From d6fa55bc87b9de6d13278b148946b9e4d54bceb9 Mon Sep 17 00:00:00 2001
From: "Jonathan C. McKinney" <pseudotensor@gmail.com>
Date: Thu, 31 Oct 2024 09:38:55 -0700
Subject: [PATCH] Protection

---
 src/gpt_langchain.py | 20 ++++++++++++--------
 src/version.py       |  2 +-
 2 files changed, 13 insertions(+), 9 deletions(-)

diff --git a/src/gpt_langchain.py b/src/gpt_langchain.py
index 07afebd64..3eb7cd920 100644
--- a/src/gpt_langchain.py
+++ b/src/gpt_langchain.py
@@ -4815,14 +4815,18 @@ def file_to_doc(file,
                 docs1.extend(docs1a)
             if len(docs1) == 0 and have_playwright or do_playwright:
                 # then something went wrong, try another loader:
-                from langchain_community.document_loaders import PlaywrightURLLoader
-                docs1a = asyncio.run(PlaywrightURLLoader(urls=final_urls).aload())
-                # docs1 = PlaywrightURLLoader(urls=[file]).load()
-                docs1a = [x for x in docs1a if
-                          x.page_content and x.page_content != '403 Forbidden' and not x.page_content.startswith(
-                              'Access Denied')]
-                add_parser(docs1a, 'PlaywrightURLLoader')
-                docs1.extend(docs1a)
+                try:
+                    from langchain_community.document_loaders import PlaywrightURLLoader
+                    docs1a = asyncio.run(PlaywrightURLLoader(urls=final_urls).aload())
+                    # docs1 = PlaywrightURLLoader(urls=[file]).load()
+                    docs1a = [x for x in docs1a if
+                              x.page_content and x.page_content != '403 Forbidden' and not x.page_content.startswith(
+                                  'Access Denied')]
+                    add_parser(docs1a, 'PlaywrightURLLoader')
+                    docs1.extend(docs1a)
+                except Exception as e0:
+                    traceback.print_exc()
+                    print("playwright failed: %s: %s" % (str(e0), traceback.print_exception(e0)), flush=True)
             if len(docs1) == 0 and have_selenium or do_selenium:
                 # then something went wrong, try another loader:
                 # but requires Chrome binary, else get: selenium.common.exceptions.WebDriverException:
diff --git a/src/version.py b/src/version.py
index 15ed4b002..f9168f1ec 100644
--- a/src/version.py
+++ b/src/version.py
@@ -1 +1 @@
-__version__ = "f5a3cf5b09f5845a7177d1fff1c97b5267804202"
+__version__ = "f3ce8a2491b1387b727280424a61680be896013b"