Alteryx Designer Desktop Ideas

It would be great if it were easier to add a comment tool to the canvas. As I think about what my workflows are lacking, it's good documentation. My biggest hurdle...is myself. I am simply too lazy to stop what I'm doing and add comment tools. What if you could either 1) double click on the canvas and it brings in a comment tool where I can start writing comments or 2) add a Keyboard short cut in order to add a comment tool (control + shift + C perhaps)? Either of these would help me build better documented workflows.

We have a need to be able to trace the lineage of fields being processed through Alteryx - and it has to be done at a field level to satisfy our regulator.   In essence, we have to be able to show exactly where a particular field came from, and demonstrate that we can trust this field.   NOTE: If we could do this - then we could also use this information to make EVERY canvas faster, by checking for unused fields and making suggestions to remove these unused fields early in the flow.

In order to do this - it would be great if there were an option to force explicit field names in the Alteryx XML so that we can trace a specific field.

• Alteryx currently only makes a note of a field name if it is changing that field which is very economical - but it makes tracing fields impossible.
• Desire is for every tool to write the field list to the XML definition for every field that it knows about (just like it looks in the UI)

cc: @AdamR_AYX @jpoz @Claje 

Example:

Here's a simple canvas:

- 3 fields coming in from an input

- Select tool with these 3

- output of three fields into a file

These three fields are not mentioned anywhere in the XML

The Tableau Hyper API supports regular SQL queries, see https://help.tableau.com/current/api/hyper_api/en-us/reference/sql/index.html and https://help.tableau.com/current/api/hyper_api/en-us/docs/hyper_api_reference.html for more information. Being able to use the In-database tools for querying Hyper would let us take advantage of Hyper's internal optimizations just like other databases.

Tableau's Hyper file structure can store multiple tables and the published Hyper API exposes a SQL interface. Therefore instead of supporting the standard file-based interface (like text, Excel, etc.) for connecting to Hyper files how about supporting the database server interface used for MS SQL Server, PostgreSQL, etc. so we can select the schema, tables, fields, or even write SQL?

Two related ideas:

Supporting alternative schema & table names: https://community.alteryx.com/t5/Alteryx-Designer-Ideas/Input-tool-Support-more-than-Extract-Extract...

Supporting multiple table extracts: https://community.alteryx.com/t5/Alteryx-Designer-Ideas/Input-tool-support-multiple-table-extracts-f... to support multiple table extracts for the Input

Jonathan

Alteryx 2019.4 introduced support for Tableau's .hyper extract format, however it only supports single table extracts. .hyper files have supported multiple tables since mid-2018, so I'd like Alteryx to support that as well.

Here are a couple of current use cases (as of February 2020) and one future one.

- We have malaria incidence data that is joined to multiple sets of spatial data. Doing all of the joins in the extract creation process to build a single table extract is not possible due to processing time & memory constraints, so we use a multiple-table extract.

- There are multiple ways to do row level security in Tableau. A common way is to have separate tables for the data & the entitlements and then use calculations at run-time to filter the data, and for that having a multiple table extract is ideal.

- In 2020 Tableau will be introducing new data modeling capabilities (this was first demoed at the 2018 Tableau Conference, there were sessions on it at the 2019 Tableau Conference) where one goal is vastly improved performance for large fact table to fact table joins where previously we'd have to do much more data preparation. This is another case where multiple table extracts would be useful.

I've attached a sample Hyper file with two tables in the extract (it's zipped because the Community site doesn't accept .hyper files).

Supporting alternative schema and table names in Hyper extracts https://community.alteryx.com/t5/Alteryx-Designer-Ideas/Input-tool-Support-more-than-Extract-Extract... is a prerequisite for this because by definition multiple table extracts have multiple table names.

A related idea is supporting multiple table extracts for the Output tool: https://community.alteryx.com/t5/Alteryx-Designer-Ideas/Support-multiple-table-extracts-in-the-Table...

Jonathan

Alteryx 2019.4 added support in the Input tool for Tableau .hyper extract files. The tables stored in the .hyper files have a schema and a table name. Tableau's old .tde files and Hyper files created by Alteryx & Tableau Desktop use "Extract.Extract" as the schema.tablename. However when using Tableau's Hyper API the default schema is "public" and the table name is arbitrarily specified by the user or application.

This has two impacts:

1) Without this support Alteryx can't open many .hyper files created by other applications. By way of example I've attached a sample .hyper file (in a .zip because the community software doesn't allow .hyper files) that has the schema.tablename "public.table1".

2) Also support for names beyond Extract.Extract is required in order to support multiple table extracts (submitted as a separate Idea).

Please update the Input tool so the user can select the particular schema and table name from the .hyper file.

Jonathan

The Select tool already has some really useful column rename options like add & remove prefix.  Please add uppercase & title case to these options.  I know I could do this with the Dynamic Rename Tool, but this will involve adding two tools if I'm already doing other select type options like removing fields.  Anyway, I'm assuming this would be an easy add for the DEV team and would make my life a little easier! 

Many thanks!

I would like to see an included select window for (nearly) all tools. Sometimes almost half of my workflow are select tools because I try to minimize my data and improve naming as much as possible.

Example:

1) Removing unnecessary columns

2) Removing Lat/Long columns

3) Removing not needed columns from distance tool.

Another example would be the typical usage of cross-tab. In 90% of my cross-tab usage, I put a select tool behind it.

I think it would be good to have an included select window in nearly all tools. There are probably tools that don't need them, but I think it would be a good addition to many tools.

I would suggest to add an included select tool for (at least) these tools:

- Input Data

- Cross-tab

- Tile

- Formula

- Multi-Row Formula

- Multi-Field Formula

- Text To Columns

- Cross Tab

- Transpose

- Create Points

- Distance

- Trade Area

- ... 

the SQL Editing screen has recently been changed (thank you @JPoz  and team!) - and now has syntax indenting and keyword coloring.

Could I ask that you make a minor change in the indenting, where the on part is indented underneath the Join?

Select

    Field1,

    Field2,

    field3

from

    Table1

    inner join table2

        On Table1.key = table2.key

        and table1.keyb = table2.keyb

    inner join table3

        on table3.key = table1.key

        and table3.date = table1.date

The In-Database tools (Connect and/or Data Stream Out) should accept SQL Syntax including comments that use "-", "--", "*/", "\*.   Comments are an important part of SQL code providing documentation and have been part of the syntax for a very long time. 

If there are other syntax omissions they should be added as well.

See attached email thread and Case #00322978

For example, allowing the letters in front of the column titles to be formatted differently to make them stand out on a report. 

As a Left-hander I use my Mouse in my Left hand and use my Right hand for keyboard short cuts.

In many windows products that means I can use the CTRL-INSERT, SHIFT+INSERT, SHIFT+DEL for copy, paste, and cut.

In Meta-info windows in Alteryx I can do this, but in the workflow windows I can not.  I'd like the functionality added.

I'd consider myself as a power user in most of the tools I use. No matter what program it is, I try to learn most of the useful shortcuts and code them into my mouse or keypad.

It's probably pretty uncommon that someone uses a mouse with 12 extra keys or a keypad, but I think many people would be happy to have the option to define shortcuts for everything. I don't really ask for shortcuts for everything by default, but a menu like Microsoft Word has it, that would be great.

For reference: 

Microsoft Word has a menu were nearly every possible action is listed and you are able to define/assign shortcuts (one or more) for every action available.

(Sorry it's German. Path: File > Options > Customize Ribbon > Customize)

My friend @jdunkerley79 posted a terse idea:  https://community.alteryx.com/t5/Alteryx-Designer-Ideas/FieldName-constant-in-Generate-Rows-Tools-an...  it is inactive, but I want to extend his thoughts.

Rephrasing his idea as mine:  The tool defaults the expressions to use [RowCount].  If you should either "Update Existing Field" or change the "Create New Field" the default expressions MUST be updated manually.  Please update the expressions to make use of the new field.

Well, that doesn't always work!  Often it will.  But if you change the TYPE to date, it certainly won't.  In fact, I see many questions about joining from within a DATE  RANGE and the technique to build date rows from the range requires the use of DateTimeAdd().  Wouldn't it be nice (like your sample workflows) to modify the default expression based on the change of data type?  I think so.

If we were thinking easy.  Suppose you could have a RANGE function (dates or numerics) where you simply selected the from, to fields and gave the user the option to select the units.  Now the tool auto-configures itself to create all of the "days" between the from and to dates or "1.0" and it creates all unit values between the two numeric amounts.

These would be "Alteryx" worthy enhancements in my opinion.

Cheers,

Mark

The following idea might not be as valuable as some of @SeanAdams posts, but it would save this user precious fractions of time.   When I leave the canvas with my mouse (point A) to go up to the pallet I select and drag the tool down to the canvas.  Sometimes I do right-click and go through the menus to add the next tool, but generally I go through that labor only when I'm inserting in-stream the tool.  So here is my idea:

Double-Click your NEXT tool and it "Alteryx-ly" appears on your canvas in proximity to the hi-lighted (last) tool.  Better yet, connect it!  Now I can move from the pallet to the configuration panel directly without having to move my mouse down to the canvas and then over to the configuration panel.

Hopefully, my friend @Hollingsworth will find this time-saving idea worthy of a star.  Speed demons like @NicoleJohnson and @BenMoss might not need this turbo boost, but at my age it is worth the ask.

Cheers,

Mark

Using the download tool is great and easy to use. However, if there was a connection problem with a request the workflow errors out. Having an option to not error out, ability to skip failed records, and retrying records that failed would be A LIFE CHANGER. Currently I have been using a Python tool to create multi-threaded requests and is proven to be time consuming.

Creation of a Multi Append Fields tool instead of having to use Multiple Append Fields to join man different columns.

I would like to suggest the functionality of converting an In-DB tool to Input Data automatically as it is done with Input Data so that it becomes an automatic Macro input.

The reverse would also be interesting, converting the Input Data, when it is connected to the database, to In-DB.

When using the equal sum option in the Tile tool, you have the choice to sort by a column. However, it's not clear from the tool whether the data is being sorted in ascending or descending order.

Confusingly, the data is sorted in descending order, compared to the default of ascending in the Sort tool. This seems counterintuitive, as a user would expect the same defaults across tools (we did! and it caused some consternation when we worked out it wasn't) and also because in most of our use cases, you want the highest tile to be the highest rank, e.g., if you were sorting households by income.

I have two possible suggestions:

     1. Update the tool properties so that it reads "Sort column (descending)". (Probably the easiest change!)

     2. Allow the user to specify a direction with another drop down that becomes active when you select a sort field (similar to the sort tool).

I know that the functionality of suggestion 2 is available by adding a Sort tool before the Tile and sorting the data in ascending order. However, having to do that every time seems to defeat the point of being able to sort in the Tile tool.

Thanks for reading!

Jason

Standard Tile configuration:

Option 1 (mock-up):

Option 2 (mock-up):