Google Apps Script跨脚本调用含dataframe-js的库时出现参数类型错误问题

This issue stems from context isolation in Google Apps Script (GAS) libraries and how dataframe-js handles type validation. Let’s break down what’s happening:

Root Cause #

When you set up script1 as a library for script2, the two scripts run in separate execution contexts. This creates a critical mismatch:

• The Array constructor in script1’s library context is not the same as the Array constructor in script2’s main context.
• dataframe-js relies on type checks like instanceof Array or custom validation that expects objects to originate from the same context as the library code.

When you pass your array [[1,2,3],['a','b','c']] from script2 to the library’s DataFrame constructor, the library’s type checker doesn’t recognize it as a valid Array. It sees the array as an instance of script2’s Array (not script1’s), so it throws the confusing error ArgumentTypeError: Array while expecting DataFrame | Array | Object.

When you paste dataframe-js directly into script2, everything works because the array and library code share the same execution context—type checks pass as intended.

Workaround Solutions #

Here are a few ways to fix this:

1. Rehydrate data in the library’s context
   Serialize and deserialize your data using the library’s JSON methods to create an array that belongs to the library’s context:

   function test(){ 
     var data = [[1,2,3],['a','b','c']] 
     // Convert data to a library-compatible array
     var libraryData = script1.JSON.parse(script1.JSON.stringify(data));
     var DataFrame = script1.dfjs.DataFrame 
     var df = new DataFrame(libraryData) 
   }
   

   The parsed array is created within the library’s context, so instanceof Array checks will return true for dataframe-js’s validation.

2. Adjust dataframe-js’s type checking logic
   If you can modify the dataframe-js code in script1, replace context-dependent checks (like value instanceof Array) with context-agnostic alternatives:

   • Use Array.isArray(value) (works across contexts)
   • Or use Object.prototype.toString.call(value) === '[object Array]' for reliable type detection

   Look for validation logic in the _construct or _build methods referenced in your error stack and update those checks.

3. Use inline code instead of a library
   As you already found, pasting dataframe-js directly into each script that needs it avoids context mismatch entirely. This is the simplest fix if you don’t need to share the library across many projects.

Why This Happens in GAS #

GAS libraries aren’t just code imports—they’re separate script projects running in isolated global scopes. This isolation prevents conflicts but also means native object constructors (like Array, Object) aren’t shared between the main script and library. Any type check relying on constructor identity (via instanceof) will fail when passing objects across this boundary.

内容的提问来源于stack exchange，提问作者jason