<labelid="sidebar-toggle"class="icon-button"for="sidebar-toggle-anchor"title="Toggle Table of Contents"aria-label="Toggle Table of Contents"aria-controls="sidebar">
<ahref="https://github.com/helix-editor/helix/edit/master/book/src/languages.md"title="Suggest an edit"aria-label="Suggest an edit">
<iid="git-edit-button"class="fa fa-edit"></i>
</a>
</div>
</div>
<divid="search-wrapper"class="hidden">
<formid="searchbar-outer"class="searchbar-outer">
<inputtype="search"id="searchbar"name="searchbar"placeholder="Search this book ..."aria-controls="searchresults-outer"aria-describedby="searchresults-header">
<tr><td><code>name</code></td><td>The name of the language</td></tr>
<tr><td><code>language-id</code></td><td>The language-id for language servers, checkout the table at <ahref="https://microsoft.github.io/language-server-protocol/specifications/lsp/3.17/specification/#textDocumentItem">TextDocumentItem</a> for the right id</td></tr>
<tr><td><code>scope</code></td><td>A string like <code>source.js</code> that identifies the language. Currently, we strive to match the scope names used by popular TextMate grammars and by the Linguist library. Usually <code>source.<name></code> or <code>text.<name></code> in case of markup languages</td></tr>
<tr><td><code>injection-regex</code></td><td>regex pattern that will be tested against a language name in order to determine whether this language should be used for a potential <ahref="https://tree-sitter.github.io/tree-sitter/syntax-highlighting#language-injection">language injection</a> site.</td></tr>
<tr><td><code>file-types</code></td><td>The filetypes of the language, for example <code>["yml", "yaml"]</code>. See the file-type detection section below.</td></tr>
<tr><td><code>shebangs</code></td><td>The interpreters from the shebang line, for example <code>["sh", "bash"]</code></td></tr>
<tr><td><code>roots</code></td><td>A set of marker files to look for when trying to find the workspace root. For example <code>Cargo.lock</code>, <code>yarn.lock</code></td></tr>
<tr><td><code>auto-format</code></td><td>Whether to autoformat this language when saving</td></tr>
<tr><td><code>diagnostic-severity</code></td><td>Minimal severity of diagnostic for it to be displayed. (Allowed values: <code>Error</code>, <code>Warning</code>, <code>Info</code>, <code>Hint</code>)</td></tr>
<tr><td><code>comment-tokens</code></td><td>The tokens to use as a comment token, either a single token <code>"//"</code> or an array <code>["//", "///", "//!"]</code> (the first token will be used for commenting). Also configurable as <code>comment-token</code> for backwards compatibility</td></tr>
<tr><td><code>block-comment-tokens</code></td><td>The start and end tokens for a multiline comment either an array or single table of <code>{ start = "/*", end = "*/"}</code>. The first set of tokens will be used for commenting, any pairs in the array can be uncommented</td></tr>
<tr><td><code>indent</code></td><td>The indent to use. Has sub keys <code>unit</code> (the text inserted into the document when indenting; usually set to N spaces or <code>"\t"</code> for tabs) and <code>tab-width</code> (the number of spaces rendered for a tab)</td></tr>
<tr><td><code>language-servers</code></td><td>The Language Servers used for this language. See below for more information in the section <ahref="#configuring-language-servers-for-a-language">Configuring Language Servers for a language</a></td></tr>
<tr><td><code>grammar</code></td><td>The tree-sitter grammar to use (defaults to the value of <code>name</code>)</td></tr>
<tr><td><code>formatter</code></td><td>The formatter for the language, it will take precedence over the lsp when defined. The formatter must be able to take the original file as input from stdin and write the formatted file to stdout</td></tr>
<tr><td><code>text-width</code></td><td>Maximum line length. Used for the <code>:reflow</code> command and soft-wrapping if <code>soft-wrap.wrap-at-text-width</code> is set, defaults to <code>editor.text-width</code></td></tr>
<tr><td><code>workspace-lsp-roots</code></td><td>Directories relative to the workspace root that are treated as LSP roots. Should only be set in <code>.helix/config.toml</code>. Overwrites the setting of the same name in <code>config.toml</code> if set.</td></tr>
<tr><td><code>persistent-diagnostic-sources</code></td><td>An array of LSP diagnostic sources assumed unchanged when the language server resends the same set of diagnostics. Helix can track the position for these diagnostics internally instead. Useful for diagnostics that are recomputed on save.</td></tr>
</tbody></table>
</div>
<h3id="file-type-detection-and-the-file-types-key"><aclass="header"href="#file-type-detection-and-the-file-types-key">File-type detection and the <code>file-types</code> key</a></h3>
<p>Helix determines which language configuration to use based on the <code>file-types</code> key
from the above section. <code>file-types</code> is a list of strings or tables, for
<p>When determining a language configuration to use, Helix searches the file-types
with the following priorities:</p>
<ol>
<li>Glob: values in <code>glob</code> tables are checked against the full path of the given
file. Globs are standard Unix-style path globs (e.g. the kind you use in Shell)
and can be used to match paths for a specific prefix, suffix, directory, etc.
In the above example, the <code>{ glob = "Makefile" }</code> config would match files
with the name <code>Makefile</code>, the <code>{ glob = ".git/config" }</code> config would match
<code>config</code> files in <code>.git</code> directories, and the <code>{ glob = ".github/workflows/*.yaml" }</code>
config would match any <code>yaml</code> files in <code>.github/workflow</code> directories. Note
that globs should always use the Unix path separator <code>/</code> even on Windows systems;
the matcher will automatically take the machine-specific separators into account.
If the glob isn't an absolute path or doesn't already start with a glob prefix,
<code>*/</code> will automatically be added to ensure it matches for any subdirectory.</li>
<li>Extension: if there are no glob matches, any <code>file-types</code> string that matches
the file extension of a given file wins. In the example above, the <code>"toml"</code>
config matches files like <code>Cargo.toml</code> or <code>languages.toml</code>.</li>
</ol>
<h2id="language-server-configuration"><aclass="header"href="#language-server-configuration">Language Server configuration</a></h2>
<p>Language servers are configured separately in the table <code>language-server</code> in the same file as the languages <code>languages.toml</code></p>
<tr><td><code>timeout</code></td><td>The maximum time a request to the language server may take, in seconds. Defaults to <code>20</code></td></tr>
<tr><td><code>environment</code></td><td>Any environment variables that will be used when starting the language server <code>{ "KEY1" = "Value1", "KEY2" = "Value2" }</code></td></tr>
<tr><td><code>required-root-patterns</code></td><td>A list of <code>glob</code> patterns to look for in the working directory. The language server is started if at least one of them is found.</td></tr>
</tbody></table>
</div>
<p>A <code>format</code> sub-table within <code>config</code> can be used to pass extra formatting options to
# pass format options according to https://github.com/typescript-language-server/typescript-language-server#workspacedidchangeconfiguration omitting the "[language].format." prefix.
<h3id="configuring-language-servers-for-a-language"><aclass="header"href="#configuring-language-servers-for-a-language">Configuring Language Servers for a language</a></h3>
<p>The <code>language-servers</code> attribute in a language tells helix which language servers are used for this language.</p>
<p>They have to be defined in the <code>[language-server]</code> table as described in the previous section.</p>
<p>Different languages can use the same language server instance, e.g. <code>typescript-language-server</code> is used for javascript, jsx, tsx and typescript by default.</p>
<p>Each requested LSP feature is prioritized in the order of the <code>language-servers</code> array.
For example, the first <code>goto-definition</code> supported language server (in this case <code>typescript-language-server</code>) will be taken for the relevant LSP request (command <code>goto_definition</code>).
The features <code>diagnostics</code>, <code>code-action</code>, <code>completion</code>, <code>document-symbols</code> and <code>workspace-symbols</code> are an exception to that rule, as they are working for all language servers at the same time and are merged together, if enabled for the language.
If no <code>except-features</code> or <code>only-features</code> is given, all features for the language server are enabled.
If a language server itself doesn't support a feature, the next language server array entry will be tried (and so on).</p>
<tr><td><code>git</code></td><td>A git remote URL from which the grammar should be cloned</td></tr>
<tr><td><code>rev</code></td><td>The revision (commit hash or tag) which should be fetched</td></tr>
<tr><td><code>subpath</code></td><td>A path within the grammar directory which should be built. Some grammar repositories host multiple grammars (for example <code>tree-sitter-typescript</code> and <code>tree-sitter-ocaml</code>) in subdirectories. This key is used to point <code>hx --grammar build</code> to the correct path for compilation. When omitted, the root of repository is used</td></tr>
