Alteryx Designer Ideas

This has probably been mentioned before, but in case it hasn't....

Right now, if the dynamic input tool skips a file (which it often does!) it just appears as a warning and continues processing. Whilst this is still useful to continue processing, could it be built as an option in the tool to select a 'error if files are skipped'? 

Right now it is either easy to miss this is happening, or in production / on server you may want this process to be stopped.

Thanks,

Andy 

Hi Alteryx community,

It would be really nice to have v_string/v_wstring and max character size as a standard for text columns.

it is countless how many times I found that the error was related to a string truncation due to string size limit from the text input.

Thumbs-up those who lost their minds after discovering that the error was that! 😄

Please could we add Qubole to supported data sources,

https://www.qubole.com/

It is possible to connect to Qubole via ODBC Connection

However, this error message is often returned on a query: InboundNamedPipe GetOverlappedResult: The pipe has been ended 

I've been advised by Alteryx Support that this is likely due to the ODBC driver. 

If it's possible to add Qubole to supported data sources it would save a lot of time committed to troubleshooting this error.

All the best

Please add official support for newer versions of Microsoft SQL Server and the related drivers.

According to the data sources article for Microsoft SQL Server (https://help.alteryx.com/current/DataSources/SQLServer.htm), and validation via a support ticket, only the following products have been tested and validated with Alteryx Designer/Server:

Microsoft SQL Server

Validated On: 2008, 2012, 2014, and 2016.

• No R versions are mentioned (2008 R2, for instance)
• SQL Server 2017, which was released in October of 2017, is notably missing from the list.
• SQL Server 2019, while fairly new (~6 months old), is also missing

This is one of the most popular data sources, and the lack of support for newer versions (especially a 2+ year old product like Sql Server 2017) is hard to fathom.

ODBC Driver for SQL Server/SQL Server Native Client

Validated on ODBC Driver: 11, 13, 13.1

Validated on SQL Server Native Client: 10,11

• ODBC Driver 17+ is not mentioned, even though it was released in February of 2018. https://docs.microsoft.com/en-us/sql/connect/odbc/windows/release-notes-odbc-sql-server-windows?view...
• SQL Server Native Client is deprecated. It is being replaced by Microsoft OLE DB Driver for SQL Server. However, there is not a mention of Microsoft OLE DB Driver for SQL Server. The latest version of this is 18.3.0. https://docs.microsoft.com/en-us/sql/connect/oledb/release-notes-for-oledb-driver-for-sql-server?vie...

I think it would be really good if we have the option to cache data for few days, as currently cached data gets deleted when you close the workflow.

It’s useful to catch data when developing reports with input data from data warehouses or big data platforms , as sometimes it can takes a while to extract the data.

If we have the option to cache data for few days or delete when it’s not required anymore, it can save a lot of time, the next time when you open the workflow to complete the development or make changes to your workflow.

When using the output data tool, it would save me and my cluttered organizational skills a lot of effort if the writing workflow was saved as part of the yxdb metadata. 

I've often had to search to find a workflow which created the yxdb. I tend to use naming conventions to help me,  but it would be easier if the file and or path was easily found. 

cheers,

 mark

When you are navigating within the text input tool, it would really help if we can navigate to next/previous cell upon pressing right/left arrow keys. If the cursor is at the end of a cell value then pressing the right arrow should go to next cell. Similarly, if the cursor is at the beginning of value of a cell, it should go to previous cell.

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

Hi Alteryx 🙂

When you set maximum records per file, the filename gets _# appended.  Great!  But in reality you get:

Filename.csv

Filename_1.csv

Filename_2.csv

The first filename doesn't get a number.  I think that it should.

Cheers,

Mark

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

Referencing the previous idea: Inputs/Output should have the option to read/write a compressed file (ZIP or GZIP)

This idea has been implemented for inputting .zip files. However, we still need to use the run command workaround for outputs. It's very common for many users to want to output their .csv, .xlsx, .pdf to a .zip. The functionality would also need to extend to Gallery.

See the following links for people that are looking for this type of functionality:

https://community.alteryx.com/t5/Alteryx-Server-Discussions/Download-Multiple-Outputs-from-the-Galle...

https://community.alteryx.com/t5/Alteryx-Designer-Discussions/Output-files-to-ZIP/td-p/163502

https://community.alteryx.com/t5/Alteryx-Designer-Discussions/Zip-files/td-p/151456

Feel free to merge this idea with the previous one for continuity.

Hi GUI Gang

At the moment, I have a lovely formatted XLS with corporate branding, logos, filled cells, borders etc.  The data from the Alteryx output needs to start in cell B6.  I have tried the output tools to this named range, but Alteryx destroys all the Excel formatted cells in the data block.

As a workaround on the forums, many Alteryx users pump out to a hidden "Output" tab, and then code =OutputA1 in the formatted sheet.  This looks messy to the users who then go hunting for the hidden tab.  Personally I end up pumping the workflow out to a temporary CSV file.  Then opening that in Excel, selecting all, and then pasting values in the pretty Excel file.

This is fine for one file, but I need to split the output report block by a country field and do this 100s of time for each month end.

Please can we have a output tool that does the same as my workaround.  Outputs directly from a workflow to a range in Excel that doesnt destroy the workbook's formatting.

Jay

Hi Alteryx -

It would be nice if we could include some of the sampling functions in the Input tool. For example, I am creating a random 10% sample from a large data set but in order to do this I need to first input the entire data set then create the sample. This takes processing time and effort.  Ideally, in the Input configuration I would like to see the option to create samples without having to load in the entire data set. Similar to Record Limit but with added functionality. 

Thanks, 

Derek 

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

Could I ask for a minor change:

- The tool seems to be doing a word-wrap even if the container doesn't need it - for example in the screenshot below, the entire on clause can fit on one line because I've expanded the window so that it doesn't need to wrap.

- Could you also default to putting the ON clause for a join 4 spaces indented underneath the join clause?   worked example below.

Wordwrap:

Indenting Structure:

Could we use a default structure for queries 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 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.

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

Hi Alteryx 🙂

When I select fields within a BROWSE or Browse Anywhere, it would be nice to be able to highlight and deselect  a range of  fields instead of having to check/uncheck fields one at a time.  Yuck!  

Cheers,

Mark

I didn't know I want the Excel "Select a range (cells, rows or columsn" feature before it dropped, but now that we have it, it's time to make it  awesome.

There are two things that really need to improve to make this tool awesome:

1) Allow to fill out only one parameter

The typical use case for this tool should look like this. You have some unnecessary header rows and columns -> the data is shifted.

We want to build a flow that only selects our data columns in the middle, but the length of these columns can change, so it's a BIG NO to select just a range from D4 to G13. Nobody knows if the next report will have the same length. We can't risk to cut-off the last rows.

This is also import for the output feature. We usually don't know if we'll write 10, 20 or hundreds of data rows. To make this feature useable, we really need to be able to specify just one cell. The cell range is a nice on top feature, but it shouldn't be the requirement.

I also see an option for something like D4:G. This would be a syntax with two fields, but the G doesn't need a number and therefore says "till data ends".

In the current state of the tool, it's better to use 'Start data Import on row' + Select tool to achieve the needed behavior. 

2) Improve usability

I see multiple ways to improve the usability of this feature. I will start with the most simple way (for the dev team) and follow up with more complex. The problem is always the same: How do we know the exact position of the cells without opening it?

2.1) Add an additional row on top of the header row

The easiest possible way is to add an row above "Header F2 F3 ..." with A, B, C ...

This really helps to pick the right one.

2.2) "Select tool" inside the preview

I could imagine a feature where we can select the wanted cells and add them to the Query. The best way is probably an auto hover like word has it. See below.

When you select a text and wait a second, it automatically pops up some options to edit the text. Another option would be a simple right click.

2.3) Show Excel Button

Time to step up the quality with an additional button inside the Input Tool. The Show Excel Button opens the Excel file with a temporary window to help you select the right cells.

2.4 Select In Excel Button

This is probably the most advanced solution. Open the file (same as in the step before) and allow to select the ranges directly inside the file and throw the selected range inside the original config windows. It's probably the most difficult solution for the problem, but let me dream for a moment.

MemSQL is a very popular high-velocity; high scalability database which is often used for analytics: https://portal.memsql.com/

However, when attempting to make a connection to MemSQL, Alteryx throws the error "Get Table List Error: Error opening "SHOW VARIABLES LIKE 'aurora_version'": No Columns Returned."

From discussions with the support team at Alteryx - it appears that this is due to Alteryx presuming that the database is a MySQL database (because MemSQL also uses the MariaDB ODBC driver) and then sending a followup query to identify system version before making a connection - however MemSQL does not support this value so the connection fails.

Request:

- Please can we add the ability to connect to MemSQL without this error (i.e. trap this error and identify the DB version)

- Please can we give an advanced view to Alteryx users so that they can see this additional traffic and change this where needed?

          NOTE: By providing an advanced view - users can understand and diagnose / fix behaviours that Alteryx assumes by default, but which are not always relevant or useful.   An example of this is that on MS SQL Alteryx tries to open a cursor on columnstore tables by default and you have to use a workaround - please can you give us the ability to see and change these under-the-covers behaviours that are presumed by default.

- It may be useful to the Alteryx corporation to reach out to MemSQL and add Alteryx to the list of applications that can be used with MemSQL:  

https://docs.memsql.com/v6.8/guides/client-and-application/third-party-integrations/introduction/

To replicate:

- Fire up a MemSQL cluster (easiest is to get a trial cluster using the link above, or use an internal DB)

- Connect to this DB using the MariaDB ODBC connector (as recommended by MemSQL)

- Test this connection works with an ODBC compliant app like Tableau; Excel; etc

- Then attempt to connect using Alteryx  - see screenshot below.