stream-markdown-parser 0.0.21 → 0.0.23

package/dist/index.js

@@ -96,19 +96,102 @@ const VOID_TAGS$1 = new Set([
 function applyFixHtmlInlineTokens(md) {
 	md.core.ruler.push("fix_html_inline_tokens", (state) => {
 		const toks = state.tokens ?? [];
+		const tagStack = [];
 		for (let i = 0; i < toks.length; i++) {
 			const t = toks[i];
-			if (!t || t.type !== "inline") continue;
-			if (/\s*<$/.test(t.content)) {
-				t.children.length = 0;
+			if (t.type === "html_block") {
+				const tag = t.content?.match(/<([^\s>/]+)/)?.[1] ?? "";
+				if (!/<\s*\/\s*[^\s>]+\s*>/.test(t.content || "")) tagStack.push([tag, i]);
+				else if (tagStack.length > 0 && tagStack[tagStack.length - 1][0] === tag) tagStack.pop();
+				continue;
+			} else if (tagStack.length > 0) {
+				if (t.type === "paragraph_open" || t.type === "paragraph_close") {
+					toks.splice(i, 1);
+					i--;
+					continue;
+				}
+				const content = t.content || "";
+				const isClosingTag = (/* @__PURE__ */ new RegExp(`<\\s*\\/\\s*${tagStack[tagStack.length - 1][0]}\\s*>`)).test(content);
+				if (content) {
+					const [, openIndex] = tagStack[tagStack.length - 1];
+					const openToken = toks[openIndex];
+					openToken.content = `${openToken.content || ""}\n${content}`;
+					if (openToken.loading !== false) openToken.loading = !isClosingTag;
+				}
+				if (isClosingTag) tagStack.pop();
+				toks.splice(i, 1);
+				i--;
+			} else continue;
+		}
+		for (let i = 0; i < toks.length; i++) {
+			const t = toks[i];
+			if (t.type === "html_block") {
+				const tag = t.content?.match(/<([^\s>/]+)/)?.[1] ?? "";
+				if ([
+					"br",
+					"hr",
+					"img",
+					"input",
+					"link",
+					"meta",
+					"div",
+					"p",
+					"ul",
+					"li"
+				].includes(tag)) continue;
+				t.type = "inline";
+				const loading = t.content?.includes(`</${tag}>`) ? false : t.loading !== void 0 ? t.loading : true;
+				t.children = [{
+					type: "html_block",
+					content: t.content,
+					tag: t.content?.match(/<([^\s>/]+)/)?.[1] ?? "",
+					loading
+				}];
 				continue;
 			}
+			if (!t || t.type !== "inline") continue;
 			if (t.children.length === 2 && t.children[0].type === "html_inline") {
+				const tag = t.children[0].content?.match(/<([^\s>/]+)/)?.[1] ?? "";
+				if ([
+					"a",
+					"span",
+					"strong",
+					"em",
+					"b",
+					"i",
+					"u"
+				].includes(tag)) {
+					t.children[0].loading = true;
+					t.children[0].tag = tag;
+					t.children.push({
+						type: "html_inline",
+						tag,
+						loading: true,
+						content: `</${tag}>`
+					});
+				} else t.children = [{
+					type: "html_block",
+					loading: true,
+					tag,
+					content: t.children[0].content + t.children[1].content
+				}];
+				continue;
+			} else if (t.children.length === 3 && t.children[0].type === "html_inline" && t.children[2].type === "html_inline") {
+				const tag = t.children[0].content?.match(/<([^\s>/]+)/)?.[1] ?? "";
+				if ([
+					"a",
+					"span",
+					"strong",
+					"em",
+					"b",
+					"i",
+					"u"
+				].includes(tag)) continue;
 				t.children = [{
 					type: "html_block",
-					content: t.children[0].content + t.children[1].content,
-					tag: t.children[0].content.match(/<([^\s>/]+)/)?.[1] ?? "",
-					loading: true
+					loading: false,
+					tag,
+					content: t.children.map((ct) => ct.content).join("")
 				}];
 				continue;
 			}
@@ -176,288 +259,300 @@ function applyFixLinkTokens(md) {
 		}
 	});
 }
-function isTextToken(t) {
-	return !!t && t.type === "text" && typeof t.content === "string";
-}
 function fixLinkToken(tokens) {
-	const tokensAny = tokens;
-	tokens = fixLinkToken4(fixLinkToken6(tokens));
-	if (tokens.length < 5) return fixLinkToken3(tokens);
-	const first = tokens[tokens.length - 5];
-	const firstAny = first;
-	const firstContent = String(firstAny.content ?? "");
-	if (first.type !== "text" || !firstContent.endsWith("[")) return fixLinkTokens2(tokens);
-	if (tokens[tokens.length - 4].tag !== "em") return fixLinkTokens2(tokens);
-	const last = tokens[tokens.length - 1];
-	const lastAny = last;
-	const lastContent = String(lastAny.content ?? "");
-	if (last?.type === "text" && !lastContent.startsWith("]")) return fixLinkTokens2(tokens);
-	const thirdAny = tokens[tokens.length - 3];
-	const thirdContent = String(thirdAny.content ?? "");
-	const href = lastContent.replace(/^\]\(*/, "");
-	const loading = !lastContent.includes(")");
-	tokensAny[tokens.length - 5].content = firstContent.replace(/\[$/, "");
-	tokens.splice(tokens.length - 3, 1, {
-		type: "link",
-		href,
-		text: thirdContent,
-		children: [{
-			type: "text",
-			content: thirdContent,
-			raw: thirdContent
-		}],
-		loading
-	});
-	tokens.splice(tokens.length - 1, 1);
-	return tokens;
-}
-function fixLinkTokens2(tokens) {
-	const tokensAny = tokens;
-	if (tokens.length < 8) return tokens;
-	let length = tokens.length;
-	let last = tokens[length - 1];
-	if (!last) return tokens;
-	if (last.type !== "link_close") {
-		length--;
-		last = tokens[length - 1];
-		if (last.type !== "link_close") return tokens;
-	}
-	if (tokens[length - 7].type !== "em_open") return tokens;
-	const third = tokens[length - 6];
-	const first = tokens[length - 8];
-	if (first.type !== "text") return tokens;
-	let href = String(tokensAny[length - 2]?.content ?? "");
-	let count = 4;
-	if (length !== tokens.length) {
-		href += String(last.content ?? "");
-		count++;
-	}
-	tokens.splice(length - 4, count);
-	const thirdAny = third;
-	const content = String(thirdAny.content ?? "");
-	length -= 4;
-	const firstAny = first;
-	tokensAny[length - 8].content = String(firstAny.content ?? "").replace(/\[$/, "");
-	tokens.splice(length - 2, 1, {
-		type: "link",
-		href,
-		text: content,
-		children: [{
-			type: "text",
-			content,
-			raw: content
-		}],
-		loading: true
-	});
-	return tokens;
-}
-function fixLinkToken3(tokens) {
-	const tokensAny = tokens;
-	const last = tokens[tokens.length - 1];
-	const preLast = tokens[tokens.length - 2];
-	const fixedTokens = [...tokens];
-	if (!last) return tokens;
-	if (last.type !== "text" || !last.content?.startsWith(")")) return fixLinkToken5(tokens);
-	if (preLast.type !== "link_close") return tokens;
-	if (isTextToken(tokens[tokens.length - 5]) && String(tokens[tokens.length - 5].content ?? "").endsWith("(")) {
-		const a = tokensAny[tokens.length - 5];
-		const b = tokensAny[tokens.length - 3];
-		const content = String(a.content ?? "") + String(b.content ?? "") + String(last.content ?? "");
-		fixedTokens.splice(tokens.length - 5, 5, {
-			type: "text",
-			content,
-			raw: content
-		});
-	} else {
-		const lc = (last.content ?? "").slice(1);
-		fixedTokens[fixedTokens.length - 1] = {
-			...last,
-			content: lc
-		};
-	}
-	return fixedTokens;
-}
-function fixLinkToken4(tokens) {
-	const tokensAny = tokens;
-	const fixedTokens = [...tokens];
-	for (let i = tokens.length - 1; i >= 3; i--) {
-		const token = tokens[i];
-		if (token && token.type === "link_close") {
-			if (tokens[i - 3]?.content?.endsWith("(")) {
-				const nextToken = tokens[i + 1];
-				if (nextToken && nextToken?.type === "text") {
-					if (tokens[i - 1].type === "text" && tokens[i - 3]?.type === "text") {
-						const nextTokenContent = String(nextToken.content ?? "");
-						const a = tokensAny[i - 3];
-						const b = tokensAny[i - 1];
-						const content = String(a.content ?? "") + String(b.content ?? "") + nextTokenContent;
-						fixedTokens.splice(i - 3, 5, {
+	if (tokens.length < 4) return tokens;
+	for (let i = 0; i <= tokens.length - 1; i++) {
+		if (!tokens[i]) break;
+		if (tokens[i]?.type === "text" && tokens[i].content?.endsWith("(") && tokens[i + 1]?.type === "link_open") {
+			const match = tokens[i].content.match(/\[([^\]]+)\]/);
+			if (match) {
+				let beforeText = tokens[i].content.slice(0, match.index);
+				const emphasisMatch = beforeText.match(/(\*+)$/);
+				const replacerTokens = [];
+				if (emphasisMatch) {
+					beforeText = beforeText.slice(0, emphasisMatch.index);
+					if (beforeText) replacerTokens.push({
+						type: "text",
+						content: beforeText,
+						raw: beforeText
+					});
+					const text = match[1];
+					const type = emphasisMatch[1].length;
+					if (type === 1) replacerTokens.push({
+						type: "em_open",
+						tag: "em",
+						nesting: 1
+					});
+					else if (type === 2) replacerTokens.push({
+						type: "strong_open",
+						tag: "strong",
+						nesting: 1
+					});
+					else if (type === 3) {
+						replacerTokens.push({
+							type: "strong_open",
+							tag: "strong",
+							nesting: 1
+						});
+						replacerTokens.push({
+							type: "em_open",
+							tag: "em",
+							nesting: 1
+						});
+					}
+					let href = tokens[i + 2]?.content || "";
+					if (tokens[i + 4]?.type === "text" && !tokens[i + 4].content?.startsWith(")")) {
+						href += tokens[i + 4]?.content || "";
+						tokens[i + 4].content = "";
+					}
+					replacerTokens.push({
+						type: "link",
+						loading: !tokens[i + 4]?.content?.startsWith(")"),
+						href,
+						title: "",
+						text,
+						children: [{
 							type: "text",
-							content,
-							raw: content
+							content: text,
+							raw: text
+						}],
+						raw: String(`[${text}](${href})`)
+					});
+					if (type === 1) replacerTokens.push({
+						type: "em_close",
+						tag: "em",
+						nesting: -1
+					});
+					else if (type === 2) replacerTokens.push({
+						type: "strong_close",
+						tag: "strong",
+						nesting: -1
+					});
+					else if (type === 3) {
+						replacerTokens.push({
+							type: "em_close",
+							tag: "em",
+							nesting: -1
+						});
+						replacerTokens.push({
+							type: "strong_close",
+							tag: "strong",
+							nesting: -1
 						});
-						i -= 3;
 					}
+					if (tokens[i + 4]?.type === "text") {
+						const afterText = tokens[i + 4].content?.replace(/^\)\**/, "");
+						if (afterText) replacerTokens.push({
+							type: "text",
+							content: afterText,
+							raw: afterText
+						});
+						tokens.splice(i, 5, ...replacerTokens);
+					} else tokens.splice(i, 4, ...replacerTokens);
 				} else {
-					if (tokens[i - 1].type === "text" && tokens[i - 3]?.type === "text") {
-						const a = tokensAny[i - 3];
-						const b = tokensAny[i - 1];
-						const content = String(a.content ?? "") + String(b.content ?? "");
-						fixedTokens.splice(i - 3, 4, {
+					if (beforeText) replacerTokens.push({
+						type: "text",
+						content: beforeText,
+						raw: beforeText
+					});
+					const text = match[1];
+					let href = tokens[i + 2]?.content || "";
+					if (tokens[i + 4]?.type === "text" && !tokens[i + 4].content?.startsWith(")")) {
+						href += tokens[i + 4]?.content || "";
+						tokens[i + 4].content = "";
+					}
+					replacerTokens.push(...[{
+						type: "link",
+						loading: !tokens[i + 4]?.content?.startsWith(")"),
+						href,
+						title: "",
+						text,
+						children: [{
+							type: "text",
+							content: text,
+							raw: text
+						}],
+						raw: String(`[${text}](${href})`)
+					}]);
+					if (tokens[i + 4]?.type === "text") {
+						const afterText = tokens[i + 4].content?.replace(/^\)/, "");
+						if (afterText) replacerTokens.push({
 							type: "text",
-							content,
-							raw: content
+							content: afterText,
+							raw: afterText
 						});
+						tokens.splice(i, 5, ...replacerTokens);
+					} else tokens.splice(i, 4, ...replacerTokens);
+				}
+				i -= replacerTokens.length - 1;
+				continue;
+			}
+		} else if (tokens[i].type === "link_open" && tokens[i].markup === "linkify" && tokens[i - 1]?.type === "text" && tokens[i - 1].content?.endsWith("(")) {
+			if (tokens[i - 2]?.type === "link_close") {
+				const replacerTokens = [];
+				const text = tokens[i - 3].content || "";
+				let href = tokens[i].attrs?.find((attr) => attr[0] === "href")?.[1] || "";
+				if (tokens[i + 3]?.type === "text") {
+					const m = (tokens[i + 3]?.content ?? "").indexOf(")");
+					const loading = m === -1;
+					if (m === -1) {
+						href += tokens[i + 3]?.content?.slice(0, m) || "";
+						tokens[i + 3].content = "";
 					}
-					i -= 3;
+					replacerTokens.push({
+						type: "link",
+						loading,
+						href,
+						title: "",
+						text,
+						children: [{
+							type: "text",
+							content: text,
+							raw: text
+						}],
+						raw: String(`[${text}](${href})`)
+					});
+					const afterText = tokens[i + 3].content?.replace(/^\)\**/, "");
+					if (afterText) replacerTokens.push({
+						type: "text",
+						content: afterText,
+						raw: afterText
+					});
+					tokens.splice(i - 4, 8, ...replacerTokens);
+				} else {
+					replacerTokens.push({
+						type: "link",
+						loading: true,
+						href,
+						title: "",
+						text,
+						children: [{
+							type: "text",
+							content: href,
+							raw: href
+						}],
+						raw: String(`[${text}](${href})`)
+					});
+					tokens.splice(i - 4, 7, ...replacerTokens);
 				}
+				continue;
 			}
 		}
-	}
-	return fixedTokens;
-}
-function fixLinkToken5(tokens) {
-	const firstToken = tokens[0];
-	const nextToken = tokens[1];
-	if (firstToken.type === "text" && firstToken.content?.endsWith("(") && nextToken.type === "link_open") {
-		const linkText = firstToken.content.match(/\[([^\]]+)\]/)?.[1] || "";
-		const href = tokens[2]?.content || "";
-		const newTokens = [{
-			type: "link",
-			loading: true,
-			href,
-			title: "",
-			text: linkText,
-			children: [{
-				type: "text",
-				content: linkText,
-				raw: linkText
-			}],
-			raw: String(`[${linkText}](${href})`)
-		}];
-		tokens.splice(0, 4, ...newTokens);
-	}
-	return tokens;
-}
-function fixLinkToken6(tokens) {
-	if (tokens.length < 4) return tokens;
-	for (let i = 0; i <= tokens.length - 4; i++) if (tokens[i]?.type === "text" && tokens[i].content?.endsWith("(") && tokens[i + 1]?.type === "link_open") {
-		const match = tokens[i].content.match(/\[([^\]]+)\]/);
-		if (match) {
-			let beforeText = tokens[i].content.slice(0, match.index);
-			const emphasisMatch = beforeText.match(/(\*+)$/);
-			const replacerTokens = [];
-			if (emphasisMatch) {
-				beforeText = beforeText.slice(0, emphasisMatch.index);
-				if (beforeText) replacerTokens.push({
+		if (tokens[i].type === "link_close" && tokens[i].nesting === -1 && tokens[i + 1]?.type === "text" && tokens[i - 1]?.type === "text" && tokens[i + 2]?.type !== "link_open") {
+			tokens[i - 2].loading = true;
+			const text = tokens[i - 1].content || "";
+			let href = tokens[i - 2].attrs?.[0]?.[1] || "";
+			let count = 3;
+			if (tokens[i].markup === "linkify" && tokens[i + 1]?.type === "text") {
+				const m = (tokens[i + 1]?.content ?? "").indexOf(")");
+				if (m === -1) {
+					href += tokens[i + 1]?.content?.slice(0, m) || "";
+					tokens[i + 1].content = "";
+				}
+				count += 1;
+			}
+			tokens.splice(i - 2, count, {
+				type: "link",
+				loading: false,
+				href,
+				title: "",
+				text,
+				children: [{
 					type: "text",
-					content: beforeText,
-					raw: beforeText
+					content: text,
+					raw: text
+				}],
+				raw: String(`[${text}](${href})`)
+			});
+		} else if (tokens[i].content?.startsWith("](") && tokens[i - 1].markup?.includes("*") && tokens[i - 4].type === "text" && tokens[i - 4].content?.endsWith("[")) {
+			const type = tokens[i - 1].markup.length;
+			const replacerTokens = [];
+			const beforeText = tokens[i - 4].content.slice(0, tokens[i - 4].content.length - 1 - type);
+			if (beforeText) replacerTokens.push({
+				type: "text",
+				content: beforeText,
+				raw: beforeText
+			});
+			if (type === 1) replacerTokens.push({
+				type: "em_open",
+				tag: "em",
+				nesting: 1
+			});
+			else if (type === 2) replacerTokens.push({
+				type: "strong_open",
+				tag: "strong",
+				nesting: 1
+			});
+			else if (type === 3) {
+				replacerTokens.push({
+					type: "strong_open",
+					tag: "strong",
+					nesting: 1
 				});
-				const text = match[1];
-				const type = emphasisMatch[1].length;
-				if (type === 1) replacerTokens.push({
+				replacerTokens.push({
 					type: "em_open",
 					tag: "em",
 					nesting: 1
 				});
-				else if (type === 2) replacerTokens.push({
-					type: "strong_open",
-					tag: "strong",
-					nesting: 1
-				});
-				else if (type === 3) {
-					replacerTokens.push({
-						type: "strong_open",
-						tag: "strong",
-						nesting: 1
-					});
-					replacerTokens.push({
-						type: "em_open",
-						tag: "em",
-						nesting: 1
-					});
-				}
-				let href = tokens[i + 2]?.content || "";
-				if (tokens[i + 4]?.type === "text" && !tokens[i + 4].content?.startsWith(")")) {
-					href += tokens[i + 4]?.content || "";
-					tokens[i + 4].content = "";
+			}
+			const text = tokens[i - 2].content || "";
+			let href = tokens[i].content.slice(2);
+			let loading = true;
+			if (tokens[i + 1]?.type === "text") {
+				const m = (tokens[i + 1]?.content ?? "").indexOf(")");
+				loading = m === -1;
+				if (m === -1) {
+					href += tokens[i + 1]?.content?.slice(0, m) || "";
+					tokens[i + 1].content = "";
 				}
+			}
+			replacerTokens.push({
+				type: "link",
+				loading,
+				href,
+				title: "",
+				text,
+				children: [{
+					type: "text",
+					content: text,
+					raw: text
+				}],
+				raw: String(`[${text}](${href})`)
+			});
+			if (type === 1) replacerTokens.push({
+				type: "em_close",
+				tag: "em",
+				nesting: -1
+			});
+			else if (type === 2) replacerTokens.push({
+				type: "strong_close",
+				tag: "strong",
+				nesting: -1
+			});
+			else if (type === 3) {
 				replacerTokens.push({
-					type: "link",
-					loading: !tokens[i + 4]?.content?.startsWith(")"),
-					href,
-					title: "",
-					text,
-					children: [{
-						type: "text",
-						content: text,
-						raw: text
-					}],
-					raw: String(`[${text}](${tokens[i + 2]?.content || ""})`)
-				});
-				if (type === 1) replacerTokens.push({
 					type: "em_close",
 					tag: "em",
 					nesting: -1
 				});
-				else if (type === 2) replacerTokens.push({
+				replacerTokens.push({
 					type: "strong_close",
 					tag: "strong",
 					nesting: -1
 				});
-				else if (type === 3) {
-					replacerTokens.push({
-						type: "em_close",
-						tag: "em",
-						nesting: -1
-					});
-					replacerTokens.push({
-						type: "strong_close",
-						tag: "strong",
-						nesting: -1
-					});
-				}
-				if (tokens[i + 4]?.type === "text") {
-					const afterText = tokens[i + 4].content?.replace(/^\)\**/, "");
-					if (afterText) replacerTokens.push({
-						type: "text",
-						content: afterText,
-						raw: afterText
-					});
-					tokens.splice(i, 5, ...replacerTokens);
-				} else tokens.splice(i, 4, ...replacerTokens);
-			} else {
-				if (beforeText) replacerTokens.push({
+			}
+			if (tokens[i + 1]?.type === "text") {
+				const afterText = tokens[i + 1].content?.replace(/^\)\**/, "");
+				if (afterText) replacerTokens.push({
 					type: "text",
-					content: beforeText,
-					raw: beforeText
+					content: afterText,
+					raw: afterText
 				});
-				const text = match[1];
-				replacerTokens.push(...[{
-					type: "link",
-					loading: true,
-					href: tokens[i + 2]?.content || "",
-					title: "",
-					text,
-					children: [{
-						type: "text",
-						content: text,
-						raw: text
-					}],
-					raw: String(`[${text}](${tokens[i + 2]?.content || ""})`)
-				}]);
-				if (tokens[i + 4]?.type === "text") {
-					const afterText = tokens[i + 4].content?.replace(/^\)/, "");
-					if (afterText) replacerTokens.push({
-						type: "text",
-						content: afterText,
-						raw: afterText
-					});
-					tokens.splice(i, 5, ...replacerTokens);
-				} else tokens.splice(i, 4, ...replacerTokens);
-			}
+				tokens.splice(i - 4, 8, ...replacerTokens);
+			} else if (tokens[i + 1]?.type === "link_open") tokens.splice(i - 4, 10, ...replacerTokens);
+			else tokens.splice(i - 4, 7, ...replacerTokens);
 			i -= replacerTokens.length - 1;
+			continue;
 		}
 	}
 	return tokens;
@@ -505,9 +600,10 @@ function fixStrongTokens(tokens) {
 	if (tokens.length < 4) return fixedTokens;
 	const i = tokens.length - 4;
 	const token = tokens[i];
+	if (!token) return fixedTokens;
 	const nextToken = tokens[i + 1];
 	const tokenContent = String(token.content ?? "");
-	if (token.type === "link_open" && tokens[i - 1].type === "em_open" && tokens[i - 2].type === "text" && tokens[i - 2].content?.endsWith("*")) {
+	if (token.type === "link_open" && tokens[i - 1]?.type === "em_open" && tokens[i - 2]?.type === "text" && tokens[i - 2].content?.endsWith("*")) {
 		const textContent = String(tokens[i - 2].content ?? "").slice(0, -1);
 		const replaceTokens = [
 			{
@@ -2125,7 +2221,8 @@ function parseInlineTokens(tokens, raw, pPreToken) {
 		}
 		if (!nextToken && /[^\]]\s*\(\s*$/.test(content)) content = content.replace(/\(\s*$/, "");
 		if (handleCheckboxLike(content)) return;
-		if (content === "[") {
+		const preToken = tokens[i - 1];
+		if (content === "[" && !nextToken?.markup?.includes("*") || content === "]" && !preToken.markup?.includes("*")) {
 			i++;
 			return;
 		}
@@ -2137,7 +2234,6 @@ function parseInlineTokens(tokens, raw, pPreToken) {
 			content
 		});
 		if (handleInlineLinkContent(content, token)) return;
-		const preToken = tokens[i - 1];
 		if (currentTextNode) {
 			currentTextNode.content += textNode.content.replace(/(\*+|\(|\\)$/, "");
 			currentTextNode.raw += textNode.raw;
@@ -2978,6 +3074,10 @@ function processTokens(tokens) {
 				result.push(parseMathBlock(tokens[i]));
 				i += 1;
 				break;
+			case "inline":
+				result.push(...parseInlineTokens(token.children || []));
+				i += 1;
+				break;
 			default:
 				i += 1;
 				break;