it looks like the singer portion of meltano is foc...
# announcements
l
it looks like the singer portion of meltano is focused on REST APIs - what about processing flat files or Excel files with merged headers, skipped rows, etc?
v
Hi Lance, do you mind creating an issue in the meltano project https://gitlab.com/meltano/meltano/issues and we will prioritize accordingly?
We would also be happy to get on a call and discuss the use cases for you and what you could get from the project
Would you be open to that? (cc: @danielle_morrill)
Currently you can extract from google sheets + APIs only really
y
@lance_dacey We are currently working on a Tutorial on how to support this use case 🙂 As long as your excel files have a first line header with the "attribute names", this is currently supported in Meltano: 1. Export your Excel files as CSV 2. Use
tap-csv
to import the files to your DB. 3. Use a custom transform to clean the data, remove unwanted rows, etc
You can check our draft, work in progress tutorial if you want to get an idea on how this works: https://gitlab.com/meltano/meltano/blob/210ecd8b7816a45d4e09b2d4ff03a2129c0b8fa0/docs/src/docs/tutorial-tap-csv.md And the
Custom Transformations and Models
Tutorial on how the raw data can be transformed: https://www.meltano.com/docs/tutorial.html#advanced-adding-custom-transformations-and-models
Meltano follows the ELT paradigm, i.e. you can transform the data if you want beforehand, but you can also Extract the raw data with minimum transformations (to remove for example sensitive data or columns with PII), Load the raw data to a RAW schema (similar to a data lake) and then clean, Transform and move the data to a final analytics schema. We use dbt for that last part.
l
ah - cool, I didn't see the tap csv option. right, I need to sometimes format the files (wrong encoding, merged headers or many sheets, lots of skipped rows or summary data) so I generally try to create a standard CSV
and it looks like tap csv would work after that point
y
Exactly 🙂