shebang handling
Hi, Shebang handling, the first line starting with #! was not done fully correct and it isn't necessary to keep track of the shebang line in the AST or HIR Crate classes. Because an inner attribute also starts with #! the first line isn't regarded as a shebang line if the #! is followed by (optional) whitespace and comments and a [. In that case the #! is seen as the start of an inner attribute. I added various testcases that hopefully show the funny things you can get when the first line starts with #!. [PATCH 1/2] Handle shebang line, plus any whitespace and comment [PATCH 2/2] Remove has_shebang flag from AST and HIR Crate classes Patches can also be found here: https://code.wildebeest.org/git/user/mjw/gccrs/log/?h=shebang Cheers, Mark -- Gcc-rust mailing list Gcc-rust@gcc.gnu.org https://gcc.gnu.org/mailman/listinfo/gcc-rust
[PATCH 1/2] Handle shebang line, plus any whitespace and comment skipping in lexer
The lexer tried to handle the shebang line but used loc directly,
instead of the current_column. And it assumed a '/' should immediately
follow the "#!". But if the "#!" is followed by whitespace and/or
comments and a '[' character, then the first line isn't see as a
shebang line (even if the kernel or shell would) but as the start of
an inner attribute.
Add various tests for when the first line starting with "#!" is seen
as a shebang line (and should be skipped). And some tests there is a
'[' character following some whitespace and/or comments and the "#!"
is seen as part of an inner attribute.
---
gcc/rust/lex/rust-lex.cc | 79 ++-
.../rust/compile/torture/not_shebang.rs | 3 +
.../torture/not_shebang_block_comment.rs | 1 +
.../compile/torture/not_shebang_comment.rs| 3 +
.../torture/not_shebang_multiline_comment.rs | 7 ++
.../compile/torture/not_shebang_spaces.rs | 6 ++
gcc/testsuite/rust/compile/torture/shebang.rs | 3 +
.../rust/compile/torture/shebang_plus_attr.rs | 3 +
.../compile/torture/shebang_plus_attr2.rs | 3 +
9 files changed, 89 insertions(+), 19 deletions(-)
create mode 100644 gcc/testsuite/rust/compile/torture/not_shebang.rs
create mode 100644
gcc/testsuite/rust/compile/torture/not_shebang_block_comment.rs
create mode 100644 gcc/testsuite/rust/compile/torture/not_shebang_comment.rs
create mode 100644
gcc/testsuite/rust/compile/torture/not_shebang_multiline_comment.rs
create mode 100644 gcc/testsuite/rust/compile/torture/not_shebang_spaces.rs
create mode 100755 gcc/testsuite/rust/compile/torture/shebang.rs
create mode 100755 gcc/testsuite/rust/compile/torture/shebang_plus_attr.rs
create mode 100755 gcc/testsuite/rust/compile/torture/shebang_plus_attr2.rs
diff --git a/gcc/rust/lex/rust-lex.cc b/gcc/rust/lex/rust-lex.cc
index d1384168731..ebd69de0fd1 100644
--- a/gcc/rust/lex/rust-lex.cc
+++ b/gcc/rust/lex/rust-lex.cc
@@ -237,28 +237,63 @@ Lexer::build_token ()
current_char = peek_input ();
skip_input ();
- // return end of file token if end of file
- if (current_char == EOF)
- return Token::make (END_OF_FILE, loc);
-
// detect shebang
- if (loc == 1 && current_line == 1 && current_char == '#')
+ // Must be the first thing on the first line, starting with #!
+ // But since an attribute can also start with an #! we don't count it as
a
+ // shebang line when after any whitespace or comments there is a [. If it
+ // is a shebang line we simple drop the line. Otherwise we don't consume
+ // any characters and fall through to the real tokenizer.
+ if (current_line == 1 && current_column == 1 && current_char == '#'
+ && peek_input () == '!')
{
- current_char = peek_input ();
-
- if (current_char == '!')
+ int n = 1;
+ while (true)
{
- skip_input ();
- current_char = peek_input ();
-
- if (current_char == '/')
+ int next_char = peek_input (n);
+ if (is_whitespace (next_char))
+ n++;
+ else if (next_char == '/' && peek_input (n + 1) == '/')
{
- // definitely shebang
-
- skip_input ();
-
- // ignore rest of line
- while (current_char != '\n')
+ // A single line comment
+ n += 2;
+ next_char = peek_input (n);
+ while (next_char != '\n' && next_char != EOF)
+ {
+ n++;
+ next_char = peek_input (n);
+ }
+ if (next_char == '\n')
+ n++;
+ }
+ else if (next_char == '/' && peek_input (n + 1) == '*')
+ {
+ // Start of a block comment
+ n += 2;
+ int level = 1;
+ while (level > 0)
+ {
+ if (peek_input (n) == EOF)
+ break;
+ else if (peek_input (n) == '/'
+ && peek_input (n + 1) == '*')
+ {
+ n += 2;
+ level += 1;
+ }
+ else if (peek_input (n) == '*'
+ && peek_input (n + 1) == '/')
+ {
+ n += 2;
+ level -= 1;
+ }
+ else
+ n++;
+ }
+ }
+ else if (next_char != '[')
+ {
+ // definitely shebang, ignore the first line
+ while (current_char != '\n' && current_char != EOF)
{
current_char = peek_input ();
skip_input ();
@@ -269,11 +304,17 @
[PATCH 2/2] Remove has_shebang flag from AST and HIR Crate classes
The lexer deals with the shebang and the parser cannot detect whether
there is or isn't a shebang line. The flag isn't relevant or useful in
the AST and HIR Crate classes.
---
gcc/rust/ast/rust-ast-full-test.cc | 5 +
gcc/rust/ast/rust-ast.h | 13 +
gcc/rust/hir/rust-ast-lower.cc | 3 +--
gcc/rust/hir/tree/rust-hir-full-test.cc | 6 +-
gcc/rust/hir/tree/rust-hir.h| 14 +-
gcc/rust/parse/rust-parse-impl.h| 12 +---
6 files changed, 18 insertions(+), 35 deletions(-)
diff --git a/gcc/rust/ast/rust-ast-full-test.cc
b/gcc/rust/ast/rust-ast-full-test.cc
index 3d339ad1aed..12ef255bcbf 100644
--- a/gcc/rust/ast/rust-ast-full-test.cc
+++ b/gcc/rust/ast/rust-ast-full-test.cc
@@ -172,13 +172,10 @@ Crate::as_string () const
rust_debug ("beginning crate recursive as-string");
std::string str ("Crate: ");
- // add utf8bom and shebang
+ // add utf8bom
if (has_utf8bom)
str += "\n has utf8bom";
- if (has_shebang)
-str += "\n has shebang";
-
// inner attributes
str += append_attributes (inner_attrs, INNER);
diff --git a/gcc/rust/ast/rust-ast.h b/gcc/rust/ast/rust-ast.h
index 0e25de2be5c..ce55e1beb5e 100644
--- a/gcc/rust/ast/rust-ast.h
+++ b/gcc/rust/ast/rust-ast.h
@@ -1551,7 +1551,6 @@ protected:
struct Crate
{
bool has_utf8bom;
- bool has_shebang;
std::vector inner_attrs;
// dodgy spacing required here
@@ -1564,17 +1563,16 @@ struct Crate
public:
// Constructor
Crate (std::vector > items,
-std::vector inner_attrs, bool has_utf8bom = false,
-bool has_shebang = false)
-: has_utf8bom (has_utf8bom), has_shebang (has_shebang),
- inner_attrs (std::move (inner_attrs)), items (std::move (items)),
+std::vector inner_attrs, bool has_utf8bom = false)
+: has_utf8bom (has_utf8bom), inner_attrs (std::move (inner_attrs)),
+ items (std::move (items)),
node_id (Analysis::Mappings::get ()->get_next_node_id ())
{}
// Copy constructor with vector clone
Crate (Crate const &other)
-: has_utf8bom (other.has_utf8bom), has_shebang (other.has_shebang),
- inner_attrs (other.inner_attrs), node_id (other.node_id)
+: has_utf8bom (other.has_utf8bom), inner_attrs (other.inner_attrs),
+ node_id (other.node_id)
{
items.reserve (other.items.size ());
for (const auto &e : other.items)
@@ -1587,7 +1585,6 @@ public:
Crate &operator= (Crate const &other)
{
inner_attrs = other.inner_attrs;
-has_shebang = other.has_shebang;
has_utf8bom = other.has_utf8bom;
node_id = other.node_id;
diff --git a/gcc/rust/hir/rust-ast-lower.cc b/gcc/rust/hir/rust-ast-lower.cc
index c7222e2e998..0f3c86dc7bf 100644
--- a/gcc/rust/hir/rust-ast-lower.cc
+++ b/gcc/rust/hir/rust-ast-lower.cc
@@ -41,7 +41,6 @@ ASTLowering::go ()
{
std::vector > items;
bool has_utf8bom = false;
- bool has_shebang = false;
for (auto it = astCrate.items.begin (); it != astCrate.items.end (); it++)
{
@@ -57,7 +56,7 @@ ASTLowering::go ()
UNKNOWN_LOCAL_DEFID);
return HIR::Crate (std::move (items), astCrate.get_inner_attrs (), mapping,
-has_utf8bom, has_shebang);
+has_utf8bom);
}
// rust-ast-lower-block.h
diff --git a/gcc/rust/hir/tree/rust-hir-full-test.cc
b/gcc/rust/hir/tree/rust-hir-full-test.cc
index 261b3af672a..051ba8736ad 100644
--- a/gcc/rust/hir/tree/rust-hir-full-test.cc
+++ b/gcc/rust/hir/tree/rust-hir-full-test.cc
@@ -73,15 +73,11 @@ std::string
Crate::as_string () const
{
std::string str ("HIR::Crate: ");
- // add utf8bom and shebang
+ // add utf8bom
if (has_utf8bom)
{
str += "\n has utf8bom";
}
- if (has_shebang)
-{
- str += "\n has shebang";
-}
// inner attributes
str += "\n inner attributes: ";
diff --git a/gcc/rust/hir/tree/rust-hir.h b/gcc/rust/hir/tree/rust-hir.h
index 35dc71ae1df..f918f2dc106 100644
--- a/gcc/rust/hir/tree/rust-hir.h
+++ b/gcc/rust/hir/tree/rust-hir.h
@@ -679,7 +679,6 @@ public:
struct Crate
{
bool has_utf8bom;
- bool has_shebang;
AST::AttrVec inner_attrs;
// dodgy spacing required here
@@ -692,17 +691,15 @@ struct Crate
public:
// Constructor
Crate (std::vector > items, AST::AttrVec inner_attrs,
-Analysis::NodeMapping mappings, bool has_utf8bom = false,
-bool has_shebang = false)
-: has_utf8bom (has_utf8bom), has_shebang (has_shebang),
- inner_attrs (std::move (inner_attrs)), items (std::move (items)),
- mappings (mappings)
+Analysis::NodeMapping mappings, bool has_utf8bom = false)
+: has_utf8bom (has_utf8bom), inner_attrs (std::move (inner_attrs)),
+ items (std::move (items)), mappings (mappings)
{}
// Copy constructor with vector clone
Crate (Crate const &other)
-: has_utf8bom (other.has_utf8bom), has_shebang (other.has_shebang),
- inner_attrs (other.inner_attrs), mappings (ot
Re: shebang handling
Mark Wielaard writes: > [PATCH 1/2] Handle shebang line, plus any whitespace and comment > [PATCH 2/2] Remove has_shebang flag from AST and HIR Crate classes Pushed to githup: https://github.com/Rust-GCC/gccrs/pull/546 Marc -- Gcc-rust mailing list Gcc-rust@gcc.gnu.org https://gcc.gnu.org/mailman/listinfo/gcc-rust
