Get certified for free when you join Fabric Data Days 2026 and dive into Fabric, Power BI, SQL, AI, and other essential data skills.
Join nowTry your skills in the Power BI Dataviz World Championship! Round one ends June 26. Join now
Hello,
I am trying to append 30 tables, 20-100MB max each, in Dataflows. I am getting this error: Evaluation resulted in a stack overflow and cannot continue.
The tables are all similar, originating from JSON files.
They all look like this:
let
Source = SharePoint.Files(SharepointSource, [ApiVersion = 15]),
importedfile = Source{[Name="FILE_NAME.Json",#"Folder Path"=SharepointSource & SharepointFolder]}[Content],
importedJSON = Json.Document(importedfile),
#"Converted to Table" = Table.FromList(importedJSON, Splitter.SplitByNothing(), null, null, ExtraValues.Error),
// Convert the first record to a table to extract column names
FirstRecord = #"Converted to Table"{0}[Column1],
FirstRecordTable = Table.FromRecords({FirstRecord}),
ColumnNames = Table.ColumnNames(FirstRecordTable),
// Expand the table using dynamic column names
#"Expanded Table" = Table.ExpandRecordColumn(#"Converted to Table", "Column1", ColumnNames, ColumnNames)
in
#"Expanded Table"
and the append is this:
let
Source = Table.Combine({TABLE1, TABLE2, ETC...}),
TRANSFORMATIONS,
TRANSFORMATIONS,
TRANSFORMATIONS
in
RESULT
- Solutions i've tried:
Buffering each table before appending
During the evaluation, the virtual memory of Dataflow is 6GB. Does that mean that Dataflow doesn't support combining 30 tables TOTALING 1.5GB? That's a huge deal breaker.
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.