Introduction

Compose for Notepad++ is a plugin for Notepad++ which implements a compose key for entering characters that are not available on the keyboard. Like Notepad++, this software is released under the GNU General Public License (either version 3 of the License, or, at your option, any later version). Some original source code files which are not dependent on Notepad++ are released under the MIT (Expat) license: see individual files for details. Compose for Notepad++ was first released by Randall Joseph Fellmy in 2025; you can find the source code on GitHub.

This plugin uses JSON for Modern C++ by Niels Lohmann, which is released under the MIT license.

This documentation, copyright ©2025 by Randy Fellmy, is provided under the Creative Commons Attribution 4.0 license.

Purpose and use

A compose key is a key that indicates the following keys are to be interpreted in a special way. Usually an easy-to-remember sequence produces a character that’s not on the keyboard, such as Compose 'a for á or Compose --- for an em-dash. Compose for Notepad++ brings this facility to Notepad++ in a flexible and customizable way, without affecting any other software on your computer.

To install Compose for Notepad++, download the x86 or x64 zip file, depending on whether you're using 32-bit or 64-bit Notepad++, from the latest release. Unzip the file to a folder named Compose (the name must be exactly this, or Notepad++ will not load the plugin) and copy that folder into the plugins directory where Notepad++ is installed (usually C:\Program Files (x86)\Notepad++\plugins for 32-bit versions or C:\Program Files\Notepad++\plugins for 64-bit versions).

How it works

Compose does nothing until you press the key or key combination chosen as the Compose key; then it watches the characters you type next. When it recognizes a meaningful sequence it substitutes the characters defined for that sequence — usually entering a special character that isn’t available on your keyboard. Examples:

• Compose AE types Æ.
• Compose -L types £.
• Compose .-a types ǡ.

Pressing Compose twice does whatever that key or key combination did originally.

When enabled, Compose for Notepad++ takes effect anywhere you press Compose in Notepad++. You can use compose key sequences in dialogs such as Find and Replace, or even in plugin dialogs, as well as when editing text.

Menu items

There are four items on the Compose menu:

• Enabled toggles the compose function on or off. Initially the plugin is disabled; click this menu item to enable it. When there is no checkmark by this item, Compose removes the hook it uses to monitor your typing and, consequently, it has no effect on Notepad++ at all.

• Compose key... opens a dialog for choosing which key or combination of keys will act as the Compose key. This works like a typical “shortcut key” — you can hold down Alt, Ctrl and/or Shift when you set the key, and only the exact combination you use will trigger the compose function. Type the combination you want into the box, then click OK. When you set a key in this dialog, composing is Enabled if it wasn’t already.

  The default (if you check Enabled without setting a key in this dialog) is the Insert key. Since most people don’t use the unmodified Insert key very often (it toggles between insert and overtype modes), this will typically be an unobtrusive choice. Another good choice, if you don’t use it often, would be the Caps Lock key.

  You can choose almost any key on your keyboard as the compose key except the Shift, Ctrl, Alt, Windows, PrintScrn/SysReq and Pause keys, and you can use them alone or with any combination of Shift, Ctrl and Alt modifiers. (You can’t use Backspace, Enter or the space bar alone, but you can use them with a combination of the Shift, Ctrl and/or Alt keys.)

• User definitions file... opens a dialog that lets you select a user definitions file. This file can add new compose sequence definitions and/or replace some of the default definitions, allowing you to customize Compose as you choose.
• Help/About provides information about the version of Compose you are running, and allows you to view the change log, license and readme for the plugin or to open the help file for the version you are running.

Typical key sequences

Many accented letters can by typed by pressing the compose key, then a key that “looks like” the accent, then the letter:

Some common letter variants:

Punctuation:

Currency signs:

Symbols:

Arbitrary Unicode character:

Character keys, dead keys and non-character keys

Most of the keys on a keyboard simply type a character. The Shift key typically changes the character produced. Many keyboards other than US English include an AltGr key which changes the character in another way, and there may be dead keys that combine with the following key to produce an accented character. In all these cases, Compose for Notepad++ considers only the character produced. If a sequence uses an ASCII tilde (~), it doesn’t matter if you produce that with Shift+` (as on a US English keyboard) or with AltGr+2 (as on a French keyboard).

A few keys, like Delete, Home and the arrow keys, do not produce a character or combine with any other key to produce a character. These keys can also be part of sequences; at present, Compose considers only the base key and ignores any modifier keys. The arrow keys are used in some of the sequences supplied with the plugin; for example, Compose ↑ 1 produces ¹ (superscript numeral one). No non-character keys other than the arrow keys are used in the supplied sequences, but you can use them in user definitions.

Explicit and implicit sequences

Unlike typical compose key implementations, Compose for Notepad++ recognizes both explicit and implicit sequences. Explicit sequences are defined individually; a long list is included with the plugin, and you can add more in a user definitions file. They are straightforward: type Compose followed by this and you get that. Explicit sequences take priority over implicit sequences.

Implicit sequences follow rules, making it possible to recognize many more combinations than it would be practical to enumerate. For example, it’s possible to add multiple accents to a single letter, or to enter any Unicode character using the hexadecimal representation of its code point. While explicit sequences always end precisely when they are complete, implicit sequences sometimes don’t end until the next character you type can no longer be part of the sequence. For example, Compose a . generates ȧ, but Compose a . - generates ǡ. The sequence doesn’t “know when to stop” until you end it by typing a character that can’t be part of the implicit sequence (a character that doesn’t represent an accent); by pressing the Enter key, which always ends an implicit sequence; or by pressing Compose, which ends the current sequence and begins a new one.

Explicit sequences

The explicit sequences supplied with Compose for Notepad++ can be grouped into several categories:

• Spacing diacritics (full list)
• Quotation marks (full list)
• Superscripts and subscripts (full list)
• Fractions (full list)
• Latin letters with strokes (full list)
• Circled numbers, letters and mathematical signs (full list)
• Latin ligatures, digraphs and extended letters (full list)
• Currency signs (full list)
• Greek alphabet (full list)
• Spaces (full list)
• Arrows (full list)
• Other symbols (full list)
• HTML named character references (full list)

Spacing diacritics (full list)

The spacing diacritics include non-combining versions of most of the diacritical marks that can be used in implicit sequences. The character that represents each mark is preceded or followed by a space. For example, either Compose space ' or Compose ' space inserts ´, the spacing acute accent.

Quotation marks (full list)

Type double angle quotation marks as Compose << for « and Compose >> for ». Other kinds of quotation marks use ' for single, " for double or . for single-angle, combined (in either order) with < for left, > for right or , for low-9. So Compose "> yields ” (right double quotation mark).

Superscripts and subscripts (full list)

An up or down arrow key before a number, a parenthesis or a plus, minus or equals sign indicates a superscript or subscript. So you can use Compose ↑ 2 to type ² (superscript two).

Fractions (full list)

Unicode only defines a few fractions as separate characters. You can type them as number / number; for example, Compose 3/8 produces ⅜.

Latin letters with strokes (full list)

To avoid conflict with use of the hypen for the macron and in some currency signs, “Latin letter with stroke” characters are generally typed with the forward slash (/) regardless of whether the stroke is horizonal or diagonal. Upper and lower case “T” are exceptions, as they have both horizontal (Compose -T yields Ŧ) and diagonal (Compose /T yields Ⱦ) stroke forms. The slash (or hyphen, for “T”) and the letter can be typed in either order.

Circled numbers, letters and mathematical signs (full list)

Circled numbers from zero to fifty, circled upper and lower case ASCII letters, and a few circled mathematical symbols can by typed by enclosing them in parentheses: Compose (12) produces ⑫.

Latin ligatures, digraphs and extended letters (full list)

A few ligatures, digraphs and extended latin characters are defined, including:

Currency signs (full list)

Most currency signs are typed with a hypen (-), an equals sign (=) or a vertical bar (|) before or after a letter:

Greek alphabet (full list)

Letters from the Greek alphabet result from typing Compose g followed by an upper or lower case letter corresponding to the position of the desired letter on the standard Greek keyboard. Most of the letters are where you would expect: Compose gp gives π, Compose gS gives Σ, and so on.

Spaces (full list)

A number of special-purpose spaces have built-in definitions, including:

Arrows (full list)

Type Compose followed by a space and an arrow key (in either order) to produce an arrow graphic; type Compose followed by the same arrow key twice to produce an arrow with a double line.

Remaining ISO-8859-1 symbols (full list), Remaining Windows-1252 symbols (full list), and More symbols (full list)

Various symbols, including all the printable characters in ISO-8859-1 and Windows-1252 not otherwise represented, have built-in definitions. Examples include:

HTML named character references (full list)

All HTML named character references are defined as explicit sequences (with the qualification that the closing semi-colon is required); so you can type Compose ℞ to get ℞.

Implicit sequences

Implicit sequences implement combining marks (accents) and entry of arbitrary Unicode characters. Implicit sequences take effect only when what was typed after the Compose key doesn’t match any explicit sequence. Explicit sequences always take priority. Occasionally this can be important to remember, since some of the characters used in implicit sequences are also used in some explicit sequences.

Implicit sequences using combining marks

Combining marks are additions to characters, such as an acute accent or a tilde. An implicit sequence can begin with the Compose key followed by one or more characters that represent combining marks and end with a character that is not a combining mark. It is also possible to begin with the Compose key and a base character (one that is not a combining mark), then add combining marks following. When the combining marks follow the base character, the sequence ends when you press the Enter key or the Compose key, or when you type a character that doesn’t represent a combining mark.

When Compose for Notepad++ recognizes an implicit combining sequence, it places the combining characters, in the same order you typed them, after the base character and converts that string to Unicode normalization form C (canonical composition). In practice, this does what you would expect: as few characters as possible — usually just one for common base and mark combinations — are used to represent what you indicated.

These are the implicit combining marks included in the default sequence definition file:

The macron presents an occasional annoyance because the character which represents it, the ASCII hyphen/minus, is also used in a number of explicit sequences. Macrons usually appear over vowels or y; there are no conflicts between these uses and the explicit sequences supplied with the plugin. In the unusual case of a macron over a consonant, conflict can be avoided by typing the letter first, then the up arrow, then the hyphen/minus; for example, Compose n ↑ - will produce n̄.

Implicit sequences for arbitrary Unicode characters

Compose for Notepad++ provides three ways to enter a Unicode character by its code point number. Two use the same format as HTML numbered entities: Compose &#N; enters a Unicode character by its code point expressed as a decimal number, while Compose &#xN; takes the number as hexadecimal.

You can also just type Compose followed by one to six hexadecimal digits that specify the Unicode code point — if there isn’t an explicit sequence that will take priority. For example, Compose ae is explicitly defined to produce æ; so to get the ® symbol by its code point you would need to type Compose 0ae. There are no pre-defined explicit sequences that begin with a decimal digit followed by a hexadecimal digit; so adding a zero before any hexadecimal code that begins with a letter will always work.

Sequences consisting of only hexadecimal digits end when an additional digit would cause the number to exceed the maximum Unicode code point (10FFFF), when you press the Enter key or the Compose key, or when you type a character that isn’t a hexadecimal digit.

User definitions

While it is possible to replace the built-in definitions file (compose-default.jsonc in the plugins\Compose folder where Notepad++ is installed), doing that is not recommended, because the replacement file would be overwritten whenever Compose for Notepad++ is updated. Instead, the menu option User definitions file... lets you designate a file that is merged with the default file to add your own explicit sequences.

You can stop using a user definitions file by clicking None in that dialog. If you are editing your file, after saving it you can cause Compose for Notepad++ to re-read the file and apply your changes by selecting User definitions file... and opening your file again.

User definition files are JSONC (JSON with comments) files. The file encoding must be UTF-8. When you choose a file in the User definitions file dialog, Compose for Notepad++ will tell you if the file is not valid JSONC, but no facilities for diagnostic error reporting are implemented or planned.

A simple user definitions file might look like this:

{
"CC"   : "\u2102",   // double-struck capital C
"rx"   : "℞",        // prescription symbol
"elr"  : "\u2107",   // Euler constant
"plnk" : "\u210E",   // Planck constant
"wf"   : "1060 W Addison St\r\nChicago, IL 60613"
}

After saving that file and selecting it in the User definitions file dialog, you can type Compose CC for ℂ, Compose rx for ℞, Compose elr for ℇ, Compose plnk for ℎ and Compose wf for the address of Wrigley Field.

When constructing your own definitions, it’s important to know that the first sequence to match hides any longer sequences. If you defined Compose gem as a sequence it could never be triggered, because Compose ge is already a sequence. (It types the Greek small letter epsilon, ε.)

But what if you really, really wanted sequences like gem and gel? When user definitions are merged with the built-in definitions, user definitions replace built-in definitions of the same sequence; but replacing with a different sequence wouldn’t resolve this problem. You can remove an existing definition by setting it to null:

{
"ge"  : null,
"gel" : "\u224B",
"gem" : "💎"
}

will let you type Compose gel for ≋ and Compose gem for 💎.

Sequences can use keys that don’t produce a character. These appear in sequence definitions as a key name enclosed in square brackets. The arrow keys are used in built-in sequences, so their names are fixed as [Up], [Down], [Left] and [Right]. The remaining keys, like Page Up or Scroll Lock, can also be used in your own sequences, but their names might vary depending on your locale. It’s easy enough to find out what they are: just type the Compose key followed by a non-character key and you’ll see the name typed immediately, since that key won’t be part of any built-in sequence.

It’s possible to change the rules for implicit combining character sequences, too; but if you want to do that, you’re on your own to look at the beginning of the built-in definitions file and try to figure it out for yourself.

List of pre-defined explicit sequences

• Spacing diacritics
• Quotation marks
• Superscripts and subscripts
• Fractions
• Latin letters with strokes
• Circled numbers, letters and mathematical signs
• Latin ligatures, digraphs and extended letters
• Currency signs
• Greek alphabet
• Spaces
• Arrows
• Remaining ISO-8859-1 symbols
• Remaining Windows-1252 symbols
• More symbols
• HTML named character references

Spacing diacritics

Quotation marks

Superscripts and subscripts

Fractions

Latin letters with strokes

Circled numbers, letters and mathematical signs

Latin ligatures, digraphs and extended letters

Currency signs

Greek alphabet

Spaces

Arrows

Remaining ISO-8859-1 symbols

Remaining Windows-1252 symbols

More symbols

HTML named character references