), + _ => return None, + }; + Some(ret) + } + + #[cfg(super_unstable)] + pub fn eq(&self, other: &Span) -> bool { + match (self, other) { + (Span::Compiler(a), Span::Compiler(b)) => a.eq(b), + (Span::Fallback(a), Span::Fallback(b)) => a.eq(b), + _ => false, + } + } + + pub fn source_text(&self) -> Option<String> { + match self { + #[cfg(not(no_source_text))] + Span::Compiler(s) => s.source_text(), + #[cfg(no_source_text)] + Span::Compiler(_) => None, + Span::Fallback(s) => s.source_text(), + } + } + + fn unwrap_nightly(self) -> proc_macro::Span { + match self { + Span::Compiler(s) => s, + Span::Fallback(_) => mismatch(line!()), + } + } +} + +impl From<proc_macro::Span> for crate::Span { + fn from(proc_span: proc_macro::Span) -> Self { + crate::Span::_new(Span::Compiler(proc_span)) + } +} + +impl From<fallback::Span> for Span { + fn from(inner: fallback::Span) -> Self { + Span::Fallback(inner) + } +} + +impl Debug for Span { + fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result { + match self { + Span::Compiler(s) => Debug::fmt(s, f), + Span::Fallback(s) => Debug::fmt(s, f), + } + } +} + +pub(crate) fn debug_span_field_if_nontrivial(debug: &mut fmt::DebugStruct, span: Span) { + match span { + Span::Compiler(s) => { + debug.field("span", &s); + } + Span::Fallback(s) => fallback::debug_span_field_if_nontrivial(debug, s), + } +} + +#[derive(Clone)] +pub(crate) enum Group { + Compiler(proc_macro::Group), + Fallback(fallback::Group), +} + +impl Group { + pub fn new(delimiter: Delimiter, stream: TokenStream) -> Self { + match stream { + TokenStream::Compiler(tts) => { + let delimiter = match delimiter { + Delimiter::Parenthesis => proc_macro::Delimiter::Parenthesis, + Delimiter::Bracket => proc_macro::Delimiter::Bracket, + Delimiter::Brace => proc_macro::Delimiter::Brace, + Delimiter::None => proc_macro::Delimiter::None, + }; + Group::Compiler(proc_macro::Group::new(delimiter, tts.into_token_stream())) + } + TokenStream::Fallback(stream) => { + Group::Fallback(fallback::Group::new(delimiter, stream)) + } + } + } + + pub fn delimiter(&self) -> Delimiter { + match self { + Group::Compiler(g) => match g.delimiter() { + proc_macro::Delimiter::Parenthesis => Delimiter::Parenthesis, + proc_macro::Delimiter::Bracket => Delimiter::Bracket, + proc_macro::Delimiter::Brace => Delimiter::Brace, + proc_macro::Delimiter::None => Delimiter::None, + }, + Group::Fallback(g) => g.delimiter(), + } + } + + pub fn stream(&self) -> TokenStream { + match self { + Group::Compiler(g) => TokenStream::Compiler(DeferredTokenStream::new(g.stream())), + Group::Fallback(g) => TokenStream::Fallback(g.stream()), + } + } + + pub fn span(&self) -> Span { + match self { + Group::Compiler(g) => Span::Compiler(g.span()), + Group::Fallback(g) => Span::Fallback(g.span()), + } + } + + pub fn span_open(&self) -> Span { + match self { + Group::Compiler(g) => Span::Compiler(g.span_open()), + Group::Fallback(g) => Span::Fallback(g.span_open()), + } + } + + pub fn span_close(&self) -> Span { + match self { + Group::Compiler(g) => Span::Compiler(g.span_close()), + Group::Fallback(g) => Span::Fallback(g.span_close()), + } + } + + pub fn set_span(&mut self, span: Span) { + match (self, span) { + (Group::Compiler(g), Span::Compiler(s)) => g.set_span(s), + (Group::Fallback(g), Span::Fallback(s)) => g.set_span(s), + (Group::Compiler(_), Span::Fallback(_)) => mismatch(line!()), + (Group::Fallback(_), Span::Compiler(_)) => mismatch(line!()), + } + } + + fn unwrap_nightly(self) -> proc_macro::Group { + match self { + Group::Compiler(g) => g, + Group::Fallback(_) => mismatch(line!()), + } + } +} + +impl From<fallback::Group> for Group { + fn from(g: fallback::Group) -> Self { + Group::Fallback(g) + } +} + +impl Display for Group { + fn fmt(&self, formatter: &mut fmt::Formatter) -> fmt::Result { + match self { + Group::Compiler(group) => Display::fmt(group, formatter), + Group::Fallback(group) => Display::fmt(group, formatter), + } + } +} + +impl Debug for Group { + fn fmt(&self, formatter: &mut fmt::Formatter) -> fmt::Result { + match self { + Group::Compiler(group) => Debug::fmt(group, formatter), + Group::Fallback(group) => Debug::fmt(group, formatter), + } + } +} + +#[derive(Clone)] +pub(crate) enum Ident { + Compiler(proc_macro::Ident), + Fallback(fallback::Ident), +} + +impl Ident { + #[track_caller] + pub fn new_checked(string: &str, span: Span) -> Self { + match span { + Span::Compiler(s) => Ident::Compiler(proc_macro::Ident::new(string, s)), + Span::Fallback(s) => Ident::Fallback(fallback::Ident::new_checked(string, s)), + } + } + + pub fn new_unchecked(string: &str, span: fallback::Span) -> Self { + Ident::Fallback(fallback::Ident::new_unchecked(string, span)) + } + + #[track_caller] + pub fn new_raw_checked(string: &str, span: Span) -> Self { + match span { + Span::Compiler(s) => Ident::Compiler(proc_macro::Ident::new_raw(string, s)), + Span::Fallback(s) => Ident::Fallback(fallback::Ident::new_raw_checked(string, s)), + } + } + + pub fn new_raw_unchecked(string: &str, span: fallback::Span) -> Self { + Ident::Fallback(fallback::Ident::new_raw_unchecked(string, span)) + } + + pub fn span(&self) -> Span { + match self { + Ident::Compiler(t) => Span::Compiler(t.span()), + Ident::Fallback(t) => Span::Fallback(t.span()), + } + } + + pub fn set_span(&mut self, span: Span) { + match (self, span) { + (Ident::Compiler(t), Span::Compiler(s)) => t.set_span(s), + (Ident::Fallback(t), Span::Fallback(s)) => t.set_span(s), + (Ident::Compiler(_), Span::Fallback(_)) => mismatch(line!()), + (Ident::Fallback(_), Span::Compiler(_)) => mismatch(line!()), + } + } + + fn unwrap_nightly(self) -> proc_macro::Ident { + match self { + Ident::Compiler(s) => s, + Ident::Fallback(_) => mismatch(line!()), + } + } +} + +impl PartialEq for Ident { + fn eq(&self, other: &Ident) -> bool { + match (self, other) { + (Ident::Compiler(t), Ident::Compiler(o)) => t.to_string() == o.to_string(), + (Ident::Fallback(t), Ident::Fallback(o)) => t == o, + (Ident::Compiler(_), Ident::Fallback(_)) => mismatch(line!()), + (Ident::Fallback(_), Ident::Compiler(_)) => mismatch(line!()), + } + } +} + +impl<T> PartialEq<T> for Ident +where + T: ?Sized + AsRef<str>, +{ + fn eq(&self, other: &T) -> bool { + let other = other.as_ref(); + match self { + Ident::Compiler(t) => t.to_string() == other, + Ident::Fallback(t) => t == other, + } + } +} + +impl Display for Ident { + fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result { + match self { + Ident::Compiler(t) => Display::fmt(t, f), + Ident::Fallback(t) => Display::fmt(t, f), + } + } +} + +impl Debug for Ident { + fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result { + match self { + Ident::Compiler(t) => Debug::fmt(t, f), + Ident::Fallback(t) => Debug::fmt(t, f), + } + } +} + +#[derive(Clone)] +pub(crate) enum Literal { + Compiler(proc_macro::Literal), + Fallback(fallback::Literal), +} + +macro_rules! suffixed_numbers { + ($($name:ident => $kind:ident,)*) => ($( + pub fn $name(n: $kind) -> Literal { + if inside_proc_macro() { + Literal::Compiler(proc_macro::Literal::$name(n)) + } else { + Literal::Fallback(fallback::Literal::$name(n)) + } + } + )*) +} + +macro_rules! unsuffixed_integers { + ($($name:ident => $kind:ident,)*) => ($( + pub fn $name(n: $kind) -> Literal { + if inside_proc_macro() { + Literal::Compiler(proc_macro::Literal::$name(n)) + } else { + Literal::Fallback(fallback::Literal::$name(n)) + } + } + )*) +} + +impl Literal { + pub unsafe fn from_str_unchecked(repr: &str) -> Self { + if inside_proc_macro() { + Literal::Compiler(proc_macro::Literal::from_str(repr).expect("invalid literal")) + } else { + Literal::Fallback(unsafe { fallback::Literal::from_str_unchecked(repr) }) + } + } + + suffixed_numbers! { + u8_suffixed => u8, + u16_suffixed => u16, + u32_suffixed => u32, + u64_suffixed => u64, + u128_suffixed => u128, + usize_suffixed => usize, + i8_suffixed => i8, + i16_suffixed => i16, + i32_suffixed => i32, + i64_suffixed => i64, + i128_suffixed => i128, + isize_suffixed => isize, + + f32_suffixed => f32, + f64_suffixed => f64, + } + + unsuffixed_integers! { + u8_unsuffixed => u8, + u16_unsuffixed => u16, + u32_unsuffixed => u32, + u64_unsuffixed => u64, + u128_unsuffixed => u128, + usize_unsuffixed => usize, + i8_unsuffixed => i8, + i16_unsuffixed => i16, + i32_unsuffixed => i32, + i64_unsuffixed => i64, + i128_unsuffixed => i128, + isize_unsuffixed => isize, + } + + pub fn f32_unsuffixed(f: f32) -> Literal { + if inside_proc_macro() { + Literal::Compiler(proc_macro::Literal::f32_unsuffixed(f)) + } else { + Literal::Fallback(fallback::Literal::f32_unsuffixed(f)) + } + } + + pub fn f64_unsuffixed(f: f64) -> Literal { + if inside_proc_macro() { + Literal::Compiler(proc_macro::Literal::f64_unsuffixed(f)) + } else { + Literal::Fallback(fallback::Literal::f64_unsuffixed(f)) + } + } + + pub fn string(t: &str) -> Literal { + if inside_proc_macro() { + Literal::Compiler(proc_macro::Literal::string(t)) + } else { + Literal::Fallback(fallback::Literal::string(t)) + } + } + + pub fn character(t: char) -> Literal { + if inside_proc_macro() { + Literal::Compiler(proc_macro::Literal::character(t)) + } else { + Literal::Fallback(fallback::Literal::character(t)) + } + } + + pub fn byte_string(bytes: &[u8]) -> Literal { + if inside_proc_macro() { + Literal::Compiler(proc_macro::Literal::byte_string(bytes)) + } else { + Literal::Fallback(fallback::Literal::byte_string(bytes)) + } + } + + pub fn span(&self) -> Span { + match self { + Literal::Compiler(lit) => Span::Compiler(lit.span()), + Literal::Fallback(lit) => Span::Fallback(lit.span()), + } + } + + pub fn set_span(&mut self, span: Span) { + match (self, span) { + (Literal::Compiler(lit), Span::Compiler(s)) => lit.set_span(s), + (Literal::Fallback(lit), Span::Fallback(s)) => lit.set_span(s), + (Literal::Compiler(_), Span::Fallback(_)) => mismatch(line!()), + (Literal::Fallback(_), Span::Compiler(_)) => mismatch(line!()), + } + } + + pub fn subspan<R: RangeBounds<usize>>(&self, range: R) -> Option<Span> { + match self { + #[cfg(proc_macro_span)] + Literal::Compiler(lit) => lit.subspan(range).map(Span::Compiler), + #[cfg(not(proc_macro_span))] + Literal::Compiler(_lit) => None, + Literal::Fallback(lit) => lit.subspan(range).map(Span::Fallback), + } + } + + fn unwrap_nightly(self) -> proc_macro::Literal { + match self { + Literal::Compiler(s) => s, + Literal::Fallback(_) => mismatch(line!()), + } + } +} + +impl From<fallback::Literal> for Literal { + fn from(s: fallback::Literal) -> Self { + Literal::Fallback(s) + } +} + +impl FromStr for Literal { + type Err = LexError; + + fn from_str(repr: &str) -> Result<Self, Self::Err> { + if inside_proc_macro() { + let literal = proc_macro::Literal::from_str(repr)?; + Ok(Literal::Compiler(literal)) + } else { + let literal = fallback::Literal::from_str(repr)?; + Ok(Literal::Fallback(literal)) + } + } +} + +impl Display for Literal { + fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result { + match self { + Literal::Compiler(t) => Display::fmt(t, f), + Literal::Fallback(t) => Display::fmt(t, f), + } + } +} + +impl Debug for Literal { + fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result { + match self { + Literal::Compiler(t) => Debug::fmt(t, f), + Literal::Fallback(t) => Debug::fmt(t, f), + } + } +} diff --git a/tests/comments.rs b/tests/comments.rs new file mode 100644 index 0000000..4f7236d --- /dev/null +++ b/tests/comments.rs @@ -0,0 +1,105 @@ +#![allow(clippy::assertions_on_result_states)] + +use proc_macro2::{Delimiter, Literal, Spacing, TokenStream, TokenTree}; + +// #[doc = "..."] -> "..." +fn lit_of_outer_doc_comment(tokens: &TokenStream) -> Literal { + lit_of_doc_comment(tokens, false) +} + +// #![doc = "..."] -> "..." +fn lit_of_inner_doc_comment(tokens: &TokenStream) -> Literal { + lit_of_doc_comment(tokens, true) +} + +fn lit_of_doc_comment(tokens: &TokenStream, inner: bool) -> Literal { + let mut iter = tokens.clone().into_iter(); + match iter.next().unwrap() { + TokenTree::Punct(punct) => { + assert_eq!(punct.as_char(), '#'); + assert_eq!(punct.spacing(), Spacing::Alone); + } + _ => panic!("wrong token {:?}", tokens), + } + if inner { + match iter.next().unwrap() { + TokenTree::Punct(punct) => { + assert_eq!(punct.as_char(), '!'); + assert_eq!(punct.spacing(), Spacing::Alone); + } + _ => panic!("wrong token {:?}", tokens), + } + } + iter = match iter.next().unwrap() { + TokenTree::Group(group) => { + assert_eq!(group.delimiter(), Delimiter::Bracket); + assert!(iter.next().is_none(), "unexpected token {:?}", tokens); + group.stream().into_iter() + } + _ => panic!("wrong token {:?}", tokens), + }; + match iter.next().unwrap() { + TokenTree::Ident(ident) => assert_eq!(ident.to_string(), "doc"), + _ => panic!("wrong token {:?}", tokens), + } + match iter.next().unwrap() { + TokenTree::Punct(punct) => { + assert_eq!(punct.as_char(), '='); + assert_eq!(punct.spacing(), Spacing::Alone); + } + _ => panic!("wrong token {:?}", tokens), + } + match iter.next().unwrap() { + TokenTree::Literal(literal) => { + assert!(iter.next().is_none(), "unexpected token {:?}", tokens); + literal + } + _ => panic!("wrong token {:?}", tokens), + } +} + +#[test] +fn closed_immediately() { + let stream = "/**/".parse::<TokenStream>().unwrap(); + let tokens = stream.into_iter().collect::<Vec<_>>(); + assert!(tokens.is_empty(), "not empty -- {:?}", tokens); +} + +#[test] +fn incomplete() { + assert!("/*/".parse::<TokenStream>().is_err()); +} + +#[test] +fn lit() { + let stream = "/// doc".parse::<TokenStream>().unwrap(); + let lit = lit_of_outer_doc_comment(&stream); + assert_eq!(lit.to_string(), "\" doc\""); + + let stream = "//! doc".parse::<TokenStream>().unwrap(); + let lit = lit_of_inner_doc_comment(&stream); + assert_eq!(lit.to_string(), "\" doc\""); + + let stream = "/** doc */".parse::<TokenStream>().unwrap(); + let lit = lit_of_outer_doc_comment(&stream); + assert_eq!(lit.to_string(), "\" doc \""); + + let stream = "/*! doc */".parse::<TokenStream>().unwrap(); + let lit = lit_of_inner_doc_comment(&stream); + assert_eq!(lit.to_string(), "\" doc \""); +} + +#[test] +fn carriage_return() { + let stream = "///\r\n".parse::<TokenStream>().unwrap(); + let lit = lit_of_outer_doc_comment(&stream); + assert_eq!(lit.to_string(), "\"\""); + + let stream = "/**\r\n*/".parse::<TokenStream>().unwrap(); + let lit = lit_of_outer_doc_comment(&stream); + assert_eq!(lit.to_string(), "\"\\r\\n\""); + + "///\r".parse::<TokenStream>().unwrap_err(); + "///\r \n".parse::<TokenStream>().unwrap_err(); + "/**\r \n*/".parse::<TokenStream>().unwrap_err(); +} diff --git a/tests/features.rs b/tests/features.rs new file mode 100644 index 0000000..073f6e6 --- /dev/null +++ b/tests/features.rs @@ -0,0 +1,8 @@ +#[test] +#[ignore] +fn make_sure_no_proc_macro() { + assert!( + !cfg!(feature = "proc-macro"), + "still compiled with proc_macro?" + ); +} diff --git a/tests/marker.rs b/tests/marker.rs new file mode 100644 index 0000000..d08fbfc --- /dev/null +++ b/tests/marker.rs @@ -0,0 +1,99 @@ +#![allow(clippy::extra_unused_type_parameters)] + +use proc_macro2::{ + Delimiter, Group, Ident, LexError, Literal, Punct, Spacing, Span, TokenStream, TokenTree, +}; + +macro_rules! assert_impl { + ($ty:ident is $($marker:ident) and +) => { + #[test] + #[allow(non_snake_case)] + fn $ty() { + fn assert_implemented<T: $($marker +)+>() {} + assert_implemented::<$ty>(); + } + }; + + ($ty:ident is not $($marker:ident) or +) => { + #[test] + #[allow(non_snake_case)] + fn $ty() { + $( + { + // Implemented for types that implement $marker. + trait IsNotImplemented { + fn assert_not_implemented() {} + } + impl<T: $marker> IsNotImplemented for T {} + + // Implemented for the type being tested. + trait IsImplemented { + fn assert_not_implemented() {} + } + impl IsImplemented for $ty {} + + // If $ty does not implement $marker, there is no ambiguity + // in the following trait method call. + <$ty>::assert_not_implemented(); + } + )+ + } + }; +} + +assert_impl!(Delimiter is Send and Sync); +assert_impl!(Spacing is Send and Sync); + +assert_impl!(Group is not Send or Sync); +assert_impl!(Ident is not Send or Sync); +assert_impl!(LexError is not Send or Sync); +assert_impl!(Literal is not Send or Sync); +assert_impl!(Punct is not Send or Sync); +assert_impl!(Span is not Send or Sync); +assert_impl!(TokenStream is not Send or Sync); +assert_impl!(TokenTree is not Send or Sync); + +#[cfg(procmacro2_semver_exempt)] +mod semver_exempt { + use proc_macro2::{LineColumn, SourceFile}; + + assert_impl!(LineColumn is Send and Sync); + + assert_impl!(SourceFile is not Send or Sync); +} + +mod unwind_safe { + use proc_macro2::{ + Delimiter, Group, Ident, LexError, Literal, Punct, Spacing, Span, TokenStream, TokenTree, + }; + #[cfg(procmacro2_semver_exempt)] + use proc_macro2::{LineColumn, SourceFile}; + use std::panic::{RefUnwindSafe, UnwindSafe}; + + macro_rules! assert_unwind_safe { + ($($types:ident)*) => { + $( + assert_impl!($types is UnwindSafe and RefUnwindSafe); + )* + }; + } + + assert_unwind_safe! { + Delimiter + Group + Ident + LexError + Literal + Punct + Spacing + Span + TokenStream + TokenTree + } + + #[cfg(procmacro2_semver_exempt)] + assert_unwind_safe! { + LineColumn + SourceFile + } +} diff --git a/tests/test.rs b/tests/test.rs new file mode 100644 index 0000000..b75cd55 --- /dev/null +++ b/tests/test.rs @@ -0,0 +1,759 @@ +#![allow( + clippy::assertions_on_result_states, + clippy::items_after_statements, + clippy::non_ascii_literal, + clippy::octal_escapes +)] + +use proc_macro2::{Ident, Literal, Punct, Spacing, Span, TokenStream, TokenTree}; +use std::iter; +use std::str::{self, FromStr}; + +#[test] +fn idents() { + assert_eq!( + Ident::new("String", Span::call_site()).to_string(), + "String" + ); + assert_eq!(Ident::new("fn", Span::call_site()).to_string(), "fn"); + assert_eq!(Ident::new("_", Span::call_site()).to_string(), "_"); +} + +#[test] +fn raw_idents() { + assert_eq!( + Ident::new_raw("String", Span::call_site()).to_string(), + "r#String" + ); + assert_eq!(Ident::new_raw("fn", Span::call_site()).to_string(), "r#fn"); +} + +#[test] +#[should_panic(expected = "`r#_` cannot be a raw identifier")] +fn ident_raw_underscore() { + Ident::new_raw("_", Span::call_site()); +} + +#[test] +#[should_panic(expected = "`r#super` cannot be a raw identifier")] +fn ident_raw_reserved() { + Ident::new_raw("super", Span::call_site()); +} + +#[test] +#[should_panic(expected = "Ident is not allowed to be empty; use Option<Ident>")] +fn ident_empty() { + Ident::new("", Span::call_site()); +} + +#[test] +#[should_panic(expected = "Ident cannot be a number; use Literal instead")] +fn ident_number() { + Ident::new("255", Span::call_site()); +} + +#[test] +#[should_panic(expected = "\"a#\" is not a valid Ident")] +fn ident_invalid() { + Ident::new("a#", Span::call_site()); +} + +#[test] +#[should_panic(expected = "not a valid Ident")] +fn raw_ident_empty() { + Ident::new("r#", Span::call_site()); +} + +#[test] +#[should_panic(expected = "not a valid Ident")] +fn raw_ident_number() { + Ident::new("r#255", Span::call_site()); +} + +#[test] +#[should_panic(expected = "\"r#a#\" is not a valid Ident")] +fn raw_ident_invalid() { + Ident::new("r#a#", Span::call_site()); +} + +#[test] +#[should_panic(expected = "not a valid Ident")] +fn lifetime_empty() { + Ident::new("'", Span::call_site()); +} + +#[test] +#[should_panic(expected = "not a valid Ident")] +fn lifetime_number() { + Ident::new("'255", Span::call_site()); +} + +#[test] +#[should_panic(expected = r#""'a#" is not a valid Ident"#)] +fn lifetime_invalid() { + Ident::new("'a#", Span::call_site()); +} + +#[test] +fn literal_string() { + assert_eq!(Literal::string("foo").to_string(), "\"foo\""); + assert_eq!(Literal::string("\"").to_string(), "\"\\\"\""); + assert_eq!(Literal::string("didn't").to_string(), "\"didn't\""); + assert_eq!( + Literal::string("a\00b\07c\08d\0e\0").to_string(), + "\"a\\x000b\\x007c\\08d\\0e\\0\"", + ); + + "\"\\\r\n x\"".parse::<TokenStream>().unwrap(); + "\"\\\r\n \rx\"".parse::<TokenStream>().unwrap_err(); +} + +#[test] +fn literal_raw_string() { + "r\"\r\n\"".parse::<TokenStream>().unwrap(); + + fn raw_string_literal_with_hashes(n: usize) -> String { + let mut literal = String::new(); + literal.push('r'); + literal.extend(iter::repeat('#').take(n)); + literal.push('"'); + literal.push('"'); + literal.extend(iter::repeat('#').take(n)); + literal + } + + raw_string_literal_with_hashes(255) + .parse::<TokenStream>() + .unwrap(); + + // https://github.com/rust-lang/rust/pull/95251 + raw_string_literal_with_hashes(256) + .parse::<TokenStream>() + .unwrap_err(); +} + +#[test] +fn literal_byte_string() { + assert_eq!(Literal::byte_string(b"").to_string(), "b\"\""); + assert_eq!( + Literal::byte_string(b"\0\t\n\r\"\\2\x10").to_string(), + "b\"\\0\\t\\n\\r\\\"\\\\2\\x10\"", + ); + assert_eq!( + Literal::byte_string(b"a\00b\07c\08d\0e\0").to_string(), + "b\"a\\x000b\\x007c\\08d\\0e\\0\"", + ); + + "b\"\\\r\n x\"".parse::<TokenStream>().unwrap(); + "b\"\\\r\n \rx\"".parse::<TokenStream>().unwrap_err(); + "b\"\\\r\n \u{a0}x\"".parse::<TokenStream>().unwrap_err(); + "br\"\u{a0}\"".parse::<TokenStream>().unwrap_err(); +} + +#[test] +fn literal_c_string() { + let strings = r###" + c"hello\x80我叫\u{1F980}" // from the RFC + cr"\" + cr##"Hello "world"!"## + c"\t\n\r\"\\" + "###; + + let mut tokens = strings.parse::<TokenStream>().unwrap().into_iter(); + + for expected in &[ + r#"c"hello\x80我叫\u{1F980}""#, + r#"cr"\""#, + r###"cr##"Hello "world"!"##"###, + r#"c"\t\n\r\"\\""#, + ] { + match tokens.next().unwrap() { + TokenTree::Literal(literal) => { + assert_eq!(literal.to_string(), *expected); + } + unexpected => panic!("unexpected token: {:?}", unexpected), + } + } + + if let Some(unexpected) = tokens.next() { + panic!("unexpected token: {:?}", unexpected); + } + + for invalid in &[r#"c"\0""#, r#"c"\x00""#, r#"c"\u{0}""#, "c\"\0\""] { + if let Ok(unexpected) = invalid.parse::<TokenStream>() { + panic!("unexpected token: {:?}", unexpected); + } + } +} + +#[test] +fn literal_character() { + assert_eq!(Literal::character('x').to_string(), "'x'"); + assert_eq!(Literal::character('\'').to_string(), "'\\''"); + assert_eq!(Literal::character('"').to_string(), "'\"'"); +} + +#[test] +fn literal_integer() { + assert_eq!(Literal::u8_suffixed(10).to_string(), "10u8"); + assert_eq!(Literal::u16_suffixed(10).to_string(), "10u16"); + assert_eq!(Literal::u32_suffixed(10).to_string(), "10u32"); + assert_eq!(Literal::u64_suffixed(10).to_string(), "10u64"); + assert_eq!(Literal::u128_suffixed(10).to_string(), "10u128"); + assert_eq!(Literal::usize_suffixed(10).to_string(), "10usize"); + + assert_eq!(Literal::i8_suffixed(10).to_string(), "10i8"); + assert_eq!(Literal::i16_suffixed(10).to_string(), "10i16"); + assert_eq!(Literal::i32_suffixed(10).to_string(), "10i32"); + assert_eq!(Literal::i64_suffixed(10).to_string(), "10i64"); + assert_eq!(Literal::i128_suffixed(10).to_string(), "10i128"); + assert_eq!(Literal::isize_suffixed(10).to_string(), "10isize"); + + assert_eq!(Literal::u8_unsuffixed(10).to_string(), "10"); + assert_eq!(Literal::u16_unsuffixed(10).to_string(), "10"); + assert_eq!(Literal::u32_unsuffixed(10).to_string(), "10"); + assert_eq!(Literal::u64_unsuffixed(10).to_string(), "10"); + assert_eq!(Literal::u128_unsuffixed(10).to_string(), "10"); + assert_eq!(Literal::usize_unsuffixed(10).to_string(), "10"); + + assert_eq!(Literal::i8_unsuffixed(10).to_string(), "10"); + assert_eq!(Literal::i16_unsuffixed(10).to_string(), "10"); + assert_eq!(Literal::i32_unsuffixed(10).to_string(), "10"); + assert_eq!(Literal::i64_unsuffixed(10).to_string(), "10"); + assert_eq!(Literal::i128_unsuffixed(10).to_string(), "10"); + assert_eq!(Literal::isize_unsuffixed(10).to_string(), "10"); +} + +#[test] +fn literal_float() { + assert_eq!(Literal::f32_suffixed(10.0).to_string(), "10f32"); + assert_eq!(Literal::f64_suffixed(10.0).to_string(), "10f64"); + + assert_eq!(Literal::f32_unsuffixed(10.0).to_string(), "10.0"); + assert_eq!(Literal::f64_unsuffixed(10.0).to_string(), "10.0"); +} + +#[test] +fn literal_suffix() { + fn token_count(p: &str) -> usize { + p.parse::<TokenStream>().unwrap().into_iter().count() + } + + assert_eq!(token_count("999u256"), 1); + assert_eq!(token_count("999r#u256"), 3); + assert_eq!(token_count("1."), 1); + assert_eq!(token_count("1.f32"), 3); + assert_eq!(token_count("1.0_0"), 1); + assert_eq!(token_count("1._0"), 3); + assert_eq!(token_count("1._m"), 3); + assert_eq!(token_count("\"\"s"), 1); + assert_eq!(token_count("r\"\"r"), 1); + assert_eq!(token_count("b\"\"b"), 1); + assert_eq!(token_count("br\"\"br"), 1); + assert_eq!(token_count("r#\"\"#r"), 1); + assert_eq!(token_count("'c'c"), 1); + assert_eq!(token_count("b'b'b"), 1); + assert_eq!(token_count("0E"), 1); + assert_eq!(token_count("0o0A"), 1); + assert_eq!(token_count("0E--0"), 4); + assert_eq!(token_count("0.0ECMA"), 1); +} + +#[test] +fn literal_iter_negative() { + let negative_literal = Literal::i32_suffixed(-3); + let tokens = TokenStream::from(TokenTree::Literal(negative_literal)); + let mut iter = tokens.into_iter(); + match iter.next().unwrap() { + TokenTree::Punct(punct) => { + assert_eq!(punct.as_char(), '-'); + assert_eq!(punct.spacing(), Spacing::Alone); + } + unexpected => panic!("unexpected token {:?}", unexpected), + } + match iter.next().unwrap() { + TokenTree::Literal(literal) => { + assert_eq!(literal.to_string(), "3i32"); + } + unexpected => panic!("unexpected token {:?}", unexpected), + } + assert!(iter.next().is_none()); +} + +#[test] +fn literal_parse() { + assert!("1".parse::<Literal>().is_ok()); + assert!("-1".parse::<Literal>().is_ok()); + assert!("-1u12".parse::<Literal>().is_ok()); + assert!("1.0".parse::<Literal>().is_ok()); + assert!("-1.0".parse::<Literal>().is_ok()); + assert!("-1.0f12".parse::<Literal>().is_ok()); + assert!("'a'".parse::<Literal>().is_ok()); + assert!("\"\n\"".parse::<Literal>().is_ok()); + assert!("0 1".parse::<Literal>().is_err()); + assert!(" 0".parse::<Literal>().is_err()); + assert!("0 ".parse::<Literal>().is_err()); + assert!("/* comment */0".parse::<Literal>().is_err()); + assert!("0/* comment */".parse::<Literal>().is_err()); + assert!("0// comment".parse::<Literal>().is_err()); + assert!("- 1".parse::<Literal>().is_err()); + assert!("- 1.0".parse::<Literal>().is_err()); + assert!("-\"\"".parse::<Literal>().is_err()); +} + +#[test] +fn literal_span() { + let positive = "0.1".parse::<Literal>().unwrap(); + let negative = "-0.1".parse::<Literal>().unwrap(); + let subspan = positive.subspan(1..2); + + #[cfg(not(span_locations))] + { + let _ = negative; + assert!(subspan.is_none()); + } + + #[cfg(span_locations)] + { + assert_eq!(positive.span().start().column, 0); + assert_eq!(positive.span().end().column, 3); + assert_eq!(negative.span().start().column, 0); + assert_eq!(negative.span().end().column, 4); + assert_eq!(subspan.unwrap().source_text().unwrap(), "."); + } + + assert!(positive.subspan(1..4).is_none()); +} + +#[cfg(span_locations)] +#[test] +fn source_text() { + let input = " 𓀕 a z "; + let mut tokens = input + .parse::<proc_macro2::TokenStream>() + .unwrap() + .into_iter(); + + let first = tokens.next().unwrap(); + assert_eq!("𓀕", first.span().source_text().unwrap()); + + let second = tokens.next().unwrap(); + let third = tokens.next().unwrap(); + assert_eq!("z", third.span().source_text().unwrap()); + assert_eq!("a", second.span().source_text().unwrap()); +} + +#[test] +fn roundtrip() { + fn roundtrip(p: &str) { + println!("parse: {}", p); + let s = p.parse::<TokenStream>().unwrap().to_string(); + println!("first: {}", s); + let s2 = s.parse::<TokenStream>().unwrap().to_string(); + assert_eq!(s, s2); + } + roundtrip("a"); + roundtrip("<<"); + roundtrip("<<="); + roundtrip( + " + 1 + 1.0 + 1f32 + 2f64 + 1usize + 4isize + 4e10 + 1_000 + 1_0i32 + 8u8 + 9 + 0 + 0xffffffffffffffffffffffffffffffff + 1x + 1u80 + 1f320 + ", + ); + roundtrip("'a"); + roundtrip("'_"); + roundtrip("'static"); + roundtrip("'\\u{10__FFFF}'"); + roundtrip("\"\\u{10_F0FF__}foo\\u{1_0_0_0__}\""); +} + +#[test] +fn fail() { + fn fail(p: &str) { + if let Ok(s) = p.parse::<TokenStream>() { + panic!("should have failed to parse: {}\n{:#?}", p, s); + } + } + fail("' static"); + fail("r#1"); + fail("r#_"); + fail("\"\\u{0000000}\""); // overlong unicode escape (rust allows at most 6 hex digits) + fail("\"\\u{999999}\""); // outside of valid range of char + fail("\"\\u{_0}\""); // leading underscore + fail("\"\\u{}\""); // empty + fail("b\"\r\""); // bare carriage return in byte string + fail("r\"\r\""); // bare carriage return in raw string + fail("\"\\\r \""); // backslash carriage return + fail("'aa'aa"); + fail("br##\"\"#"); + fail("\"\\\n\u{85}\r\""); +} + +#[cfg(span_locations)] +#[test] +fn span_test() { + check_spans( + "\ +/// This is a document comment +testing 123 +{ + testing 234 +}", + &[ + (1, 0, 1, 30), // # + (1, 0, 1, 30), // [ ... ] + (1, 0, 1, 30), // doc + (1, 0, 1, 30), // = + (1, 0, 1, 30), // "This is..." + (2, 0, 2, 7), // testing + (2, 8, 2, 11), // 123 + (3, 0, 5, 1), // { ... } + (4, 2, 4, 9), // testing + (4, 10, 4, 13), // 234 + ], + ); +} + +#[cfg(procmacro2_semver_exempt)] +#[cfg(not(nightly))] +#[test] +fn default_span() { + let start = Span::call_site().start(); + assert_eq!(start.line, 1); + assert_eq!(start.column, 0); + let end = Span::call_site().end(); + assert_eq!(end.line, 1); + assert_eq!(end.column, 0); + let source_file = Span::call_site().source_file(); + assert_eq!(source_file.path().to_string_lossy(), "<unspecified>"); + assert!(!source_file.is_real()); +} + +#[cfg(procmacro2_semver_exempt)] +#[test] +fn span_join() { + let source1 = "aaa\nbbb" + .parse::<TokenStream>() + .unwrap() + .into_iter() + .collect::<Vec<_>>(); + let source2 = "ccc\nddd" + .parse::<TokenStream>() + .unwrap() + .into_iter() + .collect::<Vec<_>>(); + + assert!(source1[0].span().source_file() != source2[0].span().source_file()); + assert_eq!( + source1[0].span().source_file(), + source1[1].span().source_file() + ); + + let joined1 = source1[0].span().join(source1[1].span()); + let joined2 = source1[0].span().join(source2[0].span()); + assert!(joined1.is_some()); + assert!(joined2.is_none()); + + let start = joined1.unwrap().start(); + let end = joined1.unwrap().end(); + assert_eq!(start.line, 1); + assert_eq!(start.column, 0); + assert_eq!(end.line, 2); + assert_eq!(end.column, 3); + + assert_eq!( + joined1.unwrap().source_file(), + source1[0].span().source_file() + ); +} + +#[test] +fn no_panic() { + let s = str::from_utf8(b"b\'\xc2\x86 \x00\x00\x00^\"").unwrap(); + assert!(s.parse::<TokenStream>().is_err()); +} + +#[test] +fn punct_before_comment() { + let mut tts = TokenStream::from_str("~// comment").unwrap().into_iter(); + match tts.next().unwrap() { + TokenTree::Punct(tt) => { + assert_eq!(tt.as_char(), '~'); + assert_eq!(tt.spacing(), Spacing::Alone); + } + wrong => panic!("wrong token {:?}", wrong), + } +} + +#[test] +fn joint_last_token() { + // This test verifies that we match the behavior of libproc_macro *not* in + // the range nightly-2020-09-06 through nightly-2020-09-10, in which this + // behavior was temporarily broken. + // See https://github.com/rust-lang/rust/issues/76399 + + let joint_punct = Punct::new(':', Spacing::Joint); + let stream = TokenStream::from(TokenTree::Punct(joint_punct)); + let punct = match stream.into_iter().next().unwrap() { + TokenTree::Punct(punct) => punct, + _ => unreachable!(), + }; + assert_eq!(punct.spacing(), Spacing::Joint); +} + +#[test] +fn raw_identifier() { + let mut tts = TokenStream::from_str("r#dyn").unwrap().into_iter(); + match tts.next().unwrap() { + TokenTree::Ident(raw) => assert_eq!("r#dyn", raw.to_string()), + wrong => panic!("wrong token {:?}", wrong), + } + assert!(tts.next().is_none()); +} + +#[test] +fn test_debug_ident() { + let ident = Ident::new("proc_macro", Span::call_site()); + + #[cfg(not(span_locations))] + let expected = "Ident(proc_macro)"; + + #[cfg(span_locations)] + let expected = "Ident { sym: proc_macro }"; + + assert_eq!(expected, format!("{:?}", ident)); +} + +#[test] +fn test_debug_tokenstream() { + let tts = TokenStream::from_str("[a + 1]").unwrap(); + + #[cfg(not(span_locations))] + let expected = "\ +TokenStream [ + Group { + delimiter: Bracket, + stream: TokenStream [ + Ident { + sym: a, + }, + Punct { + char: '+', + spacing: Alone, + }, + Literal { + lit: 1, + }, + ], + }, +]\ + "; + + #[cfg(not(span_locations))] + let expected_before_trailing_commas = "\ +TokenStream [ + Group { + delimiter: Bracket, + stream: TokenStream [ + Ident { + sym: a + }, + Punct { + char: '+', + spacing: Alone + }, + Literal { + lit: 1 + } + ] + } +]\ + "; + + #[cfg(span_locations)] + let expected = "\ +TokenStream [ + Group { + delimiter: Bracket, + stream: TokenStream [ + Ident { + sym: a, + span: bytes(2..3), + }, + Punct { + char: '+', + spacing: Alone, + span: bytes(4..5), + }, + Literal { + lit: 1, + span: bytes(6..7), + }, + ], + span: bytes(1..8), + }, +]\ + "; + + #[cfg(span_locations)] + let expected_before_trailing_commas = "\ +TokenStream [ + Group { + delimiter: Bracket, + stream: TokenStream [ + Ident { + sym: a, + span: bytes(2..3) + }, + Punct { + char: '+', + spacing: Alone, + span: bytes(4..5) + }, + Literal { + lit: 1, + span: bytes(6..7) + } + ], + span: bytes(1..8) + } +]\ + "; + + let actual = format!("{:#?}", tts); + if actual.ends_with(",\n]") { + assert_eq!(expected, actual); + } else { + assert_eq!(expected_before_trailing_commas, actual); + } +} + +#[test] +fn default_tokenstream_is_empty() { + let default_token_stream = <TokenStream as Default>::default(); + + assert!(default_token_stream.is_empty()); +} + +#[test] +fn tokenstream_size_hint() { + let tokens = "a b (c d) e".parse::<TokenStream>().unwrap(); + + assert_eq!(tokens.into_iter().size_hint(), (4, Some(4))); +} + +#[test] +fn tuple_indexing() { + // This behavior may change depending on https://github.com/rust-lang/rust/pull/71322 + let mut tokens = "tuple.0.0".parse::<TokenStream>().unwrap().into_iter(); + assert_eq!("tuple", tokens.next().unwrap().to_string()); + assert_eq!(".", tokens.next().unwrap().to_string()); + assert_eq!("0.0", tokens.next().unwrap().to_string()); + assert!(tokens.next().is_none()); +} + +#[cfg(span_locations)] +#[test] +fn non_ascii_tokens() { + check_spans("// abc", &[]); + check_spans("// ábc", &[]); + check_spans("// abc x", &[]); + check_spans("// ábc x", &[]); + check_spans("/* abc */ x", &[(1, 10, 1, 11)]); + check_spans("/* ábc */ x", &[(1, 10, 1, 11)]); + check_spans("/* ab\nc */ x", &[(2, 5, 2, 6)]); + check_spans("/* áb\nc */ x", &[(2, 5, 2, 6)]); + check_spans("/*** abc */ x", &[(1, 12, 1, 13)]); + check_spans("/*** ábc */ x", &[(1, 12, 1, 13)]); + check_spans(r#""abc""#, &[(1, 0, 1, 5)]); + check_spans(r#""ábc""#, &[(1, 0, 1, 5)]); + check_spans(r##"r#"abc"#"##, &[(1, 0, 1, 8)]); + check_spans(r##"r#"ábc"#"##, &[(1, 0, 1, 8)]); + check_spans("r#\"a\nc\"#", &[(1, 0, 2, 3)]); + check_spans("r#\"á\nc\"#", &[(1, 0, 2, 3)]); + check_spans("'a'", &[(1, 0, 1, 3)]); + check_spans("'á'", &[(1, 0, 1, 3)]); + check_spans("//! abc", &[(1, 0, 1, 7), (1, 0, 1, 7), (1, 0, 1, 7)]); + check_spans("//! ábc", &[(1, 0, 1, 7), (1, 0, 1, 7), (1, 0, 1, 7)]); + check_spans("//! abc\n", &[(1, 0, 1, 7), (1, 0, 1, 7), (1, 0, 1, 7)]); + check_spans("//! ábc\n", &[(1, 0, 1, 7), (1, 0, 1, 7), (1, 0, 1, 7)]); + check_spans("/*! abc */", &[(1, 0, 1, 10), (1, 0, 1, 10), (1, 0, 1, 10)]); + check_spans("/*! ábc */", &[(1, 0, 1, 10), (1, 0, 1, 10), (1, 0, 1, 10)]); + check_spans("/*! a\nc */", &[(1, 0, 2, 4), (1, 0, 2, 4), (1, 0, 2, 4)]); + check_spans("/*! á\nc */", &[(1, 0, 2, 4), (1, 0, 2, 4), (1, 0, 2, 4)]); + check_spans("abc", &[(1, 0, 1, 3)]); + check_spans("ábc", &[(1, 0, 1, 3)]); + check_spans("ábć", &[(1, 0, 1, 3)]); + check_spans("abc// foo", &[(1, 0, 1, 3)]); + check_spans("ábc// foo", &[(1, 0, 1, 3)]); + check_spans("ábć// foo", &[(1, 0, 1, 3)]); + check_spans("b\"a\\\n c\"", &[(1, 0, 2, 3)]); +} + +#[cfg(span_locations)] +fn check_spans(p: &str, mut lines: &[(usize, usize, usize, usize)]) { + let ts = p.parse::<TokenStream>().unwrap(); + check_spans_internal(ts, &mut lines); + assert!(lines.is_empty(), "leftover ranges: {:?}", lines); +} + +#[cfg(span_locations)] +fn check_spans_internal(ts: TokenStream, lines: &mut &[(usize, usize, usize, usize)]) { + for i in ts { + if let Some((&(sline, scol, eline, ecol), rest)) = lines.split_first() { + *lines = rest; + + let start = i.span().start(); + assert_eq!(start.line, sline, "sline did not match for {}", i); + assert_eq!(start.column, scol, "scol did not match for {}", i); + + let end = i.span().end(); + assert_eq!(end.line, eline, "eline did not match for {}", i); + assert_eq!(end.column, ecol, "ecol did not match for {}", i); + + if let TokenTree::Group(g) = i { + check_spans_internal(g.stream().clone(), lines); + } + } + } +} + +#[test] +fn whitespace() { + // space, horizontal tab, vertical tab, form feed, carriage return, line + // feed, non-breaking space, left-to-right mark, right-to-left mark + let various_spaces = " \t\u{b}\u{c}\r\n\u{a0}\u{200e}\u{200f}"; + let tokens = various_spaces.parse::<TokenStream>().unwrap(); + assert_eq!(tokens.into_iter().count(), 0); + + let lone_carriage_returns = " \r \r\r\n "; + lone_carriage_returns.parse::<TokenStream>().unwrap(); +} + +#[test] +fn byte_order_mark() { + let string = "\u{feff}foo"; + let tokens = string.parse::<TokenStream>().unwrap(); + match tokens.into_iter().next().unwrap() { + TokenTree::Ident(ident) => assert_eq!(ident, "foo"), + _ => unreachable!(), + } + + let string = "foo\u{feff}"; + string.parse::<TokenStream>().unwrap_err(); +} diff --git a/tests/test_fmt.rs b/tests/test_fmt.rs new file mode 100644 index 0000000..86a4c38 --- /dev/null +++ b/tests/test_fmt.rs @@ -0,0 +1,28 @@ +#![allow(clippy::from_iter_instead_of_collect)] + +use proc_macro2::{Delimiter, Group, Ident, Span, TokenStream, TokenTree}; +use std::iter; + +#[test] +fn test_fmt_group() { + let ident = Ident::new("x", Span::call_site()); + let inner = TokenStream::from_iter(iter::once(TokenTree::Ident(ident))); + let parens_empty = Group::new(Delimiter::Parenthesis, TokenStream::new()); + let parens_nonempty = Group::new(Delimiter::Parenthesis, inner.clone()); + let brackets_empty = Group::new(Delimiter::Bracket, TokenStream::new()); + let brackets_nonempty = Group::new(Delimiter::Bracket, inner.clone()); + let braces_empty = Group::new(Delimiter::Brace, TokenStream::new()); + let braces_nonempty = Group::new(Delimiter::Brace, inner.clone()); + let none_empty = Group::new(Delimiter::None, TokenStream::new()); + let none_nonempty = Group::new(Delimiter::None, inner); + + // Matches libproc_macro. + assert_eq!("()", parens_empty.to_string()); + assert_eq!("(x)", parens_nonempty.to_string()); + assert_eq!("[]", brackets_empty.to_string()); + assert_eq!("[x]", brackets_nonempty.to_string()); + assert_eq!("{ }", braces_empty.to_string()); + assert_eq!("{ x }", braces_nonempty.to_string()); + assert_eq!("", none_empty.to_string()); + assert_eq!("x", none_nonempty.to_string()); +} diff --git a/tests/test_size.rs b/tests/test_size.rs new file mode 100644 index 0000000..46e58db --- /dev/null +++ b/tests/test_size.rs @@ -0,0 +1,42 @@ +extern crate proc_macro; + +use std::mem; + +#[rustversion::attr(before(1.32), ignore)] +#[test] +fn test_proc_macro_span_size() { + assert_eq!(mem::size_of::<proc_macro::Span>(), 4); + assert_eq!(mem::size_of::<Option<proc_macro::Span>>(), 4); +} + +#[cfg_attr(not(all(not(wrap_proc_macro), not(span_locations))), ignore)] +#[test] +fn test_proc_macro2_fallback_span_size_without_locations() { + assert_eq!(mem::size_of::<proc_macro2::Span>(), 0); + assert_eq!(mem::size_of::<Option<proc_macro2::Span>>(), 1); +} + +#[cfg_attr(not(all(not(wrap_proc_macro), span_locations)), ignore)] +#[test] +fn test_proc_macro2_fallback_span_size_with_locations() { + assert_eq!(mem::size_of::<proc_macro2::Span>(), 8); + assert_eq!(mem::size_of::<Option<proc_macro2::Span>>(), 12); +} + +#[rustversion::attr(before(1.32), ignore)] +#[rustversion::attr( + since(1.32), + cfg_attr(not(all(wrap_proc_macro, not(span_locations))), ignore) +)] +#[test] +fn test_proc_macro2_wrapper_span_size_without_locations() { + assert_eq!(mem::size_of::<proc_macro2::Span>(), 4); + assert_eq!(mem::size_of::<Option<proc_macro2::Span>>(), 8); +} + +#[cfg_attr(not(all(wrap_proc_macro, span_locations)), ignore)] +#[test] +fn test_proc_macro2_wrapper_span_size_with_locations() { + assert_eq!(mem::size_of::<proc_macro2::Span>(), 12); + assert_eq!(mem::size_of::<Option<proc_macro2::Span>>(), 12); +} |