From 1641fff9d7c5d38af2fa297f09e4b379d8364b4f Mon Sep 17 00:00:00 2001
From: Kentaro Wada <www.kentaro.wada@gmail.com>
Date: Tue, 16 Jan 2024 22:14:44 +0900
Subject: [PATCH] Set default user-agent with giving option to override it

---
 gdown/cli.py             |  6 ++++++
 gdown/download.py        | 17 ++++++++++++-----
 gdown/download_folder.py | 10 +++++++++-
 3 files changed, 27 insertions(+), 6 deletions(-)

diff --git a/gdown/cli.py b/gdown/cli.py
index bf80deaa..76171b9e 100644
--- a/gdown/cli.py
+++ b/gdown/cli.py
@@ -121,6 +121,10 @@ def main():
         help="Format of Google Docs, Spreadsheets and Slides. "
         "Default is Google Docs: 'docx', Spreadsheet: 'xlsx', Slides: 'pptx'.",
     )
+    parser.add_argument(
+        "--user-agent",
+        help="User-Agent to use for downloading file.",
+    )
 
     args = parser.parse_args()
 
@@ -159,6 +163,7 @@ def main():
                 use_cookies=not args.no_cookies,
                 verify=not args.no_check_certificate,
                 remaining_ok=args.remaining_ok,
+                user_agent=args.user_agent,
             )
         else:
             download(
@@ -173,6 +178,7 @@ def main():
                 fuzzy=args.fuzzy,
                 resume=args.continue_,
                 format=args.format,
+                user_agent=args.user_agent,
             )
     except FileURLRetrievalError as e:
         print(e, file=sys.stderr)
diff --git a/gdown/download.py b/gdown/download.py
index bcc37675..e9b376d2 100644
--- a/gdown/download.py
+++ b/gdown/download.py
@@ -54,12 +54,10 @@ def get_url_from_gdrive_confirmation(contents):
     return url
 
 
-def _get_session(proxy, use_cookies, return_cookies_file=False):
+def _get_session(proxy, use_cookies, user_agent, return_cookies_file=False):
     sess = requests.session()
 
-    sess.headers.update(
-        {"User-Agent": "Mozilla/5.0 (Macintosh; Intel Mac OS X 10_12_6)"}
-    )
+    sess.headers.update({"User-Agent": user_agent})
 
     if proxy is not None:
         sess.proxies = {"http": proxy, "https": proxy}
@@ -91,6 +89,7 @@ def download(
     fuzzy=False,
     resume=False,
     format=None,
+    user_agent=None,
 ):
     """Download file from URL.
 
@@ -124,6 +123,8 @@ def download(
             - Google Docs: 'docx'
             - Google Spreadsheet: 'xlsx'
             - Google Slides: 'pptx'
+    user_agent: str, optional
+        User-agent to use in the HTTP request.
 
     Returns
     -------
@@ -134,11 +135,17 @@ def download(
         raise ValueError("Either url or id has to be specified")
     if id is not None:
         url = "https://drive.google.com/uc?id={id}".format(id=id)
+    if user_agent is None:
+        # We need to use different user agent for file download c.f., folder
+        user_agent = "Mozilla/5.0 (Macintosh; Intel Mac OS X 10_10_1) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/39.0.2171.95 Safari/537.36"  # NOQA: E501
 
     url_origin = url
 
     sess, cookies_file = _get_session(
-        proxy=proxy, use_cookies=use_cookies, return_cookies_file=True
+        proxy=proxy,
+        use_cookies=use_cookies,
+        user_agent=user_agent,
+        return_cookies_file=True,
     )
 
     gdrive_file_id, is_gdrive_download_link = parse_url(url, warning=not fuzzy)
diff --git a/gdown/download_folder.py b/gdown/download_folder.py
index 3153b7ec..81bee678 100644
--- a/gdown/download_folder.py
+++ b/gdown/download_folder.py
@@ -203,6 +203,7 @@ def download_folder(
     use_cookies=True,
     remaining_ok=False,
     verify=True,
+    user_agent=None,
 ):
     """Downloads entire folder from URL.
 
@@ -228,6 +229,8 @@ def download_folder(
         Either a bool, in which case it controls whether the server's TLS
         certificate is verified, or a string, in which case it must be a path
         to a CA bundle to use. Default is True.
+    user_agent: str, optional
+        User-agent to use in the HTTP request.
 
     Returns
     -------
@@ -245,8 +248,13 @@ def download_folder(
         raise ValueError("Either url or id has to be specified")
     if id is not None:
         url = "https://drive.google.com/drive/folders/{id}".format(id=id)
+    if user_agent is None:
+        # We need to use different user agent for folder download c.f., file
+        user_agent = "Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/98.0.4758.102 Safari/537.36"  # NOQA: E501
 
-    sess = _get_session(proxy=proxy, use_cookies=use_cookies)
+    sess = _get_session(
+        proxy=proxy, use_cookies=use_cookies, user_agent=user_agent
+    )
 
     if not quiet:
         print("Retrieving folder contents", file=sys.stderr)