rtdt99 opened a new issue, #7141:
URL: https://github.com/apache/hudi/issues/7141

   **_Tips before filing an issue_**
   
   - Have you gone through our [FAQs](https://hudi.apache.org/learn/faq/)?
   
   - Join the mailing list to engage in conversations and get faster support at 
[email protected].
   
   - If you have triaged this as a bug, then file an 
[issue](https://issues.apache.org/jira/projects/HUDI/issues) directly.
   
   **Describe the problem you faced**
   
   We have an existing parquet table and used DeltaStreamer to bootstrap the 
table with METADATA_ONLY and COPY_ON_WRITE option. I was able to successfully 
bootstrap it. I then performed few inserts and upserts to the hudi table. I am 
able to see the updates when I query the hudi table.
   The issue I'm seeing is that the original/source parquet file that I 
bootstrapped looks unchanged. What I mean here is that, for example, initially 
the original/source parquet file has 20 rows. After bootstrapping and 
converting it to a hudi table and performing few inserts and upserts added 5 
more rows. Total rows now that I can query from hudi table snapshot is 25 rows. 
So far looks good.
   But if I look at the original/source parquet file, it contains the same 20 
rows it had before. I was expecting that it would replace the existing source 
parquet with all the updates that took place and generate a new parquet file 
reflecting all updates.
   
   If we have any downstream consumers of the parquet file(who are not using 
hudi clients for querying) at source location, then what is the right way to 
propagate these hudi updates back to parquet file for downstream consumers so 
that they can see most up to date data.
   
   **To Reproduce**
   
   Steps to reproduce the behavior:
   
   1. Use DeltaStreamer to bootstrap an existing parquet table with 
METADATA_ONLY and COPY_ON_WRITE options
   2. Do few inserts to add few rows to the hudi table
   3. We will notice that the original/source parquet file is not 
updated/replaced.
   
   
   **Expected behavior**
   Hudi updates/replaces the original/source parquet file with all updates. 
   
   
   **Environment Description**
   
   * Hudi version : 0.12.0
   
   * Spark version : 3.3
   
   * Hive version :
   
   * Hadoop version :
   
   * Storage (HDFS/S3/GCS..) : Local Macbook Pro
   
   * Running on Docker? (yes/no) : No
   
   
   **Additional context**
   
   What is the right way to periodically export hudi tables as parquet format 
files for other external consumers who are not using hudi clients for querying ?
   Can we do incremental export of hudi tables to parquet format files ?
   
   **Stacktrace**
   
   ```Add the stacktrace of the error.```
   
   


-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: [email protected]

For queries about this service, please contact Infrastructure at:
[email protected]

Reply via email to