### Installation
```
pip install flagsup
```
### Integration
- Docs: [Integration doc](https://confluence.teko.vn/display/PS/User+guide+for+setting+up+feature+flags)
- Init client
```
import flagsup
flagsup_client = flagsup.CachedFlagsupClient("https://flagsup.dev.tekoapis.net")
```
- To get flag status (and experiment info if available) for a specific user using User IAM ID
- `batchEvaluateFlags(<flagsup.EvaluateRequest>, [<default_status>, <default_treatment>, <default_exp_id>, <default_branch_id>])`
- `evaluateFlag(<flag_key>, <iam_id>, [<default_status>, <default_treatment>, <default_exp_id>, <default_branch_id>])`
It is recommended to call batchEvaluateFlags() for all your feature flags upon getting userId from IAM service
Subsequently you can call evaluateFlag() to get flag value from in-memory cache
- To get flag status when user is not logged in (**note**: this will disable canary release, black/white list, experiment)
- `batchGetFlags(<flag_key_list>, [<default_status>])`
- `getFlag(<flag_key>, [<default_status>])`
Similar to `evaluateFlag` It is recommended to call `batchGetFlags()` for all your feature flags upon initial load
Subsequently you can call `getFlag()` to get flag value from in-memory cache
### Example
```
import flagsup
if __name__ == "__main__":
# Initialisation, provide FlagSup SDK with target domain
flagsup_client = flagsup.CachedFlagsupClient("https://flagsup.dev.tekoapis.net")
# Option 1: Get flags with User ID
# It is recommended to call batchEvaluateFlags() for all your feature flags upon getting userId from IAM service
# Subsequently you can call evaluateFlag() to get flag value from in-memory cache
# Fetch multiple flag values for the same or different users. Response is cached in memory for subsequent calls.
# To reduce network overhead, call this to fetch all flags when your app initialise
batch_response = flagsup_client.batch_evaluate_flag([
flagsup.EvaluateRequest("flag01", "5f8e4768e00e4d12ba69ed3e7bf3e893"), # pragma: whitelist secret
flagsup.EvaluateRequest("flag02", "5f8e4768e00e4d12ba69ed3e7bf3e893"), # pragma: whitelist secret
])
# Fetch a flag value for the logged in user. Response is cached in memory for subsequent calls.
flag_status = flagsup_client.evaluate_flag("flag01", "5f8e4768e00e4d12ba69ed3e7bf3e893") # pragma: whitelist secret
if flag_status:
# Flow when flag is enabled
pass
else:
# // Flow when flag is disabled
pass
# Option 2: Get flags without User ID
# Fetch multiple flag values. Response is cached in memory for subsequent calls.
# To reduce network overhead, call this to fetch all flags when your app initialise
batch_response = flagsup_client.batch_get_flag_status(["flag01", "flag02"])
# Fetch a single flag value. Response is cached in memory for subsequent calls.
flag_status = flagsup_client.get_flag_status("flag01")
```
