This article is designed to help you navigate the Datagrok platform. We will introduce core concepts and walk you through the fundamentals of working with our platform.
First of all, you need to launch Datagrok. Assuming that you are going to work with
a public instance of the platform, all you need to do is to open our website
(https://datagrok.ai/) and hit the LAUNCH button.
This will lead you to the Sign-Up/Login form.
The first thing you will see after launching the platform is Datagrok's welcome page:
Here you can find an overview of the platform's key capabilities as well as the links to our Wiki, Community Forums, Tutorials, YouTube Channel, API Samples, browse a gallery of publicly available projects and more.
The welcome page consists of several visual components: a sidebar, toolbox, main
view, property panel, and context help. Typically, the view
resides in the center and occupies all available area in the screen. The sidebar and toolbox are
placed on the left. The sidebar contains controls over the main functionality, while the toolbox
displays options specific to the current view. The property panel and context help are docked to the
right. This appearance is fully customizable, you can manage the displayed windows from the sidebar
(Windows), or turn them on/off with the following shortcuts:
| Menu Option | Shortcut | Action |
|---|---|---|
| Windows | Toolbox | Alt + X |
| Windows | Properties | F4 |
| Windows | Context Help | F1 |
In addition to the design modularity, these panels can be resized and docked.
Now, let's proceed to the sidebar icons and explore the functionality they offer.
This tab brings together the platform's main data-retrieving capabilities. There are multiple ways in which you can work with data in Datagrok: from a full in-browser experience to establishing data connections, deploying interactive applications, running data jobs, building pipelines. It all starts with a simple step of importing data.
Take a local file to begin with. To open it on the platform, drag-and-drop it into the browser.
Other options include file dialog interaction: double-click on the Data icon or press Ctrl + O.
You can access this dialog from the toolbox as well (Data | Open local file). It is important to
note that the data resides in the browser up until the moment you choose to upload it to the server.
This is always done explicitly. See the list of supported file formats
and read more about saving data to the server.
File shares are arguably an even more popular choice for working with data. Files form a hierarchy, which you can browse naturally from the interface. The File Browser operates as a file system: you can organize pieces of data into folders and subfolders, rename and move files. The tree map component indicates how much space different files occupy in a shared folder, similar to disk partitions. Once the user selects a file, this area changes to a preview where, depending on the extension, you may see a table, image, or text.
The ribbon menu on top contains icons that highlight open views (a caret pointing down), control
item selection (Select all, Select none, Invert selection), update the file view (Refresh)
and toggle previews (File preview).
Every user has a special HOME directory to store their files. For convenience, it is created by
default so that users can start working with it right away. In addition, there are several ways to
set up a new file share. First of all, users can do that from Data | Files | Actions | New File Share.
Besides, since shares are essentially data connections, they can be added in the connection
dialog (Databases | Actions | Add New Connection). If you enable indexing feature, the platform
will extract metadata from the files, which you can later apply in the search. The articles on
data access, file shares and
routing explain in more detail how to work with file shares.
In Datagrok, you can retrieve data from a variety of sources. The platform comes with data connectors for 30+ popular databases, and the list is constantly growing. In addition to that, it is possible to develop your own data connectors, and seamlessly integrate them into the platform.
Connections to data sources live in Connection Tree,
which you can reach from Data | Databases. Each branch corresponds to a provider and unfolds
to show connections to the given data source. Connections, in their turn, list data queries
and, if available, tables from the database.
The Windows icon lets you manage windows and see the currently open ones. In particular, you can
customize the visibility of the following panes: Menu, Tables,
Columns, Properties, Toolbox, Variables,
Console, Context Help. To save your preferences regarding the display
of certain windows at startup, tweak your user's general settings
(Settings | General).
This panel describes the properties of a current object. Typically, whenever the user clicks on most objects within the platform, such as a table, data connection, application, command, or any other supported object type, this object becomes a current object. So, to focus on one, simply click on it. After that, you will see the corresponding property panel as well as the context help.
F4 toggles the visibility of the property panel. Additionally, you can manage it from the sidebar
(Windows | Properties). There are several useful icons in the panel header:
| Icon | Action |
|---|---|
| Back/Forward | Navigate the history of previously viewed object properties |
| Clone and detach | Detach a copy of the property panel preserving the content |
| Collapse all | Collapse all listed property values |
| Favorites | Switch to the properties of favorite objects |
The property panel saves history of viewed objects, so you can switch to the previous selected
object if necessary. This also means that the content of the property panel is synchronized with the
current object. To display properties of another object along with the currently selected one, click
on the Clone and detach icon.
The content of a property panel is generated based on the type of a current object, as well as its value. For example, if you open a table, you can find applicable actions, predictive models, and algorithms. The panel is designed in such a way that related objects are easy to reach, e.g., having a query in front of their eyes, users can identify the corresponding data connection and switch to its properties with one click.
Some visual components, such as viewers, activate the property panel in a special way. To access the
properties of a viewer, you would click on a gear icon ⚙ in the top right corner of the viewer's
header.
What is more, the information provided for a current object can be expanded by means of info panels. They work on a number of different things, e.g., files or cells in a grid.
The toolbox located on the left shows the properties and actions available for the currently active
view. You can toggle its visibility with Alt + X or Windows | Toolbox. In addition, the icons on
the top right let you pin or close the toolbox.
The toolbox contents change in accordance with the current view and extend its functionality, e.g., for a table view it will show visualizations and filters, applicable models and layouts.
Alt + V opens a variables view, which contains all variables available in the current context, and
allows manipulating them. It is also possible to enable the variables view from the toolbox
(Windows | Variables) or from the console.
There are several ways for an object to appear as a variable:
- Direct assignment like
x = 5in the console - Drag-and-drop an object to the variables view
The view consists of four fields describing the variables: name, value, type, and semantics.
The last field contains the semantic type, or quality, of a column value detected out of the box or
as per custom semantic type detection rules.
Console can be used to call functions and record macros.
| Key | Action |
|---|---|
Tilde ~ |
Open the console |
| Tab | Complete a command |
| Up/Down | Previous/next command |
Press the tilde key ~ to open the console or enable it from the toolbox (Windows | Console).
Two icons on the top left let you clear the console, or open the variables view. Clicking on the function name will bring up its details in the property panel.
Run the Mul command (multiply two numbers) with the specified parameters:
Mul(2,3)
Edit parameters of the Mul command and evaluate it in a dialog window:
Mul
Get help for the Mul command:
Mul?
Access the current object from the console with the o variable:
o.name
Select rows with empty values in the HEIGHT column:
SelectRows("demog", IsNull("HEIGHT"))
Extract rows with empty values in the HEIGHT column into a new dataframe:
ExtractRows("demog", IsNull("HEIGHT"))
Every visual transformation within the platform is backed by a corresponding function, which provides for automatic logging. Whenever a function is executed, the call gets logged in the console. No preliminary action is required on your part, so you can seamlessly have it both ways: use the console to examine which functions got triggered by particular UI events to reproduce those steps in the future, or call functions on your data directly from the console. The first case can be used in data transformations and data pipelines. The second use case is a common practice for debugging custom functions defined in a package.
Context help appears at the bottom right and shows additional information about objects you are
currently working with. You can toggle this pane with the F1 hot key or manage it from
Windows | Context Help. It is enabled by default to help you navigate the platform and get the
hang of its core components.
While thorough explanations can be found in our documentation, the main purpose of contextual help is to provide you with concise cues along the way. To learn more about a topic of your interest, make use of the icons available at the top left:
| Icon | Action |
|---|---|
| Back/Forward | Navigate visited pages |
| Home page | Open Wiki's home page |
| Clone and extend to view | Open a page in the main view |
| Open in new tab | Redirect to the Wiki page |
Client-side settings are specific to the user, and are controlled by the user (unless the organization's IT policy overrides this).
To edit client settings, open Settings from the sidebar.
Click on your user image on the sidebar to open the user profile. Here you can find a summary of your data, read notifications, change your password,retrieve API and developer keys. The data on your activity is provided by the platform's internal audit system.
The help section at the bottom left resets the toolbox options to the ones you typically see on the welcome page.
If an error occurs during any operation, you will be notified with the corresponding error message
under the help section. In this case, you can open the Inspector tool (via an exclamation mark !
icon or by pressing Alt + I) to further investigate the issue.
If you run into any difficulties, please refer to our documentation or ask a question in our community forums. Quick access to these links, as well as other useful resources, is provided in the toolbox of the help icon.
See also: