Source maps

Choose the operation that matches what you want to recover or trace:

If you have a source map, try extracting the original files first. You may not need to decompile anything. If you only want to trace Wakaru's output, you can emit a map without having an original source map.

Recover the original sources

npx wakaru extract input.js.map -o src/

Many production maps embed the original files themselves, in a field called sourcesContent. wakaru extract writes those files straight to disk. When this works, you get the original source, not a reconstruction. Try it first whenever a map is available.

Recover names with an input map

npx wakaru input.js --source-map input.js.map -o output.js

Use an input map when you want to keep decompiling the code and improve its names. Wakaru can recover identifiers from names recorded in the map or from embedded source files.

This works in single-file mode only. --unpack rejects input maps because the split modules no longer line up with the bundle's coordinates.

Emit a map for the output

npx wakaru input.js --emit-source-map -o output.js
npx wakaru bundle.js --unpack --emit-source-map -o out/

--emit-source-map writes a .map next to each output file, mapping the decompiled code back to the input. Unlike input maps, this works with --unpack: each module's map points into the bundle.

Code that Wakaru adds, such as recovered import and export statements, has no position in the input and stays unmapped. Each map refers to its input by a path relative to the map file. Unpack maps leave the bundle's text out of the map, so keep the bundle where it was when you open the maps in a tool.

See the mapping visually

The playground has a Mapping toggle. It colors each line of the output and highlights the input it came from, so you can trace any decompiled line back to the original.

Hover any line below to see its counterpart on the other side. This is the live playground, so the code is editable: