Sometimes, TG partly crashes, there are communication issues or TG internal timeout does not work (no response after timeout time).
What I propose:
Give an ability to provide HTTP timeout option.
The existing timeout parameter still would be a TG internal timeout (no changes), and http_timout would be a new parameter, that will be passed to requests.request call as timeout.
Example:
conn.getVertices(..., timout=10, http_timeout=(10, 20))
Therefore:
in the library _req method, instead of:
requests.request(method, url, headers=_headers, json=_data, params=params, verify=verify)
would be:
requests.request(method, url, headers=_headers, json=_data, params=params, verify=verify, timeout=http_timeout)
Sometimes, TG partly crashes, there are communication issues or TG internal timeout does not work (no response after timeout time).
What I propose:
Give an ability to provide HTTP timeout option.
The existing timeout parameter still would be a TG internal timeout (no changes), and http_timout would be a new parameter, that will be passed to
requests.requestcall as timeout.Example:
conn.getVertices(..., timout=10, http_timeout=(10, 20))Therefore:
in the library _req method, instead of:
requests.request(method, url, headers=_headers, json=_data, params=params, verify=verify)would be:
requests.request(method, url, headers=_headers, json=_data, params=params, verify=verify, timeout=http_timeout)