sdi-utils

SAP Data Intelligence Utilites (sdi_utils) by thhapke

sdi_utils Package

This package includes a couple of helpers that ease the development of operators for use with SAP Data Intelligence (SAP DI).

• gensolution for generating solution packages from locally developed operators.
• textfield_parser that parses textfields that could contain lists, maps, list of maps, etc.
• set_logging for channeling logging output to a string for tap wiring it with separate monitor
• tprogress for a simple time keeping to check performance of certain operators (tasks)

Installation

pip install sdi_utils

SAP Data Intelligence Helper Operators

I created a some operators which have been reused a couple of times. Maybe they provide a quick solution for your pipelines as well. Please be aware that many of them become obsolete with the next releases in particular when the looming vtypes are introduced.

You find the source code of all operators in src/sdi_utils_operators/. They have been created outside a SAP Data Intelligence instance and with gensolution the solution has been created that can then directly been uploaded. But all generated solutions you also find in the folder sdi_utils/solution/operators/.

All operators require the modules packaged in sdi_utils, so you have to add at least this to your docker image. For additional packages have a look at the tags added to each operator.

The following list might not encompass all operators you find in the source or solutions folder. Because I am not been able to maintain this document in sync with the operators I create constantly.

Operator	Description
csv_df	Converts a csv byte/string stream into a pandas DataFrame
csv_dict	Converts a csv byte/string stream into a list of dictionaries with header as key
csv_table	Converts a csv byte/string stream into a 2 dimensional array. Used for the beta Hana Operator only.
df_csv	Converts a DataFrame into a csv-string for saving with WriteFile-operator
df_table	Converts a DataFrame into a 2 dimensional array with the according attributes used for writing in a HANA Database (Beta-operator
dict_df	Converts a dictionary into DataFrame
dict_json	Converts a dictionary into a JSON-string
dict_table	Converts a dictionary into a 2-dimensional array
filter_date	When the date is part of the filename (format: yyyy-mm-dd) then files can be selected with dates within the period stated in the configuration
count_gate	Decision gate that triggers next processing step only after the number stated in the configuration has been reached. The number can also be set dynamically either directly via the limit port or via attribute port. This operator is mostly used to terminate a pipeline.
HTTPdownload	Downloads the file given in the url
json_df	Converts a json-string into a DataFrame
json_df	Converts a json-string into a dictionary
line_array	Converts a byte-stream into an 1-dimensional array
table_csv	Converts a 2-dimensional array into a csv. Using the attributes of the beta-type table for the header.

Operator Testing and Modifying

If you like to test and modify the operators outside of Data Intelligence and create the operator-solution automatically you have to install vctl and sdi_utils. Then you can add the following code-snippet at the end of the operator-code:

if __name__ == '__main__':
    test_operator()
    subprocess.run(["rm", '-r','<projectfolder>/solution/operators/sdi_utils_operators' + api.config.version])
    gs.gensolution(os.path.realpath(__file__), api.config, inports, outports)
    solution_name = api.config.operator_name + '_' + api.config.version
    subprocess.run(["vctl", "solution", "bundle", '<projectfolder>/solution/operators/sdi_utils_operators_' + api.config.version, "-t", solution_name])
    subprocess.run(["mv", solution_name + '.zip', '../../../solution operators'])

Github

github repository