Alteryx Designer Desktop Ideas

Issue

Whenever a Summarize tool is used, it renames the output field (e.g., sales becomes SUM_sales or AVG_sales).  

Proposal

I think a reasonable compromise is to by default not rename fields in the Summarize Tool, but to include an option (in the tool, or in global settings) to allow for renaming.

Rationale

I have yet to come across a use case where automatic renaming of aggregated fields is desirable. What I have come across is the annoyance to rename the fields back to what they were with a Dynamic Rename tool, and sometimes having to do this multiple times (e.g., converting back a SUM_SUM_SUM_sales back to sales). Additionally, automatic field renaming causes workflow errors when workflows are later modified by adding/removing a Summarize tool (e.g., if you later add a Summarize tool, all downstream steps will expect the "sales" field and not know to use the "Sum_sales field). 

Automatic Renaming feels very much like historic Excel with Pivot Tables field renaming and not reflective of modern code-based workflow best practices.

I appreciate you considering this improvement.

When I create a new table in a in-Db workflow, I want to specify some contraints, especially the Primary Key/Foreign Key

For PK/FK, the UX could be either the selection of some fields of the flow or a free field (to let the user choose a constant).

From wikipedia :

In the relational model of databases, a primary key is a specific choice of a minimal set of attributes (columns) that uniquely specify a tuple (row) in a relation (table).[a] Informally, a primary key is "which attributes identify a record", and in simple cases are simply a single attribute: a unique id. 

So, basically, PK/FK helps in two ways :
1/ Check for duplicate, check if the value inserted is legit

2/ Improve query plan, especially for join

Report text tools currently only give the option to allign left, right or center. Would be great if we could have the option to have a true 'Justify' option also as it makes chunks of text look so much cleaner

Hello,

According to wikipedia :

A partition is a division of a logical database or its constituent elements into distinct independent parts. Database partitioning is normally done for manageability, performance or availability reasons, or for load balancing. It is popular in distributed database management systems, where each partition may be spread over multiple nodes, with users at the node performing local transactions on the partition. This increases performance for sites that have regular transactions involving certain views of data, whilst maintaining availability and security. 

Well, basically, you split your table in several parts, according to a field. it's very useful in term of performance when your workflows are in delta or when all your queries are based on a date. (e.g. : my table helps me to follow my sales month by month, I partition my table by month).

So the idea is to support that in Alteryx, it will add a good value, especially in In-DB workflows.

Best regards,

Simon

Directory Tool retrieves today a lot of information about a file. I must say I appreciate getting easily the size and the last write time.

But why not the owner? I have developped a macro with a powershell to do that but what a nightmare for a so little piece of information.

Hello,

Here a use case :

I work on the projects A and B with Alteryx inj IN DB mode.

My coworker works only on project B and have no rights to the data of project A.

When using temporary table in Alteryx, we both create the temporary tables in the default database. The issue is my coworker can see my temporary data of project A, which is not safe.

Solution : allow me to specify the database/schema when I create my temporary table.

Hello all,

as of today, a join in-db can only be done with an equal operator.

Example : table1.customer_id = table2.customer_id

It's sufficient most of the time. However, sometimes, you need to perform another kind of join operation, (especially with calendar, period_table, etc).

Here an example of clause you can find in existing sql

inner join calendar on calendar.id_year_month between fact.start_period and fact.end_period

helping to solve that case :

(the turnaround I use to day being : I make a full cartesian product with a join on 1=1 and then I filter the lines for the between)

or <,>, .... et caetera.

It can very useful to solve the most difficult issues. Note that a product like Tableau already offers this feature.

Best regards,

Simon

Hello all,

As specified in the title, this idea is to distinguish between Append Prefix/Suffix to File and to Table on the Output Data Tool.


For most files (csv...), the table name does not really exist. However, for at least Excel files, if you choose this option, the result will be one sheet by suffix and the only option to have one file by suffix will be to change entire file path.

Best regards,

Simon

There is no tool that exists that outputs all records that are duplicates (those sharing the selected values with at least one other record) and also outputs the records that are not duplicates (those not sharing the selected values with at least one other record).

The Unique Tool is not sufficient.  It only provides the first record of a unique duplicate group along with any non-duplicates and then provides a secondary output that only contains the additional records of a duplicate group.  Sometimes you only care about the duplicates and want to quickly see what differs between the unique groups.

For example, if there are 4 records with the City of Austin and I am looking for duplicates on City I want to see all 4 records with Austin in the output so I can quickly compare additional fields to see what might differ, or if they are all indeed truly duplicates.

Hello all,

We all love pretty much the in-memory multi-row formula tool. Easy to use, etc. However, the indb counterpart does not exist.

I see that as a wizard that would generate windowing functions like LEAD or LAG
https://mode.com/sql-tutorial/sql-window-functions/

Best regards,

Simon

There is an extensive need from customers to be able to create emails but not send them (right away at least). This would be an enhancement to the email tool to allow for checking a box to say "Save in Draft" and if that box was checked the send email functionality would not be called but the email would be created. 

Hello,

As I mentioned in this previous idea : https://community.alteryx.com/t5/Alteryx-Designer-Ideas/Generic-In-database-connection-please-stop-i...

field mapping in generic in-db connection is based on Microsoft Sql Server. Given the specificity of MSQL Server field types, I would like to change that in order to at least be able to use another database. Without that, this feature has no sense at all.

Best regards,

Simon

Hello all,

Like many softwares in the market,  Alteryx uses third-party components developed by other teams/providers/entities. This is a good thing since it means standard features for a very low price. However, these components are very regurarly upgraded (usually several times a year) while Alteryx doesn't upgrade it... this leads to lack of features, performance issues, bugs let uncorrected or worse, safety failures.

Among these third-party components :

- CURL (behind Download tool for API) : on Alteryx 7.15 (2006) while the current release is 8.0 (2023)
- Active Query Builder (behind Visual Query Builder) : several years behind

- R : on Alteryx 4.1.3 (march 2022)  while the next is 4.3 (april 2023)
- Python : on Alteryx 3.8.5 (2020) whil the current is 3.10 (april 2023)
-etc, etc....
-

of course, you can't upgrade each time but once a year seems a minimum...

Best regards,

Simon

Hello,

As of today, we can't choose exactly the file format for Hadoop when writing/creating a table. There are several file format, each wih its specificity.

Therefore I suggest the ability to choose this file format :

-by default on connection (in-db connection or in-memory alias)

-ability to choose the format for the writing tool itself.

Best regards,

Simon

Hello all,

Here the issue : when you have a lot of tables, the Visual Query Builder can be very slow. On my Hive Database, with hundreds of tables, I have the result after 15 minutes and most of the time, it crashes, which is clearly unusable.

I can change the default interface in the Visual Query Builder tool but for changing this setting, I need to load all the tables in the VQB tool.

I would like to set that in User Settings to set it BEFORE opening the Visual Query Builder.

Best regards,

Simon

I wasn't able to find an existing idea for this, so here goes...

We often put URLs in Comment boxes in order to direct workflow users to different resources, but users are then required to go into the Comment and copy the string (and then paste in their browser). It would be quite handy if the Comment tool simply turned text strings with a url format into clickable links (as seen in Word, Slack, etc).

It would be even more handy if it also had basic text formatting tools (bold, italic, underline, coloring, highlighting of specified characters/words) -- or if it could just render html (like this text tool I'm typing into right now) 😎

As of today, you can pass SQL from :
-input tool

-output tool

-connect in-db

The user interface is very limited, the kind of query you can pass also, welll, not very user-friendly. This generates a lot of frustration among users.

What do I suggest :
1/ A direct button "Query Builder" without having to open a new wf and drop an input box and then go with the presql tool and fight to build a query.

2/ Basically the same features than Dbeaver (https://dbeaver.io ) or DBvisualizer (https://www.dbvis.com/) or SQUIRREL http://squirrel-sql.sourceforge.net/ 😞
  -Ability to pass any SQL Code I want (such as update, create, truncate, etc...) when I come from the button, "protected" sql when I am in a workflow
  -autocompletion

  -color coding (The idea is not new )

 
3/ A box "Free Sql Query" that I can branch on a indb or standard wf to pass any SQL query. The Output would be the same as input, just like it would be with a block until done.

Hello all,

As of today, you can use the Dynamic Select Field with two options

-by types (you can dynamically select all, all date, etc..)
-by formula

I suggest 2 easy improvements

-from a list field. You connect a field list to a second entry with a "Field name" field
-from flow : You connect a flow to a second entry and the common fields are selected

Best regards,

Simon

Hello,

SQLite is :

-free

-open source

-easy to use
-widely used

https://en.wikipedia.org/wiki/SQLite

It also works well with Alteryx input or output tool. 🙂

However, I think a InDB SQLite would be great, especially for learning purpose : you don't have to install anything, so it's really easy to implement.

Best regards,

Simon