@oh-my-pi/pi-coding-agent 13.3.7 → 13.3.8

package/src/web/scrapers/iacr.ts

@@ -1,4 +1,4 @@
-import { parse as parseHtml } from "node-html-parser";
+import { parseHTML } from "linkedom";
 import type { RenderResult, SpecialHandler } from "./types";
 import { buildResult, loadPage } from "./types";
 import { convertWithMarkitdown, fetchBinary } from "./utils";
@@ -30,22 +30,30 @@ export const handleIacr: SpecialHandler = async (
 
 		if (!result.ok) return null;
 
-		const doc = parseHtml(result.content);
+		const doc = parseHTML(result.content).document;
 
 		// Extract metadata from the page
 		const title =
-			doc.querySelector("h3.mb-3")?.text?.trim() ||
+			doc.querySelector("h3.mb-3")?.textContent?.trim() ||
 			doc.querySelector('meta[name="citation_title"]')?.getAttribute("content");
-		const authors = doc
-			.querySelectorAll('meta[name="citation_author"]')
+		const authors = Array.from(
+			doc.querySelectorAll('meta[name="citation_author"]') as Iterable<{
+				getAttribute: (name: string) => string | null;
+			}>,
+		)
 			.map(m => m.getAttribute("content"))
-			.filter(Boolean);
+			.filter((author): author is string => Boolean(author));
 		// Abstract is in <p> after <h5>Abstract</h5>
-		const abstractHeading = doc.querySelectorAll("h5").find(h => h.text?.includes("Abstract"));
+		const abstractHeading = Array.from(
+			doc.querySelectorAll("h5") as Iterable<{
+				textContent: string | null;
+				parentElement?: { querySelector: (selector: string) => { textContent: string | null } | null } | null;
+			}>,
+		).find(h => h.textContent?.includes("Abstract"));
 		const abstract =
-			abstractHeading?.parentNode?.querySelector("p")?.text?.trim() ||
+			abstractHeading?.parentElement?.querySelector("p")?.textContent?.trim() ||
 			doc.querySelector('meta[name="description"]')?.getAttribute("content");
-		const keywords = doc.querySelector(".keywords")?.text?.replace("Keywords:", "").trim();
+		const keywords = doc.querySelector(".keywords")?.textContent?.replace("Keywords:", "").trim();
 		const pubDate = doc.querySelector('meta[name="citation_publication_date"]')?.getAttribute("content");
 
 		let md = `# ${title || "IACR ePrint Paper"}\n\n`;

package/src/web/scrapers/readthedocs.ts

@@ -1,7 +1,7 @@
 /**
  * Read the Docs handler for web-fetch
  */
-import { parse as parseHtml } from "node-html-parser";
+import { parseHTML } from "linkedom";
 import { buildResult, htmlToBasicMarkdown, loadPage, type RenderResult, type SpecialHandler } from "./types";
 
 export const handleReadTheDocs: SpecialHandler = async (
@@ -39,7 +39,7 @@ export const handleReadTheDocs: SpecialHandler = async (
 	}
 
 	// Parse HTML
-	const root = parseHtml(result.content);
+	const root = parseHTML(result.content).document;
 
 	// Extract main content from common Read the Docs selectors
 	let mainContent =
@@ -60,7 +60,7 @@ export const handleReadTheDocs: SpecialHandler = async (
 		?.querySelectorAll(
 			".headerlink, .viewcode-link, nav, .sidebar, footer, .related, .sphinxsidebar, .toctree-wrapper",
 		)
-		.forEach(el => {
+		.forEach((el: Element) => {
 			el.remove();
 		});
 

package/src/web/scrapers/twitter.ts

@@ -1,4 +1,4 @@
-import { parse as parseHtml } from "node-html-parser";
+import { type HTMLElement, parseHTML } from "linkedom";
 import { ToolAbortError } from "../../tools/tool-errors";
 import type { RenderResult, SpecialHandler } from "./types";
 import { buildResult, loadPage } from "./types";
@@ -33,14 +33,14 @@ export const handleTwitter: SpecialHandler = async (
 
 			if (result.ok && result.content.length > 500) {
 				// Parse the Nitter HTML
-				const doc = parseHtml(result.content);
+				const doc = parseHTML(result.content).document;
 
 				// Extract tweet content
-				const tweetContent = doc.querySelector(".tweet-content")?.text?.trim();
-				const fullname = doc.querySelector(".fullname")?.text?.trim();
-				const username = doc.querySelector(".username")?.text?.trim();
-				const date = doc.querySelector(".tweet-date a")?.text?.trim();
-				const stats = doc.querySelector(".tweet-stats")?.text?.trim();
+				const tweetContent = doc.querySelector(".tweet-content")?.textContent?.trim();
+				const fullname = doc.querySelector(".fullname")?.textContent?.trim();
+				const username = doc.querySelector(".username")?.textContent?.trim();
+				const date = doc.querySelector(".tweet-date a")?.textContent?.trim();
+				const stats = doc.querySelector(".tweet-stats")?.textContent?.trim();
 
 				if (tweetContent) {
 					let md = `# Tweet by ${fullname || "Unknown"} (${username || "@?"})\n\n`;
@@ -49,12 +49,12 @@ export const handleTwitter: SpecialHandler = async (
 					if (stats) md += `---\n${stats.replace(/\s+/g, " ")}\n`;
 
 					// Check for replies/thread
-					const replies = doc.querySelectorAll(".timeline-item .tweet-content");
+					const replies = Array.from(doc.querySelectorAll(".timeline-item .tweet-content")) as HTMLElement[];
 					if (replies.length > 1) {
 						md += `\n---\n\n## Thread/Replies\n\n`;
-						for (const reply of Array.from(replies).slice(1, 10)) {
-							const replyUser = reply.parentNode?.querySelector(".username")?.text?.trim();
-							md += `**${replyUser || "@?"}**: ${reply.text?.trim()}\n\n`;
+						for (const reply of replies.slice(1, 10)) {
+							const replyUser = reply.parentElement?.querySelector(".username")?.textContent?.trim();
+							md += `**${replyUser || "@?"}**: ${reply.textContent?.trim()}\n\n`;
 						}
 					}
 

package/src/web/scrapers/wikipedia.ts

@@ -1,4 +1,4 @@
-import { parse as parseHtml } from "node-html-parser";
+import { parseHTML } from "linkedom";
 import type { RenderResult, SpecialHandler } from "./types";
 import { buildResult, loadPage } from "./types";
 
@@ -45,14 +45,13 @@ export const handleWikipedia: SpecialHandler = async (
 		const contentResult = await loadPage(contentUrl, { timeout, signal });
 
 		if (contentResult.ok) {
-			const doc = parseHtml(contentResult.content);
+			const doc = parseHTML(contentResult.content).document;
 
 			// Extract main content sections
 			const sections = doc.querySelectorAll("section");
 			for (const section of sections) {
 				const heading = section.querySelector("h2, h3, h4");
-				const headingText = heading?.text?.trim();
-
+				const headingText = heading?.textContent?.trim();
 				// Skip certain sections
 				if (
 					headingText &&
@@ -68,7 +67,7 @@ export const handleWikipedia: SpecialHandler = async (
 
 				const paragraphs = section.querySelectorAll("p");
 				for (const p of paragraphs) {
-					const text = p.text?.trim();
+					const text = p.textContent?.trim();
 					if (text && text.length > 20) {
 						md += `${text}\n\n`;
 					}

package/src/utils/ignore-files.ts

@@ -1,119 +0,0 @@
-/**
- * Ignore file handling for .gitignore/.ignore/.fdignore support when scanning directories.
- */
-import * as path from "node:path";
-import ignore from "ignore";
-
-export const IGNORE_FILE_NAMES = [".gitignore", ".ignore", ".fdignore"] as const;
-
-export type IgnoreMatcher = ignore.Ignore;
-
-/**
- * Convert a path to POSIX format (forward slashes).
- */
-export function toPosixPath(p: string): string {
-	return p.split(path.sep).join("/");
-}
-
-/**
- * Prefix an ignore pattern to make it relative to a subdirectory.
- * Returns null for comments and empty lines.
- */
-export function prefixIgnorePattern(line: string, prefix: string): string | null {
-	const trimmed = line.trim();
-	if (!trimmed) return null;
-	if (trimmed.startsWith("#") && !trimmed.startsWith("\\#")) return null;
-
-	let pattern = line;
-	let negated = false;
-
-	if (pattern.startsWith("!")) {
-		negated = true;
-		pattern = pattern.slice(1);
-	} else if (pattern.startsWith("\\!")) {
-		pattern = pattern.slice(1);
-	}
-
-	if (pattern.startsWith("/")) {
-		pattern = pattern.slice(1);
-	}
-
-	const prefixed = prefix ? `${prefix}${pattern}` : pattern;
-	return negated ? `!${prefixed}` : prefixed;
-}
-
-/**
- * Read and add ignore rules from a directory to the matcher.
- */
-export async function addIgnoreRules(
-	ig: IgnoreMatcher,
-	dir: string,
-	rootDir: string,
-	readFile: (path: string) => Promise<string | null>,
-): Promise<void> {
-	const relativeDir = path.relative(rootDir, dir);
-	const prefix = relativeDir ? `${toPosixPath(relativeDir)}/` : "";
-
-	for (const filename of IGNORE_FILE_NAMES) {
-		const ignorePath = path.join(dir, filename);
-		const content = await readFile(ignorePath);
-		if (!content) continue;
-
-		const patterns = content
-			.split(/\r?\n/)
-			.map(line => prefixIgnorePattern(line, prefix))
-			.filter((line): line is string => Boolean(line));
-
-		if (patterns.length > 0) {
-			ig.add(patterns);
-		}
-	}
-}
-
-/**
- * Read and add ignore rules from a directory to the matcher (synchronous version).
- */
-export function addIgnoreRulesSync(
-	ig: IgnoreMatcher,
-	dir: string,
-	rootDir: string,
-	readFileSync: (path: string) => string | null,
-): void {
-	const relativeDir = path.relative(rootDir, dir);
-	const prefix = relativeDir ? `${toPosixPath(relativeDir)}/` : "";
-
-	for (const filename of IGNORE_FILE_NAMES) {
-		const ignorePath = path.join(dir, filename);
-		const content = readFileSync(ignorePath);
-		if (!content) continue;
-
-		const patterns = content
-			.split(/\r?\n/)
-			.map(line => prefixIgnorePattern(line, prefix))
-			.filter((line): line is string => Boolean(line));
-
-		if (patterns.length > 0) {
-			ig.add(patterns);
-		}
-	}
-}
-
-/**
- * Create a fresh ignore matcher.
- */
-export function createIgnoreMatcher(): IgnoreMatcher {
-	return ignore();
-}
-
-/**
- * Check if a path should be ignored.
- * @param ig - The ignore matcher
- * @param root - The root directory for relative path calculation
- * @param fullPath - The full path to check
- * @param isDir - Whether the path is a directory
- */
-export function shouldIgnore(ig: IgnoreMatcher, root: string, fullPath: string, isDir: boolean): boolean {
-	const relPath = toPosixPath(path.relative(root, fullPath));
-	const ignorePath = isDir ? `${relPath}/` : relPath;
-	return ig.ignores(ignorePath);
-}