XML Data Toolkit

English | 繁體中文

Understand and reuse XML data in VS Code: tree and table views, XPath, CSV/JSON export, formatting, plain-language error messages and Big5 encoding handling. No third-party dependencies.

Website: xmldatatoolkit.pylabs.app

Features

Tree and table views

• Switch views: click the tree icon in the editor title bar, press Cmd+Alt+T (Windows/Linux: Ctrl+Alt+T), or use the XML Data Toolkit ▸ context menu. Hold Option/Alt while clicking the icon to open the view to the side.
• Table: repeated elements (for example 1,000 <order> elements) become rows. Attributes (@id), child elements (customer) and deeper paths (item/name) become columns. Use the Data drop-down to switch to another group, such as the line items of every order.
  • Sort, search, filter by column, summary row with each column's type and distribution, and a detailed statistics panel.
  • Double-click a row or press Enter to select that record in the text editor next to it.
  • The table is read-only, so your XML is never changed by accident.
• Tree: each nesting level has its own color, with attributes, text and child counts on every node. Use the arrow keys to move around, and double-click or press Enter to jump to the source. The XPath of the selected node is shown at the bottom and can be copied.
• Default view: set xmlDataToolkit.defaultView to visual to always open .xml files in the tree and table view.

XPath

• Status bar: shows the XPath of the cursor position, such as /orders/order[2]/@id. Click it to copy.
• XPath Query (Cmd+Shift+Alt+X, Windows/Linux: Ctrl+Shift+Alt+X): highlights every match, lists them so you can step through with the arrow keys, or selects them all as multiple cursors. Expressions such as count(//order) or sum(//amount) show the result directly.
• Supports XPath 1.0: paths, predicates, all 12 axes, comparisons, arithmetic and the standard functions, plus ends-with, lower-case and upper-case.
• Element names are matched as written, so //order works even when the file declares a default namespace, and //x:note works for prefixed names without any setup.

Export

• Export to CSV: choose the data group and columns, then save as Excel-compatible UTF-8 with BOM or plain UTF-8. Exporting from the table's Tools ▾ menu keeps the current filter and sort order.
• Convert to JSON: the whole document, or one data group from the table. Attributes become @name, repeated elements always become arrays, and values are kept as text (so 00123 stays 00123).

Formatting and cleanup

• Pretty Print and Minify only change the whitespace between tags. Text, comments, CDATA and mixed content such as <p>Some <b>bold</b> text</p> are kept exactly as they are.
• Remove Comments: deletes every comment, including the empty lines they leave behind.
• Rainbow tags: tags are colored by nesting level in the text editor.

Errors in plain language

• Problems such as unclosed tags are underlined in the editor and listed in the Problems panel, for example: "<item> is not closed. Add </item> before </order>."

Encodings

• Opening a Big5-encoded file (or one in the encoding its declaration names) shows a prompt to reopen it correctly or convert it to UTF-8.
• If the content is UTF-8 but the declaration says something else, such as encoding="Big5", you are offered a one-click fix.
• Converting to UTF-8 also updates the <?xml encoding="..."?> declaration.

Settings

Supported languages: English, 繁體中文, 简体中文.