Types for the unified-latex
Abstract Syntax Tree (AST). These types extend the unist
AST,
but instead of a children
attribute, they have a content
attribute.
If you're working with unified-latex
ASTs.
npm install @unified-latex/unified-latex-types
This package contains both esm and commonjs exports. To explicitly access the esm export,
import the .js
file. To explicitly access the commonjs export, import the .cjs
file.
export type ArgumentParser = (
nodes: Ast.Node[],
startPos: number
) => { args: Ast.Argument[]; nodesRemoved: number };
export type Ast = Node | Argument | Node[];
export type EnvInfo = {
renderInfo?: {
/**
* Whether the body of the environment should be treated as math mode
*
* @type {boolean}
*/
inMathMode?: boolean;
/**
* Whether to align the environment contents based on `&` and `\\` delimiters
* (like a matrix or tabular environment).
*
* @type {boolean}
*/
alignContent?: boolean;
/**
* Whether the arguments should be treated as pgfkeys-type arguments.
*
* @type {boolean}
*/
pgfkeysArgs?: boolean;
/**
* A list of names to be given to each argument when processing. This list should
* be the same length as the number of arguments. `null` can appear any number of times
* for "un-named" arguments.
*
* This allows a consistent reference to macro arguments even if the macro signatures are redefined between
* packages.
*
* @type {((string|null)[])}
*/
namedArguments?: (string | null)[];
/**
* Whether the body is tikz-environment like (e.g., a `tikzpicture` or `scope`, etc.)
*
* @type {boolean}
*/
tikzEnvironment?: boolean;
};
/**
* Function to process the body of an environment. The return value of `processContent`
* is treated as the new body.
*
*/
processContent?: (ast: Ast.Node[]) => Ast.Node[];
/**
* The environment signature as an xparse argument specification string.
*
* @type {string}
*/
signature?: string;
};
export type EnvInfoRecord = Record<string, EnvInfo>;
export type GenericAst = GenericNode | GenericNode[];
export type MacroInfo = {
renderInfo?: {
/**
* Whether the macro's contents wraps along with the current
* paragraph or displays as it's own block.
*
* @type {boolean}
*/
inParMode?: boolean;
/**
* Whether the arguments should be processed as pgfkeys-type arguments.
*
* @type {boolean}
*/
pgfkeysArgs?: boolean;
/**
* Whether there should be line breaks after the macro
* (e.g., like the `\\` command.)
*
* @type {boolean}
*/
breakAfter?: boolean;
/**
* Whether there should be line breaks before and after the macro
* (e.g., like the `\section{...}` command.)
*
* @type {boolean}
*/
breakAround?: boolean;
/**
* Whether there should be line breaks before the macro.
*
* @type {boolean}
*/
breakBefore?: boolean;
/**
* Whether the contents of the macro should be assumed to be in math mode.
*
* @type {boolean}
*/
inMathMode?: boolean;
/**
* Whether the arguments should be rendered with a hanging indent when the wrap
* (like the arguments to \item in an enumerate environment.)
*
* @type {boolean}
*/
hangingIndent?: boolean;
/**
* A list of names to be given to each argument when processing. This list should
* be the same length as the number of arguments. `null` can appear any number of times
* for "un-named" arguments.
*
* This allows a consistent reference to macro arguments even if the macro signatures are redefined between
* packages.
*
* @type {((string|null)[])}
*/
namedArguments?: (string | null)[];
/**
* Whether the macro represents a tikz path command (e.g. `\draw (0,0) -- (1,1);`).
*
* @type {boolean}
*/
tikzPathCommand?: boolean;
/**
* If `\sysdelims` is present, this contains the global information about the delimiters.
*/
sysdelims?: (Ast.Node[] | null)[];
};
/**
* The macro signature as an xparse argument specification string.
*
* @type {string}
*/
signature?: string;
/**
* Some special macros like `^` and `_` don't use
* an escape token. When matching against these macros,
* care must be taken to match the macro contents and the macro's
* escape token.
*/
escapeToken?: string;
/**
* Custom argument parser. If present, function overrides the default argument
* parsing of `signature`. An array of nodes is passed as well as the position
* of the first node **after** the macro. This function is expected to _mutate_
* the input array, removing any nodes that are part of the macro's argument.
*
* This function will only be called on a macro if it has no existing `args`.
*
* Note: for stability when printing/accessing a node's arguments, this function
* should always return an argument array of the same length, regardless of
* whether optional arguments are present. For example, if this function parses
* a node with signature `o m`, it should ways return a length-two array of arguments.
* A "blank" argument (one that does not show up during printing) can be created
* with `args([], { openMark: "", closeMark: "" })`, using the `unified-latex-builder` package.
*/
argumentParser?: ArgumentParser;
};
export type MacroInfoRecord = Record<string, MacroInfo>;
export type Node =
| Root
| String
| Whitespace
| Parbreak
| Comment
| Macro
| Environment
| VerbatimEnvironment
| InlineMath
| DisplayMath
| Group
| Verb;