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): parse legacy document type #6555

Merged
merged 2 commits into from
Dec 2, 2022
Merged

fix(xml/parser): parse legacy document type #6555

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
ec285bb
fix(xml/parser): parse legacy document type
alexander-akait Nov 30, 2022
21dbc9c
Merge branch 'main' into xml-fix-legacy-doctype
swc-bot Dec 2, 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
41 changes: 41 additions & 0 deletions crates/swc_xml_parser/src/lexer/mod.rs
View file
Open in desktop
Expand Up @@ -63,6 +63,7 @@ pub enum State {
DoctypeSystemIdentifierSingleQuoted,
DoctypeSystemIdentifierDoubleQuoted,
AfterDoctypeSystemIdentifier,
DoctypeTypeInternalSubSet,
BogusDoctype,
}

Expand Down Expand Up @@ -2403,6 +2404,12 @@ where
self.state = State::Data;
self.emit_doctype_token();
}
// U+005B LEFT SQUARE BRACKET ([)
// Switch to the doctype internal subset state.
Some(c @ '[') => {
self.append_raw_to_doctype_token(c);
self.state = State::DoctypeTypeInternalSubSet;
}
// EOF
// Parse error. Switch to the data state. Emit DOCTYPE token. Reconsume the EOF
// character.
Expand Down Expand Up @@ -2983,6 +2990,12 @@ where
self.state = State::Data;
self.emit_doctype_token();
}
// U+005B LEFT SQUARE BRACKET ([)
// Switch to the doctype internal subset state.
Some(c @ '[') => {
self.append_raw_to_doctype_token(c);
self.state = State::DoctypeTypeInternalSubSet;
}
// EOF
// Parse error. Switch to the data state. Emit DOCTYPE token. Reconsume the EOF
// character.
Expand All @@ -3001,6 +3014,34 @@ where
}
}
}
State::DoctypeTypeInternalSubSet => {
// Consume the next input character:
match self.consume_next_char() {
// U+005D RIGHT SQUARE BRACKET (])
// Switch to the CDATA bracket state.
Some(c @ ']') => {
self.append_raw_to_doctype_token(c);
self.state = State::AfterDoctypeName;
}
// EOF
// Parse error. Switch to the data state. Emit DOCTYPE token. Reconsume the EOF
// character.
None => {
self.emit_error(ErrorKind::EofInDoctype);
self.state = State::Data;
self.emit_doctype_token();
self.reconsume();
}
// Anything else
// Append the current input character to the current DOCTYPE token's system
// identifier.
Some(c) => {
// TODO improve parse legacy declarations
self.validate_input_stream_character(c);
self.append_raw_to_doctype_token(c);
}
}
}
State::BogusDoctype => {
// Consume the next input character:
match self.consume_next_char() {
Expand Down
3 changes: 3 additions & 0 deletions crates/swc_xml_parser/tests/fixture/document_type/internal_subset-1/dom.txt
View file
Open in desktop
@@ -0,0 +1,3 @@
| <!DOCTYPE svg>
| <root>
| "test"
5 changes: 5 additions & 0 deletions crates/swc_xml_parser/tests/fixture/document_type/internal_subset-1/input.xml
View file
Open in desktop
@@ -0,0 +1,5 @@
<!DOCTYPE svg [
<!ENTITY Pub-Status "This is a pre-release of the
specification.">
]>
<root>test</root>
44 changes: 44 additions & 0 deletions crates/swc_xml_parser/tests/fixture/document_type/internal_subset-1/output.json
View file
Open in desktop
@@ -0,0 +1,44 @@
{
"type": "Document",
"span": {
"start": 1,
"end": 112,
"ctxt": 0
},
"children": [
{
"type": "DocumentType",
"span": {
"start": 1,
"end": 94,
"ctxt": 0
},
"name": "svg",
"publicId": null,
"systemId": null,
"raw": "<!DOCTYPE svg [\n <!ENTITY Pub-Status \"This is a pre-release of the\nspecification.\">\n]>"
},
{
"type": "Element",
"span": {
"start": 95,
"end": 112,
"ctxt": 0
},
"tagName": "root",
"attributes": [],
"children": [
{
"type": "Text",
"span": {
"start": 101,
"end": 105,
"ctxt": 0
},
"data": "test",
"raw": "test"
}
]
}
]
}
55 changes: 55 additions & 0 deletions crates/swc_xml_parser/tests/fixture/document_type/internal_subset-1/span.swc-stderr
View file
Open in desktop
@@ -0,0 +1,55 @@

x Document
,-[$DIR/tests/fixture/document_type/internal_subset-1/input.xml:1:1]
1 | ,-> <!DOCTYPE svg [
2 | | <!ENTITY Pub-Status "This is a pre-release of the
3 | | specification.">
4 | | ]>
5 | `-> <root>test</root>
`----

x Child
,-[$DIR/tests/fixture/document_type/internal_subset-1/input.xml:1:1]
1 | ,-> <!DOCTYPE svg [
2 | | <!ENTITY Pub-Status "This is a pre-release of the
3 | | specification.">
4 | `-> ]>
5 | <root>test</root>
`----

x DocumentType
,-[$DIR/tests/fixture/document_type/internal_subset-1/input.xml:1:1]
1 | ,-> <!DOCTYPE svg [
2 | | <!ENTITY Pub-Status "This is a pre-release of the
3 | | specification.">
4 | `-> ]>
5 | <root>test</root>
`----

x Child
,-[$DIR/tests/fixture/document_type/internal_subset-1/input.xml:4:1]
4 | ]>
5 | <root>test</root>
: ^^^^^^^^^^^^^^^^^
`----

x Element
,-[$DIR/tests/fixture/document_type/internal_subset-1/input.xml:4:1]
4 | ]>
5 | <root>test</root>
: ^^^^^^^^^^^^^^^^^
`----

x Child
,-[$DIR/tests/fixture/document_type/internal_subset-1/input.xml:4:1]
4 | ]>
5 | <root>test</root>
: ^^^^
`----

x Text
,-[$DIR/tests/fixture/document_type/internal_subset-1/input.xml:4:1]
4 | ]>
5 | <root>test</root>
: ^^^^
`----
3 changes: 3 additions & 0 deletions crates/swc_xml_parser/tests/fixture/document_type/internal_subset-2/dom.txt
View file
Open in desktop
@@ -0,0 +1,3 @@
| <!DOCTYPE svg>
| <root>
| "test"
4 changes: 4 additions & 0 deletions crates/swc_xml_parser/tests/fixture/document_type/internal_subset-2/input.xml
View file
Open in desktop
@@ -0,0 +1,4 @@
<!DOCTYPE svg [
<!ENTITY open-hatch SYSTEM "http://www.textuality.com/boilerplate/OpenHatch.xml">
]>
<root>test</root>
44 changes: 44 additions & 0 deletions crates/swc_xml_parser/tests/fixture/document_type/internal_subset-2/output.json
View file
Open in desktop
@@ -0,0 +1,44 @@
{
"type": "Document",
"span": {
"start": 1,
"end": 127,
"ctxt": 0
},
"children": [
{
"type": "DocumentType",
"span": {
"start": 1,
"end": 109,
"ctxt": 0
},
"name": "svg",
"publicId": null,
"systemId": null,
"raw": "<!DOCTYPE svg [\n <!ENTITY open-hatch SYSTEM \"http://www.textuality.com/boilerplate/OpenHatch.xml\">\n]>"
},
{
"type": "Element",
"span": {
"start": 110,
"end": 127,
"ctxt": 0
},
"tagName": "root",
"attributes": [],
"children": [
{
"type": "Text",
"span": {
"start": 116,
"end": 120,
"ctxt": 0
},
"data": "test",
"raw": "test"
}
]
}
]
}
52 changes: 52 additions & 0 deletions crates/swc_xml_parser/tests/fixture/document_type/internal_subset-2/span.swc-stderr
View file
Open in desktop
@@ -0,0 +1,52 @@

x Document
,-[$DIR/tests/fixture/document_type/internal_subset-2/input.xml:1:1]
1 | ,-> <!DOCTYPE svg [
2 | | <!ENTITY open-hatch SYSTEM "http://www.textuality.com/boilerplate/OpenHatch.xml">
3 | | ]>
4 | `-> <root>test</root>
`----

x Child
,-[$DIR/tests/fixture/document_type/internal_subset-2/input.xml:1:1]
1 | ,-> <!DOCTYPE svg [
2 | | <!ENTITY open-hatch SYSTEM "http://www.textuality.com/boilerplate/OpenHatch.xml">
3 | `-> ]>
4 | <root>test</root>
`----

x DocumentType
,-[$DIR/tests/fixture/document_type/internal_subset-2/input.xml:1:1]
1 | ,-> <!DOCTYPE svg [
2 | | <!ENTITY open-hatch SYSTEM "http://www.textuality.com/boilerplate/OpenHatch.xml">
3 | `-> ]>
4 | <root>test</root>
`----

x Child
,-[$DIR/tests/fixture/document_type/internal_subset-2/input.xml:3:1]
3 | ]>
4 | <root>test</root>
: ^^^^^^^^^^^^^^^^^
`----

x Element
,-[$DIR/tests/fixture/document_type/internal_subset-2/input.xml:3:1]
3 | ]>
4 | <root>test</root>
: ^^^^^^^^^^^^^^^^^
`----

x Child
,-[$DIR/tests/fixture/document_type/internal_subset-2/input.xml:3:1]
3 | ]>
4 | <root>test</root>
: ^^^^
`----

x Text
,-[$DIR/tests/fixture/document_type/internal_subset-2/input.xml:3:1]
3 | ]>
4 | <root>test</root>
: ^^^^
`----
3 changes: 3 additions & 0 deletions crates/swc_xml_parser/tests/fixture/document_type/internal_subset-3/dom.txt
View file
Open in desktop
@@ -0,0 +1,3 @@
| <!DOCTYPE svg>
| <root>
| "test"
6 changes: 6 additions & 0 deletions crates/swc_xml_parser/tests/fixture/document_type/internal_subset-3/input.xml
View file
Open in desktop
@@ -0,0 +1,6 @@
<!DOCTYPE svg [
<!ENTITY open-hatch
PUBLIC "-//Textuality//TEXT Standard open-hatch boilerplate//EN"
"http://www.textuality.com/boilerplate/OpenHatch.xml">
]>
<root>test</root>
44 changes: 44 additions & 0 deletions crates/swc_xml_parser/tests/fixture/document_type/internal_subset-3/output.json
View file
Open in desktop
@@ -0,0 +1,44 @@
{
"type": "Document",
"span": {
"start": 1,
"end": 205,
"ctxt": 0
},
"children": [
{
"type": "DocumentType",
"span": {
"start": 1,
"end": 187,
"ctxt": 0
},
"name": "svg",
"publicId": null,
"systemId": null,
"raw": "<!DOCTYPE svg [\n <!ENTITY open-hatch\n PUBLIC \"-//Textuality//TEXT Standard open-hatch boilerplate//EN\"\n \"http://www.textuality.com/boilerplate/OpenHatch.xml\">\n]>"
},
{
"type": "Element",
"span": {
"start": 188,
"end": 205,
"ctxt": 0
},
"tagName": "root",
"attributes": [],
"children": [
{
"type": "Text",
"span": {
"start": 194,
"end": 198,
"ctxt": 0
},
"data": "test",
"raw": "test"
}
]
}
]
}