Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

fix(xml/parser): fix comment parsing #6449

Merged
merged 2 commits into from Nov 18, 2022
Merged

fix(xml/parser): fix comment parsing #6449

Show file tree
Hide file tree
Changes from 1 commit
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
9850b24
fix(xml/parser): comment parsing
alexander-akait Nov 15, 2022
6c03b29
Merge branch 'main' into fix-comment-xml-parsing
swc-bot Nov 18, 2022
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Jump to
Jump to file
Failed to load files.
Diff view
Diff view
2 changes: 2 additions & 0 deletions crates/swc_xml_parser/src/error.rs
View file
Open in desktop
Expand Up @@ -81,6 +81,7 @@ impl Error {
"Missing whitespace between doctype public and system identifiers".into()
}
ErrorKind::NestedComment => "Nested comment".into(),
ErrorKind::DoubleHyphenWithInComment => "Double hyper within comment".into(),
ErrorKind::NoncharacterInInputStream => "Noncharacter in input stream".into(),
ErrorKind::SurrogateInInputStream => "Surrogate in input stream".into(),
ErrorKind::SurrogateCharacterReference => "Surrogate character reference".into(),
Expand Down Expand Up @@ -146,6 +147,7 @@ pub enum ErrorKind {
MissingWhitespaceBeforeDoctypeName,
MissingWhitespaceBetweenDoctypePublicAndSystemIdentifiers,
NestedComment,
DoubleHyphenWithInComment,
NoncharacterInInputStream,
SurrogateInInputStream,
SurrogateCharacterReference,
Expand Down
1 change: 1 addition & 0 deletions crates/swc_xml_parser/src/lexer/mod.rs
View file
Open in desktop
Expand Up @@ -1620,6 +1620,7 @@ where
// Append a U+002D HYPHEN-MINUS character (-) to the comment token's data.
Some(c @ '-') => {
self.append_to_comment_token(c, c);
self.emit_error(ErrorKind::DoubleHyphenWithInComment);
}
// EOF
// This is an eof-in-comment parse error. Emit the current comment token.
Expand Down
43 changes: 43 additions & 0 deletions crates/swc_xml_parser/tests/fixture/comments/dom.rust-debug
View file
Open in desktop
@@ -0,0 +1,43 @@
<?xml version="1.0" encoding="UTF-8">
| <root>
| "
"
| <!-- comment -->
| "
"
| <!-- <head> -->
| "
"
| <!-- <!-x -->
| "
"
| <!-- <!x -->
| "
"
| <!-- <<!x -->
| "
"
| <!-- <<!-x -->
| "
"
| <!-- <x -->
| "
"
| <!-- <> -->
| "
"
| <!-- < -->
| "
"
| <!-- <! -->
| "
"
| <!-- -->
| "
"
| <!-- -a->-a -->
| "
"
| <!-- -<!-test-> -->
| "
"
17 changes: 17 additions & 0 deletions crates/swc_xml_parser/tests/fixture/comments/input.xml
View file
Open in desktop
@@ -0,0 +1,17 @@
<?xml version="1.0" encoding="UTF-8"?>
<root>
<!--comment-->
<!--<head>-->
<!--<!-x-->
<!--<!x-->
<!--<<!x-->
<!--<<!-x-->
<!--<x-->
<!--<>-->
<!--<-->
<!--<!-->
<!---->
<!---a->-a-->
<!---<!-test->-->
</root>

302 changes: 302 additions & 0 deletions crates/swc_xml_parser/tests/fixture/comments/output.json
View file
Open in desktop
@@ -0,0 +1,302 @@
{
"type": "Document",
"span": {
"start": 1,
"end": 264,
"ctxt": 0
},
"children": [
{
"type": "ProcessingInstruction",
"span": {
"start": 1,
"end": 39,
"ctxt": 0
},
"target": "xml",
"data": "version=\"1.0\" encoding=\"UTF-8\""
},
{
"type": "Element",
"span": {
"start": 40,
"end": 262,
"ctxt": 0
},
"tagName": "root",
"attributes": [],
"children": [
{
"type": "Text",
"span": {
"start": 46,
"end": 51,
"ctxt": 0
},
"data": "\n ",
"raw": "\n "
},
{
"type": "Comment",
"span": {
"start": 51,
"end": 65,
"ctxt": 0
},
"data": "comment",
"raw": "<!--comment-->"
},
{
"type": "Text",
"span": {
"start": 65,
"end": 70,
"ctxt": 0
},
"data": "\n ",
"raw": "\n "
},
{
"type": "Comment",
"span": {
"start": 70,
"end": 83,
"ctxt": 0
},
"data": "<head>",
"raw": "<!--<head>-->"
},
{
"type": "Text",
"span": {
"start": 83,
"end": 88,
"ctxt": 0
},
"data": "\n ",
"raw": "\n "
},
{
"type": "Comment",
"span": {
"start": 88,
"end": 99,
"ctxt": 0
},
"data": "<!-x",
"raw": "<!--<!-x-->"
},
{
"type": "Text",
"span": {
"start": 99,
"end": 104,
"ctxt": 0
},
"data": "\n ",
"raw": "\n "
},
{
"type": "Comment",
"span": {
"start": 104,
"end": 114,
"ctxt": 0
},
"data": "<!x",
"raw": "<!--<!x-->"
},
{
"type": "Text",
"span": {
"start": 114,
"end": 119,
"ctxt": 0
},
"data": "\n ",
"raw": "\n "
},
{
"type": "Comment",
"span": {
"start": 119,
"end": 130,
"ctxt": 0
},
"data": "<<!x",
"raw": "<!--<<!x-->"
},
{
"type": "Text",
"span": {
"start": 130,
"end": 135,
"ctxt": 0
},
"data": "\n ",
"raw": "\n "
},
{
"type": "Comment",
"span": {
"start": 135,
"end": 147,
"ctxt": 0
},
"data": "<<!-x",
"raw": "<!--<<!-x-->"
},
{
"type": "Text",
"span": {
"start": 147,
"end": 152,
"ctxt": 0
},
"data": "\n ",
"raw": "\n "
},
{
"type": "Comment",
"span": {
"start": 152,
"end": 161,
"ctxt": 0
},
"data": "<x",
"raw": "<!--<x-->"
},
{
"type": "Text",
"span": {
"start": 161,
"end": 166,
"ctxt": 0
},
"data": "\n ",
"raw": "\n "
},
{
"type": "Comment",
"span": {
"start": 166,
"end": 175,
"ctxt": 0
},
"data": "<>",
"raw": "<!--<>-->"
},
{
"type": "Text",
"span": {
"start": 175,
"end": 180,
"ctxt": 0
},
"data": "\n ",
"raw": "\n "
},
{
"type": "Comment",
"span": {
"start": 180,
"end": 188,
"ctxt": 0
},
"data": "<",
"raw": "<!--<-->"
},
{
"type": "Text",
"span": {
"start": 188,
"end": 193,
"ctxt": 0
},
"data": "\n ",
"raw": "\n "
},
{
"type": "Comment",
"span": {
"start": 193,
"end": 202,
"ctxt": 0
},
"data": "<!",
"raw": "<!--<!-->"
},
{
"type": "Text",
"span": {
"start": 202,
"end": 207,
"ctxt": 0
},
"data": "\n ",
"raw": "\n "
},
{
"type": "Comment",
"span": {
"start": 207,
"end": 214,
"ctxt": 0
},
"data": "",
"raw": "<!---->"
},
{
"type": "Text",
"span": {
"start": 214,
"end": 219,
"ctxt": 0
},
"data": "\n ",
"raw": "\n "
},
{
"type": "Comment",
"span": {
"start": 219,
"end": 232,
"ctxt": 0
},
"data": "-a->-a",
"raw": "<!---a->-a-->"
},
{
"type": "Text",
"span": {
"start": 232,
"end": 237,
"ctxt": 0
},
"data": "\n ",
"raw": "\n "
},
{
"type": "Comment",
"span": {
"start": 237,
"end": 254,
"ctxt": 0
},
"data": "-<!-test->",
"raw": "<!---<!-test->-->"
},
{
"type": "Text",
"span": {
"start": 254,
"end": 255,
"ctxt": 0
},
"data": "\n",
"raw": "\n"
}
]
}
]
}