Skip to main content

Cell-level Options

Runme badget

When running code blocks or commands with Runme, you have two execution options: document-level or cell-level options.

The cell-level options allow you to run your Markdown file by cell rather than execute the entire file at once.

In this section, we will explain the various features of the cell-level option and show you how to run your Markdown file by cell.

Let’s dive in!

Different Types of Commands

Not all commands are equal, and expectations of how execution works can differ! For example, code blocks can be:

  1. File-watchers for compilers & bundlers should run as background tasks.
  2. Executors for Interactive and Non-interactive programs & scripts.
  3. Human-centric output: JSON, text, images, etc.
  4. Terminal visibility when no errors occur.

However, Runme makes running such commands easy. It has sophisticated features that ensure commands and code are executed inside your Markdown file. In this section, we will explore how to run commands and code in one click and some features that make running commands at the cell level easier with Runme.

How to Run a Cell In One Click

To run a cell block in your Markdown file:

  • Navigate to the specific cell you wish to run.
  • Click the Run icon on the left side of the code block.

cell-level

This action will run your command and return an executed output.

You can configure how your cell should run using the configuration options. In the next section, we will be explaining how to configure your cell using the different cell-level options in Runme.

Features of the Cell-Level Options

The cell-level option is designed with unique sub-features which makes it efficiently and gives users the flexibility to modify each cell to their preference. Some of these features include:

Configuration of Cell

Runme provides two distinct ways you can configure your cell.

  • Using the Configure button

To modify the cell execution behavior of any cell you choose, click Configure on the respective cell.

configure

A new view will open, allowing you to select “general” or “advanced” cell configuration.

general-page

  • Using the Markdown Syntax

If you feel more comfortable editing the Markdown file directly, you can do it using the following configuration options schema:

Let us consider this example:

```sh {"id":"01HPM33PJFH2Q9R5S123X0HCSC","interactive":"false"}
echo "hello world"

The above example uses the echo command to display the text hello world on the terminal. The context {"interactive":"false","name":"echo-hello-world"} indicates that the command is intended for non-interactive execution and has been labeled accordingly.

Another example you can look at is this

```sh { interactive=true name=example mimeType=text/plain closeTerminalOnSuccess=false background=false }echo hello world

To edit the Markdown syntax of your cell in Runme, click on the button indicated in the image below

double view

Your VS Code view will split into two. One side has Runme View, and the other has the raw Markdown view. On the Markdown view, locate the cell you want to configure and make the necessary configuration edit.

Sidenote

Take a look at more examples available inside the VS Code extension repo for a reference on how to apply these code block attributes in different use cases!

Specify Language in Blocks

Runme, just like most Markdown viewers, will work best when a script's language is contained inside fenced code blocks.

Runme provides two ways you can specify language in code blocks.

  • Using our Shebang Feature

Runme has an awesome Shebang feature allows users to specify each cell's programming language inside the Markdown file.

interpreter

The image above shows that our first code block is in Python, and the next is in bash script.

  • The use of Markdown Standard If possible, always specify the language according to the Markdown standard as illustrated below:
  ```sh {"id":"01HPMBVRXTRHSD52YVAX4K648P"}
echo "language identifier in fenced code block"

Warning!

Out of the box, Runme will leverage the Guesslang ML/AI model with a bias towards Shell to detect the languages for unidentified code blocks.

While this works well in a lot of cases, the accuracy is not perfect.

Handle long-running processes

It is common to use file-watcher-enabled compilers/bundlers (e.g., npm start devwatchexec, etc.) in the background during development. For any cell containing an instance of these commands, tick the background cell setting. This will prevent execution from permanently blocking the notebook UX.

background running task

Once ticked notice the "Background Task" label shows up in the cell status bar!

Default: false

Example

```sh {"background":"true","id":"01HPMBT8BF57MJ36JYDKAMZG1K"}
npm run watch

background

Cell's current working directory

In most cases, you should set the current working directory at the document level; however, you can also set it per cell.

Click on configure to change the current work directory cwd:

checking directory


```sh { cwd=.. "id":"01HP475WXX5PVV658023KQJFRE"}
npm run watch #relative path


```sh {"cwd":"/tmp","id":"01HPQ8F2307NFJR3WV1EJRJ0B8"}
echo "absolute path" > dummy_file

Warning!

Please note that if both cwd is set for doc-level and cell, they don't overwrite. They combine. For example, cwd: /tmp/dummy (doc) and cwd: .. (cell) will result in /tmp.

Interactive vs non-interactive cells

If a cell's commands do not require any input from a reader it might be a good fit to include the cell's output inside the notebook. This is useful if the resulting output could be useful as input in a downstream cell. This is what interactive=false is for, and it defaults to true.

interactive execution in vs code

Default: true

Example

openssl rand -base64 32

Sidenote

Please note that the Runme team is currently working on making output in both notebook & terminal default behavior.

Set environment variables

If a cell has exported variables, the user will be prompted to set these variables. This can be useful to have a parameterized cell while not needing to manually modify the cell.

prompt user in vscode

Example

export SLEEP_SECS="0"
sleep $SLEEP_SECS

There are two prompt modes:

  • Use a placeholder
  • A prompt message.

Here's the difference:

  • Using a Placeholder:

Using quotes or double-qoutes will prompt the use the confirm the placeholder value. This is convient when a valid default value is known ahead of time.

export PROJECT_ID="your-project-7f412a"
cli make-call --project-id $PROJECT_ID describe
  • Using a Prompt Message:

When you don't know the default value, you can instead use a prompt message to ask the user for the value. The key difference here is that the message does not act as value placeholder.

export PROJECT_ID=Enter a valid project ID
cli make-call --project-id $PROJECT_ID describe

Terminal visibility post-execution

A cell's execution terminal is auto-hidden unless it fails. This default behavior can be overwritten if keeping the terminal open is in the interest of the Runme notebook reader. Just untick closeTerminalOnSuccess (false).

Default: true

Example

  ```sh {"closeTerminalOnSuccess":"false","id":"01HPM7MC8MAJB2QCRVPVCN1FTT"}
docker ps | grep runme/demo:latest

Human-friendly output

Not all cells’ output is plain text. For example, you can have JSON, text, images, etc, all in your Markdown file.

Using the optional mimeType specifier, you can specify the expected output type. Runme Notebooks have a variety of renderers that will display them as human-friendly. The MIME type defaults to text/plain.

When the interactive setting is turned off, and the mimeType value is empty, Runme attempt to auto-detect the MIME Type. This is useful for automatically displaying a cell's output according to the renderers available in your VS Code installation. If you specify a specific MIME Type, it will be used instead of auto-detection.

Human-centric output

See in the reference page for the list of supported MIME types!

Terminal Row

On Runme outputs are saved in lines also known as rows. The number of lines or rows in which an output should be rendered is defined by a setting known as Terminal row. Terminal row allows you to set the number of rows with which your output should be displayed under a cell. Where the terminal row is not set or defined, Runme displayed the output in 10 rows by default.

How To Set Up Terminal Row

The terminal row is a Runme VS Code feature. Therefore, you can set up using your code editor. If you haven’t installed Runme on your VS Code yet, see our installation guide to install Runme in your VS Code editor.

To set up the terminal row on your code editor, follow the steps below:

  • In your .md file, click on the “Configure” button at the bottom left of the file.

terminalrow-1

We have provided a list of configuration settings to upgrade your experience using the Runme extension. You can configure how your Markdown is executed on your code editor using these settings.

  • Navigate to “Advanced” > terminalrow.

terminalrow-2

  • Lastly, set the number of rows you wish your output to be rendered in.

Unnamed vs Named cells

On Runme cells are unnamed by default. However, you can name a cell directly in your notebook. This will enable you to easily identify the cell using the provided cell name. On the cell you wish to name, simply click on the “Add Name” button on the cell.

add name

You will be directed to enter a preferred name for the cell.

cell-name

Where a cell is not named, a generated name is provided for each cell. We recommend you DO NOT name a cell when using the CLI mode, you can use the VS Code extension to name the file in the configuration settings.

These examples showcase improved ways you can use Runme in your documentation to enhance your workflow and overall documentation process.

If you feel more comfortable editing the Markdown file directly, you can do it by using the following configuration options schema:

```sh {"id":"01HPM33PJFH2Q9R5S123X0HCSC","interactive":"false"}
echo "hello world"

The above example uses the echo command to display the text "hello world" on the terminal. The context {"interactive":"false","name":"echo-hello-world"} indicates that the command is intended for non-interactive execution and has been labeled accordingly.


Try out the previous command

```sh {"id":"01HPMBXN8PNCMJ87Y1BGQ1NKN3"}
echo hello world

The entire configuration schema: as an example


```sh {"background":"false","closeTerminalOnSuccess":"false","id":"01HPM33PJFH2Q9R5SNAQX0HCSC","interactive":"true","name":"example"}
echo "hello world"

Sidenote

Take a look at more examples available inside the VS Code extension repo for a reference on how to apply these code block attributes in different use cases!

Exclude Cell from Run All

Every VS Code notebook allows users to run all available cells. This can be useful if you define a complete runbook in your Markdown file, allowing developers to click the Run All button to get set up and running.

However, sometimes certain cells should be excluded from this workflow. You can configure this behavior by clicking the excludeFromRunAll option.

Default: false

Example

 ```sh {"excludeFromRunAll":"true","id":"01HPM7NRQYMD5T06M4KS6S2DWG"}
# Do something optional here

Run All Cells by Category

If you have multiple workflows in a single Markdown file you can categorize them and allow your developers to run all cells by a certain category. To enable that you can add a category as a cell option. A cell can have one or multiple categories that are comma-separated.

Default: ""

Example

  ```sh {"category":"build","id":"01HPM7Q4JQ5HVJ5KS2FJ7Q31SQ"}
# Do something here