@promptbook/website-crawler 0.100.0-1 → 0.100.0-3

package/esm/index.es.js

@@ -29,7 +29,7 @@ const BOOK_LANGUAGE_VERSION = '1.0.0';
  * @generated
  * @see https://github.com/webgptorg/promptbook
  */
-const PROMPTBOOK_ENGINE_VERSION = '0.100.0-1';
+const PROMPTBOOK_ENGINE_VERSION = '0.100.0-3';
 /**
  * TODO: string_promptbook_version should be constrained to the all versions of Promptbook engine
  * Note: [💞] Ignore a discrepancy between file name and entity name
@@ -3503,7 +3503,36 @@ async function makeKnowledgeSourceHandler(knowledgeSource, tools, options) {
         if (fileContent.length > DEFAULT_MAX_FILE_SIZE /* <- TODO: Allow to pass different value to remote server */) {
             throw new LimitReachedError(`File is too large (${Math.round(fileContent.length / 1024 / 1024)}MB). Maximum allowed size is ${Math.round(DEFAULT_MAX_FILE_SIZE / 1024 / 1024)}MB.`);
         }
-        await tools.fs.writeFile(join(rootDirname, filepath), fileContent);
+        // Note: Try to cache the downloaded file, but don't fail if the filesystem is read-only
+        try {
+            await tools.fs.writeFile(join(rootDirname, filepath), fileContent);
+        }
+        catch (error) {
+            // Note: If we can't write to cache, we'll process the file directly from memory
+            //       This handles read-only filesystems like Vercel
+            if (error instanceof Error && (error.message.includes('EROFS') ||
+                error.message.includes('read-only') ||
+                error.message.includes('EACCES') ||
+                error.message.includes('EPERM'))) {
+                // Return a handler that works directly with the downloaded content
+                return {
+                    source: name,
+                    filename: null,
+                    url,
+                    mimeType,
+                    async asJson() {
+                        return JSON.parse(fileContent.toString('utf-8'));
+                    },
+                    async asText() {
+                        return fileContent.toString('utf-8');
+                    },
+                };
+            }
+            else {
+                // Re-throw other unexpected errors
+                throw error;
+            }
+        }
         // TODO: [💵] Check the file security
         // TODO: [🧹][🧠] Delete the file after the scraping is done
         return makeKnowledgeSourceHandler({ name, knowledgeSourceContent: filepath }, tools, {
@@ -6270,7 +6299,22 @@ class WebsiteScraper {
             extension: 'html',
             isVerbose,
         });
-        await this.tools.fs.writeFile(cacheFilehandler.filename, html, 'utf-8');
+        // Note: Try to cache the scraped content, but don't fail if the filesystem is read-only
+        try {
+            await this.tools.fs.writeFile(cacheFilehandler.filename, html, 'utf-8');
+        }
+        catch (error) {
+            // Note: If we can't write to cache, we'll continue without caching
+            //       This handles read-only filesystems like Vercel
+            if (error instanceof Error && (error.message.includes('EROFS') ||
+                error.message.includes('read-only') ||
+                error.message.includes('EACCES') ||
+                error.message.includes('EPERM'))) ;
+            else {
+                // Re-throw other unexpected errors
+                throw error;
+            }
+        }
         const markdown = this.showdownConverter.makeMarkdown(html, jsdom.window.document);
         return { ...cacheFilehandler, markdown };
     }