<![CDATA[
# DeviceAtlas Cloud Client API Configuration #
This section deals with configuring the DeviceAtlas Cloud API.
### Caching ###
The API can cache the returned data after a call to the DeviceAtlas Cloud service, this will speed up subsequent requests.
The API has a file caching mechanism. It is recommended to always use the file cache if possible.
The file cache stores the returned properties on your server disk. The cache location is typically the system temp directory but this can be overridden with the "CUSTOM_CACHE_DIR" property. Items in the cache expire after a set time period to ensure the data is up-to-date. This cache is enabled by default. It is recommended to always have file caching enabled even when you are using cookie cache.
Items in the cache expire after a set time period to ensure the data is up-to-date.
### DeviceAtlas Cloud Service End-points ###
The DeviceAtlas Cloud Service is powered by independent clusters of servers
spread around the world. This ensures optimum speed and reliability. The API is
able to automatically switch to a different end-point if the current end-point
becomes unavailable. It can also (optionally) auto-rank all of the service
end-points to choose the end-point with the lowest latency for your location.
The Cloud service end-points are defined in the SERVERS variable at the top of
the "Client.py" file.
```python
SERVERS = [
{'host': 'server1-url', 'port': server1-port},
{'host': 'server2-url', 'port': server2-port},
{'host': 'server3-url', 'port': server3-port},
]
```
By default the API will analyze the end-points from time to time to rank them by
their stability and response speed. The ranked list is then cached and used
whenever the Client API needs to query the DeviceAtlas Cloud Service. If an end-
point fails, the Client API will automatically switch to the next end-point on
the list.
There is no need to set the servers array if auto-ranking is turned on. If you
wish, you may re-order the array and turn auto-ranking off. In this case the API
will respect your preferred order of end-points and only switch to a different
end-point should the primary one fail to resolve.
#### Notes ####
* With the default auto-ranking settings, the ranking is done every 24 hours.
The actual time may be more than 24 hours as the ranking is only triggered by
a request to the Client API and the cached server list is older than value
set to AUTO_SERVER_RANKING_LIFETIME.
* During end-point analysis a number of requests are made to each end-point.
Please note that these requests count towards your total hits to
the Cloud service.
e.g:
```
if
# Server list contains 3 servers
# AND Client.AUTO_SERVER_RANKING_LIFETIME == 1440
# AND Client.AUTO_SERVER_RANKING_NUM_REQUESTS == 3
then
# Auto ranking will add 9 (3x3) hits per day
```
#### Methods ####
* Get the ranked server list:
```python
rankedServerList = client.getServers()
```
The first end-point in the list will be used to make a request to the cloud, if
it fails the next end-point will be take it's place.
* Get the end-point used for the last request:
```python
server = client.getCloudUrl()
```
Note that if the data comes from cache this method will return "None".
* Get end-point info. This is useful when you want to manually rank the server list:
```python
server = client.getServersLatencies()
```
Please see https://deviceatlas.com/resources/cloud-service-end-points for more
information.
#### Cloud Server end-point settings ####
##### AUTO_SERVER_RANKING = True #####
To turn auto ranking on/off. To manually rank the servers set to "false"
and edit the $SERVERS array to set your preferred order of end-points.
The API will not rank the servers and will use the SERVERS list items
directly with the topmost server used first to get device data. On fail-
over the next end-point in the list will be used.
##### CLOUD_SERVICE_TIMEOUT = 2 #####
Time in seconds. If an end-point fails to respond in this amount of time
the API will fail-over to the next end-point on the list.
##### AUTO_SERVER_RANKING_MAX_FAILURE = 1 #####
When auto ranking servers, if a server fails more than this number of
times it will not be included in the list.
##### AUTO_SERVER_RANKING_NUM_REQUESTS = 3 #####
When auto ranking servers, number of requests to perform for service
speed calculation.
##### AUTO_SERVER_RANKING_LIFETIME = 1440 #####
Time in minutes. How often to auto rank servers.
0 = servers will be ranked and cached only once and this list will not be
updated automatically. You can update this list manually:
client::rankServers()
Note: AUTO_SERVER_RANKING must be set "true" so this cached server list
will be used by the API, even if AUTO_SERVER_RANKING_LIFETIME is set to 0.
If AUTO_SERVER_RANKING = False then the cached server list will be
totally ignored.
##### SERVER_PHASEOUT_LIFETIME = 1440 #####
Used when auto ranking is OFF. Specifies how long to use the fail-over
endpoints before the preferred end-point is re-checked. If the preferred
end-point is available it will be added back into the list of end-points
and used for future requests.
#### Usage with Client Side Component ####
This script gathers the properties and
creates a cookie containing them. This cookie is sent to the server on the next
request. Both client side and server side properties are merged and additional
logic is used to determine other properties such iPhone and iPad models which
are normally not detectable.
By default, if the cookie exists it will be used by the API. To disable using
the client side cookie:
```python
client = Client()
client.LICENCE_KEY = 'ENTER-YOUR-LICENCE-KEY'
client.USE_CLIENT_COOKIE = False
```
### Extra Tools ###
This package comes with extra tools that can help you enhance your mobile websites.
#### Latency Checker ####
There are two command line tools included in this package to help you tweak the
DeviceAtlas Cloud API end-point selecting.
##### Check Server Latency #####
This tool can be used to get info about the DeviceAtlas Cloud Service end-points.
The info shown can be used for manually setting up the DeviceAtlas Cloud service
end-points in the API. To access the cloud servers a valid DeviceAtlas licence
would be required.
Usage (command line):
```shell
$ python deviceatlas-cloud-python-2.2.1/ExtraTools/LatencyChecker/CheckServerLatency.py LICENCE_KEY
```
##### Update Server Ranking Cache #####
When AUTO_RANKING is set ON you can use this tool to re rank the DeviceAtlas Cloud
end-points. All you need to do is to run this script in regular intervals which
are smaller than the "AUTO_SERVER_RANKING_LIFETIME".
Usage (command line):
```shell
$ python deviceatlas-cloud-python-2.2.1/ExtraTools/LatencyChecker/UpdateServerRankingCache.py LICENCE_KEY
```
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
_ Copyright (c) DeviceAtlas Limited 2025. All Rights Reserved. _
_ https://deviceatlas.com _
<!-- HTML+JS for document formatting when opened in browser -->
<div class="btn-group" id="main-menu" style="float:right"><a class="btn dropdown-toggle" data-toggle="dropdown" href="#">Menu<span class="caret"></span></a><ul class="dropdown-menu"><li><a href="README.html">Cloud Client API</a></li><li><a href="README.Installation.html">Cloud Client API Installation</a></li><li><a href="README.Upgrade.html">Cloud Client API Upgrade</a></li><li class="disabled"><a href="README.Config.html">Cloud Client API Configuration</a></li><li><a href="./ApiDocs/index.html">DeviceAtlas Cloud Client API doc</a></li></ul></div>]]>