"The credentials provided for the Sharepoint source are invalid" when using custom functions

Hi dragonhood,

as of my knowledge Power BI Dataflows treat each distinct URL and each distinct connector function as a separate data source.

Maybe have a try by using exactly the same root reference in every query and function and ensure your sample file query uses the exact same site URL.
If this still does not work consider inside the transform function, to reference only the binary passed in.

Hope this helps!
Best regards!

Do you get the error while editing? If yes, at what step does it trigger? Or do you get it when you try to validate or refresh?

Can you confirm that the filepath used in the Source step for #"ISF reports - New" and #"Sample File (9)" are the same? These should be identical; otherwise, you are mixing data sources, which could cause issues (although, usually you'll get errors about forumla firewall rather credential errors).

Did you create this by going through the combine files UI? Or did you perhaps copy/paste from previous combine files queries? A bit tedious, but may be worth creating a new blank test dataflow where you build the query from scratch to confirm it's reproducible and not an anomaly.

Some quick answers to your "is this caused by" questions:

• Using Binary parameter for SharePoint?
  Maybe? This is ootb behavior of combine files UI, though, so I would assume it shouldn't so easily be breaking things.
  
• A function referencing a different authentication scope?
  Most likely, your issue has nothing to do with the function, which just does: 1) take any binary as input, 2) parse it as Excel, 3) drill into 'Charts' sheet, 4) promote headers. Nothing about it introduces a different data source that would complicate your authentication.
  
• A mismatch between the Dataflow’s linked connection and the function source?
  See above. There isn't really a concept of "function source" in this context - maybe if there were steps in the function that queried some other data source.
  

Hi dragonhood,

if you want to completely go crazy you can add the SharePoint folder as a shortcut in a Fabric Lakehouse and process the data from there. Find here the link to the docs:

https://learn.microsoft.com/en-us/fabric/onelake/create-onedrive-sharepoint-shortcut

Best regards!

Hi, dragonhood 

The "Invalid Credentials" error in Dataflows often happens because the Power Query engine in the Service struggles to validate credentials inside a separate "Custom Function" object.

The most robust way to fix this is to move the transformation logic directly into your main query (the "Inline" approach). This bypasses the need for the Invoke Custom Function step entirely.

Here is the solution based on your data structure:

Step 1: Create your Parameters In your Dataflow, create two text parameters to handle the connection dynamically:

1.            Name: SharepointSite

o             Type: Text

o             Current Value: https://yourdomain.sharepoint.com/sites/YourSite

o             (Enter the root URL of your SharePoint site here)

1.            Name: FolderPath

o             Type: Text

o             Current Value: Shared Documents/YourFolder/

o             (Note: Copy the exact value from the "Folder Path" column in the preview to ensure it matches)

Step 2: Use this M Code Create a blank query and paste the code below. This script connects, filters the specific folder using the parameters above, extracts the "Charts" sheet from the Excel files, and performs the cleanup/splitting logic without calling external functions.

let
  // 1. Load SharePoint files using the parameter
  Source = SharePoint.Files(SharepointSite, [ApiVersion = 15]),

  // 2. Keep only the required folder
  FilteredRows = Table.SelectRows(Source, each [Folder Path] = FolderPath),

  // 3. Exclude hidden files
  VisibleFiles = Table.SelectRows(FilteredRows, each [Attributes]?[Hidden]? <> true),

  // 4. Read Excel content directly (Inline transformation)
  ExcelContent = Table.AddColumn(VisibleFiles, "Excel", each Excel.Workbook([Content], null, true)),

  // 5. Extract specific sheet "Charts"
  ExtractCharts = Table.AddColumn(
    ExcelContent,
    "ChartsSheet",
    each Table.SelectRows([Excel], each [Item] = "Charts" and [Kind] = "Sheet"){0}[Data]
  ),

  // 6. Promote headers inside each nested table
  PromoteHeaders = Table.TransformColumns(
    ExtractCharts,
    {"ChartsSheet", each Table.PromoteHeaders(_, [PromoteAllScalars = true])}
  ),

  // 7. Rename file column to keep source traceability
  RenameSource = Table.RenameColumns(PromoteHeaders, {{"Name", "SourceName"}}),

  // 8. Expand data from the nested tables
  ExpandedData = Table.ExpandTableColumn(
    RenameSource,
    "ChartsSheet",
    Table.ColumnNames(PromoteHeaders{0}[ChartsSheet])
  ),

  // 9. Set Standard types
  ChangeType = Table.TransformColumnTypes(
    ExpandedData,
    {
      {"SourceName", type text},
      {"Period", type date},
      {"Total ISFs Filed", Int64.Type},
      {"Total On Time ISFs", Int64.Type},
      {"Total Late ISFs", Int64.Type},
      {"Percentage On Time", Percentage.Type},
      {"Percentage Late", Percentage.Type}
    }
  ),

  // 10. Remove unneeded columns
  RemoveExtraCols = Table.RemoveColumns(ChangeType, {"Column7", "Period_id"}),

  // 11. Filter only valid rows
  FilterValid = Table.SelectRows(
    RemoveExtraCols,
    each ([Percentage On Time] <> null and [Percentage Late] <> null)
  ),

  // 12. Duplicate SourceName for split logic
  DuplicateSource = Table.DuplicateColumn(FilterValid, "SourceName", "SourceName_Copy1"),

  // 13. Split importer info (Transition from Text to Number)
  SplitImporter = Table.SplitColumn(
    DuplicateSource,
    "SourceName_Copy1",
    Splitter.SplitTextByCharacterTransition({"c"}, each not List.Contains({"0" .. "9"}, _)),
    {"Importer", "ImporterCodePart1", "ImporterCodePart2"}
  ),

  // 14. Clean Importer Name
  CleanImporter = Table.ReplaceValue(
    SplitImporter,
    "ISF REPORT - ",
    "",
    Replacer.ReplaceText,
    {"Importer"}
  ),

  // 15. Merge importer code parts
  MergeImporterCode = Table.CombineColumns(
    CleanImporter,
    {"ImporterCodePart1", "ImporterCodePart2"},
    Combiner.CombineTextByDelimiter("", QuoteStyle.None),
    "ImporterCode"
  ),

  // 16. Clean file extension from code
  CleanImporterCode = Table.ReplaceValue(
    MergeImporterCode,
    ".xlsx",
    "",
    Replacer.ReplaceText,
    {"ImporterCode"}
  ),

  // 17. Reorder final columns
  FinalReorder = Table.ReorderColumns(
    CleanImporterCode,
    {
      "SourceName",
      "Importer",
      "ImporterCode",
      "Period",
      "Total ISFs Filed",
      "Total On Time ISFs",
      "Total Late ISFs",
      "Percentage On Time",
      "Percentage Late"
    }
  )
in
  FinalReorder

This approach is much more stable in Dataflows because all lineage is contained within a single query context.

Let me know if this works for you!

Hi dragonhood,

I would also take a moment to thank rodrigosan , Mauro89, MarkLaf  for actively participating in the community forum and for the solutions you’ve been sharing in the community forum. Your contributions make a real difference.
 

I wanted to check if you had the opportunity to review the information provided. Please feel free to contact us if you have any further questions.

Regards,
Community Support Team.