html-json-extractor 0.0.1 → 0.2.0

package/README.md

@@ -1,10 +1,10 @@
 # html-json-extractor
 
-Fast, forgiving extraction of `<script type="application/ld+json">` blocks from an HTML string.
+Fast, forgiving extraction of `<script type="application/json">` and `<script type="application/ld+json">` blocks from an HTML string.
 
 - No DOM parser or runtime dependencies
 - Returns one result per matching script block
-- Malformed JSON-LD blocks do not break the rest
+- Malformed JSON blocks do not break the rest
 
 ## Install
 
@@ -15,31 +15,53 @@ npm install html-json-extractor
 ## Usage
 
 ```ts
-import { extractJsonLd, extractJsonLdStrings } from 'html-json-extractor';
+import {
+  extractJson,
+  extractJsonStrings,
+  getJsonLdItems,
+  getJsonLdRecords
+} from 'html-json-extractor';
 
 const html = `
+  <script type="application/json">{"featureFlags":{"search":true}}</script>
   <script type="application/ld+json">{"@type":"WebSite","name":"Example"}</script>
-  <script type="application/ld+json">{"broken":</script>
+  <script type="application/json">{"broken":</script>
   <script type="application/ld+json">[{"@type":"Person","name":"Ada"}]</script>
 `;
 
-const raw = extractJsonLdStrings(html);
-// ['{"@type":"WebSite","name":"Example"}', '{"broken":', '[{"@type":"Person","name":"Ada"}]']
+const raw = extractJsonStrings(html);
+// ['{"featureFlags":{"search":true}}', '{"@type":"WebSite","name":"Example"}', '{"broken":', '[{"@type":"Person","name":"Ada"}]']
 
-const parsed = extractJsonLd(html);
-// [{ '@type': 'WebSite', name: 'Example' }, [{ '@type': 'Person', name: 'Ada' }]]
+const parsed = extractJson(html);
+// [{ featureFlags: { search: true } }, { '@type': 'WebSite', name: 'Example' }, [{ '@type': 'Person', name: 'Ada' }]]
+
+const items = parsed.flatMap(getJsonLdItems);
+// [{ featureFlags: { search: true } }, { '@type': 'WebSite', name: 'Example' }, { '@type': 'Person', name: 'Ada' }]
+
+const records = parsed.flatMap(getJsonLdRecords);
+// [{ featureFlags: { search: true } }, { '@type': 'WebSite', name: 'Example' }, { '@type': 'Person', name: 'Ada' }]
 ```
 
 ## API
 
-### `extractJsonLdStrings(html: string): string[]`
+### `extractJsonStrings(html: string): string[]`
 
-Returns normalized JSON-LD script contents as strings.
+Returns normalized `application/json` and `application/ld+json` script contents as strings.
 
-### `extractJsonLd<T = JsonValue>(html: string): T[]`
+### `extractJson<T = JsonValue>(html: string): T[]`
 
 Parses the extracted strings with `JSON.parse`. Entries that fail to parse are skipped.
 
+### `getJsonLdItems(value: unknown): unknown[]`
+
+Normalizes a single parsed entry or parsed array entry into a flat list of items.
+
+### `getJsonLdRecords(value: unknown): Record<string, unknown>[]`
+
+Returns record-shaped items and follows nested `@graph` content.
+
+`getJsonLdItems` and `getJsonLdRecords` remain JSON-LD-oriented helpers for structured-data use cases on top of the generic extractor.
+
 ## License
 
 MIT

package/dist/index.cjs

@@ -1,7 +1,7 @@
 Object.defineProperty(exports, Symbol.toStringTag, { value: "Module" });
 //#region src/scan.ts
 const SCRIPT_TAG_NAME = "script";
-const APPLICATION_LD_JSON_MIME = "application/ld+json";
+const SUPPORTED_JSON_SCRIPT_MIME_TYPES = ["application/ld+json", "application/json"];
 const CHAR_TAB = 9;
 const CHAR_LINE_FEED = 10;
 const CHAR_FORM_FEED = 12;
@@ -45,7 +45,7 @@ const TEXT_LITERAL_CONTAINERS = [
 		isTerminal: true
 	}
 ];
-function extractJsonLdStrings(html) {
+function extractJsonStrings(html) {
 	if (html.length === 0) return [];
 	const results = [];
 	let cursor = 0;
@@ -57,8 +57,8 @@ function extractJsonLdStrings(html) {
 		const contentStart = openTag.tagEnd + 1;
 		const closeTagStart = findNextCloseTag(html, contentStart, SCRIPT_TAG_NAME);
 		const contentEnd = closeTagStart === -1 ? html.length : closeTagStart;
-		if (openTag.isJsonLd) {
-			const content = normalizeJsonLdContent(html.slice(contentStart, contentEnd));
+		if (openTag.hasSupportedMime) {
+			const content = normalizeJsonScriptContent(html.slice(contentStart, contentEnd));
 			if (content.length > 0) results.push(content);
 		}
 		if (closeTagStart === -1) break;
@@ -105,13 +105,13 @@ function findNextScriptOpenTag(html, start) {
 }
 function parseScriptOpenTag(html, openTagStart) {
 	let cursor = openTagStart + 1 + 6;
-	let isJsonLd = false;
+	let hasSupportedMime = false;
 	while (cursor < html.length) {
 		cursor = skipLeadingHtmlWhitespace(html, cursor, html.length);
 		if (cursor >= html.length) break;
 		const code = html.charCodeAt(cursor);
 		if (code === CHAR_GREATER_THAN) return {
-			isJsonLd,
+			hasSupportedMime,
 			tagEnd: cursor
 		};
 		if (isHtmlWhitespace(code) || code === CHAR_SLASH) {
@@ -147,11 +147,11 @@ function parseScriptOpenTag(html, openTagStart) {
 			}
 			valueEnd = cursor;
 		}
-		if (isTypeAttribute && isApplicationLdJsonMime(html, valueStart, valueEnd)) isJsonLd = true;
+		if (isTypeAttribute && isSupportedJsonScriptMime(html, valueStart, valueEnd)) hasSupportedMime = true;
 	}
 	return null;
 }
-function normalizeJsonLdContent(content) {
+function normalizeJsonScriptContent(content) {
 	let start = 0;
 	let end = content.length;
 	let changed = false;
@@ -225,13 +225,17 @@ function findNextCloseTag(html, start, tagName) {
 	}
 	return -1;
 }
-function isApplicationLdJsonMime(value, start, end) {
+function isSupportedJsonScriptMime(value, start, end) {
 	start = skipLeadingHtmlWhitespace(value, start, end);
 	end = skipTrailingHtmlWhitespace(value, start, end);
-	if (end - start < 19) return false;
-	if (!matchesAsciiLiteral(value, start, APPLICATION_LD_JSON_MIME)) return false;
-	if (end - start === 19) return true;
-	const suffixStart = start + 19;
+	for (const mime of SUPPORTED_JSON_SCRIPT_MIME_TYPES) if (matchesMimeType(value, start, end, mime)) return true;
+	return false;
+}
+function matchesMimeType(value, start, end, expectedMime) {
+	if (end - start < expectedMime.length) return false;
+	if (!matchesAsciiLiteral(value, start, expectedMime)) return false;
+	if (end - start === expectedMime.length) return true;
+	const suffixStart = start + expectedMime.length;
 	const nextCode = value.charCodeAt(suffixStart);
 	if (nextCode === CHAR_SEMICOLON) return true;
 	if (!isHtmlWhitespace(nextCode)) return false;
@@ -306,8 +310,8 @@ function isTagStartChar(code) {
 }
 //#endregion
 //#region src/parse.ts
-function extractJsonLd(html) {
-	const rawEntries = extractJsonLdStrings(html);
+function extractJson(html) {
+	const rawEntries = extractJsonStrings(html);
 	const results = [];
 	for (const rawEntry of rawEntries) try {
 		results.push(JSON.parse(rawEntry));
@@ -315,5 +319,29 @@ function extractJsonLd(html) {
 	return results;
 }
 //#endregion
-exports.extractJsonLd = extractJsonLd;
-exports.extractJsonLdStrings = extractJsonLdStrings;
+//#region src/helpers.ts
+const isRecord = (value) => {
+	return typeof value === "object" && value !== null && !Array.isArray(value);
+};
+const asRecord = (value) => {
+	return isRecord(value) ? value : void 0;
+};
+const asArray = (value) => {
+	return Array.isArray(value) ? value : void 0;
+};
+function getJsonLdRecords(value) {
+	const items = asArray(value);
+	if (items) return items.flatMap(getJsonLdRecords);
+	const record = asRecord(value);
+	if (!record) return [];
+	return [record, ...getJsonLdRecords(record["@graph"])];
+}
+function getJsonLdItems(value) {
+	if (value === void 0) return [];
+	return asArray(value) ?? [value];
+}
+//#endregion
+exports.extractJson = extractJson;
+exports.extractJsonStrings = extractJsonStrings;
+exports.getJsonLdItems = getJsonLdItems;
+exports.getJsonLdRecords = getJsonLdRecords;

package/dist/index.d.cts

@@ -7,9 +7,13 @@ type JsonArray = JsonValue[];
 type JsonValue = JsonArray | JsonObject | JsonPrimitive;
 //#endregion
 //#region src/parse.d.ts
-declare function extractJsonLd<T extends JsonValue = JsonValue>(html: string): T[];
+declare function extractJson<T extends JsonValue = JsonValue>(html: string): T[];
 //#endregion
 //#region src/scan.d.ts
-declare function extractJsonLdStrings(html: string): string[];
+declare function extractJsonStrings(html: string): string[];
 //#endregion
-export { type JsonArray, type JsonObject, type JsonPrimitive, type JsonValue, extractJsonLd, extractJsonLdStrings };
+//#region src/helpers.d.ts
+declare function getJsonLdRecords(value: unknown): Record<string, unknown>[];
+declare function getJsonLdItems(value: unknown): unknown[];
+//#endregion
+export { type JsonArray, type JsonObject, type JsonPrimitive, type JsonValue, extractJson, extractJsonStrings, getJsonLdItems, getJsonLdRecords };

package/dist/index.d.ts

@@ -7,9 +7,13 @@ type JsonArray = JsonValue[];
 type JsonValue = JsonArray | JsonObject | JsonPrimitive;
 //#endregion
 //#region src/parse.d.ts
-declare function extractJsonLd<T extends JsonValue = JsonValue>(html: string): T[];
+declare function extractJson<T extends JsonValue = JsonValue>(html: string): T[];
 //#endregion
 //#region src/scan.d.ts
-declare function extractJsonLdStrings(html: string): string[];
+declare function extractJsonStrings(html: string): string[];
 //#endregion
-export { type JsonArray, type JsonObject, type JsonPrimitive, type JsonValue, extractJsonLd, extractJsonLdStrings };
+//#region src/helpers.d.ts
+declare function getJsonLdRecords(value: unknown): Record<string, unknown>[];
+declare function getJsonLdItems(value: unknown): unknown[];
+//#endregion
+export { type JsonArray, type JsonObject, type JsonPrimitive, type JsonValue, extractJson, extractJsonStrings, getJsonLdItems, getJsonLdRecords };

package/dist/index.js

@@ -1,6 +1,6 @@
 //#region src/scan.ts
 const SCRIPT_TAG_NAME = "script";
-const APPLICATION_LD_JSON_MIME = "application/ld+json";
+const SUPPORTED_JSON_SCRIPT_MIME_TYPES = ["application/ld+json", "application/json"];
 const CHAR_TAB = 9;
 const CHAR_LINE_FEED = 10;
 const CHAR_FORM_FEED = 12;
@@ -44,7 +44,7 @@ const TEXT_LITERAL_CONTAINERS = [
 		isTerminal: true
 	}
 ];
-function extractJsonLdStrings(html) {
+function extractJsonStrings(html) {
 	if (html.length === 0) return [];
 	const results = [];
 	let cursor = 0;
@@ -56,8 +56,8 @@ function extractJsonLdStrings(html) {
 		const contentStart = openTag.tagEnd + 1;
 		const closeTagStart = findNextCloseTag(html, contentStart, SCRIPT_TAG_NAME);
 		const contentEnd = closeTagStart === -1 ? html.length : closeTagStart;
-		if (openTag.isJsonLd) {
-			const content = normalizeJsonLdContent(html.slice(contentStart, contentEnd));
+		if (openTag.hasSupportedMime) {
+			const content = normalizeJsonScriptContent(html.slice(contentStart, contentEnd));
 			if (content.length > 0) results.push(content);
 		}
 		if (closeTagStart === -1) break;
@@ -104,13 +104,13 @@ function findNextScriptOpenTag(html, start) {
 }
 function parseScriptOpenTag(html, openTagStart) {
 	let cursor = openTagStart + 1 + 6;
-	let isJsonLd = false;
+	let hasSupportedMime = false;
 	while (cursor < html.length) {
 		cursor = skipLeadingHtmlWhitespace(html, cursor, html.length);
 		if (cursor >= html.length) break;
 		const code = html.charCodeAt(cursor);
 		if (code === CHAR_GREATER_THAN) return {
-			isJsonLd,
+			hasSupportedMime,
 			tagEnd: cursor
 		};
 		if (isHtmlWhitespace(code) || code === CHAR_SLASH) {
@@ -146,11 +146,11 @@ function parseScriptOpenTag(html, openTagStart) {
 			}
 			valueEnd = cursor;
 		}
-		if (isTypeAttribute && isApplicationLdJsonMime(html, valueStart, valueEnd)) isJsonLd = true;
+		if (isTypeAttribute && isSupportedJsonScriptMime(html, valueStart, valueEnd)) hasSupportedMime = true;
 	}
 	return null;
 }
-function normalizeJsonLdContent(content) {
+function normalizeJsonScriptContent(content) {
 	let start = 0;
 	let end = content.length;
 	let changed = false;
@@ -224,13 +224,17 @@ function findNextCloseTag(html, start, tagName) {
 	}
 	return -1;
 }
-function isApplicationLdJsonMime(value, start, end) {
+function isSupportedJsonScriptMime(value, start, end) {
 	start = skipLeadingHtmlWhitespace(value, start, end);
 	end = skipTrailingHtmlWhitespace(value, start, end);
-	if (end - start < 19) return false;
-	if (!matchesAsciiLiteral(value, start, APPLICATION_LD_JSON_MIME)) return false;
-	if (end - start === 19) return true;
-	const suffixStart = start + 19;
+	for (const mime of SUPPORTED_JSON_SCRIPT_MIME_TYPES) if (matchesMimeType(value, start, end, mime)) return true;
+	return false;
+}
+function matchesMimeType(value, start, end, expectedMime) {
+	if (end - start < expectedMime.length) return false;
+	if (!matchesAsciiLiteral(value, start, expectedMime)) return false;
+	if (end - start === expectedMime.length) return true;
+	const suffixStart = start + expectedMime.length;
 	const nextCode = value.charCodeAt(suffixStart);
 	if (nextCode === CHAR_SEMICOLON) return true;
 	if (!isHtmlWhitespace(nextCode)) return false;
@@ -305,8 +309,8 @@ function isTagStartChar(code) {
 }
 //#endregion
 //#region src/parse.ts
-function extractJsonLd(html) {
-	const rawEntries = extractJsonLdStrings(html);
+function extractJson(html) {
+	const rawEntries = extractJsonStrings(html);
 	const results = [];
 	for (const rawEntry of rawEntries) try {
 		results.push(JSON.parse(rawEntry));
@@ -314,4 +318,26 @@ function extractJsonLd(html) {
 	return results;
 }
 //#endregion
-export { extractJsonLd, extractJsonLdStrings };
+//#region src/helpers.ts
+const isRecord = (value) => {
+	return typeof value === "object" && value !== null && !Array.isArray(value);
+};
+const asRecord = (value) => {
+	return isRecord(value) ? value : void 0;
+};
+const asArray = (value) => {
+	return Array.isArray(value) ? value : void 0;
+};
+function getJsonLdRecords(value) {
+	const items = asArray(value);
+	if (items) return items.flatMap(getJsonLdRecords);
+	const record = asRecord(value);
+	if (!record) return [];
+	return [record, ...getJsonLdRecords(record["@graph"])];
+}
+function getJsonLdItems(value) {
+	if (value === void 0) return [];
+	return asArray(value) ?? [value];
+}
+//#endregion
+export { extractJson, extractJsonStrings, getJsonLdItems, getJsonLdRecords };

package/package.json

@@ -1,9 +1,17 @@
 {
   "name": "html-json-extractor",
-  "version": "0.0.1",
-  "description": "Fast, forgiving extraction of application/ld+json script blocks from HTML strings.",
+  "version": "0.2.0",
+  "description": "Fast, forgiving extraction of application/json and application/ld+json script blocks from HTML strings.",
   "license": "MIT",
   "author": "VastBlast",
+  "repository": {
+    "type": "git",
+    "url": "git+https://github.com/VastBlast/html-json-extractor.git"
+  },
+  "bugs": {
+    "url": "https://github.com/VastBlast/html-json-extractor/issues"
+  },
+  "homepage": "https://github.com/VastBlast/html-json-extractor#readme",
   "type": "module",
   "sideEffects": false,
   "files": [
@@ -11,6 +19,8 @@
   ],
   "keywords": [
     "html",
+    "json",
+    "application/json",
     "json-ld",
     "ld+json",
     "schema",
@@ -42,10 +52,9 @@
   },
   "main": "./dist/index.cjs",
   "module": "./dist/index.js",
-  "types": "./dist/index.d.ts",
+  "types": "./dist/index.d.cts",
   "exports": {
     ".": {
-      "types": "./dist/index.d.ts",
       "import": "./dist/index.js",
       "require": "./dist/index.cjs"
     },