Failed to merge incompatible data types LongType and StringType

[tassiodahora]

Guys, good morning!

I am writing the results of a json in a delta table, only the json structure is not always the same, if the field does not list in the json it generates type incompatibility when I append

(dfbrzagend.write

 .format("delta")

 .mode("append")

 .option("inferSchema", "true")

 .option("path",brzpath)

 .option("schema",defaultschema)

 .saveAsTable(brzbdtable))

Failed to merge fields 'age_responsavelnotafiscalpallet' and 'age_responsavelnotafiscalpallet'. Failed to merge incompatible data types LongType and StringType

[Anonymous · Accepted solution]

Hi @Tássio Santos​ 

The delta table performs schema validation of every column, and the source dataframe column data types must match the column data types in the target table. If they don’t match, an exception is raised.

For reference-

https://docs.databricks.com/delta/delta-batch.html#schema-validation-1

you can cast the column explicitly before writing it to target table to avoid this

[Kaniz_Fatma]

Hi @Tássio Santos​ , We haven’t heard from you on the last response from @Chetan Kardekar​ , and I was checking back to see if you have a resolution yet. If you have any solution, please share it with the community as it can be helpful to others. Otherwise, we will respond with more details and try to help.

[ifun]

The following example shows changing a column type:

(spark.read.table(...)
  .withColumn("birthDate", col("birthDate").cast("date"))
  .write
  .mode("overwrite")
  .option("overwriteSchema", "true")
  .saveAsTable(...)
)

Details see https://docs.databricks.com/delta/update-schema.html