Datafile Inputs¶
Whilst all your testcases, sections, data & AEtest features can be coded, set & leveraged directly within your testscript, altering their values between runs (aside from dynamic parameters and script arguments) require physical script changes, which may not always be desirable.
The datafile feature in AEtest allows users to run their testscript with an additional YAML based input file, allowing dynamic updates and/or overwrites to the current test script module with more information. It is an optional augmentation to regular script runs, allowing users to easily change script values and various features without modifying the original testscript.
Usages & Behavior¶
To leverage this feature, run your testscript and provide the datafile
Standard Arguments with full path/name to a YAML data file
written in accordance to the Datafile Schema below.
# Example
# -------
#
# example usages of datafile input
# (pseudo-code, snippet)
# as a standalone run argument
if __name__ == '__main__':
aetest.main(datafile = '/path/to/your/datafile.yaml')
# through jobfile run/Task argument
def main():
run(script, datafile = '/path/to/your/datafile.yaml')
# as command-line argument
# python script.py -datafile=/path/to/your/datafile.yaml
Once loaded, the content of this YAML file is then used to dynamically update your testscript objects. The following describes this behavior:
The content of this datafile updates the script’s module and classes directly, after the script is imported, before execution starts.
Only module & classes level attributes and features may be provided via the datafile (eg,
CommonSetup,Testcases,CommonCleanup). Function based sections such astest,subsectionetc are not affected.Testcases & common sections defined in the datafile must each match up to a corresponding class:
common_setup:andcommon_cleanup:to script’s uniqueCommonSetupandCommonCleanupsubclasses; testcases block’sclassname:to the actual testcase class definition name.If a testcase is looping (using Looping Sections feature), the base class definition that is being looped is updated. Datafiles cannot assign and/or remove testcase loop feature: it may only update the base class’s attributes & parameters.
Only a single datafile may be provided. However, each datafile may extend one or more datafiles, creating a chaining relationship effect. When a datafile extends another datafile, the other datafile forms the basis, and contents of the current datafile is then applied on top using recursive dictionary update.
Test Parameters provided via the various
parameters:block in the YAML file are updated into corresponding section’s base parameters usingdict.update()mechanism.all other key/value pairs, including Section Processors provided via the various
processors:block in the YAML file replace any existing values & settings.
Tip
to better understand this feature, see the datafile example script provided in `GitHub example repository<https://github.com/CiscoTestAutomation/examples/tree/master/datafiles>`_.
Hint
as YAML naturally loads in Python nested dictionaries, it is possible to
provide a dictionary that respects the Datafile Schema instead
of an input file. However, this only works for job file execution and/or
standalone execution through aetest.main(), and does not work over the
command line.
Datafile Schema¶
The input datafile must satisfy the following schema. Do not be discouraged by this long structure: most likely you will only need few of these fields. Beware of YAML’s sensitivity to indentation and whitespaces.
# Datafile Schema
# ---------------
extends: # Datafile(s) to extend/build on.
# Use this field to extend an existing datafile.
# Allows datafiles to be chained together in extension
# relationships.
# Supports full path/names or name of file in the same dir.
# The content of the last file on the list forms the base and
# is updated with the preceding file, and so on,
# until the existing file content is updated last.
# (optional)
parameters: # testscript parameters
# all key/values here becomes the testscript's base parameters
# (optional)
processors: # global processors
# pre/post processors to be used as part of this script run
# (optional)
pre: # list of global pre-processors
# eg: mylib.mymodule.preprocessor_func
#
# or, list of global pre-processors with arguments
# eg: - processor: mylib.mymodule.preprocessor_func
# args: <list of positional arguments>
# kwargs:
# <key>: <value>
# (optional)
post: # list of global post-processors
# eg: mylib.mymodule.postprocessor_func
#
# or, list of global post-processors with arguments
# eg: - processor: mylib.mymodule.postprocessor_func
# args: <list of positional arguments>
# kwargs:
# <key>: <value>
# (optional)
common_setup: # common_setup block
# everything related to script's common_setup section
# (optional)
parameters: # common_setup parameters
# key/values becomes parameters belonging to common_setup
# section.
# (optional)
processors: # common_setup local processors
# pre/post processors to be used on common_setup
# (optional)
pre: # list of pre-processors for common_setup section
# eg: mylib.mymodule.preprocessor_func
#
# or, list of common_setup pre-processors with arguments
# eg: - processor: mylib.mymodule.preprocessor_func
# args: <list of positional arguments>
# kwargs:
# <key>: <value>
# (optional)
post: # list of post-processors for common_setup section
# eg: mylib.mymodule.postprocessor_func
#
# or, list of common_setup post-processors with arguments
# eg: - processor: mylib.mymodule.postprocessor_func
# args: <list of positional arguments>
# kwargs:
# <key>: <value>
# (optional)
# any custom key/value pairs to be set as data (attributes) to
# your script's common_setup section
<key>: <value>
testcases: # testcases block
# all testcase related info gets defined under here
# (optional)
<name>: # testcase class name
# this needs to match the testcase's class definition.
# do not confuse with the testcase's uid
# eg: MyTestcase
# (mandatory)
uid: # testcase's string uid
# use this to alter the testcase's reported uid
# (optional)
groups: # testcase grouping
# list of groups this testcase belongs to. See testcase
# grouping feature under flow control documentation.
# (optional)
name: # testcase name
# define a testcase's descriptive name. Use this to
# give your testcase a more descriptive name
# (useful only when run in Easypy mode)
# (optional)
description: # testcase description
# string describing what this testcase does
# (optional)
processors: # testcase's local processors
# pre/post processors to be used in this testcase
# (optional)
pre: # list of pre-processors for this testcase
# eg: mylib.mymodule.preprocessor_func
#
# or, list of testcase pre-processors with arguments
# eg: - processor: mylib.mymodule.preprocessor_func
# args: <list of positional arguments>
# kwargs:
# <key>: <value>
# (optional)
post: # list of post-processors for this testcase
# eg: mylib.mymodule.postprocessor_func
#
# or, list of testcase post-processors with arguments
# eg: - processor: mylib.mymodule.postprocessor_func
# args: <list of positional arguments>
# kwargs:
# <key>: <value>
# (optional)
# any custom key/value pairs to be set as data (attributes) to
# this testcase class
<key>: <value>
common_cleanup: # common_cleanup block
# everything related to script's common_cleanup section
# (optional)
parameters: # common_cleanup parameters
# key/values becomes parameters belonging to
# common_cleanup section.
# (optional)
processors: # common_cleanup local processors
# pre/post processors to be used on common_cleanup
# (optional)
pre: # list of pre-processors for common_cleanup section
# eg: mylib.mymodule.preprocessor_func
#
# or, list of common_cleanup pre-processors with args
# eg: - processor: mylib.mymodule.preprocessor_func
# args: <list of positional arguments>
# kwargs:
# <key>: <value>
# (optional)
post: # list of post-processors for common_cleanup section
# eg: mylib.mymodule.postprocessor_func
#
# or, list of common_cleanup post-processors with args
# eg: - processor: mylib.mymodule.postprocessor_func
# args: <list of positional arguments>
# kwargs:
# <key>: <value>
# (optional)
# any custom key/value pairs to be set as data (attributes) to
# your script's common_cleanup section
<key>: <value>
# any other key/value pairs to be set as variables/attributes directly
# into your testscript module
<key>: <value>
Example Datafile¶
# Example
# -------
#
# the following is an example datafile yaml file
extends: sanity_data.yaml
parameters:
ip_seed: 1.1.1.1
vlan: 4382
traffic_streams: 50
processors:
pre:
- cflow.init_instrumentation
- router_health.reset
post:
- cflow.collect_results
- router_health.collect_health_info
testcases:
MyTestcase_One:
uid: alternative_uid_1
groups: [sanity, regression, ha]
parameters:
input_one: 1000
input_two: 2000
expected_routes: 35
MyTestcase_Two:
uid: alternative_uid_2
groups: [sanity, regression, ha, stability]
parameters:
input_x: 2000
input_y: 3000
Example Run¶
The following is a short script designed to be run with datafiles. Notice how many parameters and values are not defined directly in the script.
# Example
# -------
#
# short script designed to be run with a datafile
# (notice many expected values/parameters undefined)
import logging
from pyats import aetest
logger = logging.getLogger(__name__)
class MyTestcase(aetest.Testcase):
@aetest.test
def uid_and_groups(self):
logger.info('notice how testcase uid/groups are modified')
logger.info(' uid = %s' % self.uid)
logger.info(' groups = %s' % self.groups)
@aetest.test
def script_params(self, script_param_a, script_param_b):
logger.info('the following parameters are script-level')
logger.info(' script_param_a = %s' % script_param_a)
logger.info(' script_param_b = %s' % script_param_b)
@aetest.test
def testcase_params(self, tc_param_a, tc_param_b):
logger.info('the following parameters are local to this testcase')
logger.info(' tc_param_a = %s' % tc_param_a)
logger.info(' tc_param_b = %s' % tc_param_b)
@aetest.test
def module_variables(self):
logger.info('the following variables are defined at module level')
logger.info(' module_var_a = %s' % module_var_a)
logger.info(' module_var_b = %s' % module_var_b)
@aetest.test
def class_attributes(self):
logger.info('the following attributes are defined at class level')
logger.info(' class_var_a = %s' % self.class_var_a)
logger.info(' class_var_b = %s' % self.class_var_b)
if __name__ == '__main__':
aetest.main()
Let’s use the datafile below to provide these much-needed values:
# Example
# -------
#
# yaml datafile
module_var_a: some string value
module_var_b: 99999
parameters:
script_param_a: 3.1415926
script_param_b: 2016-01-01
testcases:
MyTestcase:
uid: customized_uid_from_datafile
groups: [demo, datafile, awesomeness]
parameters:
tc_param_a: 100
tc_param_b: 200
class_var_a: [1,2,3,4,5]
class_var_b: datafile feature is just that awesome
Running the above together, here is the expected output:
# Example
# -------
#
# running the above
(pyats) [tony@jarvis:pyats]$ python testscript.py -datafile=datafile.yaml
INFO: +------------------------------------------------------------------------------+
INFO: | Starting testcase customized_uid_from_datafile |
INFO: +------------------------------------------------------------------------------+
INFO: +------------------------------------------------------------------------------+
INFO: | Starting section uid_and_groups |
INFO: +------------------------------------------------------------------------------+
INFO: notice how testcase uid/groups are modified
INFO: uid = customized_uid_from_datafile
INFO: groups = ['demo', 'datafile', 'awesomeness']
INFO: The result of section uid_and_groups is => PASSED
INFO: +------------------------------------------------------------------------------+
INFO: | Starting section script_params |
INFO: +------------------------------------------------------------------------------+
INFO: the following parameters are script-level
INFO: script_param_a = 3.1415926
INFO: script_param_b = 2016-01-01
INFO: The result of section script_params is => PASSED
INFO: +------------------------------------------------------------------------------+
INFO: | Starting section testcase_params |
INFO: +------------------------------------------------------------------------------+
INFO: the following parameters are local to this testcase
INFO: tc_param_a = 100
INFO: tc_param_b = 200
INFO: The result of section testcase_params is => PASSED
INFO: +------------------------------------------------------------------------------+
INFO: | Starting section module_variables |
INFO: +------------------------------------------------------------------------------+
INFO: the following variables are defined at module level
INFO: module_var_a = some string value
INFO: module_var_b = 99999
INFO: The result of section module_variables is => PASSED
INFO: +------------------------------------------------------------------------------+
INFO: | Starting section class_attributes |
INFO: +------------------------------------------------------------------------------+
INFO: the following attributes are defined at class level
INFO: class_var_a = [1, 2, 3, 4, 5]
INFO: class_var_b = datafile feature is just that awesome
INFO: The result of section class_attributes is => PASSED
INFO: The result of testcase customized_uid_from_datafile is => PASSED
INFO: +------------------------------------------------------------------------------+
INFO: | Detailed Results |
INFO: +------------------------------------------------------------------------------+
INFO: SECTIONS/TESTCASES RESULT
INFO: --------------------------------------------------------------------------------
INFO: .
INFO: `-- customized_uid_from_datafile PASSED
INFO: |-- uid_and_groups PASSED
INFO: |-- script_params PASSED
INFO: |-- testcase_params PASSED
INFO: |-- module_variables PASSED
INFO: `-- class_attributes PASSED
INFO: +------------------------------------------------------------------------------+
INFO: | Summary |
INFO: +------------------------------------------------------------------------------+
INFO: Number of ABORTED 0
INFO: Number of BLOCKED 0
INFO: Number of ERRORED 0
INFO: Number of FAILED 0
INFO: Number of PASSED 1
INFO: Number of PASSX 0
INFO: Number of SKIPPED 0
INFO: --------------------------------------------------------------------------------