Changelogs

Novembre 10th - More customization, and several bug fixes

This version brings more options to resync data sources (directly from the Data Explorer, or scheduled), a new setting to enforce some locale for the whole organization, and several bug fixes.

Do not hesitate to share some feedback using the dedicated bottom-left menu item.

More options to resync data sources

Your data warehouse is constantly evolving, and as soon as you've created a new table or model, you want to run queries on it, and find it in Husprey Data Explorer. We've got your back 😌 with these 2 new features on data source syncs 👇

First, if allowed by their admin, analysts can now directly resync a data source from the Data Explorer.

Note: this was already available for organizations using data sources with personal credentials.

Second, admins can schedule such resyncs from the Settings page, with the frequency of their choice.

Click the new "Schedule sync" entry in the data source menu:

Pick schedule options:

At anytime, check when the next sync is scheduled from the Data Sources pages in Settings:

Organization-wide locale parameter

By default, Husprey uses the locale provided by the user's browser, in particular to format numbers.

If you want to provide strict consistency to your whole organization, you can now define a value that will apply for all users, whatever their own browser-level parameter:

Other improvements

• Improve UX on cost control for BigQuery data sources

• Redesign the data source item in the Settings page, with more readable status

Bug fixes

• Fix app freezing at load with large personal data source

• Fix scheduled execution on ClickHouse returning only 200 rows

• Fix keyboard shortcut to run a single query running several of them in some cases

• Fix unexpected save call being triggered on empty notebooks in some cases

• Fix gzip compression not being enabled for frontend assets

October 20th - Cost control for BigQuery, smarter autocomplete, and more

This version brings cost control for BigQuery data sources, smarter autocomplete, and performance improvements in the notebook page.

Do not hesitate to share some feedback using the dedicated bottom-left menu item.

Cost control for BigQuery

Data teams with BigQuery data sources can now more easily control the costs incurred by the queries they run in Husprey. This is made possible via 3 new elements detailed below 👇

For every query, get the estimated number of bytes that will be processed:

Admins can also define a threshold on these bytes processed from the Settings page. A warning will then be displayed before executing a query that will process more data than this threshold.

Finally, displaying details about a specific column in the Data Explorer triggers a few queries to the data source. Admins can now disable this feature from the Settings page to prevent them from being run.

Read more about cost control in the BigQuery documentation.

Smarter autocomplete

Autocomplete now provides more relevant hints as you type a query by better taking into account the particular part you are currently writing.

Performance improvements

Several elements about performance have been improved, including:

• Speed up the notebook page (in particular when there are many queries), by optimizing how the "Local tables" and "Queries" lists are computed

• Speed up several elements over the whole app, by "memoizing" several util functions

Other improvements

• Use new (and nice!) illustrations in many "empty state" screens

• Support dbt manifest v7

• Improve default settings for scatter plots

Bug fixes

• Fix "auto scroll to comment thread when clicking a comment notification" being broken

• Fix "Run" shortcut (Cmd/Ctrl + Enter) not being available from the query handle/title

• Fix loop of NotFoundError after trashing a notebook in some cases

• Fix topbar not being properly updated when notebook is detected as trashed

• Fix skeletons in home and folder pages not being the same size as the cards they represent on very large screens

• Fix column menu being available in result table when the notebook is published

October 6th - Scatter charts, new schedule options, and more

This version brings scatter charts, new schedule options, and performance improvements in the Data Explorer.

Do not hesitate to share some feedback using the dedicated bottom-left menu item.

Scatter charts

Visualize a population via the new Scatter charts. Pick one field for horizontal axis, one for left axis, and map individuals in this 2D representation.

Optionally choose a category field as "Split by" to group these individuals:

New schedule options

Use notebooks for near real-time monitoring with the 2 new schedule options (available on demand): "Hourly", and "Every 15 minutes"

Performance improvements

Several elements about performance have been improved, including:

• Speed up the Data Explorer (browsing and searching) by refactoring how it's built, and improving the rendering of its elements,

• Speed up loading the whole app, by removing some unused dependencies, and reducing the size of the code bundles

Other improvements

• Adapt folder and notebook cards to larger screens

• Always make the Data Explorer searchable (published notebook, or viewers)

• Remove the ambiguous "(i)" button on table items in the data explorer

Bug fixes

• Fix results not currently in view not being loaded when printing

• Fix NULL not being displayed in a "category" axis in charts

• Ignore ResizeObserver errors earlier

• Fix loop of NotFoundError when on a newly-trashed notebook

September 22nd - Horizontal bar charts, performance improvements, and more

This version brings horizontal bar charts, and performance improvements in the notebook page.

Do not hesitate to share some feedback using the dedicated bottom-left menu item.

Horizontal bar charts

Instantly rotate your bar chart from vertical to horizontal bars by selecting the new "Horizontal bar chart" type in the chart editor:

As their vertical counterpart, these charts support standard stacking, 100% stacking, multiple axes, and combos:

Performance improvements

Several elements about performance have been improved, including:

• Speed up using notebooks (browsing, typing, etc.) by improving the rendering of surrounding elements, and unmounting specific nodes when they are not in view.

• Speed up loading the whole app, by removing some little-used dependencies, and reducing the size of the code bundles

Other improvements

• Add missing links to docs and guides in the data source connection modals

• Display 10 data catalogs in the explorer by default, add a "See more" button to display the next ones

Bug fixes

• Fix "Print notebook" only printing 1 page

• Fix dbt freshness tooltip not being displayed anymore

• Fix execution of ClickHouse queries failing when last line is commented

• Fix disabled state of the "IconButton" not using the right style

• Fix data source sync writing too many logs in some cases

• Fix Snowflake sync failing entirely when cannot list columns of a single database

• Fix main action still being triggered when discarding a notification

• Fix notification description being improperly (not) wrapped

September 8th - Performance improvements, additional documentation on BigQuery tables, and more

This version brings major performance improvements in the whole application, and specifically on the notebook page.

It also includes improvements on the Documentation panel for BigQuery data sources, and redesigned data source connection modals.

Do not hesitate to share some feedback using the dedicated bottom-left menu item.

Performance improvements

Several elements about performances have been improved in this release, including:

• Speed up loading, browsing, and using large notebooks. This is particularly significant for notebooks with large SQL queries, displayed in full width.

• Speed up rendering all elements in the applications generally

Please feel free to reach out would you have specific performance issues left with your notebooks. We'd be very much interested in understanding your setup and use case.

Documentation panel for BigQuery data sources

New information have been added to BigQuery tables, namely:

• the actual object type: BASE TABLE, CLONE, SNAPSHOT, VIEW, MATERIALIZED VIEW, EXTERNAL (read more in BigQuery documentation)

• the corresponding SQL definition (i.e. the SQL instruction that created this object)

Redesigned data source connection modals

Data source connection modals now embed a first level of documentation and help, as well as links to more detailed connection guides.

It's never been easier to connect your data source to Husprey!

Update the default Snowflake warehouse

As an admin, you can now edit an existing Snowflake data source and change its default warehouse.

Note: Using different warehouses can help you balance query speed and costs. Learn more about Snowflake warehouses with this Getting Started video from Snowflake.

Warehouse selection is performed in Snowflake with the following priority:

• At the query level. Enforce a warehouse for a specific query using USE <warehouse> SQL statement.

• At the connection level. Define a default warehouse in Husprey connection form

• At the user account level. Default warehouse for a user. This can be edited from Snowflake settings. (Suggested default usage)

Read more about warehouse usage in Snowflake documentation.

Other improvements

• Update the application color palette, fix inconsistencies, and improve dark theme

• Set position of the "Trash" button as fixed in the navigation panel, so that it's always visible

Bug fixes

• Fix silent authentication loop in some cases

• Fix "Opened Dialog" event being sent for each char typed in search bar

August 30th - dbt Cloud integration, and performance improvements

Husprey now fully integrates with dbt to sync transformation metadata, whatever your setup (dbt Core or dbt Cloud). Descriptions, dbt models, freshness and columns constraints are available directly in your notebooks.

This version also brings major performance improvements in home, folder, and notebook pages.

Do not hesitate to share some feedback using the dedicated bottom-left menu item.

dbt Cloud integration

As an admin, you can sync metadata from dbt Cloud in a matter of minutes. Our integration relies on the powerful dbt Metadata API to sync both sources and models metadata.

Generate and paste credentials of the dbt Cloud job you’d like to use and that’s it!

And access the transformation metadata directly in the Documentation Panel:

To keep these metadata up-to-date, the synchronization of your dbt Cloud project is scheduled regularly.

Read more on our integration with dbt Cloud in our docs.

Using dbt Core? No worries, we have you covered!

Performance improvements 🚀

Several elements about performances have been improved in this release. This includes:

• Speed up rendering of folder and notebook cards (displayed in home and folder pages)

• Speed up rendering of result tables, and more generally all notebook elements

More mouse shortcuts 🤓

Organize your workspace at light speed with our new keyboard shortcuts! Whenever you hover a folder or notebook card, press the following keys to perform the indicated action:

Think you won't remember them? No worries, they are all displayed in the related menus:

Other improvements

• Editor can rename notebooks from card and navigation item

• Add "Sort by" option for notebooks in a folder page

• Add direct actions on items in trash

• Add a confirmation modal when trashing a public notebook

Bug fixes

• Fix rename folder crashing in some cases when in Home view

• Fix table preview raising an error in some cases

• Fix tutorial not being displayed after onboarding

• Fix "Copy row" in result table raising an error in some cases

• Fix "Create" button at the top of the notebook lists being displayed (with no effect) to viewers

• Fix path of trashed elements showing empty strings instead of "Untitled"

June 21st - Workspace color palette

This version adds custom workspace color palette with smart default rules for charts, as well as several other improvements and bug fixes.

Do not hesitate to share some feedback using the dedicated bottom-left menu item.

Workspace color palette

As an admin, you can customize the default color palette to be used in your entire workspace.

Additionally, associate a color to a set of values, and it will be automatically picked for series with matching names.

It's never been that easy to bring consistency to your analyses and reports!

Other improvements

• Improve open/close conditions of the mention popper for a smoother experience

• Improve performance when editing a notebook

• Improve design of the Settings pages

• Redesign "Notebook cards", used on home page (and in upcoming pages for new feature 🤫!)

• Revamp public page header, and add a footer to make it clearer and prevent misclick

• Remove limit on rows for charts everywhere

• Accept dbt artifacts with non-compiled models

Bug fixes

• Fix edges of items in the user menu not being clickable

• Fix job items in the "Schedule" panel not behaving properly (scroll to query)

• Fix redirect from public to private notebook page not keeping hashes and search params

• Fix move nodes up/down with keyboard shortcut that only moves 1 node instead of all selected ones

• Fix height of result loader not being correctly computed in some cases

May 26th - User mentions, and full Snowflake experience

This version adds user mentions in notebooks and comments, brings the full Husprey experience for Snowflake users, as well as a few other improvements and bug fixes.

Do not hesitate to share some feedback using the dedicated bottom-left menu item.

User mentions

When you want to get another person in your workspace to look at something, you can mention them anywhere inside a notebook, or comment.

Type @ followed by the person's name. A pop up menu will search for them in real-time. Press Enter to pick the one you want.

When someone mentions you, you'll see a red dot appear next to the Notification Center icon in your home page.

From there, see all your notifications, and access the appropriate notebook in one click:

If you are not using Husprey at that time, and do not see this notification, no worries! A recap email will be sent to you within the next 15 minutes.

Inline formatting in comments

Add inline formatting in your comments (bold, italic, underlined). Available from the toolbar upon text selection, or via usual keyboard shortcuts.

Snowflake connector

Our Snowflake connector is now out of the beta phase. Any Snowflake database can hence be added to your workspace, with a full-featured Husprey experience, including:

• Explore the data source (databases, tables, columns) in the "Data Explorer" panel;

• Easily get a preview of any table, and statistics of any column from the "Documentation" panel;

• Get autocomplete hints when typing a query

Copy full row or result table

From time to time, you need to extract some results from Husprey to paste them in an email or a spreadsheet (we know you do! 🙈). To ease this operation, you can now copy a whole row or table from the dedicated buttons, and properly paste them wherever needed!

Other improvements

• The bottom-left user menu has been redesigned to clean the navigation panel (and prepare some great upcoming feature! 🤫)

• The width of resized columns is now saved within the notebook, so that it's persisted across reloads, and shared with your peers and readers.

• Charts in full-width notebooks can also be resized

• The limit on the number of rows for charts has been removed (it was previously set to 1k rows)

• Highlight error location for ClickHouse when possible

• A simple viewer can access a draft notebook from its specific URL (in read-only), and be able to react to mentions, and post / reply to comments.

• Sort data catalogs in the data tree by data source name first, and catalog display name second

Bug fixes

• Fix format of first element being lost during copy/paste within a notebook in some cases

• Fix icon of query items in the "Queries" panel not showing the waiting / running status in some cases

• Fix "Go to query" buttons behaving as external links

• Fix BigQuery records not being parsed and displayed properly in some cases

• Fix column name being cropped in table details

• Fix large white area being displayed for a data query with both SQL and result hidden

• Fix active marks (bold / italic / underlined) not being highlighted in the toolbar anymore

• Fix extraneous scrollbars being displayed in some cases

May 6th - Snowflake connector (beta)

This version mostly brings Snowflake integration, as well as a few improvements and bug fixes.

Do not hesitate to share some feedback using the dedicated bottom-left menu item.

Snowflake connector (beta)

Any Snowflake data source can now be added to your workspace and used in Notebooks: sync data source; browse Data Tree; see table Preview; write (with appropriate autocomplete), run and cancel queries.

This connector currently supports the SNOWFLAKE authenticator only i.e., you will need to provide a username and password.

Want to test it? Just ask us to join the beta!

Note: column statistics in the Documentation are missing in this beta version.

Other improvements

• View the Result of a local table from the Documentation panel, to ease articulating several queries within a single Notebook:

• Update unfurl image to match the one of our website

• Update connector logos

Bug fixes

• Fix local tables opened in the Documentation panel not being remembered

• Fix syntax coloring of local tables not being updated when selecting another data catalog

• Fix some cases where notebook elements overlap each other when printing

• Fix "Copy link" always copying the public URL, and leading to an error screen for private notebooks

April 29th - "Big Value" chart, column statistics, public sharing, and more.

This version brings many new features and improvements to ease your experience with notebooks, more specifically around charts, the SQL editor, and the "Documentation" panel.

Do not hesitate to share some feedback using the dedicated bottom-left menu item.

"Big Value" chart

Thanks to the new "Big Value" chart type, make what matters most stand out in your Notebooks within a nice and large visualization.

For instance, highlight specific metrics and their variations:

Column statistics

In the Documentation panel, get an overview of the data stored in any given column of your data warehouse. This includes basic statistics and, depending on its type, the Top values and Distribution.

No longer wonder if a given "string" column should be considered as an "enum" with only a few values!

Share to web

The next step after writing a great analysis is to share it with your coworkers and business users. While signing up to Husprey is already quick and simple (especially if you have added your domain name in the Settings), you can now also allow anyone with the link to access a read-only version of a given Notebook, without creating any account:

ClickHouse connector

Our ClickHouse connector is now out of the beta phase. Any ClickHouse database can hence be added to your workspace, with a full-featured Husprey experience, including:

• Explore the data source (databases, tables, columns) in the "Data Explorer" panel;

• Easily get a preview of any table from the "Documentation" panel;

• Get autocomplete hints when typing a query

Redesigned autocomplete menu

The autocomplete menu has been redesigned, now showing icons and data types of the suggested columns:

Other improvements

• View the SQL code of a local table from the Documentation panel, to ease articulating several queries within a single Notebook:

• Add a new "Notation" customizable property for numbers in tables and charts

• Highlight the location of a SQL error if we know it for postgres

• In the SQL editor, hit "Alt+Up" or "Alt+Down" to move the selected lines up or down

• Improve experience for multiple personal data sources

• Add logout suggestion and user menu when user cannot register

• Allow editors to remove or edit an existing schedule config on a published notebook

• Improve performance at application load by better caching assets

Bug fixes

• Fix sync for Redshift late binding views

• Fix number formatting for the y-axis ticks

• Fix unhandled error when invalid refresh token

• Fix zIndex of result actions and autocomplete menu

• Fix infinite loader in some cases when printing a notebook

• Fix result pagination for BigQuery in some cases

• Fix never-executed local tables missing the corresponding hint in Data Explorer and Documentation panels

• Fix Execution status not being properly selectable

April 15th - Span gaps option for charts, more keyboard shortcuts, and SSL certificates for postgres data sources

This version brings improvements in several key moments of the Husprey experience (connecting a data source, writing a query, drawing a chart), as well as a few bug fixes.

Do not hesitate to share some feedback using the dedicated bottom-left menu item.

Charts: Specify span gaps behavior

Optionally draw lines between points with no or "null" data (a.k.a. "Span gaps"), using a solid or dash line:

This option is available for line and area visualizations, in each series menu:

SQL editor: Comment/uncomment from keyboard shortcut

When writing a SQL query, simply hit Cmd+/ or Ctrl+/ to toggle comment for the selected lines.

PostgreSQL data sources: Optionally provide SSL certificates

When adding a PostgreSQL data source, you can now provide an SSL root certificate, along with client certificate and key. These are available in the "Advanced SSL options":

Read more about SSL certificates for PostgreSQL in general (https://www.postgresql.org/docs/current/ssl-tcp.html), and how they can be setup and used with Google Cloud SQL (https://cloud.google.com/sql/docs/postgres/configure-ssl-instance).

Bug fixes

• Fix autocomplete not being triggered in some cases

• Properly handle pagination when listing many BigQuery projects/datasets

• Properly handle BigQuery datasets which names start with a number

• Better handle unexpected errors when saving a notebook

• Fix label of the "Color" selector for chart series

April 8th - Full Husprey experience for ClickHouse data sources (beta)

This version brings the experience for ClickHouse data sources up to the level of all other connectors, as well as other improvements.

ClickHouse connector (beta)

Any ClickHouse database can be added to your workspace. In addition to sync this data source and run queries, this version brings the full-featured Husprey experience, including:

• Explore the data source (databases, tables, columns) in the "Data Explorer" panel

• Easily get a preview of any table from the "Documentation" panel

• Get autocomplete hints when typing a query

• Cancel a query from the 3-dots menu

Other improvements

• To avoid any confusion or misinterpretation, display a warning on the chart if there are exactly 100 rows in the result set and rows limit is enabled:

• The in-app tutorial has been improved for an easier onboarding

April 1st - Multi data sources, and conditional formatting

You can now connect several data sources to your workspace!

This version also brings conditional formatting in result table, as well as some other improvements, and fixes a few bugs.

Do not hesitate to share some feedback using the dedicated bottom-left menu item.

Multi data sources

You can now connect as many data sources as you want! There are numerous use cases when this proves to be useful: compare some staging and production environments; monitor and check a database / data warehouse migration; write queries on data not (yet) integrated into your main data warehouse; etc.

To add a new data source, go to the Settings > Data Sources page, and follow the instructions on screen.

Then, when writing a query in a notebook, pick the right data source that it should run onto from the dedicated select:

Conditional formatting in result tables

Apply conditional formatting to result tables, with progress bars or color scales, and get a better view of your data in the blink of an eye!

New option in the table column menu:

Progress bars and color scale:

ClickHouse connector (beta)

Any ClickHouse data source can now be added to your workspace. Some elements are missing for now, but you can already sync your data source, and run queries.

Want to test it? Just ask us to join the beta!

Other improvements

• Customize line thickness in line charts:

• Editors can decide to not show SQL in a notebook when it is published

Bug fixes

• Move the result action buttons down so that it does not overlap with the table or chart anymore

• Fix infinite save loop of a notebook in some cases

• Fix sticky query header being hidden or hiding other elements in some cases

March 8th - Charts and autocomplete improvements

This version brings several charts, autocomplete, and performance improvements, and fixes a few bugs.

Do not hesitate to share some feedback using the dedicated bottom-left menu item.

Charts: Full-screen mode, Download as image, 100% stacked

The previous release brought resizable charts for higher customization and better readability. Now, users can display any result in full-screen mode from the new action bar:

This new action bar also allows users to download any chart as a PNG image:

Last but not least, a new "100% Stacking" chart option has been added for easier data manipulation directly within Husprey:

Autocomplete

Several fixes and improvements have been made, including:

• Provide autocomplete hints when working with aliases

• Fix context being lost in some cases for BigQuery

• Better handle mixed-case identifiers

Other improvements

• Move elements of a notebook up & down from simple keyboard shortcuts: Alt-Shift-Up, and Alt-Shift-Down

• For a smoother experience, do not fetch all results of a notebook at page load, a.k.a. lazy loading

• Locally define a display name to any result field, further used in table and chart:

• Better handle very long queries. In particular, make their header stick to the top of screen, and set a max height to the SQL editor

• Sometimes it's easier to start over from scratch: analysts can now reset the whole chart configuration (type, series, colors, etc.) from a single button in the editor:

• In query results, fetch up to 100k rows from the data source, instead of 10k (onprem only)

Bug fixes

• Fix table preview unavailable in some cases

• Do no run only the selection when a query is collapsed

• Fix label formatting using the wrong format for "Split by" / "Stack by" series

• Fix crash in result table in some cases

• Silently ignore unsupported notifications for better retro-compatibility

• Fix clickable notebook footer not being stretched to the end

• Fix right panel not being remembered to be closed

• Fix very long queries not being properly received by some components

• Fix BigQuery connection parameters to be able to query GSheet-based tables

February 21st - New roles, chart resizing and color pickers

This version brings the support of a new User role and chart improvements. Some performance improvements and bugfixes are part of the release as always.

Do not hesitate to share some feedback using the dedicated bottom-left menu item.

New role - User

A new role has been introduced: User. Users can browse and read published notebooks. They cannot create or edit content. They cannot access raw data - only see query results (charts and tables) that are part of published notebooks. Very often, this role particularly fits Business Users.

Tip: Users can still leave comments on your notebooks!

You can choose the role of a user directly when inviting a user (or you can change it later from the Settings menu)

You'll find all details about roles and exact permissions here.

Chart Improvements - resize and colors

Sometimes charts need some space. We give them the space you want them to have with a simple resize. Top asked feature over the past month!

And you can pick any color for any series! Our picker also remembers last used colors so your notebook is easy to style.

Analysts can also pick the horizontal axis type between: Time series, linear or category.

Queries improvements

• Analysts can cancel an execution from query menu

• New status icon in query footer to ease readability

• Analysts can run a selected sub-part of a query to ease debugging

• Analysts can easily duplicate or move up and down a query

• Line numbers are displayed in queries

• Errors are highlighted (Big Query only)

Fixes and performances improvements

• Search-specific database indexes for faster autocomplete in SQL editor

• Handle dbt artifacts up to version v4

• Improved SQL parsing for local tables

• Improved error messages and actual database errors when it was not the case

• Do not fail a scheduled job when a query has been deleted

February 4th - Multiple BigQuery projects, and more.

This version brings the support of multiple BigQuery projects, major performance improvements 🚀, and fixes a few bugs🐛.

Do not hesitate to share some feedback using the dedicated bottom-left menu item.

Support multiple BigQuery projects

Teams with a multi-projects data warehouse in BigQuery now get the full Husprey experience!

All BigQuery projects available to the service account used to connect Husprey to your data source are now synced with Husprey, and made available in the Data Explorer and autocomplete.

New user notification 💬

For easier administration and monitoring, admins receive a dedicated notification when a new user joins their workspace:

Any user can add tags 🏷️

We've heard your feedback about tags! Now any user in the organization can access the Tags management page in the Settings

Performance improvements 🚀

Several elements about performances have been improved in this release. This includes:

• Revamp autocomplete suggestions for faster rendering

• Speed up publish / unpublish operations

Other improvements

• Directly get a view on the Schedule config and last execution from the About panel, for faster check and better discoverability:

• Full-width preference is now made an attribute of the Notebook, hence saved and shared (not anymore a local toggle)

Bug fixes

• Fix hint "Ctrl + Enter to run your query" only being displayed for a brief moment

• Fix dbt description not being cropped on Safari 15

• Fix crash on Safari for iOS < 13.4

• Fix pattern "2022w01" in notebook title not being auto-incremented to "2022w02" when duplicating

• Fix data source sync for BigQuery datasets which name starts with some special characters

• Fix extraneous save operations on notebooks in some cases

January 18th - Notebook scheduling, chart improvements, and more.

This version brings important new features: notebook scheduling for automatic updates ⏰ , "Queries" panel from which you can run all of them in one click ⏩, and new chart capabilities 📊. It also improves performances 🚀 , and fixes a few bugs 🐛 .

Do not hesitate to share some feedback using the dedicated bottom-left menu item.

Notebook scheduling

Schedule automatic executions of a notebook from the dedicated panel, and use it for e.g:

• Feature Launch - you launched a new feature and want to follow impact while taking notes. Data is refreshed daily and after about a month you can finish up and freeze both the data and your conclusions.

• Weekly insights - you can prepare weekly sales, marketing, or product reports with scheduled data execution. All that is left for you is understanding the results, maybe deep-diving in a specific area, and sharing decisions.

The history of all scheduled executions will then be available in the right panel:

"Queries" panel

That same right panel now includes a new "Queries" section, from which you can get an overview of all queries in the current notebook with their execution status and duration, updated in real time.

With the "Run all" button, refresh the results of a whole notebook in one click, and better focus on what matters: rationale, results, and comments.

Stack by / Split by option for charts

Break your current metric out by an additional dimension by choosing it as "Stack by" or "Split by" option. It comes in handy for SQL queries with "double group by" (e.g. on both time and a category).

This feature comes with default rules for a smoother experience:

Note that you cannot then pick multiple series / metric to be displayed.

Dark mode

Turn Husprey to dark mode from your personal settings. Choose one option manually, or pick the "Auto" one to make Husprey follow your system preferences.

Other improvements

• Improve rendering performance of the whole notebook page, with a focus on Safari 🚀

• Display labels on data points for any chart type.

• Get notified when a new version is available

• Improve invalid query handle error location for better discoverability and clarity

Bug fixes

• Fix memory leak due to local cache not being cleaned up

• Fix SQL parsing errors in some cases

December 10th - Mostly bug fixes, and new real time sync for comments and data results.

This version mostly fixes bugs 🐛, and brings real-time sync for comments and data results 🔃.

Do not hesitate to share some feedback using the dedicated bottom-left menu item.

Real-time sync for comments and data results

Comments and notifications are synced in real time. Actual conversation can now happen! 💬

Also, if a SQL query takes a long time to run, the application now receives its result in real time as it comes out of the database. While this bring little visible improvement for now, it is part of the foundations of our next big feature 🌟

For the tech savvy readers, we are now using websockets!

Other improvements

• Properly set the theme-color meta tag for nicer surrounding user interface ✨

• Display the string representation of a postgres interval in the result table ⏳

• Allow (back again) series to be hidden/displayed directly from the chart legend 📈

• Improve the authentication refresh process for a smoother experience 🚀

• Improve rendering performance of data query elements 🎨

Bug fixes

• Fix table columns not being displayed in the Documentation panel with some versions of Safari

• Fix SQL parsing for some cases, including elements about the lag function, the on clause, and the cast operator.

November 26th - New design for data queries, view notebook in full width, and more

This version brings a new design for data queries ✨, a "full-width" option for notebooks ↔️, and fixes some bugs 🐛.

Do not hesitate to share some feedback using the dedicated bottom-left menu item.

New design for data queries

The data query component has been fully redesigned to improve its usability and readability.

Also, you can now customize 2 particular elements:

• the query handle: the name of the corresponding local table that you can use as a reference in subsequent queries (organizations_by_plan in the example above), and

• the result title: meant to be written in plain natural language, as a table or chart legend, for your final readers ("Distribution of organizations by plan"')

New full-width option

Want to make the most of your large monitor? Need space to write / read long and complex SQL queries?

We've got your back with the new full-width option available from the top-right notebook menu!

Other improvements

The right panel has been slightly redesigned to prepare great upcoming features 😄

Bug fixes

• Fix some columns being duplicated in the autocomplete suggestions

• Fix SQL parsing for some cases

November 22nd - Combo charts, source tables of dbt models, and more

This version brings combo charts 📈, and extends (further) the dbt integration.

Do not hesitate to share some feedback using the dedicated bottom-left menu item.

Combo charts

We have heard your feedback and feature requests: combo charts are here!

They come with several new customization options:

• Associate each series to a specific vertical axis

• Customize the visualization of each individual series

• Smart rules automatically distribute series over the 2 vertical axes.

• The top-level "Combo chart" option comes with a handful of default values to smooth your experience.

Underlying source tables of dbt models

We believe in "build once and query anywhere". Right now it means helping Analytics teams leverage the Transform part of their workflow.

With this release, we deepen our dbt integration to bring Transform metadata where you write queries and reports.

Now, in addition to models and source tables being identified and enriched with information from your dbt project (description, tests, etc.), we make use of the dependency graph to display the underlying source tables a model depends on.

This comes handy when wondering on what specific data service/provider a given model is built upon (embedded analytics? CRM? application database?)

Note: source tables will be displayed after the next upload of dbt artifacts.

If these source tables have freshness information, it is furthermore used to compute a model-level aggregated status, very much useful to assess the reliability of a given model.

Also, this integration is now available to everyone!

Other improvements

• Better home page for new users that still haven't provided their database credentials.

November 8th - Extended dbt integration, improved UX for data panel, and more

This version extends the dbt integration, improves the data panel UX, brings several polished UI elements 🌟, and fixes some bugs 🐛.

Do not hesitate to share some feedback using the dedicated bottom-left menu item.

Extended dbt integration

We believe in "build once and query anywhere". Right now it means helping Analytics teams leverage the Transform part of their workflow.

We are shipping the second part of our dbt integration to bring Transform metadata where you write queries and reports.

Now, in addition to descriptions and SQL being displayed for dbt models, new elements are available from the Documentation panel:

• description and freshness information (if available) for source tables

• tests on columns for models and source tables:

Also, any admin can activate this integration from the Settings > Workspace page:

Current limitations: For now, it's only available for dbtCore (not dbtCloud)

Improved data panel

For a better navigation, details of a table now open in tabs within the "Documentation" panel. Both "Data Explorer" and "Documentation" panels can be collapsed.

Polished UI

Several elements have been polished to make them easier on the eye, starting with a new font, and new navigation items:

Other improvements

• Smart naming rules when duplicating a notebook are now implemented for additional title patterns (week number, month, quarter, year):

Bug fixes

• Fix crash when searching in the data tree or in the list of columns

• Fix SQL parsing for some cases

• Fix websocket connection status being tracked at a too high level of the app, leading to extraneous renders

• Properly handle errors that can occur in specific cases during data source sync

• Reload the app if code chunk is not found during lazy loading

October 15th - Metadata panel revamped and initial dbt integration available on demand

New data panel 👨‍🎨

We wanted to help our users to make the most of the data tree. It's now a Metadata panel with information you need to write your SQL queries.

Once again do not hesitate to share feedback about this redesign. It would mean the world to us. You can use the bottom-left menu item for that.

You can navigate your schemas and tables as before. But now a table panel opens when needed so you can list and search columns from a table (even when hundreds of them). A preview of your table is available with sample rows.

These changes drastically improve your knowledge about what can be done.

dbt integration

We believe in "build once and query anywhere". Right now it means helping Analytics teams leverage the Transform part of their workflow.

We are shipping the first parts of our dbt integration to bring Transform metadata where you write queries and reports. If you want to try it out, please contact us.

Description (text and markdown) are displayed but also dbt models SQL so you can better understand how a table was built.

Current limitations: For now, it's only available for one project per workspace and for dbt Core (not Cloud). We have more to come on this in the coming weeks.

Other improvements and fixes

• Add Husprey as application_name in our data base connections to ease monitoring

• Local tables use a snake-cased version of query title as local table identifier

• Fix edge cases where SQL generation failed when using local tables

• Fix search not expanding all nodes when using personal credentials

• Fix a crash when selecting a text format in the Editor toolbar on Safari

September 30th - Local tables, improvements on comments, and bug fixes

This version brings a fundamental piece of our vision 🔭 : our notebooks are the heart of our SQL workspace for Data Analysts and they get a lot more powerful with local tables 🚀 ! Read on for more details 👇

Do not hesitate to share your thoughts using the dedicated "Feedback" button right from the app!

Local tables

You can now directly run:

SELECT *
FROM my_awesome_previous_query

All previous queries now generate a local table that can be referenced in a notebook. You can then quickly see how intermediate results look like while writing something that is closer to your actual workflow.

If you are using CTEs ("with my_favorite_cte as"), then it might sound familiar to you. Actually, behind the hood we use CTEs but we drastically improved the analysts' experience.

This feature is still currently considered in beta. Any admin can activate it from the Workspace settings.

When activated globally, you can still disable this feature for a given notebook if needed

Comments and notifications

🔗 Easily get a direct link to any thread from the new top-right menu

(Cloud only feature) ✉️ Receive unread notifications by email, and do not miss anything in your Workspace.

You can deactivate these email notifications from the settings page:

Other improvements

Here are some additional improvements coming into this release:

• Split the application code bundle in several chunks for a faster initial load

Bug fixes

• Fix crash on mobile due to Notification API

• Fix button to expand the right sidebar in Notebooks

• Fix Notebook action menu that could fly around in some cases

September 20th - Various improvements, and a surprise 🎁

This new version improves the experience around comments and notifications 💬, drastically improves performances of the text and SQL editors 🚀, and more. Read on for more details 👇

Do not hesitate to share your thoughts using the dedicated "Feedback" button right from the app!

Comments and notifications

🔍 Unseen comments of interest are now highlighted to more easily follow conversations.

⚙️ Desktop notifications can be activated / deactivated from...

... the notification center

... and the Settings page

🎁 Reply to a comment with an emoji (who can live without emojis, huh? 🤷)

Reminder: Comments are available for new notebooks only. If you work on one that does not support the new feature, you'll be advised to duplicate it or create a new one.

Note: This "Comments" feature uses websockets to get real-time updates. Please check that your setup supports them.

Text and SQL editors

• Improve performances when typing text in a notebook body, and when typing SQL in a query.

• Improve auto-scroll behavior when going directly to a given query

Other improvements

Here are some additional improvements coming into this release:

• Redesign the left panel to prepare upcoming great features 😊

• Autocomplete can be disabled from the Settings page

September 10th - Comments, notification center, and more

This new version brings real-time notebook comments 💬, a global notification center 🏠 , and more. Read on for more details 👇

Do not hesitate to share your thoughts using the dedicated "Feedback" button right from the app!

Comment on a notebook

Collaborate with your coworkers with real-time notebook comments 💬

You can post global comments from the header:

Or comment on specific elements:

Working on another notebook? No worries, notifications are here to keep you up-to-date.

Note: Comments are available for new notebooks only. If you work on one that does not support the new feature, you'll be advised to duplicate it or create a new one.

Notification center

Stay on top of the latest events in your workspace with the notification center available from the home page 🏠

Improvements

Here a some additional improvements coming into this release:

• After signup, new users get redirected to the page they tried to first visit (handy when sharing notebooks!)

• New pre-login screen to give a warmer welcome to our users 🤗

Bug fixes

• Fix notebook duplication and CSV export with large result sets

• Fix crash on pie charts in some cases

• Display BigQuery objects in result table

August 16th - Export CSV, duplicate notebooks, and more

📣 Much requested features ahead! 😃

With this version, you can export query results in CSV, duplicate an existing notebook, and more. Read on for more details 👇

Do not hesitate to share your thoughts using the dedicated "Feedback" button right from the app!

Export query results in CSV

Easily export specific query results for further sharing and use in external tools (did someone say spreadsheets?).

To do so, use the new button available in the header of a query when in table view:

Duplicate an existing notebook

Reproduce a previous analysis or update a report by duplicating an existing notebook, right from the top-right 3-dots menu:

Improvements

Here a some additional improvements coming into this feature:

• Copy direct link to a given query. No more manual scrolling to the right section!

• Run a collapsed query directly from its header:

• Admins can invite several new users at once.

• Mobile users are gently informed that the mobile experience is mostly read-only.

• Tooltips, modals and welcome screens have been polished for better readability.

Bug fixes

The page number of notebook list is reset when filters have changed.

July 21st - Charts and UI improvements

This version improves the chart experience 📈, brings a few more polished UI elements 🌟, and fixes some bugs 🐛.

Do not hesitate to share your thoughts using the dedicated "Feedback" button right from the app!

Charts

Customize the x-axis ticks rotation via the dedicated drop-down menu in the chart editor.

More vivid colors for nicer charts:

Improvements

The tutorial has a nice surprise in store for users who complete it (will you find it? 🔍 ).

Onboarding screens, buttons and dialogs have been polished.

Bug fixes

No more clutter with poppers and SQL code lines.

Copying a JSON value from a result table... actually copies that JSON.

July 9th - Tutorial, date formats, and more

This version gives a warm welcome to new users 👶 with a step-by-step tutorial, and charms data aficionados 🤓 with date formats!

We've also polished a few elements of the UI to make them easier on the eye 🌟

Tutorial

New users are now welcomed in Husprey with a dedicated tutorial and revamped home screen.

Date formatting

As you always wanted to, you can now nicely format dates in tables and charts.

Some default rules are applied to smooth your data experience!

Autocomplete for BigQuery

Autocomplete is now also available for BigQuery data sources.

June 21 - Pie charts, filters, and more

We've heard your various feedbacks (thanks a lot for sharing them 🙏 - do not hesitate to send new ones using the dedicated "Feedback" button right from the app), and here are 3 new features and improvements in this release!

Pie & donut charts

Use a pie or donut chart to display a single metric split into a (small) number of groups (i.e. the result of a GROUP BY aggregation).

The options let you choose which field to use as your "Series", and which one to use for the "View By" dimension (i.e. the pie slices). You can also customize the chart’s legend, and whether or not to show the total in the middle of the donut.

Filter Notebooks by tags

Filter the list of all Notebooks by tags, and more easily browse your company's data knowledge.

As a shortcut, simply click on any tag displayed in that list to add it to the filters.

Autocomplete for joins

Getting tired of manually writing these tedious join conditions? 😫

Autocomplete is now supercharged with elements from your data model, and directly suggests appropriate joins, just as you type.

May 27 - Autocomplete

No more silly typos and headaches trying to remember these 🤯 column names, autocomplete is here! 😅 🎉

• Automatic. As you type a SQL query, the suggestion menu will pop up if there are possible completions.

• Convenient. If you continue typing characters, the list of members is filtered to only include members containing your typed characters. Pressing Tab or Enter will insert the selected member.

• Bound to your data model. Go through schemas, tables and columns within a few key strokes.

• Handy. You can trigger autocomplete in any editor by typing Ctrl+Space.

  The editor also supports "tab completion" which inserts the first matching completion when pressing Tab. This works regardless of the suggest widget showing or not.

May 21 - Default 100-rows limit that can be extended to 10k rows

• You are not limited to 50 rows anymore. You can query up to 10k rows and chart up to 1k rows.

• All new queries have a 100-row limit by default. You can edit that from a query menu or hovering the query execution line!

  

April 27th, 2021 - Use your write rights (Personal data source only)

• When using personal credentials, you can now write into your database. It's limited to current user's rights.

Example:

• Also, you can refresh your data model when you need to see your latest changes. After you created a table for example.

  

April 19th, 2021 - Personal credentials

• Share a data source but not your credentials. When setting up a data source, you can now make sure each teammate use their own credentials to make requests.

• More space to write. Notebook metadata are now displayed in a collapsable right panel.

• Never lose work. A Notebook is locked if concurrent updates happen. You can force save if needed or restore a previous version

• (SaaS only - contact us to activate on an on-premise instance). Online viewers of a Notebook are displayed in the metadata right panel.