AWS Glue crawler

Hi,

how the Crawler is extracting the metadata from the files in the buckets depends on a few factors that you can review in this knowledge base article.

For example if the files are in subfolders/prefixes and are very similar they could be detected as partitions. More information can be found here.

To have more precise feedback for your specific situation could you please provide an example of the structure in S3 and the tables that are and are not created by the crawler?

Thank you, and hope this helps