User Guide

Latest Version: v1.8

Created 2017-06-12. This User Manual contains an in-depth description of all the features in the RCloud Graphical User Interface (GUI). It is currently maintained by Spencer Seidel on a voluntary basis.

Note that v1.8 of this user guide incorporates changes made in the v1.7 release.

Getting Started

RCloud supports Chrome, Chromium (the open-source version of Chrome), and Firefox web browsers. Download the latest version of any of these to get started.

RCloud stores information about users and their RCloud notebooks using GitHub technology. Therefore, new users must either register with their installation's instance of GitHub or create a public GitHub account. Please refer to your installation's documentation about creating a new GitHub user or see our Tutorial, Anonymous User versus Data Scientist Access, for information about creating an RCloud/GitHub account on the public RCloud instance.

Since every installation's networking architecture will be different, please refer to your installation's documentation for any relevant environment information.

Graphical User Interface (GUI) overview

Parts of the GUI

The RCloud GUI consists of:

  • A header bar located along the top of the browser window
  • Left and right windowshade panels
  • Prompt and markdown cells located in the center of the screen

GUI navigation

Opening and closing panels

Clicking on the title of a panel, such as Notebooks and Search, opens, closes, and in some cases, resizes it.

Opening and closing all panels

When all RCloud panels are minimized in a column, a "+" sign will appear, as shown here:

Left Windowshade Panel

Left Windowshade Panel

Clicking the "+" sign will restore all panels in the column to their previous sizes. Now, a "-" sign will appear, shown below, which will minimize all open panels in a column.

Right Windowshade Panel

Right Windowshade Panel

Changing panel width

To change the width of a panel, hover your mouse over the edge that touches the cells in the middle. When the mouse cursor changes into a double arrow, click and drag left and right to resize.

Header bar

The RCloud header bar is located at the top of the browser window and contains several notebook control icons:

: Please see the Sharing Your Notebooks section for more information.

: Click the star icon to add the current notebook to your "Notebooks I Starred" folder, shown in the Notebooks panel. The number next to the star represents the number of RCloud users who have starred the notebook.

: Click the fork icon to make a new copy of a notebook for editing.

: Whenever you run a notebook, RCloud automatically saves a version in GitHub. If you need to close your RCloud session immediately without waiting for a lengthy run time, click the save icon to save a version immediately.

: Click the play icon to run all markdown and prompt cells in the current notebook. Cells are executed asynchronously. RCloud displays the output as it becomes available.

: This is the title of the current notebook. Click in the title to change it.

Note that you can create subdirectories by adding any number of forward slashes ('/') to your title, similar to a Unix command line. E.g. "Cookbook for R/1 - Basics/1.1 - Indexing into a Data Structure." If the subdirectories don't already exist, RCloud will create them on the fly, so there is no need to create individual subdirectories before changing the name of your notebook. Subdirectories allow you to group your notebooks and will be displayed in a tree-like hierarchy in the Notebooks section of the left sidebar.

If a notebook was created as a result of Forking another notebook, the forked notebook name is displayed in a smaller font beneath the current notebook's title.

: Click the Advanced drop-down menu to access more features.

Header Bar: Advanced Menu

Header Bar: Advanced Menu

  • Open in GitHub: Internally, notebooks are stored as GitHub "gists." Depending on your installation, you can view your notebook gist within GitHub directly. Some installations, however, will use the RCloud Gist Service, in which case the menu item will be disabled. Regardless, RCloud stores every revision of your notebooks, so you can retrieve them later, if necessary.
  • Load Notebook by ID: Replace the current notebook with another via URL or gist ID.
  • Pull and Replace Notebook: Opens a dialog box where you can tell RCloud to copy the contents of an existing notebook (within the same RCloud instance) via URL, file, or ID and replace the contents of the current notebook.
  • Import External Notebooks: Opens a dialog where you can import multiple notebooks stored in another GitHub repository. You'll need the source GitHub repository API URL and a list of notebook IDs, newline separated. In addition, you can supply a prefix that will cause all the imported notebooks to go into a folder. E.g. "myfolder/". Note that a trailing '/' character is required for this to work.
  • Export Notebook to File: Your browser will automatically save a copy of the current notebook in JSON format in whatever directory you've designated for downloads. The file name will be the same as your notebook with a .gist extension.
  • Import Notebook from File: In order to import a notebook, it must be in either .gist (see above) or .R (see below) format. You can validate the notebook by clicking the word "Validate." To import, click the Import button.
  • Export Notebook as R Source File: Your browser will automatically save a copy of the current notebook as an R source text file in whatever directory you've designated for downloads. The file name will be the same as your notebook with a .R extension.
  • Manage Groups: Opens the Notebook Permissions / Group Management dialog, where you can manage your groups.
  • Publish Notebook: By default, users who wish to view your notebooks must be logged into RCloud. If the Publish Notebook box is checked, any user who has network access to the notebook's URL will be able to view the notebook. Editing features will be turned off for these users.
  • Import Rmarkdown File: Imports Rmarkdown containined in a .Rmd file.
  • Export Rmarkdown File: Your browser will automatically save a copy of the current notebook as an Rmd text file in whatever directory you've designated for downloads. The file name will be the same as your notebook with a .Rmd extension.

: Click the Discover link in the header bar to view the most recent and most popular notebooks.

You can create a mini-view of your notebook for the Discover view by creating an asset called thumb.png. thumb.png will be displayed in the Discover view if present.

: Click Logout to end your RCloud session.


There are two types of cells in RCloud. The first is the prompt cell, which allows you to interact with RCloud in more-or-less command-line fashion. Prompt cells are useful for quick, interactive sessions.

Prompt cells support R, Python, and Bash, which you can specify in the pull-down menu to the right of each cell:

Prompt / Markdown Cell Selection

Prompt / Markdown Cell Selection

The second type is the markdown cell. Markdown cells are better suited for cutting and pasting chunks of R code and adding simple formatted documentation. Note that Markdown cells do not currently support Python or Bash.

We'll get to the difference between Markdown and RMarkdown cells in a moment.

"Data marshalling," or using objects between cells of different languages, is not supported at this time. Also, each "shell" cell represents a separate Unix shell, so environment variables cannot be passed across shell cells. However, R environment variables defined in R cells are inherited by shell cells automatically. Shell cell commands are executed in Bash.

Prompt cells

Prompt cells mimic an interactive R, Python, or Bash shell. Type a line of code, press Enter/Return, and the command is immediately executed. When RCloud is finished, it presents the result and a new prompt cell.

For example, here's the result of pressing Enter/Return after typing a command:

Prompt Cell Result

Prompt Cell Result

Note that after executing a command, RCloud presents several icons that allow you to interact with the cell:

: Run the code in the cell.

: Edit the code in the cell (clicking on your code also enables editing).

: Split the cell into two parts at the cursor.

: Delete the cell.

Another way to interact in multi-line mode with prompt cells is to cut and paste multiple lines of code into the cell. When you do, you'll see something like this:

To execute the code, simply put your cursor at the end of the last line and press Enter/Return.

Note that you cannot insert a prompt cell above an existing prompt cell. The only way to add new prompt cells is by executing the current prompt cell. When you do, a new cell is created under the existing one.


Throughout this documentation, prompt cells are sometimes referred to as R cells, Python cells, or Bash cells.

Markdown cells

Adding Code

Markdown cells are where you enter and edit blocks of multi-line R markdown. Markdown is a plain-text formatting syntax used to create simple formatted documents. In order to differentiate your R code from text, surround your code with the following (back ticks, brackets and "r"):

## R code goes here
print("Hello World!")

For additional information about Markdown, please refer to the full documentation of Markdown syntax.

When you're done editing a markdown cell, click the icon to the right of the cell to render the contents of the cell.

Editing and viewing results

If you find a typo or would like to otherwise edit your code, click the icon. You can also click on any code portions of the output to enter edit mode.

At the beginning or end of a cell's code, use the up and down arrow keys to jump to the next or previous cell.

Adding and deleting cells

To insert a cell above, click the icon. To insert a cell below, click the icon. To delete a cell altogether, click the icon.

Cell run-state indicator

The run state of each cell is displayed via an icon in between the gutter and cell name:

: The cell has not been run. This could also mean that the cell ran successfully, but the output may not be consistent with the code in the cell because the code was modified after RCloud initiated a run of your notebook.

: The cell is queued to be run.

: RCloud initiated a run of your notebook, but a cell's code was modified after execution was initiated.

: The cell is running.

: The cell is running, but because the code was modified after notebook execution was initiated, the output may not be consistent with the code.

: The cell ran successfully.

: The cell ran but had errors.

: The cell's run was cancelled.

Stopping cell execution

When you run a notebook, you can stop running cells and prevent queued cells from running by pressing the stop button, located in the header at the top of the screen:

Nonpreemptive Stop Icon

Nonpreemptive Stop Icon

This sends an interrupt to the R process and terminates execution if possible.

Rearranging cells

To rearrange your cells, click and drag the blank status area above the cell or the icon to the left of the cell.

Joining cells

To join cells of the same flavor, click the join icon at the right of the cell. This will combine the contents of the cell with the cell immediately above it.

Join Cells

Join Cells

Markdown versus RMarkdown cells

Behind the scenes, RCloud uses several different R packages to render output. Markdown cells use the Markdown and knitr packages directly for output. RMarkdown cells, on the other hand, use rmarkdown (also known as R Markdown v2).

Currently, RMarkdown support should be considered experimental.

Saving plots

Hover the mouse over a plot created in an R cell to make the disk icon appear in the upper right corner (see 1), which contains a list of available image formats. A widget at the lower-right corner can be used to resize the image (see 2).

Note that you can only save plots created in R cells.


RCloud supports the standard R locator() function. When the locator is active, RCloud adds a blue border around the plot and changes the cursor to a crosshair.

To add points, left click anywhere on the plot. To end or abort a locator request, press the Esc key. Once you have selected your locations, locator() returns the points clicked:

Locators only work in R cells.


RCloud notebooks are simply collections of prompt and Markdown cells (as well as comments and assets, both of which we'll get to later). Everything in your public notebooks is searchable by every other user of the system. This encourages reuse and makes learning how to use the hundreds of available R packages somewhat easier.

You can also browse everyone else's notebooks by opening the Notebooks section on the left sidebar. To do this, simply click on Notebooks at the top of the panel. This toggles the panel, opening or closing it:

RCloud User / Notebook Directory

RCloud User / Notebook Directory

To load a notebook into the current session, click on the name. After it loads, you can examine the source code or click in the header bar to execute all the cells on the page.

Creating a notebook

To create a new, blank notebook, click the + sign at the right of the Notebooks panel header area:

New Notebook Creation Icon

New Notebook Creation Icon

RCloud will automatically choose a title for your new notebook, Notebook N, where N is the next available number among your notebooks. To give your notebook a more meaningful title, click on the title in the header bar.

To change the default name of new notebooks, see the New Notebook Prefixes section in the Settings panel.

Running a notebook

To run all the cells in your notebook, click the icon in the header bar.

RCloud notebooks are executed asynchronously. RCloud will show individual cell results as the results are ready to display.

To run only selected cells, hold down the Ctrl/Cmd key when you click the icon in the header bar. See the Multi-Cell Selection section for more information about multiple-cell selection.

Partial notebook runs

Pressing the icon in a cell while holding down the Shift key will run that cell and every cell after in order.

Long-running notebooks

Notebooks that run longer than a few seconds will cause the browser screen to dim and a please-wait message to be displayed.

Currently, RCloud has no explicit mechanism to stop a long-running notebook. If you mistakenly launch a long-running notebook, you can simply reload the notebook in another browser tab or reload the page. This doesn't stop the execution behind the scenes, but the output of the previous run will not interrupt your current session. Be careful of side-effects, like changing the contents of a file in your local directory in such a way that it affects the output of the notebook.

Forking (copying) a notebook

To copy another user's notebook, first navigate to it using the left sidebar and then click to load it into your current session. Now, you're running another user's public notebook in your own session. This is sufficient for running reports or performing other read-only activities. If you want to edit the notebook, you'll need to make your own copy, or, "fork" it.

After you've loaded the notebook you want to fork, click the Fork icon in the header bar at the top of the screen:

Fork Icon

Fork Icon

After forking a notebook, you'll own your own copy and be able to edit it.

The fork icon is always available, which means you can fork your own notebooks. If you are viewing a previous version of a notebook, you can fork a copy of that version. Caution: currently when you fork your own notebook, the history is lost; we hope to fix this soon.

Saving your work

There are two mechanisms by which your work is saved in RCloud.

Manual Saving

You can save your notebook at any time by clicking the icon in the header bar.


RCloud keeps track of your notebook versions automatically and frequently. Every time you save, create, or run a Markdown or Prompt cell, the newest version of your notebook is saved. To browse different versions of your notebook, which are stored chronologically with the latest version on top, hover over the name of your notebook in the left sidebar and click the clock icon. A drop-down list of versions will appear:

Notebook History / Version Icon

Notebook History / Version Icon

To change they way dates and times are displayed next to your notebook versions, see the show terse version dates setting.

Version Tagging

To tag a notebook version, click twice on a version name to edit it in place.

Version Tagging

Version Tagging

Now, rather than referring to a specific notebook version with &version=hash in a URL, you can refer to a specific notebook tag:

Version Tagging URL

Version Tagging URL


This is useful when you want to share a version of a notebook but plan to continue developing it. For example, you can tag a version as the "LatestProductionVersion," and then apply that tag to another version when you're ready to share your new work. This way, existing URLs (perhaps stored in someone's bookmarks) won't break as you update your notebooks.

Reverting to a previous version

Should you decide that a previous version of your notebook is the "best" version, you can make that version the current version by first loading the desired previous version of the notebook and then clicking the revert icon.

Revert Icon

Revert Icon

Hidden notebooks

By default, all RCloud notebooks are visible to all RCloud users. If you'd like to toggle the Show/Hide flag on a notebook, hover over the name of your notebook on the left sidebar and click the eye icon. Note that hidden notebook titles are grayed out for owners and invisible to other users.

Hidden notebooks are only invisible within the RCloud interface. Hidden notebooks are still visible within the gists stored in your GitHub instance.

Toggle hidden

Clicking the icon will hide your notebook from other RCloud users.

Toggle show

Clicking the icon will make your notebook readable by other RCloud users.

Protecting your notebooks

Protected notebooks are readable only by the owner and (optionally) a select group of users and will not show up in search results (although previously unprotected versions might).

Notebook permissions

View or modify notebook protection by clicking the notebook "info" button next to the notebook name in the notebooks tree:

Notebook Information Icon

Notebook Information Icon

If you own the notebook, click the "public" link (or "no group" if that displays):

Protection Dialog Box; No Group

Protection Dialog Box; No Group

This opens the notebook protection dialog:

Protection Dialog Box: Permissions

Protection Dialog Box: Permissions

Here, you can assign the notebook to any group you are a member of, make it entirely private (readable only by you), or make it public (readable by anyone).


Use the second tab of the protection dialog to create or rename groups. Additionally, you can designate other users as group administrators or members. Alternatively, you can select Manage Groups from the Advanced menu item on the header bar — note that the Notebook tab will be grayed out in that case, as Manage Groups is not notebook specific.

Protection Dialog Box: Group Information

Protection Dialog Box: Group Information

Unlike a hidden notebook, protected notebooks are not readable by anyone without permission, even within your GitHub instance.

Deleting Notebooks

To delete a notebook, hover over the name of your notebook in the left sidebar and click the icon.

RCloud will ask for a confirmation:

Confirm Notebook Deletion Dialog Box

Confirm Notebook Deletion Dialog Box

Click OK. The notebook will disappear from the left sidebar. The last-viewed available notebook will automatically load into the current session.

Sharing your notebooks

There are several ways you can share/view your notebooks. When you click on the downward arrow next to the share icon in the header bar , a popup menu will appear:

Header Bar: Notebook Share Type / View Mode

Header Bar: Notebook Share Type / View Mode

Here, you select the kind of URL you'd like to share. Make your selection using the popup menu and then right click on the icon to copy the hyperlink.

Clicking on a selection in the sharing menu will open a new browser tab and load the selected link.

Note that if you have a tagged version of your notebook currently loaded, where appropriate, RCloud will populate the shared URL with the tag instead of the version. This is beneficial because then you can tag future versions with the same tag and not break existing URLs.


This is the simplest method. This will create a link that will allow someone to see the notebook code and execute the notebook within the RCloud IDE. Users who do not own the notebook will see the play and share icons in the header. Clicking the play icon will execute all cells in the notebook. Clicking the edit icon will return to the normal header, allowing a user to fork the notebook, etc.

Hiding UI Elements

To hide all UI elements, add &quiet=1 to the URL. Note that this works only with view.html.


View the notebook as a Slidy Powerpoint-like slideshow. (TBD: more information needed)


View a notebook that uses the flexdashboard package. (TBD: more information needed)


This option is intended for FastRweb notebooks. Loading the URL (from anywhere, including other notebooks, a perl script, etc.) executes a notebook "behind the scenes" by opening a one-time R session, running the defined "function" within, shutting down the R session, and finally, returning the result. FastRweb notebooks MUST have a function named function() defined, as this is what notebook.R tries to execute upon instantiation. Output from notebook.R can be anything (e.g. text, binary data). This information will ultimately be processed by whatever mechanism called notebook.R.

notebook.R allows trailing paths to be processed by the notebook code if they start with /.self/. The subsequent path portion is passed to the run function as the argument. This allows notebooks to handle a "full tree" argument to the notebook on top of a single notebook URL.

E.g.: will call the notebook with set to /foo/bar. Note that the .self part distinguishes asset look up from a path info call.

See the notebook.R URLs section of the documentation for more detailed information about notebook.R URLs.


Unlike notebook.R, mini.html URLs open an R session via a WebSocket and keeps it open. mini.html notebooks must have a function named rcw.result() defined, as that is what mini.html tries to execute upon instantiation. Because the R session is kept open, users or processes can interact with the R session while the websocket is open.


RCloud supports the RStudio Shiny web application framework. To share Shiny-enabled notebooks, select this option.

Who starred my notebook?

To find out which users starred your notebook, click the notebook information icon:

Notebook Information Icon

Notebook Information Icon

Multi-cell selection

You can use RCloud's multi-cell selection features to perform various actions on many cells at once.

Multi-cell Feature

Multi-cell Feature

  • At the top of your loaded notebook, you'll find a header bar. Click the checkbox to select/deselect every cell in your notebook (1).
  • Click a cell's header to select that cell. Note that clicking the header of a selected cell does not deselect the cell (2).
  • Click a cell's checkbox to select/deselect that cell (3).
  • Click the icon to hide the output of selected cells. Note that this feature is only available with your own notebooks (4).
  • Click the icon to remove every cell that isn't selected (4).
  • Click the icon to delete all selected cells. Note that RCloud will indicate how many cells are selected out of the total number of cells (4).

Use the checkbox dropdown menu to fine-tune your selection:

Multi-call Feature Dropdown Menu

Multi-call Feature Dropdown Menu

  • All selects every cell in your notebook.
  • None deselects every cell in your notebook.
  • Invert changes every selected cell to a deselected cell and every deselected cell to a selected cell.

Multi-cell selection supports common keyboard extensions:

  • Shift-clicking a cell's checkbox will select all cells between the last selected checkbox and the current checkbox.
  • Ctrl/command-clicking a cell's checkbox is functionaly equivalent to checking a cell's checkbox and is noted here because it is a common selection activity.
  • Pressing the delete key will delete all selected cells.
  • Ctrl/command-k will crop (remove) all unselected cells.
  • Ctrl/command-Shift-i will invert the selection (select deselected cells and deselect selected cells)

Find and find replace

To find text within your notebook, type Ctrl-F (Win/Linux) or Cmd-F (Mac) to open a find dialog at the top of your notebook:

Find Text Dialog

Find Text Dialog

To find and replace text within your notebook, type Ctrl-H (Win/Linux) or Cmd-Option-F (Mac) to open a find and replace dialog at the top of your notebook:

Find and Replace Text Dialog

Find and Replace Text Dialog

Recent Notebooks

Access your recently opened notebooks via the Recent link in the Notebooks titlebar:

Recently Opened Notebooks Link

Recently Opened Notebooks Link

Data access

There are many ways to access data in RCloud. We'll cover the options available in the GUI here, but be sure to browse the "RCloud Sample Notebooks" section of the Notebooks panel for ideas.

File upload

Open the File Upload panel on the right sidebar by clicking on the heading. Then, select "Choose Files" and browse to a local file.

File Upload: Example 1

File Upload: Example 1

Next, click the "Upload" button. If the file exists, RCloud prompts with a notice that it will be overwritten. If all goes well, RCloud confirms the upload.

The file now exists in your installation server's home directory and is accessible from within RCloud. For example:

File Upload: Example 2

File Upload: Example 2

The first line of R code loads the contents of the CSV file into an object called "mydata." The next line tells R to output a string version of the object.

Upload to notebook

The Upload to Notebook checkbox changes the way File Upload works. Rather than uploading your file to your home directory, RCloud will store the file inside your notebook as an "asset." To view notebook assets, click on the Aassets panel on the right sidebar. Please see the Notebook Assets section for more information.

Notebook assets

Notebooks can contain "assets," which are files that can be used within your notebooks or simply for keeping track of unused code (as in the the case of scratch.R, which is a text file where you can keep bits of code while working on your notebook).

Right Windowshade Panel; Asset Area

Right Windowshade Panel; Asset Area

Data as an asset

For example, here is the content of a file called distrib.csv:


We can store this data in a new asset called distrib.csv simply by clicking New Asset and typing distrib.csv as a name.

Now, this data is accessible in your R code:

Data (CSV file) as an RCloud Asset

Data (CSV file) as an RCloud Asset

Uploading assets

In addition to manually entering asset text, you can also drag and drop files into the Assets panel to upload them:

Drag and Drop to Upload Data

Drag and Drop to Upload Data

RCloud automatically generates asset links and displays them in the lower left-hand corner of the Assets panel. To copy the URL, right-click on it.

Automatically Generated Link (URL) to RCloud Asset

Automatically Generated Link (URL) to RCloud Asset

Binary Assets

Assets can be binary (e.g. an image). RCloud transparently encodes and decodes using base-64 encoding. When possible, the content is displayed in its native format in the asset panel.

Asset size

Assets are limited to 2.5MB each.

Cascading style sheets

Assets can contain CSS formatting information. This changes the way information is presented when your notebook is executed. For example, here is a bit of CSS that defines a paragraph style:

p.mystyle {
  font-size: 20px;
  color: red;

To use this CSS as an asset, it needs to have a special name that begins with "rcloud-" and ends with ".css". In the example below, the name is "rcloud-mystyle.css".

RCloud automatically uses CSS asset files with this file pattern and ignores others, so you can save bits of CSS in other files without worrying about overlap.

To use the mystyle paragraph style, simply reference it in Markdown using HTML:

# My Header

Here is a bit of red text:

<p class="mystyle">This is red.</p>
Using or Adding CSS as an Asset

Using or Adding CSS as an Asset

Note that you must reload your notebook to apply the CSS.


Assets can also contain JavaScript. When editing JavaScript (files must have the .js extension), RCloud automatically uses a JavaScript editing mode, which has built-in syntax checking.

Javascript Files and Syntax Checking

Javascript Files and Syntax Checking


When editing HTML (files must have the .html or .htm extension), RCloud automatically uses an HTML editing mode, which has built-in syntax checking and tag completion.

HTML Files and Syntax Checking

HTML Files and Syntax Checking

Renaming assets

To rename an asset, simply click on the file name on the asset's tab.

Renaming an RCloud Asset

Renaming an RCloud Asset

notebook.R URLs

It's possible to construct a URL for a notebook asset by selecting Open in GitHub in the Advanced menu, locating your asset and right-clicking the "View Raw" icon, which looks like <>, next to your asset. The URLs look like this:

This is cumbersome, however. A better and more powerful way to access assets is via the HTTP entry point, notebook.R.

notebook.R allows you to access your notebook in the following ways:<notebook-id><notebook-id>/<version-hash><notebook-id>/<filename><notebook-id>/<version-hash>/<filename><user>/<notebook-name><user>/<notebook-name>/<filename>

URLs ending in "<filename>" will return the given asset (file). To illustrate how it works using the previous example, any of these give HTTP access to the latest version of DummyData.csv:

Access to assets isn't the only thing you can do with notebook.R. Notice that in the list of ways to access your notebook above, not all methods reference a filename. If you reference a notebook or revision of a notebook, the URL will return the result of the evaluated notebook.

notebook.R is intended to be a general-purpose Remote Procedure Call (RPC) in R. RPCs in RCloud should always contain some Markdown to document what the RPC does, what the arguments are, etc. This way other users can simply view your notebook in RCloud to understand how to use it. This isn't enforced but is encouraged to promote reuse. The Markdown is only visible when users visit your notebook in RCloud. The Markdown is not output when called remotely.

Please see the following for more notebook.R features and information.

shared.R URLs

Files can be served from user R libraries via shared.R/<user>/<package>. Users can develop packages that use shared.R in their own development library until it is ready to be released.

Note that users are still responsible for setting the permissions on the library path — default permissions deny access, including to the web server, to user libraries.


To conduct a global text search in all public RCloud notebooks, open up the search panel by clicking on Search on the left sidebar. Here are the results for a search on "markdown." RCloud searches code, comments, notebook names, assets, everything.

Search Results

Search Results

1: Results can be sorted by the number of stars a notebook has, the author of the notebook (User), the notebook's name, or by the date a notebook was created.

2: Sorts the results in descending (Desc) or ascending (Asc) order.

3: When too many results are returned for any one notebook, RCloud will display a Show me more link that will toggle the rest of the results.

4: Search results are paginated. Click on any page number below the results or use the back or next page to page through the results.

Complex searches

RCloud supports Lucene's feature-rich query parser syntax for more complex searches. Features include wildcard, fuzzy, and proximity searches, boolean operators, grouping, and much more.

Please see the official query parser syntax documentation for more information.


Various aspects of your RCloud environment parameters may be changed in the Settings section of the left Windowshade panel, including the default new notebook name.

Settings Section of the Left Windowshade Panel

Settings Section of the Left Windowshade Panel

Show command prompt

This setting toggles the appearance of the default prompt cell that appears at the bottom of the currently loaded notebook.

Show terse version dates

This controls how RCloud displays dates when viewing notebook versions. When selected, RCloud will display dates and times only when they're different from the version before it.

Show cell numbers

Toggles "Cell 1," "Cell 2," etc. in the cells panel.

Arrange panels by size

In some situations, larger panels on the left-hand side of the GUI "crowd-out" the currently loaded notebook and force the user to resize. In order to minimize this, select the "Arrange Panel by Size" checkbox in the Settings panel. RCloud will then rearrange panels in order to reduce the need for resizing the notebook.

Clear R Session when entire notebook is run

Runs notebooks as though you'd just logged in.

Export only selected cells

In the various export commands available in the Advanced header menu item, only export selected cells.


Power users can extend RCloud's user interface with global and per-user extensions. Although this functionality is outside the scope of this document, more information is available here.

You can enable and disable extensions using the "Enable Extensions" and "Disable Extensions" text boxes. These set the user options addons and skip-addons, respectively. On starting the session, any extensions listed in skip-addons are not loaded.

Enter a list of extensions, comma delimited, and press enter. You will then have to reload the page.

New notebook prefixes

Use the New Notebook Prefix setting to change how RCloud names new notebooks.

Suppose you were working on a project, Foo. You might choose "Foo " for your new notebook prefix. New notebooks would be named "Foo 1," "Foo 2," and so on.

You can also include a folder. RCloud then places new notebooks within that folder in the notebook tree. For example, you may choose to make "Foo/Notebook " your new notebook prefix name. RCloud would create new notebooks "Foo/Notebook 1," "Foo/Notebook 2," and so on.


Anyone can leave comments about anyone's notebooks in the Comments panel in the right sidebar, as demonstrated below:

Notebook Comments

Notebook Comments

Comments are included in search results.

To submit a comment, simply input your text in the text area and click the comment icon . You can also submit your comment by pressing Ctrl-Enter (Win/Linux) or Cmd-Enter (Mac).

Editing comments

To edit a comment, click on the comment text to make changes. Then click Ctrl- or Cmd-Enter to update the comment.

Editing Notebook Comments

Editing Notebook Comments

Deleting comments

To delete a comment, hover your mouse over the comment. Click the X next to the comment to delete it.

Deleting Notebook Comments

Deleting Notebook Comments


It's often difficult to remember arguments to functions and what they mean, so RCloud has an inline help feature. In a new prompt cell, type a question mark immediately followed by an R function name and then press Enter/Return to execute the cell:


If help is available, RCloud will open the Help panel:

Alternatively, you can enter a function name in the text box at the top of the panel and click the help icon .


In the course of working with your notebook, variables are defined and assigned. Often, it is convenient to know the value of a variable without issuing a command to display it. The Workspace panel displays every variable you've defined along with its value.

Workspace Section in the Right Windowshade Panel

Workspace Section in the Right Windowshade Panel

In the case of dataframe variables, a link is displayed, which will open up the Dataframe panel.


Dataframe objects are displayed here after you click a dataframe link in the Workspace panel or use the View(object) command in a cell:

Dataframe Section in the Right Windowshade Panel

Dataframe Section in the Right Windowshade Panel


The session pane displays critical system information when available, such as session timeouts. The panel will automatically open when messages are available.

RCloud Session Information in the Right Windowshade Panel

RCloud Session Information in the Right Windowshade Panel

Click the icon to dismiss a session notification.

Note that Session panel output is limited to 10K in order to improve responsiveness when there is a large amount of output.

RStudio Shiny Support

RCloud is 100% compatible with the RStudio Shiny web application framework via the rcloud.shiny package. rcloud.shiny emulates a network connection to run Shiny on an RCloud server and client instead of a Shiny server.

Although comprehensive documentation of rcloud.shiny is beyond the scope of this document, the implementation is simple enough to get many users already familiar with Shiny started.

# Include rcloud.shiny library
# Include all the libraries from ui.R and server.R

# The rcloud.shinyApp function is the equivalent of the shinyApp
# (
# function in shiny library.
    # Pass the ui.R code something as shown below. 
  ui = fluidPage(

    # Application title
    headerPanel("Word Cloud"),

To learn more about how to use Shiny in RCloud, please see the rcloud.shiny example notebooks under RCloud Sample Notebooks in the Notebooks panel and the official Shiny documentation.

Keyboard Shortcuts

RCloud supports many keyboard shortcuts. To see an exhaustive list, click in a blank area of the RCloud GUI (so that the cursor focus is not in a cell, for example) and then type ?.

Keyboard Shortcuts

Keyboard Shortcuts