def get_channel_id_with_timeout(integration, name, timeout):
"""
Fetches the internal slack id of a channel.
:param integration: The slack integration
:param name: The name of the channel
:param timeout: Our self-imposed time limit.
:return: a tuple of three values
1. prefix: string (`"#"` or `"@"`)
2. channel_id: string or `None`
3. timed_out: boolean (whether we hit our self-imposed time limit)
"""
token_payload = {"token": integration.metadata["access_token"]}
# Look for channel ID
payload = dict(token_payload, **{
"exclude_archived": False,
"exclude_members": True
})
# workspace tokens are the only tokens that don't works with the conversations.list endpoint,
# once eveyone is migrated we can remove this check and usages of channels.list
if get_integration_type(integration) == "workspace_app":
list_types = LEGACY_LIST_TYPES
else:
list_types = LIST_TYPES
payload = dict(payload, **{"types": "public_channel,private_channel"})
time_to_quit = time.time() + timeout
client = SlackClient()
id_data = None
found_duplicate = False
for list_type, result_name, prefix in list_types:
cursor = ""
while True:
endpoint = "/%s.list" % list_type
try:
# Slack limits the response of `<list_type>.list` to 1000 channels
items = client.get(endpoint,
params=dict(payload,
cursor=cursor,
limit=1000))
except ApiError as e:
logger.info("rule.slack.%s_list_failed" % list_type,
extra={"error": six.text_type(e)})
return (prefix, None, False)
for c in items[result_name]:
# The "name" field is unique (this is the username for users)
# so we return immediately if we find a match.
# convert to lower case since all names in Slack are lowercase
if c["name"].lower() == name.lower():
return (prefix, c["id"], False)
# If we don't get a match on a unique identifier, we look through
# the users' display names, and error if there is a repeat.
if list_type == "users":
profile = c.get("profile")
if profile and profile.get("display_name") == name:
if id_data:
found_duplicate = True
else:
id_data = (prefix, c["id"], False)
cursor = items.get("response_metadata",
{}).get("next_cursor", None)
if time.time() > time_to_quit:
return (prefix, None, True)
if not cursor:
break
if found_duplicate:
raise DuplicateDisplayNameError(name)
elif id_data:
return id_data
return (prefix, None, False)
def get_channel_id_with_timeout(
integration: Integration,
name: Optional[str],
timeout: int,
) -> Tuple[str, Optional[str], bool]:
"""
Fetches the internal slack id of a channel.
:param integration: The slack integration
:param name: The name of the channel
:param timeout: Our self-imposed time limit.
:return: a tuple of three values
1. prefix: string (`"#"` or `"@"`)
2. channel_id: string or `None`
3. timed_out: boolean (whether we hit our self-imposed time limit)
"""
headers = {"Authorization": f"Bearer {integration.metadata['access_token']}"}
payload = {
"exclude_archived": False,
"exclude_members": True,
"types": "public_channel,private_channel",
}
list_types = LIST_TYPES
time_to_quit = time.time() + timeout
client = SlackClient()
id_data: Optional[Tuple[str, Optional[str], bool]] = None
found_duplicate = False
prefix = ""
for list_type, result_name, prefix in list_types:
cursor = ""
while True:
endpoint = f"/{list_type}.list"
try:
# Slack limits the response of `<list_type>.list` to 1000 channels
items = client.get(
endpoint, headers=headers, params=dict(payload, cursor=cursor, limit=1000)
)
except ApiRateLimitedError as e:
logger.info(f"rule.slack.{list_type}_list_rate_limited", extra={"error": str(e)})
raise e
except ApiError as e:
logger.info(f"rule.slack.{list_type}_list_rate_limited", extra={"error": str(e)})
return (prefix, None, False)
if not isinstance(items, dict):
continue
for c in items[result_name]:
# The "name" field is unique (this is the username for users)
# so we return immediately if we find a match.
# convert to lower case since all names in Slack are lowercase
if name and c["name"].lower() == name.lower():
return (prefix, c["id"], False)
# If we don't get a match on a unique identifier, we look through
# the users' display names, and error if there is a repeat.
if list_type == "users":
profile = c.get("profile")
if profile and profile.get("display_name") == name:
if id_data:
found_duplicate = True
else:
id_data = (prefix, c["id"], False)
cursor = items.get("response_metadata", {}).get("next_cursor", None)
if time.time() > time_to_quit:
return (prefix, None, True)
if not cursor:
break
if found_duplicate:
raise DuplicateDisplayNameError(name)
elif id_data:
return id_data
return (prefix, None, False)
