Skip to content

Tune Alteryx Imports

PlaidCloud imports Alteryx workflows into Advanced workflows and reports conversion details during import. Use this guide to quickly complete dependency packages, review generated macros, tune variables, and interpret validation comparisons.

Symptoms:

  • A converted input step cannot find a file.
  • Dynamic input resolves to no files.
  • Validation row counts are lower than expected.

Actions:

  1. Confirm the file was included in the import package.
  2. Confirm the file was uploaded to the selected Document path.
  3. Confirm dynamic file patterns still match after upload.
  4. Re-import or update the converted step to use the correct Document path.

Symptoms:

  • Spatial input needs a companion file.
  • Geometry fields are empty.
  • Spatial output differs from the expected result.

Actions:

  1. Confirm all shapefile or MapInfo sidecars were included.
  2. Keep sidecars in the same Document folder.
  3. Confirm projection files are present when the workflow depends on coordinate reference behavior.
  4. Rerun the workflow after correcting the package.

Symptoms:

  • A macro call cannot resolve.
  • Macro inputs or outputs are not connected.
  • A workflow imports structurally and a macro output needs to be connected.

Actions:

  1. Include the .yxmc file for each referenced macro.
  2. Include nested macros called by those macros.
  3. Avoid duplicate macro names unless the package intentionally includes versioned macros.
  4. Confirm the generated PlaidCloud macro has macro input and macro output steps.

Symptoms:

  • A .yxdb file fails to import while other files from the same package import fine.
  • The file’s first bytes read Alteryx e2 Database file instead of Alteryx Database File.

The file was written by the Alteryx AMP engine in a proprietary format that PlaidCloud cannot read. Rewrite it with Alteryx Designer while you still have it — re-save files individually, or batch-convert a whole folder to CSV with a downloadable workflow. See Convert .yxdb Files for Import.

Symptoms:

  • A converted Calgary step stops and names a calgary_<database> table that does not exist yet.
  • The source is a .cydb file — often a licensed demographic or reference database Alteryx ships (Experian, Census, and similar).

The Alteryx Calgary database is a proprietary indexed store with no open reader, so PlaidCloud cannot load a .cydb directly. The Calgary tools themselves convert — each .cydb becomes an ordinary stand-in table named calgary_<database> — but you supply that table’s data yourself:

  1. In Alteryx, add a Calgary Input reading the .cydb, then re-export its records to an open format — an Output Data tool to CSV or Parquet, or a warehouse table.
  2. Import that file into PlaidCloud as the calgary_<database> table the converted workflow expects.
  3. For a read limited by Skip Records or Max Records, include an order column in the export — the stand-in table carries no inherent record order.

For licensed reference databases you cannot re-export, bring your own licensed copy of the same data and load it into the named table.

Symptoms:

  • A converted app input is unset.
  • A condition fires unexpectedly.
  • A file or folder variable points to a desktop path.

Actions:

  1. Review workflow variables before running.
  2. Set required values.
  3. Replace desktop file paths with Document file or folder paths.
  4. Confirm controlled choices match the expected app selection.
  5. Rerun the workflow.

Symptoms:

  • Schema differs.
  • Row count differs.
  • Row values differ.
  • Artifact output has a difference to review.

Actions:

  1. Confirm the same input data was used in both runs.
  2. Confirm variable values match.
  3. Confirm source systems are from the same snapshot date.
  4. Review null handling, date handling, numeric precision, and string collation.
  5. For spatial outputs, review geometry format and coordinate reference behavior.
  6. For artifacts, confirm the intended cloud-native output and compare the business content.

Specialized operations such as fuzzy matching, spatial processing, PDF extraction, OCR, NLP, and reporting can run through managed executors. (Machine-learning pipelines convert to the native ML: Train Model and ML: Score steps instead.)

If an executor reports a note:

  1. Open the workflow run details.
  2. Review the step note.
  3. Confirm required inputs and parameters are present.
  4. Confirm expected output fixtures are available if parity validation is required.
  5. Rerun after correcting inputs or settings.

Re-import when:

  • Important files or macros were added after the original package.
  • The wrong Document path was selected.
  • A newer workflow version should replace the imported version.
  • The workflow package needs a cleaner dependency layout.

If only a variable value, credential, or Document path changed, updating the converted workflow may be enough.