Rhema

XML-based markdown format for tracking translation/internationalization files. Aims to unify translations into a more singular code base. From the rhema project/files, language files can be exported for JSON/JavaScript based, iOS/Mac .strings, and Android XML strings.xml.

MASSIVELY UNDER WORK, NOT ALL FEATURES IMPLEMENTED

Why, or the use-case

When building an app/project for cross-platform deployment, tracking down and keeping the translations in order can be a burden. Especially if they're all in different formats. It's easy enough for a developer to read JSON or XML files, but getting someone outside this community to work with them can be less ideal. So I came up with this. Heavily inspired by the XLIFF 2.0 format, but with a bit more of the "machine" feel taken out. Instead of a single file spec, I've gone for two, which is just a project file to join together sub-files for translations. With this project, I can generate and export the needed files for which-ever platform I'm building for. And by keeping them together it becomes a one-stop-shop for the project.

Pros

Single-format exports to many other formats (XML, JSON, Properties, XLIFF(maybe) )
Tracking a "base language" file fills in the blanks and notifies of errors
Use's IDs for entries to minify the lookup
Supports groups and sub-groups for organizing
Platform switch can add/remove keys from being exported per the platform they target
Dataset is provided in the file, so variable and template replacement is easier
Dataset provides helpful hints to what it is, ie. Number or Text; with support for grammar type (expect a proper-noun for instance).
Dataset variables will be converted on export to their proper fashion. (ie. Android variables become %1$s)
Context, Description, and Note elements for explaining the meaning (acts as comments and not exported)

Example markdown

Check the examples folder for the files that reflect this

<?rhema version="1.0" encoding="UTF-8" ?>
<language xmlns="rhema.100" source="en">
    <notes>
        <note>Scrabble note here</note>
        <note op="chris" time="2018-01-01T12-00-00">Labeled note</note>
    </notes>
    <desc>Description of this object; if at root level, describes the file then</desc>
    <group id="general" platform="all">
        <entry id="hello_world" platform="ios,android">
            <dataset>
                <text id="name" sizeMax=128 grammar="proper-noun"  />
            </dataset>
            <source>Hello [name]</source>
        </entry>
    </group>
</language>

Exports to JSON as:

{
    "en": {
        "general": {
            "hello_world": "Hello ${name}"
        }
    }
}

Nodes, and Format

Each file is prefixed with the XML processor instruction: <?rhema version="1.0" ?>

Project file (*.RLP)

Element	Required?	Description
`project`	Y	Declares this is a project outline, must be first node
`files`	Y	Starts the list of `file` objects
`base`	Y	Declares the 'base' or master file, inherits from `file`, all attributes apply
`file`		Describes the locations and language of a file this project references

Translation file (*.RLF)

Element	Required?	Description
`language`	Y	Declares this is a language outline, must be first node
`group`		Group block of child groups or entries, requires an ID
`entry`		Specifies a specific language entry requires an ID
`source`	If no target	The source string, or one to be translated
`target`	If no source	The source string, translated into the target language
`dataset`		Groups the data entries for an entry, to help with variables
`text`		Data variable of a textual type, can declare it's grammar hint, requires ID
`number`		Data variable of a numeric type, requires ID
`notes`		A block of notes designed for developer<->translator communication
`desc`		Describes it's block, meant as a context explanation