Atom Packages Directory

a package directory for a text editor of the 21st Century

Linter Autocomplete Jing by Alexander Erhard (

Install with:
    apm install linter-autocomplete-jing


Travis CI Status Appveyor CI Status Circle CI Status Dependencies devDependencies Status

Autocomplete and on-the-fly validation of XML documents in Atom.

Supported schema types:

You can find a basic set of common XML schemata and corresponding rules for linter-autocomplete-jing in the xml-common-schemata package.

XML document processing is handled in Java using Jing, Xerces and Saxon HE. For the source code of the Java part see


Select the package in Atom’s Settings View (Ctrl-,) / Install Packages or run apm install linter-autocomplete-jing from the command line.

The package depends on a Java Runtime Environment (JRE) v1.6 or above. If running java -version on the command line returns an appropriate version number, you should be set. Otherwise, install a recent JRE and provide the path to the Java executable either in the PATH environment variable or in the linter-autocomplete-jing package settings in Atom.

Package Settings

Settings Form Atom Config Property Description
Java Executable Path javaExecutablePath The path to the Java executable (java).
JVM Arguments jvmArguments Space-separated list of arguments to get passed to the Java Virtual Machine on which the validation server is run.
Schema Cache Size schemaCacheSize The maximum number of schemata retained simultaneously in memory.
Display Schema Parser Warnings displaySchemaWarnings Whether or not to display warning messages from the schema parser.
XML Catalog xmlCatalog The path to the XML Catalog file to be used in validation.
DTD Validation dtdValidation Determines under which circumstances DTDs should be used in validation. Possible values: ‘always’, ‘never’ or ‘only as fallback’. When ‘only as fallback’ is selected, documents get validated against DTDs only if no other schemata are available for validation.
Autocomplete Priority autocompletePriority The inclusion priority of the Autocomplete Plus provider. In order to exclude other autocomplete providers, the number must be larger than the other providers’ priorities. Defaults to 1 (In order to suppresses the default tag snippets provided by the language-xml package, set autocomplete priority to 2.)
Autocomplete Scope autocompleteScope The schema types which should be used for autocomplete.
Wildcard Suggestions wildcardSuggestions Inclusion of wildcards in autocomplete suggestions. Possible values: ‘all’, ‘localparts’, ‘none’.

(In order to edit the settings, press Ctrl-, or select “Packages” / “Settings View” / “Open” in the main menu. Then either navigate to the settings form (“Packages” tab, search for “linter-autocomplete-jing” and click the “Settings” button) or click the link to Atom’s config file.

File types

Validation and autocomplete get activated when the current file’s grammar scope includes an item starting with text.xml. The Atom core package language-xml (installed by default) assigns a large set of common XML file extensions to text.xml and text.xml.xsl. XML property lists (text.xml.plist) are supported by another core package, language-property-list.

If a file extension you’re working with isn’t included in language-xml or any grammar derived from text.xml you can either request adding the extension at or create a local association in your config.cson:

  1. Open config.cson by pressing Ctrl-Shift-P and then entering Open Your Config
  2. Add or extend the customFileTypes property of core

The following example assigns the .tei, .mei and .odd extensions to text.xml:

core: customFileTypes: "text.xml": [ "tei" "mei" "odd" ]

A second way of supporting custom file extensions is creating a new grammar package based on text.xml and specifying the extensions in the grammar definition. An example can be found in the demo package.

Specifying Schemata

Schema References in Source Documents

You can specify schemata by providing a DOCTYPE declaration, XSI schema instance attributes or xml-model processing instructions; see for a collection of examples).
If your documents contain references to remote schemata, you can improve performance and reduce network traffic by using an XML catalog file to map remote resources to local files. Projects like the Text Encoding Initiative (TEI) provide packages which include catalog files that can be added to the linter settings.

Schema Rules

If you’d like to avoid providing schema hints in each individual source document you can specify ‘validation rules’ in Atom’s config.cson or in a separate Atom package.

Here are two demo rules, the first of which associates files ending with .rng with an RNG schema at /path/to/relaxng.rng, the second associating the root element namespace urn:oasis:names:tc:entity:xmlns:xml:catalog with the RNC schema at /path/to/catalog.rnc:

"linter-autocomplete-jing": rules: [ { priority: 1 test: pathRegex: "\\.rng$" outcome: schemaProps: [ { lang: "rng" path: "/path/to/relaxng.rng" } ] } { priority: 1 test: rootNs: "urn:oasis:names:tc:entity:xmlns:xml:catalog" outcome: schemaProps: [ { lang: "rnc" path: "/path/to/catalog.rnc" } ] } ]

Each rule must contain a test, an outcome and a priority property. test contains the criteria which need to be fulfilled for that rule to apply. outcome contains the schema / catalog information to be used when the rule is matched. When there are multiple rules, the outcome of the first matched rule gets applied (order is significant). Rules with higher priority get evaluated first, rules with the same priority get evaluated in the order in which they are specified. Rules in config.cson always take precedence over rules from packages.

Possible properties of test:

Possible properties of outcome:

For an example of an Atom package containing a schema and schema association rules see demo package.



If you intend to adjust only the JavaScript part of the code, uninstall then linter-autocomplete-jing package in Atom, checkout from GitHub and link it to Atom by running apm link from the repository root (undo with apm unlink). The ES6 source code is in the src folder of the project. npm run build and npm run watch transpile the content of src to a single CoffeeScript file in lib which gets exposed to Atom. Both commands depend on Rollup (npm i rollup -g). Press Ctrl-Alt-R in Atom to reload the updated module.

In order to make adjustments to the JavaScript as well as as the Java parts, checkout and follow the instructions in the README file.

Keywords: xml, relaxng, schematron, xsd, validation, autocomplete Suggest keywords
Fork me on GitHub