refactor: Improve code coverage (#517)

inikulin · Apr 30, 2022 · 6e57110 · 6e57110
1 parent c1f64b8
commit 6e57110
Show file tree

Hide file tree

Showing 11 changed files with 128 additions and 29 deletions.
diff --git a/README.md b/README.md
@@ -13,6 +13,7 @@
   <a href="https://www.npmjs.com/package/parse5"><img alt="NPM Version" src="https://img.shields.io/npm/v/parse5.svg"></a>
   <a href="https://npmjs.org/package/parse5"><img alt="Downloads" src="http://img.shields.io/npm/dm/parse5.svg"></a>
   <a href="https://npmjs.org/package/parse5"><img alt="Downloads total" src="http://img.shields.io/npm/dt/parse5.svg"></a>
+  <a href="https://coveralls.io/github/inikulin/parse5"><img alt="Coverage" src="https://img.shields.io/coveralls/github/inikulin/parse5/master"></a>
 </p>
 
 <p align="center">

diff --git a/package.json b/package.json
@@ -68,6 +68,11 @@
             "^(parse5[^/]*)/dist/(.*?)(?:\\.js)?$": "<rootDir>/packages/$1/lib/$2",
             "^(parse5[^/]*)$": "<rootDir>/packages/$1/lib/index.ts",
             "^(.*)\\.js$": "$1"
-        }
+        },
+        "coveragePathIgnorePatterns": [
+            "node_modules",
+            "bench",
+            "test"
+        ]
     }
 }
diff --git a/packages/parse5-html-rewriting-stream/test/rewriting-stream.test.ts b/packages/parse5-html-rewriting-stream/test/rewriting-stream.test.ts
@@ -202,6 +202,33 @@ describe('RewritingStream', () => {
         })
     );
 
+    it(
+        'rewrite doctype (no public id)',
+        createRewriterTest({
+            src: srcHtml,
+            expected: outdent`
+              <!DOCTYPE html SYSTEM "hey">
+              <html>
+                  <!-- comment1 -->
+                  <head /// 123>
+                  </head>
+                  <!-- comment2 -->
+                  <body =123>
+                      <div>Hey ya</div>
+                  </body>
+              </html>
+            `,
+            assignTokenHandlers: (rewriter) => {
+                rewriter.on('doctype', (token) => {
+                    token.publicId = null;
+                    token.systemId = 'hey';
+
+                    rewriter.emitDoctype(token);
+                });
+            },
+        })
+    );
+
     it(
         'emit multiple',
         createRewriterTest({
@@ -210,7 +237,7 @@ describe('RewritingStream', () => {
               <!DOCTYPE html "">
               <wrap><html></wrap>
                   <!-- comment1 -->
-                  <wrap><head 123=""></wrap>
+                  <wrap><head 123=""/></wrap>
                   </head>
                   <!-- comment2 -->
                   <wrap><body =123=""></wrap>
@@ -221,6 +248,11 @@ describe('RewritingStream', () => {
             assignTokenHandlers: (rewriter) => {
                 rewriter.on('startTag', (token) => {
                     rewriter.emitRaw('<wrap>');
+
+                    if (token.tagName === 'head') {
+                        token.selfClosing = true;
+                    }
+
                     rewriter.emitStartTag(token);
                     rewriter.emitRaw('</wrap>');
                 });

diff --git a/packages/parse5-htmlparser2-tree-adapter/lib/index.ts b/packages/parse5-htmlparser2-tree-adapter/lib/index.ts
@@ -149,9 +149,9 @@ export const adapter: TreeAdapter<Htmlparser2TreeAdapterMap> = {
             adapter.appendChild(document, doctypeNode);
         }
 
-        doctypeNode['x-name'] = name ?? undefined;
-        doctypeNode['x-publicId'] = publicId ?? undefined;
-        doctypeNode['x-systemId'] = systemId ?? undefined;
+        doctypeNode['x-name'] = name;
+        doctypeNode['x-publicId'] = publicId;
+        doctypeNode['x-systemId'] = systemId;
     },
 
     setDocumentMode(document: Document, mode: html.DOCUMENT_MODE): void {

diff --git a/packages/parse5-parser-stream/test/utils/parse-chunked.ts b/packages/parse5-parser-stream/test/utils/parse-chunked.ts
@@ -17,9 +17,6 @@ export function parseChunked<T extends TreeAdapterTypeMap>(
     parserStream.parser.tokenizer.preprocessor.bufferWaterline = 8;
 
     for (let i = 0; i < chunks.length - 1; i++) {
-        if (typeof chunks[i] !== 'string') {
-            throw new TypeError('Expected chunk to be a string');
-        }
         parserStream.write(chunks[i]);
     }
 

diff --git a/packages/parse5-sax-parser/test/sax-parser.test.ts b/packages/parse5-sax-parser/test/sax-parser.test.ts
@@ -137,4 +137,21 @@ describe('SAX parser', () => {
 
         assert.throws(() => stream.write(buf), TypeError);
     });
+
+    it('Should treat NULL characters as normal text', async () => {
+        const parser = new SAXParser();
+        let foundText = false;
+
+        parser.on('text', ({ text }) => {
+            foundText = true;
+            assert.strictEqual(text, '\0');
+        });
+
+        parser.write('\0');
+        parser.end();
+
+        await finished(parser);
+
+        assert.strictEqual(foundText, true);
+    });
 });
diff --git a/packages/parse5/lib/parser/formatting-element-list.test.ts b/packages/parse5/lib/parser/formatting-element-list.test.ts
@@ -142,6 +142,10 @@ generateTestsForEachTreeAdapter('FormattingElementList', (treeAdapter) => {
         list.clearToLastMarker();
 
         assert.strictEqual(list.entries.length, 2);
+
+        list.clearToLastMarker();
+
+        assert.strictEqual(list.entries.length, 0);
     });
 
     test('Remove entry', () => {

diff --git a/packages/parse5/lib/parser/formatting-element-list.ts b/packages/parse5/lib/parser/formatting-element-list.ts
@@ -127,6 +127,11 @@ export class FormattingElementList<T extends TreeAdapterTypeMap> {
         }
     }
 
+    /**
+     * Clears the list of formatting elements up to the last marker.
+     *
+     * @see https://html.spec.whatwg.org/multipage/parsing.html#clear-the-list-of-active-formatting-elements-up-to-the-last-marker
+     */
     clearToLastMarker(): void {
         const markerIdx = this.entries.indexOf(MARKER);
 

diff --git a/packages/parse5/lib/parser/open-element-stack.test.ts b/packages/parse5/lib/parser/open-element-stack.test.ts
@@ -317,6 +317,8 @@ generateTestsForEachTreeAdapter('open-element-stack', (treeAdapter) => {
     test('Has numbered header in scope', () => {
         const stack = new OpenElementStack(treeAdapter.createDocument(), treeAdapter, stackHandler);
 
+        assert.ok(stack.hasNumberedHeaderInScope());
+
         stack.push(createElement(TN.HTML), $.HTML);
         stack.push(createElement(TN.DIV), $.DIV);
         assert.ok(!stack.hasNumberedHeaderInScope());
@@ -337,6 +339,8 @@ generateTestsForEachTreeAdapter('open-element-stack', (treeAdapter) => {
     test('Has element in list item scope', () => {
         const stack = new OpenElementStack(treeAdapter.createDocument(), treeAdapter, stackHandler);
 
+        assert.ok(stack.hasInListItemScope($.P));
+
         stack.push(createElement(TN.HTML), $.HTML);
         stack.push(createElement(TN.DIV), $.DIV);
         assert.ok(!stack.hasInListItemScope($.P));
@@ -353,6 +357,8 @@ generateTestsForEachTreeAdapter('open-element-stack', (treeAdapter) => {
     test('Has element in button scope', () => {
         const stack = new OpenElementStack(treeAdapter.createDocument(), treeAdapter, stackHandler);
 
+        assert.ok(stack.hasInButtonScope($.P));
+
         stack.push(createElement(TN.HTML), $.HTML);
         stack.push(createElement(TN.DIV), $.DIV);
         assert.ok(!stack.hasInButtonScope($.P));
@@ -406,6 +412,8 @@ generateTestsForEachTreeAdapter('open-element-stack', (treeAdapter) => {
     test('Has element in select scope', () => {
         const stack = new OpenElementStack(treeAdapter.createDocument(), treeAdapter, stackHandler);
 
+        assert.ok(stack.hasInSelectScope($.P));
+
         stack.push(createElement(TN.HTML), $.HTML);
         stack.push(createElement(TN.DIV), $.DIV);
         assert.ok(!stack.hasInSelectScope($.P));

diff --git a/packages/parse5/lib/tokenizer/index.test.ts b/packages/parse5/lib/tokenizer/index.test.ts
@@ -1,9 +1,51 @@
 import { Tokenizer } from 'parse5';
 import { generateTokenizationTests } from 'parse5-test-utils/utils/generate-tokenization-tests.js';
+import * as assert from 'node:assert';
 
 const dataPath = new URL('../../../../test/data/html5lib-tests/tokenizer', import.meta.url);
 const tokenizerOpts = {
     sourceCodeLocationInfo: true,
 };
 
 generateTokenizationTests('Tokenizer', dataPath.pathname, (handler) => new Tokenizer(tokenizerOpts, handler));
+
+function noop(): void {
+    // Noop
+}
+
+describe('Tokenizer methods', () => {
+    it('should pause and resume', () => {
+        let count = 0;
+        const tokenizer = new Tokenizer(tokenizerOpts, {
+            onComment(t): void {
+                assert.strictEqual(t.data, 'INIT');
+                assert.strictEqual(count++, 0);
+
+                tokenizer.pause();
+                tokenizer.write('<!doctype foo>', false);
+            },
+            onDoctype(t): void {
+                assert.strictEqual(t.name, 'foo');
+                assert.strictEqual(count++, 2);
+
+                expect(() => tokenizer.resume()).toThrow('Parser was already resumed');
+                tokenizer.write('<next>', true);
+            },
+            onStartTag(t): void {
+                assert.strictEqual(count++, 3);
+                assert.strictEqual(t.tagName, 'next');
+            },
+            onEndTag: noop,
+            onEof: noop,
+            onCharacter: noop,
+            onNullCharacter: noop,
+            onWhitespaceCharacter: noop,
+        });
+
+        tokenizer.write('<!--INIT-->', false);
+        assert.strictEqual(count++, 1);
+        expect(tokenizer).toHaveProperty('paused', true);
+
+        tokenizer.resume();
+    });
+});
diff --git a/packages/parse5/lib/tokenizer/index.ts b/packages/parse5/lib/tokenizer/index.ts
@@ -132,7 +132,6 @@ const enum State {
     AMBIGUOUS_AMPERSAND,
     NUMERIC_CHARACTER_REFERENCE,
     HEXADEMICAL_CHARACTER_REFERENCE_START,
-    DECIMAL_CHARACTER_REFERENCE_START,
     HEXADEMICAL_CHARACTER_REFERENCE,
     DECIMAL_CHARACTER_REFERENCE,
     NUMERIC_CHARACTER_REFERENCE_END,
@@ -993,10 +992,6 @@ export class Tokenizer {
                 this._stateHexademicalCharacterReferenceStart(cp);
                 break;
             }
-            case State.DECIMAL_CHARACTER_REFERENCE_START: {
-                this._stateDecimalCharacterReferenceStart(cp);
-                break;
-            }
             case State.HEXADEMICAL_CHARACTER_REFERENCE: {
                 this._stateHexademicalCharacterReference(cp);
                 break;
@@ -3029,9 +3024,16 @@ export class Tokenizer {
 
         if (cp === $.LATIN_SMALL_X || cp === $.LATIN_CAPITAL_X) {
             this.state = State.HEXADEMICAL_CHARACTER_REFERENCE_START;
+        }
+        // Inlined decimal character reference start state
+        else if (isAsciiDigit(cp)) {
+            this.state = State.DECIMAL_CHARACTER_REFERENCE;
+            this._stateDecimalCharacterReference(cp);
         } else {
-            this.state = State.DECIMAL_CHARACTER_REFERENCE_START;
-            this._stateDecimalCharacterReferenceStart(cp);
+            this._err(ERR.absenceOfDigitsInNumericCharacterReference);
+            this._flushCodePointConsumedAsCharacterReference($.AMPERSAND);
+            this._flushCodePointConsumedAsCharacterReference($.NUMBER_SIGN);
+            this._reconsumeInState(this.returnState);
         }
     }
 
@@ -3050,20 +3052,6 @@ export class Tokenizer {
         }
     }
 
-    // Decimal character reference start state
-    //------------------------------------------------------------------
-    private _stateDecimalCharacterReferenceStart(cp: number): void {
-        if (isAsciiDigit(cp)) {
-            this.state = State.DECIMAL_CHARACTER_REFERENCE;
-            this._stateDecimalCharacterReference(cp);
-        } else {
-            this._err(ERR.absenceOfDigitsInNumericCharacterReference);
-            this._flushCodePointConsumedAsCharacterReference($.AMPERSAND);
-            this._flushCodePointConsumedAsCharacterReference($.NUMBER_SIGN);
-            this._reconsumeInState(this.returnState);
-        }
-    }
-
     // Hexademical character reference state
     //------------------------------------------------------------------
     private _stateHexademicalCharacterReference(cp: number): void {