Data Quality & Catalog ⚙️

I’ve been working on a use case recently which I think may be of interest to the community here. Some reference data is stored in a single table DATA_KEY_ID column tells us what reference data set it is DATA_VAL column stores the actual reference data in the format of a json array Goal is to extract the values from the json array, and use it as a lookup item in Ataccama ONE, which in turn is used in Data Quality Validation rules To start with, you need to have some preview access to the data and understand the schema of the json. For the example in this post, I have very simple “id” and “name” properties for my json. You will need to analyse that data before starting this, so you know what to put into the Json parser, or which values you want to get out from the json object. Solution 1: Virtual Catalog Item (VCI) Bring the reference data table into Ataccama ONE Build a VCI, with a Catalog Item Reader of the reference data table Use a filter step to filter the exact set I need (it will

I’m having this import error note, how to fix this? thank you.

I tried to run monitoring project at morning, but it is forever running. Please let me a know a way to stop it. Thanks! @anna.spakova @Catherine

Introduction This post was created to answer the following questions: When moving data through data pipelines, how can the Ataccama ONE DQ platform support Data Integration platforms other than its own. How do other Integration Platforms access the DQ rules? How do other Integration Platforms flag records for Data Stewards to review? DQ platform in context of integration processes There are three key patterns that are typically used for integration of the ETL process with Data Quality: Option 1 The integration platform calls Ataccama DQ rules exposed as real-time API, sending in batches or micro-batches of data for evaluation. Option 2 The Integration platform loads data into a staging/landing layer and triggers (via API call) a batch process in Ataccama, Ataccama evaluates/improves the data and populates the cleansed layer in the data lake, including DQ scores and flags. Option 3 Ataccama can be connected to a data stream, such as Kafka, and on the fly evaluating data in the stream -

Hello, community I’m currently working on a project which requires a workflow on the runtime server to be triggered after the monitoring project on Ataccama ONE is finished. Can anyone recommend a solution to achieve this? Thank you in advance

I see column level quality score in monitoring project. Do we have row level quality score?

I want to export DQ Evaluation Rules from Ataccama ONE and import them onto another Ataccama environment. To my knowledge, this will require an export plan. However, I’m unsure how to configure the JSON file for the plan. Could anyone guide me on this, please? Thank you in advance

Hi, We'd like to compare two string-fields with several words. Is there a fuzzy matching or approximate string matching step for a plan? In this case it is not reference of lookup data. They are name fields. One is in the DB the other in a file. Kind regards, Marnix

I’m running a monitoring project and it is failing. Here is what I am receiving as a message: Caused by: java.io.IOException: strange number of lookup header sections: 0 What does this mean?

You can optimize your plans by simple things. In this case I would like to show how the number of columns/fields influences the performance of the processing. Test conditions local environment, stable conditions 14M records generated by Random Record Generator 78 fields containing fixed strings or dates No parallelism level set Test Cases One stream – simple reading - BASELINE Overall progress: 117k/s Duration: 3:08min Split into 4 streams, same number of fields – you really need ALL fields in ALL streams Overall progress: 91k/s Duration: 3:42min Split into 4 streams, reduced number of field to 1 – maybe you just need 1 field in each stream Overall progress: 109k/s Duration: 3:19min Split into 4 streams, 3 of them with reduced number of fields to 1 – you might not need all fields in other streams Overall progress: 101k/s Duration: 3:33min In this example it’s just about simple transformation flow however in case you use JOINs, RepresentativeCreators, Aggregators etc. and you reduce the

I need to connect to an AS400 database. I have an odbc connection working for that but I can use ODBC with DQA. No OBDC driver is available. it is not like what is displayed on the FAQ support page when they explain how to connect to MS access using ODBC. it seems the driver has disapeared :-(

Australian Tax Files Numbers, commonly known as TFNs, are 9-digit numbers where the last digit is a validation check-digit. How would we go about setting up Detection and DQ Evaluation Rules for TFNs? The following article applies to ONE version 13.x. The screenshots are from v13.9. For the Detection Rule, we might want to ignore any extraneous characters in the data, for example, the data that is profiled might contain, TFN:123456782, that could be easily transformed into a valid TFN by removing the prefix. For the DQ Evaluation Rule, we might want to consider values that contain the TFN: prefix like in previous example as invalid, to indicate that further processing/cleansing of the data is required. The good news is that we can use almost the same condition for detecting and validating the numeric part of the TFN. For the DQ Evaluation Rule, we’ll use an extra condition to check that only numbers are present. Detection Rule To configure the Detection Rule in ONE, head to the Rules s

I have a plan file in which there are multiple flows which generate multiple result csv files having different number/size of column . I want to store the output of each flow into a single csv file by appending the rows. I want to this happen every time I run the plan file Example - On 1st run of the plan file each result csv file have 10 rows and there are 5 flows . This should lead to a total of 50 rows in the appended file . Now on the 2nd run of the plan file the 10 rows are generated again for each flow , this should lead to a total of 100 rows( 50(1st run) + 50 (2nd run) ) in appended file .

"When one column have countries, another column have city" This city should be in that country else it is invalid. Please help.

Hello everyone! Did anyone enable and configure Lineage via by configuring Lineage Database with 3 tables? I don't know where to create, how to fill in the table and how to start setting up. What conditions are needed? What needs to be created? I will be very grateful for your help

How one can create a counter with rule for e.g. I am trying to add validation for a column value which will have two rules. Based on three inputs column value,range lower limit and range uper limit. 1) If the column value is within the specified range of value (input). Count it's sum how many passed the rule ? 2)If the column is out of the specified range of value (input).Count it's sum how many failing the rule?.

I created a lookup file from existing attributes in the table but some values are missing so I need to update the look up file. How to do that? How to create lkp file outside of web interface and upload to ataccama?

I have rule having 2 attributes but it cannot be linked to business term. How can make use of this rule to validate the atrributes?

I have to parse the below JSON GET response to get “email” and "is_valid_format.value” as columns in Json Call step with Data Streams, How can I perform this? { "email": "xxxx@abc.com", "autocorrect": "", "deliverability": "DELIVERABLE", "quality_score": "0.99", "is_valid_format": { "value": true, "text": "TRUE" }, "is_free_email": { "value": false, "text": "FALSE" }, "is_disposable_email": { "value": false, "text": "FALSE" }, "is_role_email": { "value": false, "text": "FALSE" }, "is_catchall_email": { "value": null, "text": "UNKNOWN" }, "is_mx_found": { "value": true, "text": "TRUE" }, "is_smtp_valid": { "value": true, "text": "TRUE" } }

Based on data sensitivity classification of the data source, we would like to validate if the data is encrypted or in clear. If the classification is confidential and the data is in the clear (unecrypted) we want to generate a report for the data custodian. Requirements: workflow, scheduler, alert notification

INTRO Sometimes you need to enrich your entity view with the combination of attribute values or attributes that belong to some other entity… and there is a way to do it using computed content feature. Following are the two use cases I happen to encounter most often. Show parent attribute in a child entity You have report and reportColumn entities. Entity reportColumn is an array of embedded objects inside report. Entity report has a reportName attribute. The goal is to add report.reportName to reportColumn entity as reportName attribute. High level steps: Create a new entity named reportNameCC, save it, **DO NOT PUBLISH**. In the Computed Content tab of the entity add parameters and SQL. Edit reportColumn entity add embedded object with type reportNameCC; add delegate property to embedded reportNameCC. Computed Content Entity Goto Organization → Metadata Model Press Create entity and create empty reportNameCC entity. Save but **DO NOT PUBLISH!** Open Computed Content tab and fill in: I

I tried to profile flat file in ADLS but profiling failed with the below error : {"status":"FAILURE"} Detail: org/apache/hadoop/conf/Configurationorg.apache.hadoop.conf.Configuration Please help

Hi, Have gotten some good answers here. So here is a case. It is a requirements to merge data. This involves steps: Compare to sets of data - say Target and Source. Any record not in Target but in Source will be added (insert). Any record in Target but not in Source will be flagged ‘deleted’ or deleted (delete) Any record in both sets will be updated - say a timestamp will be updated (update) In SQL it is known as MERGE. One way would be an full outer join and then use filters, with three branches in the plan, but I hope I miss an easier way ;)

Hi all! There is a table in the source that contains about 100 million records and is updated every day. Is there any way to profile the table so that it adds only new data (for example, those that have appeared today) while not profiling all 100 million rows, but conditionally adding 1 million new records to the 99 million existing ones?

Hello, How to write the following condition in Ataccama Desktop (Join step): left join anl.productsjur p on (asd.creditpurpose = p.cp or coalesce(asd.creditpurpose, p.cp) is null) and (asd.creditpurpose2 = p.cp2 or coalesce(asd.creditpurpose2, p.cp2) is null) and (asd.creditobject = p.co or coalesce(asd.creditobject, p.co) is null) The difficulty is that in the join step only one field for left key and right key is available for selection..