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

JS: Improve JSX lexing (support fragment syntax and embedded comments) #5765

Merged
merged 2 commits into from
Apr 7, 2023
Merged

JS: Improve JSX lexing (support fragment syntax and embedded comments) #5765

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
d33be40
Improve JSX parsing (fragment syntax)
matthiasblaesing Mar 26, 2023
1eb0cfb
Improve JSX parsing (empty assignments with only comment as comment)
matthiasblaesing Mar 26, 2023
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
559 changes: 278 additions & 281 deletions webcommon/javascript2.lexer/src/org/netbeans/modules/javascript2/lexer/JsColoringLexer.java
View file
Open in desktop

Large diffs are not rendered by default.

1 change: 1 addition & 0 deletions webcommon/javascript2.lexer/test/unit/data/testfiles/jsx/jsxComment.js
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1 @@
const a = <table>{/* Test */ /* Test */ /* Test */}{ a = 3 }</table>;
28 changes: 28 additions & 0 deletions webcommon/javascript2.lexer/test/unit/data/testfiles/jsx/jsxComment.js.tokens.txt
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,28 @@
<Unnamed test>
KEYWORD_CONST "const", la=1, st=LexerState{canFollowLiteral=false, canFollowKeyword=true, braceBalances=[], jsxBalances=[]}
WHITESPACE " ", la=1, st=LexerState{canFollowLiteral=false, canFollowKeyword=true, braceBalances=[], jsxBalances=[]}
IDENTIFIER "a", la=1, st=LexerState{canFollowLiteral=false, canFollowKeyword=true, braceBalances=[], jsxBalances=[]}
WHITESPACE " ", la=1, st=LexerState{canFollowLiteral=false, canFollowKeyword=true, braceBalances=[], jsxBalances=[]}
OPERATOR_ASSIGNMENT "=", la=1, st=LexerState{canFollowLiteral=true, canFollowKeyword=true, braceBalances=[], jsxBalances=[]}
WHITESPACE " ", la=1, st=LexerState{canFollowLiteral=true, canFollowKeyword=true, braceBalances=[], jsxBalances=[]}
JSX_TEXT "<table>", la=1, st=LexerState{canFollowLiteral=false, canFollowKeyword=true, braceBalances=[], jsxBalances=[1]}
JSX_EXP_BEGIN "{", st=LexerState{canFollowLiteral=false, canFollowKeyword=true, braceBalances=[JSX], jsxBalances=[1]}
BLOCK_COMMENT "/* Test */", st=LexerState{canFollowLiteral=false, canFollowKeyword=true, braceBalances=[JSX], jsxBalances=[1]}
WHITESPACE " ", la=1, st=LexerState{canFollowLiteral=false, canFollowKeyword=true, braceBalances=[JSX], jsxBalances=[1]}
BLOCK_COMMENT "/* Test */", st=LexerState{canFollowLiteral=false, canFollowKeyword=true, braceBalances=[JSX], jsxBalances=[1]}
WHITESPACE " ", la=1, st=LexerState{canFollowLiteral=false, canFollowKeyword=true, braceBalances=[JSX], jsxBalances=[1]}
BLOCK_COMMENT "/* Test */", st=LexerState{canFollowLiteral=false, canFollowKeyword=true, braceBalances=[JSX], jsxBalances=[1]}
JSX_EXP_END "}", st=LexerState{canFollowLiteral=false, canFollowKeyword=true, braceBalances=[], jsxBalances=[1]}
JSX_EXP_BEGIN "{", st=LexerState{canFollowLiteral=false, canFollowKeyword=true, braceBalances=[JSX], jsxBalances=[1]}
WHITESPACE " ", la=1, st=LexerState{canFollowLiteral=false, canFollowKeyword=true, braceBalances=[JSX], jsxBalances=[1]}
IDENTIFIER "a", la=1, st=LexerState{canFollowLiteral=false, canFollowKeyword=true, braceBalances=[JSX], jsxBalances=[1]}
WHITESPACE " ", la=1, st=LexerState{canFollowLiteral=false, canFollowKeyword=true, braceBalances=[JSX], jsxBalances=[1]}
OPERATOR_ASSIGNMENT "=", la=1, st=LexerState{canFollowLiteral=true, canFollowKeyword=true, braceBalances=[JSX], jsxBalances=[1]}
WHITESPACE " ", la=1, st=LexerState{canFollowLiteral=true, canFollowKeyword=true, braceBalances=[JSX], jsxBalances=[1]}
NUMBER "3", la=1, st=LexerState{canFollowLiteral=false, canFollowKeyword=true, braceBalances=[JSX], jsxBalances=[1]}
WHITESPACE " ", la=1, st=LexerState{canFollowLiteral=false, canFollowKeyword=true, braceBalances=[JSX], jsxBalances=[1]}
JSX_EXP_END "}", st=LexerState{canFollowLiteral=false, canFollowKeyword=true, braceBalances=[], jsxBalances=[1]}
JSX_TEXT "</table>", st=LexerState{canFollowLiteral=false, canFollowKeyword=true, braceBalances=[], jsxBalances=[]}
OPERATOR_SEMICOLON ";", st=LexerState{canFollowLiteral=true, canFollowKeyword=true, braceBalances=[], jsxBalances=[]}
----- EOF -----

1 change: 1 addition & 0 deletions webcommon/javascript2.lexer/test/unit/data/testfiles/jsx/jsxFragment.js
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1 @@
const a = <></>;
11 changes: 11 additions & 0 deletions webcommon/javascript2.lexer/test/unit/data/testfiles/jsx/jsxFragment.js.tokens.txt
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,11 @@
<Unnamed test>
KEYWORD_CONST "const", la=1, st=LexerState{canFollowLiteral=false, canFollowKeyword=true, braceBalances=[], jsxBalances=[]}
WHITESPACE " ", la=1, st=LexerState{canFollowLiteral=false, canFollowKeyword=true, braceBalances=[], jsxBalances=[]}
IDENTIFIER "a", la=1, st=LexerState{canFollowLiteral=false, canFollowKeyword=true, braceBalances=[], jsxBalances=[]}
WHITESPACE " ", la=1, st=LexerState{canFollowLiteral=false, canFollowKeyword=true, braceBalances=[], jsxBalances=[]}
OPERATOR_ASSIGNMENT "=", la=1, st=LexerState{canFollowLiteral=true, canFollowKeyword=true, braceBalances=[], jsxBalances=[]}
WHITESPACE " ", la=1, st=LexerState{canFollowLiteral=true, canFollowKeyword=true, braceBalances=[], jsxBalances=[]}
JSX_TEXT "<></>", st=LexerState{canFollowLiteral=false, canFollowKeyword=true, braceBalances=[], jsxBalances=[]}
OPERATOR_SEMICOLON ";", st=LexerState{canFollowLiteral=true, canFollowKeyword=true, braceBalances=[], jsxBalances=[]}
----- EOF -----

10 changes: 10 additions & 0 deletions ...avascript2.lexer/test/unit/src/org/netbeans/modules/javascript2/lexer/JsLexerJsxTest.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -93,4 +93,14 @@ public void testIncLess() throws Exception {
LexerTestUtilities.checkTokenDump(this, "testfiles/jsx/incLess.js",
JsTokenId.javascriptLanguage());
}

public void testJsxFragment() throws Exception {
LexerTestUtilities.checkTokenDump(this, "testfiles/jsx/jsxFragment.js",
JsTokenId.javascriptLanguage());
}

public void testJsxComment() throws Exception {
LexerTestUtilities.checkTokenDump(this, "testfiles/jsx/jsxComment.js",
JsTokenId.javascriptLanguage());
}
}
2 changes: 1 addition & 1 deletion webcommon/javascript2.lexer/tools/JsColoringScanner.flex
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -712,7 +712,7 @@ RegexpFirstCharacter = [^*\x5b/\r\n\\] | {RegexpBackslashSequence} | {RegexpClas
}

<JSX> {
"/>" | "</"{JSXCharacter}+">"
"/>" | "</"{JSXCharacter}*">"
{
Integer balance = jsxBalances.isEmpty() ? 0 : jsxBalances.pop() - 1;
if (balance <= 0) {
Expand Down
18 changes: 17 additions & 1 deletion webcommon/libs.nashorn/src/com/oracle/js/parser/Parser.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -5896,7 +5896,7 @@ private Expression jsxElement(long startToken) {
}
break;
case LBRACE:
if (T(k + 1) == TokenType.RBRACE) {
if (! lookaheadIsJsxAssignment()) {
next();
next();
} else {
Expand All @@ -5912,6 +5912,19 @@ private Expression jsxElement(long startToken) {
return new JsxElementNode(name, attributes, children, realStart, finish);
}

private boolean lookaheadIsJsxAssignment() {
assert type == LBRACE;
for (int i = 1;; i++) {
TokenType t = T(k + i);
switch (t) {
case COMMENT:
continue;
default:
return t != TokenType.RBRACE;
}
}
}

private Expression jsxAttribute() {
Expression attribute = null;
if (type == LBRACE) {
Expand Down Expand Up @@ -5958,6 +5971,9 @@ private Expression jsxAttribute() {
}

private String jsxElementName() {
if (type == TokenType.JSX_ELEM_END) {
return "";
}
expectDontAdvance(TokenType.JSX_IDENTIFIER);
StringBuilder name = new StringBuilder((String) getValue(token));
next();
Expand Down
14 changes: 12 additions & 2 deletions webcommon/libs.nashorn/test/unit/src/com/oracle/js/parser/DumpingVisitor.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -27,6 +27,7 @@
import com.oracle.js.parser.ir.FunctionNode;
import com.oracle.js.parser.ir.IdentNode;
import com.oracle.js.parser.ir.IndexNode;
import com.oracle.js.parser.ir.JsxElementNode;
import com.oracle.js.parser.ir.LexicalContext;
import com.oracle.js.parser.ir.LiteralNode;
import com.oracle.js.parser.ir.Node;
Expand All @@ -51,7 +52,9 @@ protected Node leaveDefault(Node node) {
@Override
protected boolean enterDefault(Node node) {
if (node instanceof IdentNode) {
System.out.println(indent() + node.getClass().getName() + " [" + ((IdentNode) node).getName() + "]");
System.out.printf("%s%s [%d-%d, name=%s]%n", indent(),
node.getClass().getName(), node.getStart(),
node.getFinish(), ((IdentNode) node).getName());
} else if (node instanceof LiteralNode) {
System.out.println(indent() + node.getClass().getName() + " [" + ((LiteralNode) node).getValue() + "]");
} else if (node instanceof FunctionNode) {
Expand Down Expand Up @@ -89,8 +92,15 @@ protected boolean enterDefault(Node node) {
} else if (node instanceof PropertyNode) {
PropertyNode pn = (PropertyNode) node;
System.out.println(indent() + node.getClass().getName() + " [static=" + pn.isStatic() + "]");
} else if (node instanceof JsxElementNode) {
JsxElementNode jen = (JsxElementNode) node;
System.out.printf("%s%s [%d-%d, name=%s]%n", indent(),
node.getClass().getName(), node.getStart(),
node.getFinish(), jen.getName());
} else {
System.out.println(indent() + node.getClass().getName());
System.out.printf("%s%s [%d-%d]%n", indent(),
node.getClass().getName(), node.getStart(),
node.getFinish());
}
indent += INDENT_PER_LEVEL;
return super.enterDefault(node);
Expand Down
7 changes: 1 addition & 6 deletions webcommon/libs.nashorn/test/unit/src/com/oracle/js/parser/ManualTest.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -38,12 +38,7 @@ public static void main(String[] args) {
// Source source = Source.sourceFor("dummy.js", "var a = import('test');");
// Source source = Source.sourceFor("dummy.js", "try {} catch (e) {}");
// Source source = Source.sourceFor("dummy.js", "function a() {}; async function b() {}; class x { y(){} async z(){} }");
Source source = Source.sourceFor("dummy.js", "class Polygon {\n" +
" constructor(height, width) {\n" +
" this.height = height;\n" +
" this.width = width;\n" +
" }\n" +
"} ");
Source source = Source.sourceFor("dummy.js", "const a = <table>{/* Test */ /* Test */ /* Test */}{ a = 3 }</table>");
ScriptEnvironment.Builder builder = ScriptEnvironment.builder();
Parser parser = new Parser(
builder.emptyStatements(true).ecmacriptEdition(13).jsx(true).build(),
Expand Down
8 changes: 8 additions & 0 deletions webcommon/libs.nashorn/test/unit/src/com/oracle/js/parser/ParserTest.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -266,6 +266,14 @@ public void testFor() {
assertParses(13, "async function dummy() { for await (const num of foo()) {}}");
}

@Test
public void testJsx() {
// JSX Fragments are parsed as JSX Elements without name and attributes
assertParses(13, "const a = <></>;");
assertParses(13, "const a = <table style={{ border: '2px solid black', borderRadius: '.5em'}}></table>");
assertParses(13, "const a = <table>{/* Test */ /* Test */ /* Test */}{ a = 3 }</table>");
}

private Predicate<Node> functionNodeWithName(String name) {
return n -> n instanceof FunctionNode && name.equals(((FunctionNode) n).getName());
}
Expand Down