What is @rushstack/terminal?
@rushstack/terminal is a library designed to provide a standardized way to handle terminal output in Node.js applications. It offers features for writing colored text, handling different log levels, and managing output streams in a consistent manner.
What are @rushstack/terminal's main functionalities?
Colored Text Output
This feature allows you to write colored text to the terminal. The `formatColor` method is used to apply color to the text.
const { Terminal } = require('@rushstack/terminal');
const terminal = new Terminal(process.stdout);
terminal.writeLine('This is a normal line.');
terminal.writeLine(terminal.formatColor('red', 'This is a red line.'));
terminal.writeLine(terminal.formatColor('green', 'This is a green line.'));
Log Levels
This feature allows you to write messages with different log levels such as verbose, warning, and error. This helps in categorizing the output based on its importance.
const { Terminal, ConsoleTerminalProvider } = require('@rushstack/terminal');
const terminalProvider = new ConsoleTerminalProvider();
const terminal = new Terminal(terminalProvider);
terminal.writeVerboseLine('This is a verbose message.');
terminal.writeWarningLine('This is a warning message.');
terminal.writeErrorLine('This is an error message.');
Output Streams
This feature allows you to manage output streams. In this example, the output is written to a string buffer instead of directly to the console, which can be useful for testing or logging purposes.
const { Terminal, StringBufferTerminalProvider } = require('@rushstack/terminal');
const stringBuffer = new StringBufferTerminalProvider();
const terminal = new Terminal(stringBuffer);
terminal.writeLine('This is a line in the buffer.');
console.log(stringBuffer.getOutput());
Other packages similar to @rushstack/terminal
chalk
Chalk is a popular library for styling terminal strings in Node.js. It offers a wide range of colors and styles, making it a versatile choice for terminal output. Compared to @rushstack/terminal, Chalk focuses more on styling text rather than providing a comprehensive terminal management solution.
winston
Winston is a versatile logging library for Node.js that supports multiple transports (e.g., console, file, HTTP). It offers features like log levels, custom formats, and more. While @rushstack/terminal focuses on terminal output, Winston provides a more comprehensive logging solution.
ora
Ora is a library for creating elegant terminal spinners. It is useful for indicating ongoing processes in the terminal. While @rushstack/terminal provides general terminal output management, Ora specializes in creating and managing spinners.
@rushstack/terminal
This library implements a system for processing human readable text that
will be output by console applications.
The design is based loosely on the WritableStream
and TransformStream
classes from
the system Streams API,
except that instead of asynchronous byte streams, the TerminalWritable
system synchronously transmits
human readable messages intended to be rendered on a text console or log file.
Consider a console application whose output may need to be processed in different ways
before finally being output. The conceptual block diagram might look like this:
[Terminal API]
|
V
[normalize newlines]
|
V
+----[splitter]-------+
| |
V V
[shell console] [remove ANSI colors]
|
V
[write to build.log]
The application uses the Terminal
API to print stdout
and stderr
messages, for example with standardized
formatting for errors and warnings, and ANSI escapes to make nice colors. Maybe it also includes text
received from external processes, whose newlines may be inconsistent. Ultimately we want to write the
output to the shell console and a build.log
file, but we don't want to put ANSI colors in the build log.
For the above example, [shell console]
and [write to build.log]
would be modeled as subclasses of
TerminalWritable
. The [normalize newlines]
and [remove ANSI colors]
steps are modeled as subclasses
of TerminalTransform
, because they output to a "destination" object. The [splitter]
would be
implemented using SplitterTransform
.
The stream of messages are {@link ITerminalChunk} objects, which can represent both stdout
and stderr
channels. The pipeline operates synchronously on each chunk, but by processing one chunk at a time,
it avoids storing the entire output in memory. This means that operations like [remove ANSI colors]
cannot be simple regular expressions -- they must be implemented as state machines (TextRewriter
subclasses)
capable of matching substrings that span multiple chunks.
API documentation for this package: https://rushstack.io/pages/api/terminal/