)]}'
{"/PATCHSET_LEVEL":[{"author":{"_account_id":39146,"name":"Nathaniel Martes","display_name":"Nate Martes","email":"nmartes@NVIDIA.com","username":"nmartes"},"change_message_id":"a08a277edb46750723da597f0161c7884353bdd8","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":6,"id":"049b7c3e_b625f464","updated":"2026-06-02 21:50:24.000000000","message":"Looks good to me! we need this","commit_id":"db1331733e21a199c09b870c9363c1217befed81"},{"author":{"_account_id":38496,"name":"Andressa Cabistani","display_name":"Andressa","email":"acabistani@gmail.com","username":"andressadotpy","status":"I\u0027m a Software Engineer at Red Hat and I love Open Source and connect with people! Feel free to DM through IRC, I\u0027ll be delighted to chat"},"change_message_id":"6c42d3888b65a24d779802a2fe12f8b284a8e50d","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":10,"id":"022b58c3_25fc87c1","updated":"2026-07-09 15:47:02.000000000","message":"I reviewed this patch as part of a dependency chain review starting from patch 981747.\n\nThis is looking really good and my testing confirmed the performance improvements. I would like to see the performance benefit documented for future reference.","commit_id":"ad7607b043e8de14a468d95f864f9da1a5d14b99"},{"author":{"_account_id":39146,"name":"Nathaniel Martes","display_name":"Nate Martes","email":"nmartes@NVIDIA.com","username":"nmartes"},"change_message_id":"4d443bce4af9349451e5609ef61047ab34a6c199","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":11,"id":"a0cec6a2_b761ab93","updated":"2026-07-08 15:36:21.000000000","message":"LGTM","commit_id":"7c79eefdeeb15eab9f65994cace593876337c23e"}],"swift/common/wsgi_gunicorn.py":[{"author":{"_account_id":39146,"name":"Nathaniel Martes","display_name":"Nate Martes","email":"nmartes@NVIDIA.com","username":"nmartes"},"change_message_id":"a08a277edb46750723da597f0161c7884353bdd8","unresolved":false,"context_lines":[{"line_number":150,"context_line":"    # self.reader.read(1024) call to use the actual requested size. This avoids"},{"line_number":151,"context_line":"    # useless copying of data and speeds up chunked transfers significantly."},{"line_number":152,"context_line":"    # Remove when https://github.com/benoitc/gunicorn/issues/2596 is closed"},{"line_number":153,"context_line":"    def swift_read(self, size\u003dNone):"},{"line_number":154,"context_line":"        size \u003d self.getsize(size)"},{"line_number":155,"context_line":"        if size \u003d\u003d 0:"},{"line_number":156,"context_line":"            return b\"\""}],"source_content_type":"text/x-python","patch_set":6,"id":"5c73f3a6_d1b66693","line":153,"range":{"start_line":153,"start_character":0,"end_line":153,"end_character":4},"updated":"2026-06-02 21:50:24.000000000","message":"Really good here!","commit_id":"db1331733e21a199c09b870c9363c1217befed81"},{"author":{"_account_id":38496,"name":"Andressa Cabistani","display_name":"Andressa","email":"acabistani@gmail.com","username":"andressadotpy","status":"I\u0027m a Software Engineer at Red Hat and I love Open Source and connect with people! Feel free to DM through IRC, I\u0027ll be delighted to chat"},"change_message_id":"6c42d3888b65a24d779802a2fe12f8b284a8e50d","unresolved":true,"context_lines":[{"line_number":190,"context_line":"            return ret"},{"line_number":191,"context_line":""},{"line_number":192,"context_line":"        while size \u003e self.buf.tell():"},{"line_number":193,"context_line":"            data \u003d self.reader.read(size)  # changed to size from 1024"},{"line_number":194,"context_line":"            if not data:"},{"line_number":195,"context_line":"                break"},{"line_number":196,"context_line":"            self.buf.write(data)"}],"source_content_type":"text/x-python","patch_set":11,"id":"3971fc5a_d658d3b6","line":193,"updated":"2026-07-09 15:47:02.000000000","message":"I did some poking around to verify the memory implications with large chunk sizes.\n\n```\nChunk Size      Old Memory      New Memory      Difference\n64KB            1,024 B         65,536 B        +64,512 B\n256KB           1,024 B         262,144 B       +261,120 B\n512KB           1,024 B         524,288 B       +523,264 B\n1MB             1,024 B         1,048,576 B     +1,047,552 B\n```\n\nI\u0027m not sure if in a realistic scenario this is valid.\n\nQuestion: Have you tested with realistic Swift workloads?","commit_id":"7c79eefdeeb15eab9f65994cace593876337c23e"}]}