From 2acbe9c7438478b022fa52e54ba3ceb8ca570c20 Mon Sep 17 00:00:00 2001 From: Nicholas Nethercote Date: Fri, 3 May 2024 09:26:34 +1000 Subject: [PATCH] Move some tests from `rustc_expand` to `rustc_parse`. There are some test cases involving `parse` and `tokenstream` and `mut_visit` that are located in `rustc_expand`. Because it used to be the case that constructing a `ParseSess` required the involvement of `rustc_expand`. However, since #64197 merged (a long time ago) `rust_expand` no longer needs to be involved. This commit moves the tests into `rustc_parse`. This is the optimal place for the `parse` tests. It's not ideal for the `tokenstream` and `mut_visit` tests -- they would be better in `rustc_ast` -- but they still rely on parsing, which is not available in `rustc_ast`. But `rustc_parse` is lower down in the crate graph and closer to `rustc_ast` than `rust_expand`, so it's still an improvement for them. The exact renaming is as follows: - rustc_expand/src/mut_visit/tests.rs -> rustc_parse/src/parser/mut_visit/tests.rs - rustc_expand/src/tokenstream/tests.rs -> rustc_parse/src/parser/tokenstream/tests.rs - rustc_expand/src/tests.rs + rustc_expand/src/parse/tests.rs -> compiler/rustc_parse/src/parser/tests.rs The latter two test files are combined because there's no need for them to be separate, and having a `rustc_parse::parser::parse` module would be weird. This also means some `pub(crate)`s can be removed. --- Cargo.lock | 2 +- compiler/rustc_expand/Cargo.toml | 1 - compiler/rustc_expand/src/lib.rs | 20 - compiler/rustc_expand/src/parse/tests.rs | 382 ----------------- compiler/rustc_parse/Cargo.toml | 4 + compiler/rustc_parse/src/parser/mod.rs | 14 + .../src/parser}/mut_visit/tests.rs | 3 +- .../src => rustc_parse/src/parser}/tests.rs | 401 +++++++++++++++++- .../src/parser}/tokenstream/tests.rs | 3 +- 9 files changed, 405 insertions(+), 425 deletions(-) delete mode 100644 compiler/rustc_expand/src/parse/tests.rs rename compiler/{rustc_expand/src => rustc_parse/src/parser}/mut_visit/tests.rs (97%) rename compiler/{rustc_expand/src => rustc_parse/src/parser}/tests.rs (61%) rename compiler/{rustc_expand/src => rustc_parse/src/parser}/tokenstream/tests.rs (98%) diff --git a/Cargo.lock b/Cargo.lock index a6b6f3340dc..e5b5543b367 100644 --- a/Cargo.lock +++ b/Cargo.lock @@ -3976,7 +3976,6 @@ dependencies = [ "rustc_session", "rustc_span", "smallvec", - "termcolor", "thin-vec", "tracing", ] @@ -4464,6 +4463,7 @@ dependencies = [ "rustc_macros", "rustc_session", "rustc_span", + "termcolor", "thin-vec", "tracing", "unicode-normalization", diff --git a/compiler/rustc_expand/Cargo.toml b/compiler/rustc_expand/Cargo.toml index 63247f9d051..ce014364b0d 100644 --- a/compiler/rustc_expand/Cargo.toml +++ b/compiler/rustc_expand/Cargo.toml @@ -25,7 +25,6 @@ rustc_serialize = { path = "../rustc_serialize" } rustc_session = { path = "../rustc_session" } rustc_span = { path = "../rustc_span" } smallvec = { version = "1.8.1", features = ["union", "may_dangle"] } -termcolor = "1.2" thin-vec = "0.2.12" tracing = "0.1" # tidy-alphabetical-end diff --git a/compiler/rustc_expand/src/lib.rs b/compiler/rustc_expand/src/lib.rs index c2c3b777699..4222c9fe906 100644 --- a/compiler/rustc_expand/src/lib.rs +++ b/compiler/rustc_expand/src/lib.rs @@ -34,24 +34,4 @@ pub mod module; #[allow(rustc::untranslatable_diagnostic)] pub mod proc_macro; -// HACK(Centril, #64197): These shouldn't really be here. -// Rather, they should be with their respective modules which are defined in other crates. -// However, since for now constructing a `ParseSess` sorta requires `config` from this crate, -// these tests will need to live here in the interim. - -#[cfg(test)] -mod tests; -#[cfg(test)] -mod parse { - mod tests; -} -#[cfg(test)] -mod tokenstream { - mod tests; -} -#[cfg(test)] -mod mut_visit { - mod tests; -} - rustc_fluent_macro::fluent_messages! { "../messages.ftl" } diff --git a/compiler/rustc_expand/src/parse/tests.rs b/compiler/rustc_expand/src/parse/tests.rs deleted file mode 100644 index 066afd7a41d..00000000000 --- a/compiler/rustc_expand/src/parse/tests.rs +++ /dev/null @@ -1,382 +0,0 @@ -use crate::tests::{ - matches_codepattern, psess, string_to_stream, with_error_checking_parse, - with_expected_parse_error, -}; - -use ast::token::IdentIsRaw; -use rustc_ast::ptr::P; -use rustc_ast::token::{self, Delimiter, Token}; -use rustc_ast::tokenstream::{DelimSpacing, DelimSpan, Spacing, TokenStream, TokenTree}; -use rustc_ast::visit; -use rustc_ast::{self as ast, PatKind}; -use rustc_ast_pretty::pprust::item_to_string; -use rustc_errors::PResult; -use rustc_parse::new_parser_from_source_str; -use rustc_parse::parser::ForceCollect; -use rustc_session::parse::ParseSess; -use rustc_span::create_default_session_globals_then; -use rustc_span::symbol::{kw, sym, Symbol}; -use rustc_span::{BytePos, FileName, Pos, Span}; -use std::path::PathBuf; - -/// Parses an item. -/// -/// Returns `Ok(Some(item))` when successful, `Ok(None)` when no item was found, and `Err` -/// when a syntax error occurred. -fn parse_item_from_source_str( - name: FileName, - source: String, - psess: &ParseSess, -) -> PResult<'_, Option>> { - new_parser_from_source_str(psess, name, source).parse_item(ForceCollect::No) -} - -// Produces a `rustc_span::span`. -fn sp(a: u32, b: u32) -> Span { - Span::with_root_ctxt(BytePos(a), BytePos(b)) -} - -/// Parses a string, return an expression. -fn string_to_expr(source_str: String) -> P { - with_error_checking_parse(source_str, &psess(), |p| p.parse_expr()) -} - -/// Parses a string, returns an item. -fn string_to_item(source_str: String) -> Option> { - with_error_checking_parse(source_str, &psess(), |p| p.parse_item(ForceCollect::No)) -} - -#[test] -fn bad_path_expr_1() { - // This should trigger error: expected identifier, found keyword `return` - create_default_session_globals_then(|| { - with_expected_parse_error( - "::abc::def::return", - "expected identifier, found keyword `return`", - |p| p.parse_expr(), - ); - }) -} - -// Checks the token-tree-ization of macros. -#[test] -fn string_to_tts_macro() { - create_default_session_globals_then(|| { - let stream = string_to_stream("macro_rules! zip (($a)=>($a))".to_string()); - let tts = &stream.trees().collect::>()[..]; - - match tts { - [ - TokenTree::Token( - Token { kind: token::Ident(name_macro_rules, IdentIsRaw::No), .. }, - _, - ), - TokenTree::Token(Token { kind: token::Not, .. }, _), - TokenTree::Token(Token { kind: token::Ident(name_zip, IdentIsRaw::No), .. }, _), - TokenTree::Delimited(.., macro_delim, macro_tts), - ] if name_macro_rules == &kw::MacroRules && name_zip.as_str() == "zip" => { - let tts = ¯o_tts.trees().collect::>(); - match &tts[..] { - [ - TokenTree::Delimited(.., first_delim, first_tts), - TokenTree::Token(Token { kind: token::FatArrow, .. }, _), - TokenTree::Delimited(.., second_delim, second_tts), - ] if macro_delim == &Delimiter::Parenthesis => { - let tts = &first_tts.trees().collect::>(); - match &tts[..] { - [ - TokenTree::Token(Token { kind: token::Dollar, .. }, _), - TokenTree::Token( - Token { kind: token::Ident(name, IdentIsRaw::No), .. }, - _, - ), - ] if first_delim == &Delimiter::Parenthesis && name.as_str() == "a" => { - } - _ => panic!("value 3: {:?} {:?}", first_delim, first_tts), - } - let tts = &second_tts.trees().collect::>(); - match &tts[..] { - [ - TokenTree::Token(Token { kind: token::Dollar, .. }, _), - TokenTree::Token( - Token { kind: token::Ident(name, IdentIsRaw::No), .. }, - _, - ), - ] if second_delim == &Delimiter::Parenthesis - && name.as_str() == "a" => {} - _ => panic!("value 4: {:?} {:?}", second_delim, second_tts), - } - } - _ => panic!("value 2: {:?} {:?}", macro_delim, macro_tts), - } - } - _ => panic!("value: {:?}", tts), - } - }) -} - -#[test] -fn string_to_tts_1() { - create_default_session_globals_then(|| { - let tts = string_to_stream("fn a(b: i32) { b; }".to_string()); - - let expected = TokenStream::new(vec![ - TokenTree::token_alone(token::Ident(kw::Fn, IdentIsRaw::No), sp(0, 2)), - TokenTree::token_joint_hidden( - token::Ident(Symbol::intern("a"), IdentIsRaw::No), - sp(3, 4), - ), - TokenTree::Delimited( - DelimSpan::from_pair(sp(4, 5), sp(11, 12)), - // `JointHidden` because the `(` is followed immediately by - // `b`, `Alone` because the `)` is followed by whitespace. - DelimSpacing::new(Spacing::JointHidden, Spacing::Alone), - Delimiter::Parenthesis, - TokenStream::new(vec![ - TokenTree::token_joint( - token::Ident(Symbol::intern("b"), IdentIsRaw::No), - sp(5, 6), - ), - TokenTree::token_alone(token::Colon, sp(6, 7)), - // `JointHidden` because the `i32` is immediately followed by the `)`. - TokenTree::token_joint_hidden( - token::Ident(sym::i32, IdentIsRaw::No), - sp(8, 11), - ), - ]) - .into(), - ), - TokenTree::Delimited( - DelimSpan::from_pair(sp(13, 14), sp(18, 19)), - // First `Alone` because the `{` is followed by whitespace, - // second `Alone` because the `}` is followed immediately by - // EOF. - DelimSpacing::new(Spacing::Alone, Spacing::Alone), - Delimiter::Brace, - TokenStream::new(vec![ - TokenTree::token_joint( - token::Ident(Symbol::intern("b"), IdentIsRaw::No), - sp(15, 16), - ), - // `Alone` because the `;` is followed by whitespace. - TokenTree::token_alone(token::Semi, sp(16, 17)), - ]) - .into(), - ), - ]); - - assert_eq!(tts, expected); - }) -} - -#[test] -fn parse_use() { - create_default_session_globals_then(|| { - let use_s = "use foo::bar::baz;"; - let vitem = string_to_item(use_s.to_string()).unwrap(); - let vitem_s = item_to_string(&vitem); - assert_eq!(&vitem_s[..], use_s); - - let use_s = "use foo::bar as baz;"; - let vitem = string_to_item(use_s.to_string()).unwrap(); - let vitem_s = item_to_string(&vitem); - assert_eq!(&vitem_s[..], use_s); - }) -} - -#[test] -fn parse_extern_crate() { - create_default_session_globals_then(|| { - let ex_s = "extern crate foo;"; - let vitem = string_to_item(ex_s.to_string()).unwrap(); - let vitem_s = item_to_string(&vitem); - assert_eq!(&vitem_s[..], ex_s); - - let ex_s = "extern crate foo as bar;"; - let vitem = string_to_item(ex_s.to_string()).unwrap(); - let vitem_s = item_to_string(&vitem); - assert_eq!(&vitem_s[..], ex_s); - }) -} - -fn get_spans_of_pat_idents(src: &str) -> Vec { - let item = string_to_item(src.to_string()).unwrap(); - - struct PatIdentVisitor { - spans: Vec, - } - impl<'a> visit::Visitor<'a> for PatIdentVisitor { - fn visit_pat(&mut self, p: &'a ast::Pat) { - match &p.kind { - PatKind::Ident(_, ident, _) => { - self.spans.push(ident.span); - } - _ => { - visit::walk_pat(self, p); - } - } - } - } - let mut v = PatIdentVisitor { spans: Vec::new() }; - visit::walk_item(&mut v, &item); - return v.spans; -} - -#[test] -fn span_of_self_arg_pat_idents_are_correct() { - create_default_session_globals_then(|| { - let srcs = [ - "impl z { fn a (&self, &myarg: i32) {} }", - "impl z { fn a (&mut self, &myarg: i32) {} }", - "impl z { fn a (&'a self, &myarg: i32) {} }", - "impl z { fn a (self, &myarg: i32) {} }", - "impl z { fn a (self: Foo, &myarg: i32) {} }", - ]; - - for src in srcs { - let spans = get_spans_of_pat_idents(src); - let (lo, hi) = (spans[0].lo(), spans[0].hi()); - assert!( - "self" == &src[lo.to_usize()..hi.to_usize()], - "\"{}\" != \"self\". src=\"{}\"", - &src[lo.to_usize()..hi.to_usize()], - src - ) - } - }) -} - -#[test] -fn parse_exprs() { - create_default_session_globals_then(|| { - // just make sure that they parse.... - string_to_expr("3 + 4".to_string()); - string_to_expr("a::z.froob(b,&(987+3))".to_string()); - }) -} - -#[test] -fn attrs_fix_bug() { - create_default_session_globals_then(|| { - string_to_item( - "pub fn mk_file_writer(path: &Path, flags: &[FileFlag]) - -> Result, String> { -#[cfg(windows)] -fn wb() -> c_int { - (O_WRONLY | libc::consts::os::extra::O_BINARY) as c_int -} - -#[cfg(unix)] -fn wb() -> c_int { O_WRONLY as c_int } - -let mut fflags: c_int = wb(); -}" - .to_string(), - ); - }) -} - -#[test] -fn crlf_doc_comments() { - create_default_session_globals_then(|| { - let psess = psess(); - - let name_1 = FileName::Custom("crlf_source_1".to_string()); - let source = "/// doc comment\r\nfn foo() {}".to_string(); - let item = parse_item_from_source_str(name_1, source, &psess).unwrap().unwrap(); - let doc = item.attrs.iter().filter_map(|at| at.doc_str()).next().unwrap(); - assert_eq!(doc.as_str(), " doc comment"); - - let name_2 = FileName::Custom("crlf_source_2".to_string()); - let source = "/// doc comment\r\n/// line 2\r\nfn foo() {}".to_string(); - let item = parse_item_from_source_str(name_2, source, &psess).unwrap().unwrap(); - let docs = item.attrs.iter().filter_map(|at| at.doc_str()).collect::>(); - let b: &[_] = &[Symbol::intern(" doc comment"), Symbol::intern(" line 2")]; - assert_eq!(&docs[..], b); - - let name_3 = FileName::Custom("clrf_source_3".to_string()); - let source = "/** doc comment\r\n * with CRLF */\r\nfn foo() {}".to_string(); - let item = parse_item_from_source_str(name_3, source, &psess).unwrap().unwrap(); - let doc = item.attrs.iter().filter_map(|at| at.doc_str()).next().unwrap(); - assert_eq!(doc.as_str(), " doc comment\n * with CRLF "); - }); -} - -#[test] -fn ttdelim_span() { - fn parse_expr_from_source_str( - name: FileName, - source: String, - psess: &ParseSess, - ) -> PResult<'_, P> { - new_parser_from_source_str(psess, name, source).parse_expr() - } - - create_default_session_globals_then(|| { - let psess = psess(); - let expr = parse_expr_from_source_str( - PathBuf::from("foo").into(), - "foo!( fn main() { body } )".to_string(), - &psess, - ) - .unwrap(); - - let ast::ExprKind::MacCall(mac) = &expr.kind else { panic!("not a macro") }; - let span = mac.args.tokens.trees().last().unwrap().span(); - - match psess.source_map().span_to_snippet(span) { - Ok(s) => assert_eq!(&s[..], "{ body }"), - Err(_) => panic!("could not get snippet"), - } - }); -} - -// This tests that when parsing a string (rather than a file) we don't try -// and read in a file for a module declaration and just parse a stub. -// See `recurse_into_file_modules` in the parser. -#[test] -fn out_of_line_mod() { - create_default_session_globals_then(|| { - let item = parse_item_from_source_str( - PathBuf::from("foo").into(), - "mod foo { struct S; mod this_does_not_exist; }".to_owned(), - &psess(), - ) - .unwrap() - .unwrap(); - - let ast::ItemKind::Mod(_, mod_kind) = &item.kind else { panic!() }; - assert!(matches!(mod_kind, ast::ModKind::Loaded(items, ..) if items.len() == 2)); - }); -} - -#[test] -fn eqmodws() { - assert_eq!(matches_codepattern("", ""), true); - assert_eq!(matches_codepattern("", "a"), false); - assert_eq!(matches_codepattern("a", ""), false); - assert_eq!(matches_codepattern("a", "a"), true); - assert_eq!(matches_codepattern("a b", "a \n\t\r b"), true); - assert_eq!(matches_codepattern("a b ", "a \n\t\r b"), true); - assert_eq!(matches_codepattern("a b", "a \n\t\r b "), false); - assert_eq!(matches_codepattern("a b", "a b"), true); - assert_eq!(matches_codepattern("ab", "a b"), false); - assert_eq!(matches_codepattern("a b", "ab"), true); - assert_eq!(matches_codepattern(" a b", "ab"), true); -} - -#[test] -fn pattern_whitespace() { - assert_eq!(matches_codepattern("", "\x0C"), false); - assert_eq!(matches_codepattern("a b ", "a \u{0085}\n\t\r b"), true); - assert_eq!(matches_codepattern("a b", "a \u{0085}\n\t\r b "), false); -} - -#[test] -fn non_pattern_whitespace() { - // These have the property 'White_Space' but not 'Pattern_White_Space' - assert_eq!(matches_codepattern("a b", "a\u{2002}b"), false); - assert_eq!(matches_codepattern("a b", "a\u{2002}b"), false); - assert_eq!(matches_codepattern("\u{205F}a b", "ab"), false); - assert_eq!(matches_codepattern("a \u{3000}b", "ab"), false); -} diff --git a/compiler/rustc_parse/Cargo.toml b/compiler/rustc_parse/Cargo.toml index f5623323168..b33896154f2 100644 --- a/compiler/rustc_parse/Cargo.toml +++ b/compiler/rustc_parse/Cargo.toml @@ -21,3 +21,7 @@ tracing = "0.1" unicode-normalization = "0.1.11" unicode-width = "0.1.4" # tidy-alphabetical-end + +[dev-dependencies] +termcolor = "1.2" + diff --git a/compiler/rustc_parse/src/parser/mod.rs b/compiler/rustc_parse/src/parser/mod.rs index 7486da33b21..f1476fe02c5 100644 --- a/compiler/rustc_parse/src/parser/mod.rs +++ b/compiler/rustc_parse/src/parser/mod.rs @@ -45,6 +45,20 @@ use crate::errors::{ self, IncorrectVisibilityRestriction, MismatchedClosingDelimiter, NonStringAbiLiteral, }; +#[cfg(test)] +mod tests; + +// Ideally, these tests would be in `rustc_ast`. But they depend on having a +// parser, so they are here. +#[cfg(test)] +mod tokenstream { + mod tests; +} +#[cfg(test)] +mod mut_visit { + mod tests; +} + bitflags::bitflags! { #[derive(Clone, Copy)] struct Restrictions: u8 { diff --git a/compiler/rustc_expand/src/mut_visit/tests.rs b/compiler/rustc_parse/src/parser/mut_visit/tests.rs similarity index 97% rename from compiler/rustc_expand/src/mut_visit/tests.rs rename to compiler/rustc_parse/src/parser/mut_visit/tests.rs index 8974d45b4d8..b3cb28af657 100644 --- a/compiler/rustc_expand/src/mut_visit/tests.rs +++ b/compiler/rustc_parse/src/parser/mut_visit/tests.rs @@ -1,5 +1,4 @@ -use crate::tests::{matches_codepattern, string_to_crate}; - +use crate::parser::tests::{matches_codepattern, string_to_crate}; use rustc_ast as ast; use rustc_ast::mut_visit::MutVisitor; use rustc_ast_pretty::pprust; diff --git a/compiler/rustc_expand/src/tests.rs b/compiler/rustc_parse/src/parser/tests.rs similarity index 61% rename from compiler/rustc_expand/src/tests.rs rename to compiler/rustc_parse/src/parser/tests.rs index 2b0b58eb1d9..a31e350541a 100644 --- a/compiler/rustc_expand/src/tests.rs +++ b/compiler/rustc_parse/src/parser/tests.rs @@ -1,25 +1,30 @@ -use rustc_ast as ast; -use rustc_ast::tokenstream::TokenStream; -use rustc_parse::{new_parser_from_source_str, parser::Parser, source_file_to_stream}; -use rustc_session::parse::ParseSess; -use rustc_span::create_default_session_globals_then; -use rustc_span::source_map::{FilePathMapping, SourceMap}; -use rustc_span::{BytePos, Span}; - +use crate::parser::ForceCollect; +use crate::{new_parser_from_source_str, parser::Parser, source_file_to_stream}; +use ast::token::IdentIsRaw; +use rustc_ast::ptr::P; +use rustc_ast::token::{self, Delimiter, Token}; +use rustc_ast::tokenstream::{DelimSpacing, DelimSpan, Spacing, TokenStream, TokenTree}; +use rustc_ast::visit; +use rustc_ast::{self as ast, PatKind}; +use rustc_ast_pretty::pprust::item_to_string; use rustc_data_structures::sync::Lrc; use rustc_errors::emitter::HumanEmitter; use rustc_errors::{DiagCtxt, MultiSpan, PResult}; -use termcolor::WriteColor; - +use rustc_session::parse::ParseSess; +use rustc_span::create_default_session_globals_then; +use rustc_span::source_map::{FilePathMapping, SourceMap}; +use rustc_span::symbol::{kw, sym, Symbol}; +use rustc_span::{BytePos, FileName, Pos, Span}; use std::io; use std::io::prelude::*; use std::iter::Peekable; use std::path::{Path, PathBuf}; use std::str; use std::sync::{Arc, Mutex}; +use termcolor::WriteColor; -pub(crate) fn psess() -> ParseSess { - ParseSess::new(vec![crate::DEFAULT_LOCALE_RESOURCE, rustc_parse::DEFAULT_LOCALE_RESOURCE]) +fn psess() -> ParseSess { + ParseSess::new(vec![crate::DEFAULT_LOCALE_RESOURCE, crate::DEFAULT_LOCALE_RESOURCE]) } /// Map string to parser (via tts). @@ -31,7 +36,7 @@ fn create_test_handler() -> (DiagCtxt, Lrc, Arc>>) { let output = Arc::new(Mutex::new(Vec::new())); let source_map = Lrc::new(SourceMap::new(FilePathMapping::empty())); let fallback_bundle = rustc_errors::fallback_fluent_bundle( - vec![crate::DEFAULT_LOCALE_RESOURCE, rustc_parse::DEFAULT_LOCALE_RESOURCE], + vec![crate::DEFAULT_LOCALE_RESOURCE, crate::DEFAULT_LOCALE_RESOURCE], false, ); let emitter = HumanEmitter::new(Box::new(Shared { data: output.clone() }), fallback_bundle) @@ -44,7 +49,7 @@ fn create_test_handler() -> (DiagCtxt, Lrc, Arc>>) { /// Returns the result of parsing the given string via the given callback. /// /// If there are any errors, this will panic. -pub(crate) fn with_error_checking_parse<'a, T, F>(s: String, psess: &'a ParseSess, f: F) -> T +fn with_error_checking_parse<'a, T, F>(s: String, psess: &'a ParseSess, f: F) -> T where F: FnOnce(&mut Parser<'a>) -> PResult<'a, T>, { @@ -56,7 +61,7 @@ where /// Verifies that parsing the given string using the given callback will /// generate an error that contains the given text. -pub(crate) fn with_expected_parse_error(source_str: &str, expected_output: &str, f: F) +fn with_expected_parse_error(source_str: &str, expected_output: &str, f: F) where F: for<'a> FnOnce(&mut Parser<'a>) -> PResult<'a, T>, { @@ -150,8 +155,8 @@ struct SpanLabel { label: &'static str, } -pub(crate) struct Shared { - pub data: Arc>, +struct Shared { + data: Arc>, } impl WriteColor for Shared { @@ -1053,3 +1058,365 @@ error: foo "#, ); } + +/// Parses an item. +/// +/// Returns `Ok(Some(item))` when successful, `Ok(None)` when no item was found, and `Err` +/// when a syntax error occurred. +fn parse_item_from_source_str( + name: FileName, + source: String, + psess: &ParseSess, +) -> PResult<'_, Option>> { + new_parser_from_source_str(psess, name, source).parse_item(ForceCollect::No) +} + +// Produces a `rustc_span::span`. +fn sp(a: u32, b: u32) -> Span { + Span::with_root_ctxt(BytePos(a), BytePos(b)) +} + +/// Parses a string, return an expression. +fn string_to_expr(source_str: String) -> P { + with_error_checking_parse(source_str, &psess(), |p| p.parse_expr()) +} + +/// Parses a string, returns an item. +fn string_to_item(source_str: String) -> Option> { + with_error_checking_parse(source_str, &psess(), |p| p.parse_item(ForceCollect::No)) +} + +#[test] +fn bad_path_expr_1() { + // This should trigger error: expected identifier, found keyword `return` + create_default_session_globals_then(|| { + with_expected_parse_error( + "::abc::def::return", + "expected identifier, found keyword `return`", + |p| p.parse_expr(), + ); + }) +} + +// Checks the token-tree-ization of macros. +#[test] +fn string_to_tts_macro() { + create_default_session_globals_then(|| { + let stream = string_to_stream("macro_rules! zip (($a)=>($a))".to_string()); + let tts = &stream.trees().collect::>()[..]; + + match tts { + [ + TokenTree::Token( + Token { kind: token::Ident(name_macro_rules, IdentIsRaw::No), .. }, + _, + ), + TokenTree::Token(Token { kind: token::Not, .. }, _), + TokenTree::Token(Token { kind: token::Ident(name_zip, IdentIsRaw::No), .. }, _), + TokenTree::Delimited(.., macro_delim, macro_tts), + ] if name_macro_rules == &kw::MacroRules && name_zip.as_str() == "zip" => { + let tts = ¯o_tts.trees().collect::>(); + match &tts[..] { + [ + TokenTree::Delimited(.., first_delim, first_tts), + TokenTree::Token(Token { kind: token::FatArrow, .. }, _), + TokenTree::Delimited(.., second_delim, second_tts), + ] if macro_delim == &Delimiter::Parenthesis => { + let tts = &first_tts.trees().collect::>(); + match &tts[..] { + [ + TokenTree::Token(Token { kind: token::Dollar, .. }, _), + TokenTree::Token( + Token { kind: token::Ident(name, IdentIsRaw::No), .. }, + _, + ), + ] if first_delim == &Delimiter::Parenthesis && name.as_str() == "a" => { + } + _ => panic!("value 3: {:?} {:?}", first_delim, first_tts), + } + let tts = &second_tts.trees().collect::>(); + match &tts[..] { + [ + TokenTree::Token(Token { kind: token::Dollar, .. }, _), + TokenTree::Token( + Token { kind: token::Ident(name, IdentIsRaw::No), .. }, + _, + ), + ] if second_delim == &Delimiter::Parenthesis + && name.as_str() == "a" => {} + _ => panic!("value 4: {:?} {:?}", second_delim, second_tts), + } + } + _ => panic!("value 2: {:?} {:?}", macro_delim, macro_tts), + } + } + _ => panic!("value: {:?}", tts), + } + }) +} + +#[test] +fn string_to_tts_1() { + create_default_session_globals_then(|| { + let tts = string_to_stream("fn a(b: i32) { b; }".to_string()); + + let expected = TokenStream::new(vec![ + TokenTree::token_alone(token::Ident(kw::Fn, IdentIsRaw::No), sp(0, 2)), + TokenTree::token_joint_hidden( + token::Ident(Symbol::intern("a"), IdentIsRaw::No), + sp(3, 4), + ), + TokenTree::Delimited( + DelimSpan::from_pair(sp(4, 5), sp(11, 12)), + // `JointHidden` because the `(` is followed immediately by + // `b`, `Alone` because the `)` is followed by whitespace. + DelimSpacing::new(Spacing::JointHidden, Spacing::Alone), + Delimiter::Parenthesis, + TokenStream::new(vec![ + TokenTree::token_joint( + token::Ident(Symbol::intern("b"), IdentIsRaw::No), + sp(5, 6), + ), + TokenTree::token_alone(token::Colon, sp(6, 7)), + // `JointHidden` because the `i32` is immediately followed by the `)`. + TokenTree::token_joint_hidden( + token::Ident(sym::i32, IdentIsRaw::No), + sp(8, 11), + ), + ]) + .into(), + ), + TokenTree::Delimited( + DelimSpan::from_pair(sp(13, 14), sp(18, 19)), + // First `Alone` because the `{` is followed by whitespace, + // second `Alone` because the `}` is followed immediately by + // EOF. + DelimSpacing::new(Spacing::Alone, Spacing::Alone), + Delimiter::Brace, + TokenStream::new(vec![ + TokenTree::token_joint( + token::Ident(Symbol::intern("b"), IdentIsRaw::No), + sp(15, 16), + ), + // `Alone` because the `;` is followed by whitespace. + TokenTree::token_alone(token::Semi, sp(16, 17)), + ]) + .into(), + ), + ]); + + assert_eq!(tts, expected); + }) +} + +#[test] +fn parse_use() { + create_default_session_globals_then(|| { + let use_s = "use foo::bar::baz;"; + let vitem = string_to_item(use_s.to_string()).unwrap(); + let vitem_s = item_to_string(&vitem); + assert_eq!(&vitem_s[..], use_s); + + let use_s = "use foo::bar as baz;"; + let vitem = string_to_item(use_s.to_string()).unwrap(); + let vitem_s = item_to_string(&vitem); + assert_eq!(&vitem_s[..], use_s); + }) +} + +#[test] +fn parse_extern_crate() { + create_default_session_globals_then(|| { + let ex_s = "extern crate foo;"; + let vitem = string_to_item(ex_s.to_string()).unwrap(); + let vitem_s = item_to_string(&vitem); + assert_eq!(&vitem_s[..], ex_s); + + let ex_s = "extern crate foo as bar;"; + let vitem = string_to_item(ex_s.to_string()).unwrap(); + let vitem_s = item_to_string(&vitem); + assert_eq!(&vitem_s[..], ex_s); + }) +} + +fn get_spans_of_pat_idents(src: &str) -> Vec { + let item = string_to_item(src.to_string()).unwrap(); + + struct PatIdentVisitor { + spans: Vec, + } + impl<'a> visit::Visitor<'a> for PatIdentVisitor { + fn visit_pat(&mut self, p: &'a ast::Pat) { + match &p.kind { + PatKind::Ident(_, ident, _) => { + self.spans.push(ident.span); + } + _ => { + visit::walk_pat(self, p); + } + } + } + } + let mut v = PatIdentVisitor { spans: Vec::new() }; + visit::walk_item(&mut v, &item); + return v.spans; +} + +#[test] +fn span_of_self_arg_pat_idents_are_correct() { + create_default_session_globals_then(|| { + let srcs = [ + "impl z { fn a (&self, &myarg: i32) {} }", + "impl z { fn a (&mut self, &myarg: i32) {} }", + "impl z { fn a (&'a self, &myarg: i32) {} }", + "impl z { fn a (self, &myarg: i32) {} }", + "impl z { fn a (self: Foo, &myarg: i32) {} }", + ]; + + for src in srcs { + let spans = get_spans_of_pat_idents(src); + let (lo, hi) = (spans[0].lo(), spans[0].hi()); + assert!( + "self" == &src[lo.to_usize()..hi.to_usize()], + "\"{}\" != \"self\". src=\"{}\"", + &src[lo.to_usize()..hi.to_usize()], + src + ) + } + }) +} + +#[test] +fn parse_exprs() { + create_default_session_globals_then(|| { + // just make sure that they parse.... + string_to_expr("3 + 4".to_string()); + string_to_expr("a::z.froob(b,&(987+3))".to_string()); + }) +} + +#[test] +fn attrs_fix_bug() { + create_default_session_globals_then(|| { + string_to_item( + "pub fn mk_file_writer(path: &Path, flags: &[FileFlag]) + -> Result, String> { +#[cfg(windows)] +fn wb() -> c_int { + (O_WRONLY | libc::consts::os::extra::O_BINARY) as c_int +} + +#[cfg(unix)] +fn wb() -> c_int { O_WRONLY as c_int } + +let mut fflags: c_int = wb(); +}" + .to_string(), + ); + }) +} + +#[test] +fn crlf_doc_comments() { + create_default_session_globals_then(|| { + let psess = psess(); + + let name_1 = FileName::Custom("crlf_source_1".to_string()); + let source = "/// doc comment\r\nfn foo() {}".to_string(); + let item = parse_item_from_source_str(name_1, source, &psess).unwrap().unwrap(); + let doc = item.attrs.iter().filter_map(|at| at.doc_str()).next().unwrap(); + assert_eq!(doc.as_str(), " doc comment"); + + let name_2 = FileName::Custom("crlf_source_2".to_string()); + let source = "/// doc comment\r\n/// line 2\r\nfn foo() {}".to_string(); + let item = parse_item_from_source_str(name_2, source, &psess).unwrap().unwrap(); + let docs = item.attrs.iter().filter_map(|at| at.doc_str()).collect::>(); + let b: &[_] = &[Symbol::intern(" doc comment"), Symbol::intern(" line 2")]; + assert_eq!(&docs[..], b); + + let name_3 = FileName::Custom("clrf_source_3".to_string()); + let source = "/** doc comment\r\n * with CRLF */\r\nfn foo() {}".to_string(); + let item = parse_item_from_source_str(name_3, source, &psess).unwrap().unwrap(); + let doc = item.attrs.iter().filter_map(|at| at.doc_str()).next().unwrap(); + assert_eq!(doc.as_str(), " doc comment\n * with CRLF "); + }); +} + +#[test] +fn ttdelim_span() { + fn parse_expr_from_source_str( + name: FileName, + source: String, + psess: &ParseSess, + ) -> PResult<'_, P> { + new_parser_from_source_str(psess, name, source).parse_expr() + } + + create_default_session_globals_then(|| { + let psess = psess(); + let expr = parse_expr_from_source_str( + PathBuf::from("foo").into(), + "foo!( fn main() { body } )".to_string(), + &psess, + ) + .unwrap(); + + let ast::ExprKind::MacCall(mac) = &expr.kind else { panic!("not a macro") }; + let span = mac.args.tokens.trees().last().unwrap().span(); + + match psess.source_map().span_to_snippet(span) { + Ok(s) => assert_eq!(&s[..], "{ body }"), + Err(_) => panic!("could not get snippet"), + } + }); +} + +// This tests that when parsing a string (rather than a file) we don't try +// and read in a file for a module declaration and just parse a stub. +// See `recurse_into_file_modules` in the parser. +#[test] +fn out_of_line_mod() { + create_default_session_globals_then(|| { + let item = parse_item_from_source_str( + PathBuf::from("foo").into(), + "mod foo { struct S; mod this_does_not_exist; }".to_owned(), + &psess(), + ) + .unwrap() + .unwrap(); + + let ast::ItemKind::Mod(_, mod_kind) = &item.kind else { panic!() }; + assert!(matches!(mod_kind, ast::ModKind::Loaded(items, ..) if items.len() == 2)); + }); +} + +#[test] +fn eqmodws() { + assert_eq!(matches_codepattern("", ""), true); + assert_eq!(matches_codepattern("", "a"), false); + assert_eq!(matches_codepattern("a", ""), false); + assert_eq!(matches_codepattern("a", "a"), true); + assert_eq!(matches_codepattern("a b", "a \n\t\r b"), true); + assert_eq!(matches_codepattern("a b ", "a \n\t\r b"), true); + assert_eq!(matches_codepattern("a b", "a \n\t\r b "), false); + assert_eq!(matches_codepattern("a b", "a b"), true); + assert_eq!(matches_codepattern("ab", "a b"), false); + assert_eq!(matches_codepattern("a b", "ab"), true); + assert_eq!(matches_codepattern(" a b", "ab"), true); +} + +#[test] +fn pattern_whitespace() { + assert_eq!(matches_codepattern("", "\x0C"), false); + assert_eq!(matches_codepattern("a b ", "a \u{0085}\n\t\r b"), true); + assert_eq!(matches_codepattern("a b", "a \u{0085}\n\t\r b "), false); +} + +#[test] +fn non_pattern_whitespace() { + // These have the property 'White_Space' but not 'Pattern_White_Space' + assert_eq!(matches_codepattern("a b", "a\u{2002}b"), false); + assert_eq!(matches_codepattern("a b", "a\u{2002}b"), false); + assert_eq!(matches_codepattern("\u{205F}a b", "ab"), false); + assert_eq!(matches_codepattern("a \u{3000}b", "ab"), false); +} diff --git a/compiler/rustc_expand/src/tokenstream/tests.rs b/compiler/rustc_parse/src/parser/tokenstream/tests.rs similarity index 98% rename from compiler/rustc_expand/src/tokenstream/tests.rs rename to compiler/rustc_parse/src/parser/tokenstream/tests.rs index 78795e86fd5..9be00a14791 100644 --- a/compiler/rustc_expand/src/tokenstream/tests.rs +++ b/compiler/rustc_parse/src/parser/tokenstream/tests.rs @@ -1,5 +1,4 @@ -use crate::tests::string_to_stream; - +use crate::parser::tests::string_to_stream; use rustc_ast::token::{self, IdentIsRaw}; use rustc_ast::tokenstream::{TokenStream, TokenTree}; use rustc_span::create_default_session_globals_then;