2021-07-30 17:53:27 +00:00
|
|
|
package php8_test
|
|
|
|
|
|
|
|
import (
|
|
|
|
"testing"
|
|
|
|
|
2021-07-30 18:01:34 +00:00
|
|
|
"github.com/VKCOM/php-parser/internal/php8"
|
|
|
|
"github.com/VKCOM/php-parser/internal/tester"
|
|
|
|
"github.com/VKCOM/php-parser/pkg/conf"
|
php8: fixed a bug with `#` comments (#22)
Example from issue:
```php
<?php
#
# Comment
#
$a = 100;
```
The problem with the example from the issue is that `#` is immediately followed by a line break.
And since the rule in the lexer for such comments was changed, this case was handled incorrectly.
```
(('#' ^'[') | '//') any_line* when is_not_comment_end => {
lex.ungetStr("?>")
lex.addFreeFloatingToken(tkn, token.T_COMMENT, lex.ts, lex.te)
};
```
This rule has one problem, it checks two characters at once, first for the match `#`, and
then for the mismatch `[`, which leads to the fact that in the case of an empty comment, the first
matcher will capture `#`, and the second line break (`\n`), which will lead to the fact that `any_line`
matcher will not work and will not increase the line number.
The next rule added is specifically for this case.
```
'#' newline when is_not_comment_end => {
lex.ungetStr("?>")
lex.addFreeFloatingToken(tkn, token.T_COMMENT, lex.ts, lex.te)
};
```
2021-08-02 09:37:08 +00:00
|
|
|
"github.com/VKCOM/php-parser/pkg/position"
|
2021-07-30 18:01:34 +00:00
|
|
|
"github.com/VKCOM/php-parser/pkg/token"
|
2021-07-30 17:53:27 +00:00
|
|
|
"gotest.tools/assert"
|
|
|
|
)
|
|
|
|
|
|
|
|
func TestNullsafeMethodCallTokens(t *testing.T) {
|
|
|
|
suite := tester.NewLexerTokenStringTestSuite(t)
|
|
|
|
suite.UsePHP8()
|
|
|
|
suite.Code = "<?php $a?->foo();"
|
|
|
|
suite.Expected = []string{
|
|
|
|
"$a",
|
|
|
|
"?->",
|
|
|
|
"foo",
|
|
|
|
"(",
|
|
|
|
")",
|
|
|
|
";",
|
|
|
|
}
|
|
|
|
suite.Run()
|
|
|
|
}
|
|
|
|
|
|
|
|
func TestNullsafePropertyFetchTokens(t *testing.T) {
|
|
|
|
suite := tester.NewLexerTokenStringTestSuite(t)
|
|
|
|
suite.UsePHP8()
|
|
|
|
suite.Code = "<?php $a?->prop;"
|
|
|
|
suite.Expected = []string{
|
|
|
|
"$a",
|
|
|
|
"?->",
|
|
|
|
"prop",
|
|
|
|
";",
|
|
|
|
}
|
|
|
|
suite.Run()
|
|
|
|
}
|
|
|
|
|
|
|
|
func TestNullsafePropertyFetchInStringTokens(t *testing.T) {
|
|
|
|
suite := tester.NewLexerTokenStringTestSuite(t)
|
|
|
|
suite.UsePHP8()
|
|
|
|
suite.Code = "<?php \"$a?->prop\";"
|
|
|
|
suite.Expected = []string{
|
|
|
|
"\"",
|
|
|
|
"$a",
|
|
|
|
"?->",
|
|
|
|
"prop",
|
|
|
|
"\"",
|
|
|
|
";",
|
|
|
|
}
|
|
|
|
suite.Run()
|
|
|
|
}
|
|
|
|
|
|
|
|
func TestNullsafeMethodCallTokensFreeFloating(t *testing.T) {
|
|
|
|
suite := tester.NewLexerTokenFreeFloatingTestSuite(t)
|
|
|
|
suite.UsePHP8()
|
|
|
|
suite.Code = `<?php
|
|
|
|
$a ?-> bar ( '' ) ;`
|
|
|
|
|
|
|
|
suite.Expected = [][]*token.Token{
|
|
|
|
{
|
|
|
|
{
|
|
|
|
ID: token.T_OPEN_TAG,
|
|
|
|
Value: []byte("<?php"),
|
|
|
|
},
|
|
|
|
{
|
|
|
|
ID: token.T_WHITESPACE,
|
|
|
|
Value: []byte("\n\t"),
|
|
|
|
},
|
|
|
|
},
|
|
|
|
{
|
|
|
|
{
|
|
|
|
ID: token.T_WHITESPACE,
|
|
|
|
Value: []byte(" "),
|
|
|
|
},
|
|
|
|
},
|
|
|
|
{
|
|
|
|
{
|
|
|
|
ID: token.T_WHITESPACE,
|
|
|
|
Value: []byte(" "),
|
|
|
|
},
|
|
|
|
},
|
|
|
|
{
|
|
|
|
{
|
|
|
|
ID: token.T_WHITESPACE,
|
|
|
|
Value: []byte(" "),
|
|
|
|
},
|
|
|
|
},
|
|
|
|
{
|
|
|
|
{
|
|
|
|
ID: token.T_WHITESPACE,
|
|
|
|
Value: []byte(" "),
|
|
|
|
},
|
|
|
|
},
|
|
|
|
{
|
|
|
|
{
|
|
|
|
ID: token.T_WHITESPACE,
|
|
|
|
Value: []byte(" "),
|
|
|
|
},
|
|
|
|
},
|
|
|
|
{
|
|
|
|
{
|
|
|
|
ID: token.T_WHITESPACE,
|
|
|
|
Value: []byte(" "),
|
|
|
|
},
|
|
|
|
},
|
|
|
|
}
|
|
|
|
|
|
|
|
suite.Run()
|
|
|
|
}
|
|
|
|
|
|
|
|
func TestMatchStringTokens(t *testing.T) {
|
|
|
|
suite := tester.NewLexerTokenStringTestSuite(t)
|
|
|
|
suite.UsePHP8()
|
|
|
|
suite.Code = "<?php match($a) {}"
|
|
|
|
suite.Expected = []string{
|
|
|
|
"match",
|
|
|
|
"(",
|
|
|
|
"$a",
|
|
|
|
")",
|
|
|
|
"{",
|
|
|
|
"}",
|
|
|
|
}
|
|
|
|
suite.Run()
|
|
|
|
}
|
|
|
|
|
|
|
|
func TestMatchWithConditionStringTokens(t *testing.T) {
|
|
|
|
suite := tester.NewLexerTokenStringTestSuite(t)
|
|
|
|
suite.UsePHP8()
|
|
|
|
suite.Code = "<?php match($a) { 10 => 100 }"
|
|
|
|
suite.Expected = []string{
|
|
|
|
"match",
|
|
|
|
"(",
|
|
|
|
"$a",
|
|
|
|
")",
|
|
|
|
"{",
|
|
|
|
"10",
|
|
|
|
"=>",
|
|
|
|
"100",
|
|
|
|
"}",
|
|
|
|
}
|
|
|
|
suite.Run()
|
|
|
|
}
|
|
|
|
|
|
|
|
func TestMatchWithDefaultStringTokens(t *testing.T) {
|
|
|
|
suite := tester.NewLexerTokenStringTestSuite(t)
|
|
|
|
suite.UsePHP8()
|
|
|
|
suite.Code = "<?php match($a) { default => 10 }"
|
|
|
|
suite.Expected = []string{
|
|
|
|
"match",
|
|
|
|
"(",
|
|
|
|
"$a",
|
|
|
|
")",
|
|
|
|
"{",
|
|
|
|
"default",
|
|
|
|
"=>",
|
|
|
|
"10",
|
|
|
|
"}",
|
|
|
|
}
|
|
|
|
suite.Run()
|
|
|
|
}
|
|
|
|
|
|
|
|
func TestAttributeTokens(t *testing.T) {
|
|
|
|
suite := tester.NewLexerTokenStringTestSuite(t)
|
|
|
|
suite.UsePHP8()
|
|
|
|
suite.Code = "<?php #[ FooAttribute]"
|
|
|
|
suite.Expected = []string{
|
|
|
|
"#[",
|
|
|
|
"FooAttribute",
|
|
|
|
"]",
|
|
|
|
}
|
|
|
|
suite.Run()
|
|
|
|
}
|
|
|
|
|
|
|
|
func TestCommentEnd2(t *testing.T) {
|
|
|
|
src := `<?php //`
|
|
|
|
|
|
|
|
expected := []*token.Token{
|
|
|
|
{
|
|
|
|
ID: token.T_OPEN_TAG,
|
|
|
|
Value: []byte("<?php"),
|
|
|
|
},
|
|
|
|
|
|
|
|
{
|
|
|
|
ID: token.T_WHITESPACE,
|
|
|
|
Value: []byte(" "),
|
|
|
|
},
|
|
|
|
{
|
|
|
|
ID: token.T_COMMENT,
|
|
|
|
Value: []byte("//"),
|
|
|
|
},
|
|
|
|
}
|
|
|
|
|
|
|
|
lexer := php8.NewLexer([]byte(src), conf.Config{})
|
|
|
|
|
|
|
|
tkn := lexer.Lex()
|
|
|
|
|
|
|
|
actual := tkn.FreeFloating
|
|
|
|
for _, v := range actual {
|
|
|
|
v.Position = nil
|
|
|
|
}
|
|
|
|
|
|
|
|
assert.DeepEqual(t, expected, actual)
|
|
|
|
}
|
|
|
|
|
|
|
|
func TestCommentEnd3(t *testing.T) {
|
|
|
|
src := `<?php #`
|
|
|
|
|
|
|
|
expected := []*token.Token{
|
|
|
|
{
|
|
|
|
ID: token.T_OPEN_TAG,
|
|
|
|
Value: []byte("<?php"),
|
|
|
|
},
|
|
|
|
|
|
|
|
{
|
|
|
|
ID: token.T_WHITESPACE,
|
|
|
|
Value: []byte(" "),
|
|
|
|
},
|
|
|
|
{
|
|
|
|
ID: token.T_COMMENT,
|
|
|
|
Value: []byte("#"),
|
|
|
|
},
|
|
|
|
}
|
|
|
|
|
|
|
|
lexer := php8.NewLexer([]byte(src), conf.Config{})
|
|
|
|
|
|
|
|
tkn := lexer.Lex()
|
|
|
|
|
|
|
|
actual := tkn.FreeFloating
|
|
|
|
for _, v := range actual {
|
|
|
|
v.Position = nil
|
|
|
|
}
|
|
|
|
|
|
|
|
assert.DeepEqual(t, expected, actual)
|
|
|
|
}
|
|
|
|
|
|
|
|
func TestAttribute(t *testing.T) {
|
|
|
|
src := `<?php #[ FooAttribute]`
|
|
|
|
|
|
|
|
expected := []*token.Token{
|
|
|
|
{
|
|
|
|
ID: token.T_OPEN_TAG,
|
|
|
|
Value: []byte("<?php"),
|
|
|
|
},
|
|
|
|
{
|
|
|
|
ID: token.T_WHITESPACE,
|
|
|
|
Value: []byte(" "),
|
|
|
|
},
|
|
|
|
}
|
|
|
|
|
|
|
|
lexer := php8.NewLexer([]byte(src), conf.Config{})
|
|
|
|
|
|
|
|
tkn := lexer.Lex()
|
|
|
|
|
|
|
|
actual := tkn.FreeFloating
|
|
|
|
for _, v := range actual {
|
|
|
|
v.Position = nil
|
|
|
|
}
|
|
|
|
|
|
|
|
assert.DeepEqual(t, expected, actual)
|
|
|
|
|
|
|
|
assert.DeepEqual(t, token.T_ATTRIBUTE, tkn.ID)
|
|
|
|
assert.DeepEqual(t, "#[", string(tkn.Value))
|
|
|
|
}
|
|
|
|
|
|
|
|
func TestNamespaceFullyQualifiedTokens(t *testing.T) {
|
|
|
|
suite := tester.NewLexerTokenStructTestSuite(t)
|
|
|
|
suite.UsePHP8()
|
|
|
|
suite.Code = `<?php use \Foo;`
|
|
|
|
suite.Expected = []*token.Token{
|
|
|
|
{
|
|
|
|
ID: php8.T_USE,
|
|
|
|
Value: []byte("use"),
|
|
|
|
},
|
|
|
|
{
|
|
|
|
ID: php8.T_NAME_FULLY_QUALIFIED,
|
|
|
|
Value: []byte(`\Foo`),
|
|
|
|
},
|
|
|
|
{
|
|
|
|
ID: ';',
|
|
|
|
Value: []byte(";"),
|
|
|
|
},
|
|
|
|
}
|
|
|
|
suite.Run()
|
|
|
|
}
|
|
|
|
|
|
|
|
func TestNamespaceFullyQualifiedWithKeywordsTokens(t *testing.T) {
|
|
|
|
suite := tester.NewLexerTokenStructTestSuite(t)
|
|
|
|
suite.UsePHP8()
|
|
|
|
suite.Code = `<?php use \Foo\match\fn;`
|
|
|
|
suite.Expected = []*token.Token{
|
|
|
|
{
|
|
|
|
ID: php8.T_USE,
|
|
|
|
Value: []byte("use"),
|
|
|
|
},
|
|
|
|
{
|
|
|
|
ID: php8.T_NAME_FULLY_QUALIFIED,
|
|
|
|
Value: []byte(`\Foo\match\fn`),
|
|
|
|
},
|
|
|
|
{
|
|
|
|
ID: ';',
|
|
|
|
Value: []byte(";"),
|
|
|
|
},
|
|
|
|
}
|
|
|
|
suite.Run()
|
|
|
|
}
|
|
|
|
|
|
|
|
func TestNamespaceQualifiedTokens(t *testing.T) {
|
|
|
|
suite := tester.NewLexerTokenStructTestSuite(t)
|
|
|
|
suite.UsePHP8()
|
|
|
|
suite.Code = `<?php namespace Boo\Foo;`
|
|
|
|
suite.Expected = []*token.Token{
|
|
|
|
{
|
|
|
|
ID: php8.T_NAMESPACE,
|
|
|
|
Value: []byte("namespace"),
|
|
|
|
},
|
|
|
|
{
|
|
|
|
ID: php8.T_NAME_QUALIFIED,
|
|
|
|
Value: []byte(`Boo\Foo`),
|
|
|
|
},
|
|
|
|
{
|
|
|
|
ID: ';',
|
|
|
|
Value: []byte(";"),
|
|
|
|
},
|
|
|
|
}
|
|
|
|
suite.Run()
|
|
|
|
}
|
|
|
|
|
|
|
|
func TestNamespaceRelativeTokens(t *testing.T) {
|
|
|
|
suite := tester.NewLexerTokenStructTestSuite(t)
|
|
|
|
suite.UsePHP8()
|
|
|
|
suite.Code = `<?php namespace\match;`
|
|
|
|
suite.Expected = []*token.Token{
|
|
|
|
{
|
|
|
|
ID: php8.T_NAME_RELATIVE,
|
|
|
|
Value: []byte(`namespace\match`),
|
|
|
|
},
|
|
|
|
{
|
|
|
|
ID: ';',
|
|
|
|
Value: []byte(";"),
|
|
|
|
},
|
|
|
|
}
|
|
|
|
suite.Run()
|
|
|
|
}
|
php8: fixed a bug with `#` comments (#22)
Example from issue:
```php
<?php
#
# Comment
#
$a = 100;
```
The problem with the example from the issue is that `#` is immediately followed by a line break.
And since the rule in the lexer for such comments was changed, this case was handled incorrectly.
```
(('#' ^'[') | '//') any_line* when is_not_comment_end => {
lex.ungetStr("?>")
lex.addFreeFloatingToken(tkn, token.T_COMMENT, lex.ts, lex.te)
};
```
This rule has one problem, it checks two characters at once, first for the match `#`, and
then for the mismatch `[`, which leads to the fact that in the case of an empty comment, the first
matcher will capture `#`, and the second line break (`\n`), which will lead to the fact that `any_line`
matcher will not work and will not increase the line number.
The next rule added is specifically for this case.
```
'#' newline when is_not_comment_end => {
lex.ungetStr("?>")
lex.addFreeFloatingToken(tkn, token.T_COMMENT, lex.ts, lex.te)
};
```
2021-08-02 09:37:08 +00:00
|
|
|
|
|
|
|
func TestSharpCommentTokens(t *testing.T) {
|
|
|
|
suite := tester.NewLexerTokenStructTestSuite(t)
|
|
|
|
suite.UsePHP8()
|
|
|
|
suite.WithPosition()
|
|
|
|
suite.WithFreeFloating()
|
|
|
|
suite.Code = `<?php
|
|
|
|
#
|
|
|
|
# Hello
|
|
|
|
#
|
|
|
|
|
|
|
|
$a;
|
|
|
|
`
|
|
|
|
suite.Expected = []*token.Token{
|
|
|
|
{
|
|
|
|
ID: php8.T_VARIABLE,
|
|
|
|
Value: []byte(`$a`),
|
|
|
|
FreeFloating: []*token.Token{
|
|
|
|
{
|
|
|
|
ID: php8.T_OPEN_TAG,
|
|
|
|
Value: []byte("<?php"),
|
2023-03-26 00:54:00 +00:00
|
|
|
Position: &position.Position{StartLine: 1, EndLine: 1, EndPos: 5, StartCol: 0, EndCol: 5},
|
php8: fixed a bug with `#` comments (#22)
Example from issue:
```php
<?php
#
# Comment
#
$a = 100;
```
The problem with the example from the issue is that `#` is immediately followed by a line break.
And since the rule in the lexer for such comments was changed, this case was handled incorrectly.
```
(('#' ^'[') | '//') any_line* when is_not_comment_end => {
lex.ungetStr("?>")
lex.addFreeFloatingToken(tkn, token.T_COMMENT, lex.ts, lex.te)
};
```
This rule has one problem, it checks two characters at once, first for the match `#`, and
then for the mismatch `[`, which leads to the fact that in the case of an empty comment, the first
matcher will capture `#`, and the second line break (`\n`), which will lead to the fact that `any_line`
matcher will not work and will not increase the line number.
The next rule added is specifically for this case.
```
'#' newline when is_not_comment_end => {
lex.ungetStr("?>")
lex.addFreeFloatingToken(tkn, token.T_COMMENT, lex.ts, lex.te)
};
```
2021-08-02 09:37:08 +00:00
|
|
|
},
|
|
|
|
{
|
|
|
|
ID: php8.T_WHITESPACE,
|
|
|
|
Value: []byte("\n"),
|
2023-03-26 00:54:00 +00:00
|
|
|
Position: &position.Position{StartLine: 1, EndLine: 1, StartPos: 5, EndPos: 6, StartCol: 5, EndCol: 6},
|
php8: fixed a bug with `#` comments (#22)
Example from issue:
```php
<?php
#
# Comment
#
$a = 100;
```
The problem with the example from the issue is that `#` is immediately followed by a line break.
And since the rule in the lexer for such comments was changed, this case was handled incorrectly.
```
(('#' ^'[') | '//') any_line* when is_not_comment_end => {
lex.ungetStr("?>")
lex.addFreeFloatingToken(tkn, token.T_COMMENT, lex.ts, lex.te)
};
```
This rule has one problem, it checks two characters at once, first for the match `#`, and
then for the mismatch `[`, which leads to the fact that in the case of an empty comment, the first
matcher will capture `#`, and the second line break (`\n`), which will lead to the fact that `any_line`
matcher will not work and will not increase the line number.
The next rule added is specifically for this case.
```
'#' newline when is_not_comment_end => {
lex.ungetStr("?>")
lex.addFreeFloatingToken(tkn, token.T_COMMENT, lex.ts, lex.te)
};
```
2021-08-02 09:37:08 +00:00
|
|
|
},
|
|
|
|
{
|
|
|
|
ID: php8.T_COMMENT,
|
|
|
|
Value: []byte("#\n"),
|
2023-03-26 00:54:00 +00:00
|
|
|
Position: &position.Position{StartLine: 2, EndLine: 2, StartPos: 6, EndPos: 8, StartCol: 0, EndCol: 2},
|
php8: fixed a bug with `#` comments (#22)
Example from issue:
```php
<?php
#
# Comment
#
$a = 100;
```
The problem with the example from the issue is that `#` is immediately followed by a line break.
And since the rule in the lexer for such comments was changed, this case was handled incorrectly.
```
(('#' ^'[') | '//') any_line* when is_not_comment_end => {
lex.ungetStr("?>")
lex.addFreeFloatingToken(tkn, token.T_COMMENT, lex.ts, lex.te)
};
```
This rule has one problem, it checks two characters at once, first for the match `#`, and
then for the mismatch `[`, which leads to the fact that in the case of an empty comment, the first
matcher will capture `#`, and the second line break (`\n`), which will lead to the fact that `any_line`
matcher will not work and will not increase the line number.
The next rule added is specifically for this case.
```
'#' newline when is_not_comment_end => {
lex.ungetStr("?>")
lex.addFreeFloatingToken(tkn, token.T_COMMENT, lex.ts, lex.te)
};
```
2021-08-02 09:37:08 +00:00
|
|
|
},
|
|
|
|
{
|
|
|
|
ID: php8.T_COMMENT,
|
|
|
|
Value: []byte("# Hello\n"),
|
2023-03-26 00:54:00 +00:00
|
|
|
Position: &position.Position{StartLine: 3, EndLine: 3, StartPos: 8, EndPos: 16, StartCol: 0, EndCol: 8},
|
php8: fixed a bug with `#` comments (#22)
Example from issue:
```php
<?php
#
# Comment
#
$a = 100;
```
The problem with the example from the issue is that `#` is immediately followed by a line break.
And since the rule in the lexer for such comments was changed, this case was handled incorrectly.
```
(('#' ^'[') | '//') any_line* when is_not_comment_end => {
lex.ungetStr("?>")
lex.addFreeFloatingToken(tkn, token.T_COMMENT, lex.ts, lex.te)
};
```
This rule has one problem, it checks two characters at once, first for the match `#`, and
then for the mismatch `[`, which leads to the fact that in the case of an empty comment, the first
matcher will capture `#`, and the second line break (`\n`), which will lead to the fact that `any_line`
matcher will not work and will not increase the line number.
The next rule added is specifically for this case.
```
'#' newline when is_not_comment_end => {
lex.ungetStr("?>")
lex.addFreeFloatingToken(tkn, token.T_COMMENT, lex.ts, lex.te)
};
```
2021-08-02 09:37:08 +00:00
|
|
|
},
|
|
|
|
{
|
|
|
|
ID: php8.T_COMMENT,
|
|
|
|
Value: []byte("#\n"),
|
2023-03-26 00:54:00 +00:00
|
|
|
Position: &position.Position{StartLine: 4, EndLine: 4, StartPos: 16, EndPos: 18, StartCol: 0, EndCol: 2},
|
php8: fixed a bug with `#` comments (#22)
Example from issue:
```php
<?php
#
# Comment
#
$a = 100;
```
The problem with the example from the issue is that `#` is immediately followed by a line break.
And since the rule in the lexer for such comments was changed, this case was handled incorrectly.
```
(('#' ^'[') | '//') any_line* when is_not_comment_end => {
lex.ungetStr("?>")
lex.addFreeFloatingToken(tkn, token.T_COMMENT, lex.ts, lex.te)
};
```
This rule has one problem, it checks two characters at once, first for the match `#`, and
then for the mismatch `[`, which leads to the fact that in the case of an empty comment, the first
matcher will capture `#`, and the second line break (`\n`), which will lead to the fact that `any_line`
matcher will not work and will not increase the line number.
The next rule added is specifically for this case.
```
'#' newline when is_not_comment_end => {
lex.ungetStr("?>")
lex.addFreeFloatingToken(tkn, token.T_COMMENT, lex.ts, lex.te)
};
```
2021-08-02 09:37:08 +00:00
|
|
|
},
|
|
|
|
{
|
|
|
|
ID: php8.T_WHITESPACE,
|
|
|
|
Value: []byte("\n"),
|
2023-03-26 00:54:00 +00:00
|
|
|
Position: &position.Position{StartLine: 5, EndLine: 5, StartPos: 18, EndPos: 19, StartCol: 0, EndCol: 1},
|
php8: fixed a bug with `#` comments (#22)
Example from issue:
```php
<?php
#
# Comment
#
$a = 100;
```
The problem with the example from the issue is that `#` is immediately followed by a line break.
And since the rule in the lexer for such comments was changed, this case was handled incorrectly.
```
(('#' ^'[') | '//') any_line* when is_not_comment_end => {
lex.ungetStr("?>")
lex.addFreeFloatingToken(tkn, token.T_COMMENT, lex.ts, lex.te)
};
```
This rule has one problem, it checks two characters at once, first for the match `#`, and
then for the mismatch `[`, which leads to the fact that in the case of an empty comment, the first
matcher will capture `#`, and the second line break (`\n`), which will lead to the fact that `any_line`
matcher will not work and will not increase the line number.
The next rule added is specifically for this case.
```
'#' newline when is_not_comment_end => {
lex.ungetStr("?>")
lex.addFreeFloatingToken(tkn, token.T_COMMENT, lex.ts, lex.te)
};
```
2021-08-02 09:37:08 +00:00
|
|
|
},
|
|
|
|
},
|
2023-03-26 00:54:00 +00:00
|
|
|
Position: &position.Position{StartLine: 6, EndLine: 6, StartPos: 19, EndPos: 21, StartCol: 0, EndCol: 2},
|
php8: fixed a bug with `#` comments (#22)
Example from issue:
```php
<?php
#
# Comment
#
$a = 100;
```
The problem with the example from the issue is that `#` is immediately followed by a line break.
And since the rule in the lexer for such comments was changed, this case was handled incorrectly.
```
(('#' ^'[') | '//') any_line* when is_not_comment_end => {
lex.ungetStr("?>")
lex.addFreeFloatingToken(tkn, token.T_COMMENT, lex.ts, lex.te)
};
```
This rule has one problem, it checks two characters at once, first for the match `#`, and
then for the mismatch `[`, which leads to the fact that in the case of an empty comment, the first
matcher will capture `#`, and the second line break (`\n`), which will lead to the fact that `any_line`
matcher will not work and will not increase the line number.
The next rule added is specifically for this case.
```
'#' newline when is_not_comment_end => {
lex.ungetStr("?>")
lex.addFreeFloatingToken(tkn, token.T_COMMENT, lex.ts, lex.te)
};
```
2021-08-02 09:37:08 +00:00
|
|
|
},
|
|
|
|
{
|
|
|
|
ID: ';',
|
|
|
|
Value: []byte(";"),
|
2023-03-26 00:54:00 +00:00
|
|
|
Position: &position.Position{StartLine: 6, EndLine: 6, StartPos: 21, EndPos: 22, StartCol: 2, EndCol: 3},
|
php8: fixed a bug with `#` comments (#22)
Example from issue:
```php
<?php
#
# Comment
#
$a = 100;
```
The problem with the example from the issue is that `#` is immediately followed by a line break.
And since the rule in the lexer for such comments was changed, this case was handled incorrectly.
```
(('#' ^'[') | '//') any_line* when is_not_comment_end => {
lex.ungetStr("?>")
lex.addFreeFloatingToken(tkn, token.T_COMMENT, lex.ts, lex.te)
};
```
This rule has one problem, it checks two characters at once, first for the match `#`, and
then for the mismatch `[`, which leads to the fact that in the case of an empty comment, the first
matcher will capture `#`, and the second line break (`\n`), which will lead to the fact that `any_line`
matcher will not work and will not increase the line number.
The next rule added is specifically for this case.
```
'#' newline when is_not_comment_end => {
lex.ungetStr("?>")
lex.addFreeFloatingToken(tkn, token.T_COMMENT, lex.ts, lex.te)
};
```
2021-08-02 09:37:08 +00:00
|
|
|
},
|
|
|
|
}
|
|
|
|
suite.Run()
|
|
|
|
}
|