From bfa87980ed0cc73f2a72cbc2dfc051d50699dc6c Mon Sep 17 00:00:00 2001
From: Samuel Miserendino <samuelgm@openai.com>
Date: Mon, 31 Mar 2025 22:06:39 -0700
Subject: [PATCH 1/2] add utility to download videos

---
 utils/download_videos.py | 62 ++++++++++++++++++++++++++++++++++++++++
 1 file changed, 62 insertions(+)
 create mode 100644 utils/download_videos.py

diff --git a/utils/download_videos.py b/utils/download_videos.py
new file mode 100644
index 00000000..88afad0e
--- /dev/null
+++ b/utils/download_videos.py
@@ -0,0 +1,62 @@
+import os
+import re
+import requests
+
+def fetch_issue(issue_number):
+    url = f"https://api.github.com/repos/Expensify/App/issues/{issue_number}"
+    headers = {
+        "Accept": "application/vnd.github.v3+json",
+        "User-Agent": "ExpensifyVideoDownloader"
+    }
+    try:
+        response = requests.get(url, headers=headers)
+        response.raise_for_status()
+        return response.json()
+    except Exception as e:
+        print(f"Error fetching issue {issue_number}: {e}")
+        return None
+
+def download_issue_videos(issue):
+    title = issue.get('title', '<No Title>')
+    body = issue.get('body', '')
+    
+    print("Issue Title:", title)
+    print("Issue Body:", body)
+
+    video_urls = re.findall(r'(https?://[^\s]+?\.(?:mp4|mov))', body)
+    if not video_urls:
+        print("No .mp4 or .mov files found in the issue body.")
+        return
+
+    issue_id = str(issue.get('number') or issue.get('id', 'unknown'))
+    destination_dir = os.path.join("issue_videos", issue_id)
+    os.makedirs(destination_dir, exist_ok=True)
+
+    for url in video_urls:
+        video_name = os.path.basename(url.split('?')[0])
+        destination_path = os.path.join(destination_dir, video_name)
+        print(f"Downloading {url} to {destination_path} ...")
+
+        try:
+            video_response = requests.get(url, stream=True)
+            video_response.raise_for_status()
+            with open(destination_path, 'wb') as out_file:
+                for chunk in video_response.iter_content(chunk_size=8192):
+                    if chunk:
+                        out_file.write(chunk)
+            print(f"Downloaded {video_name} successfully.")
+        except Exception as e:
+            print(f"Error downloading {url}: {e}")
+
+def fetch_and_download_issue_videos(issue_number):
+    issue = fetch_issue(issue_number)
+    if issue is not None:
+        download_issue_videos(issue)
+
+if __name__ == '__main__':
+    import sys
+    if len(sys.argv) < 2:
+        print("Usage: python utils/download_videos.py <issue_number>")
+    else:
+        issue_number = sys.argv[1]
+        fetch_and_download_issue_videos(issue_number)

From a4b2e5cdfe592d7280bd7cb7c7146f8f4f879df6 Mon Sep 17 00:00:00 2001
From: Samuel Miserendino <samuelgm@openai.com>
Date: Mon, 31 Mar 2025 22:08:22 -0700
Subject: [PATCH 2/2] update readme

---
 README.md | 6 ++++++
 1 file changed, 6 insertions(+)

diff --git a/README.md b/README.md
index 197cdaac..08765c62 100644
--- a/README.md
+++ b/README.md
@@ -183,3 +183,9 @@ For a complete example of a ComputerInterface implementation, you can refer to t
       url={https://arxiv.org/abs/2502.12115}, 
 }
 ```
+
+## Utilities 
+
+We include the following utilities to facilitate future research:
+
+- `download_videos.py` allows you to download the videos attached to an Expensify GitHub issue if your model supports video input
\ No newline at end of file