How to make the most of the config file
Requirements specification
[optional] - Any package specified will be checked.
Use cases:
A new feature is introduced and used in the test
Breaking change introduced with a new release
Specific package used in a test that does not belong to pykiso
#__________________________ Requirements section ________________________
# FEATURE: Check the environment before running the tests
#
# The version can be:
# - specified alone (minimum version accepted)
# - conditioned using <, <=, >, >=, == or !=
# - no specified using 'any' (accept any version)
#
# /!\: If the check fail, the tests will not start and the mismatch
# displayed
#_______________________________________________________________________
requirements:
- pykiso: '>=0.10.1, <1.0.0'
- robotframework: 3.2.2
- pyyaml: any
Real-World Configuration File
1auxiliaries:
2 DUT:
3 connectors:
4 com: uart
5 config: null
6 type: pykiso.lib.auxiliaries.example_test_auxiliary:ExampleAuxiliary
7connectors:
8 uart:
9 config:
10 serialPort: COM3
11 type: pykiso.lib.connectors.cc_uart:CCUart
12test_suite_list:
13- suite_dir: test_suite_1
14 test_filter_pattern: '*.py'
15 test_suite_id: 1
Activation of specific loggers
By default, every logger that does not belong to the pykiso
package or that is not an auxiliary
logger will see its level set to WARNING even if you have in the command line pykiso --log-level DEBUG
.
This aims to reduce redundant logs from additional modules during the test execution.
For keeping specific loggers to the set log-level, it is possible to set the activate_log
parameter in the auxiliary
config.
The following example activates the jlink
logger from the pylink
package, imported in cc_rtt_segger.py
:
auxiliaries:
aux1:
connectors:
com: rtt_channel
config:
activate_log:
# only specifying pylink will include child loggers
- pylink.jlink
- my_pkg
type: pykiso.lib.auxiliaries.dut_auxiliary:DUTAuxiliary
connectors:
rtt_channel:
config: null
type: pykiso.lib.connectors.cc_rtt_segger:CCRttSegger
Based on this example, by specifying my_pkg
, all child loggers will also be set to the set log-level.
Note
If e.g. only the logger my_pkg.module_1
should be set to the level, it should be entered as such.
Ability to use environment variables
It is possible to replace any value by an environment variable in the YAML files.
When using environment variables, the following format should be respected: ENV{my-env-var}
.
In the following example, an environment variable called TEST_SUITE_1
contains the path to the test suite 1 directory.
- suite_dir: ENV{TEST_SUITE_1}
test_filter_pattern: '*.py'
test_suite_id: 1
- suite_dir: ENV{TEST_SUITE_2=./test_suite_2}
It is also possible to set a default value in case the environment variable is not found.
The following format should be used: ENV{my-env-var=my_default_value}
.
In the following example, an environment variable called TEST_SUITE_2
would contain the path to
the test_suite_2 directory. If the variable is not set, the default value will be taken instead.
config: null
Specify files and folders
To specify files and folders you can use absolute or relative paths. Relative paths are always given relative to the location of the yaml file.
According to the YAML specification, values enclosed in single quotes are enforced as strings, and will not be parsed.
example_config:
# this relative path will not be made absolute
rel_script_path_unresolved: './script_folder/my_awesome_script.py'
# this one will
rel_script_path: ./script_folder/my_awesome_script.py
abs_script_path_win: C:/script_folder/my_awesome_script.py
abs_script_path_unix: /home/usr/script_folder/my_awesome_script.py
Warning
Relative path or file locations must always start with ./
.
If not, it will still be resolved but unexpected behaviour can result from it.
Include sub-YAMLs
Frequently used configuration parts can be stored in a separate YAML file. To include this configuration file in the main one, the path to the sub-configuration file has to be provided, preceded with the !include tag.
Relative paths in the sub-YAML file are then resolved relative to the sub-YAML’s location.
#__________________________ Connectors section _________________________
# FEATURE : yaml in yaml
#
# In order to call a yaml file inside a other yaml file, the special
# tag !include has to be used. The path could be in a relative or
# absolute form.
#_______________________________________________________________________
connectors:
<chan>: !include ./channel_config/my_channel_config.yaml
Delay an auxiliary start-up
All threaded auxiliaries are capable to delay their start-up (not starting at import level). This means, from user point of view, it’s possible to start it on demand and especially where it’s really needed.
Warning
in an explicitely defined proxy setup (for shared communication channels) be sure to
always start the proxy auxiliary last. Otherwise, an error will occur due to the specific
ProxyAuxiliary
import rules.
To avoid such corner cases, consider switching to an implicit definition of shared communication channels.
In order to achieved that, a parameter was added at the auxiliary configuration level.
auxiliaries:
proxy_aux:
connectors:
com: can_channel
config:
aux_list: [aux1, aux2]
activate_trace: True
trace_dir: ./suite_proxy
trace_name: can_trace
# if False create the auxiliary instance but don't start it, an
# additional call of start method has to be performed.
# By default, auto_start flag is set to True and "normal" ITF aux
# creation mechanism is used.
auto_start: False
type: pykiso.lib.auxiliaries.proxy_auxiliary:ProxyAuxiliary
aux1:
connectors:
com: proxy_com1
config:
auto_start: False
type: pykiso.lib.auxiliaries.communication_auxiliary:CommunicationAuxiliary
aux2:
connectors:
com: proxy_com2
config:
auto_start: False
type: pykiso.lib.auxiliaries.communication_auxiliary:CommunicationAuxiliary
In user’s script simply call the related auxiliary start method:
used for sending and the other one for the reception
"""
def setUp(self):
"""If a fixture is not use just override it like below."""
logging.info(
f"--------------- SETUP: {self.test_suite_id}, {self.test_case_id} ---------------"
Make a proxy auxiliary trace
Proxy auxiliary is capable of creating a trace file, where all received messages at connector
level are written. This feature is useful when proxy auxiliary is associated with a communication channel
that doesn’t have any tracing capability (in contrast to CCPcanCan
or CCRttSegger
for example).
Everything is handled at configuration level within the YAML test configuration file:
proxy_aux:
connectors:
# communication channel alias
com: <channel-alias>
config:
# Auxiliaries alias list bound to proxy auxiliary
aux_list : [<aux alias 1>, <aux alias 2>, <aux alias 3>]
# activate trace at proxy level, sniff everything received at
# connector level and write it in .log file.
activate_trace : True
# by default the trace is placed where pykiso is launched
# otherwise user should specify his own path
# (absolute and relative)
trace_dir: ./suite_proxy
# by default the trace file's name is :
# YY-MM-DD_hh-mm-ss_proxy_logging.log
# otherwise user should specify his own name
trace_name: can_trace
type: pykiso.lib.auxiliaries.proxy_auxiliary:ProxyAuxiliary
Note
This feature is only available with an explicit proxy definition as shown above.