Processes a CSV file and outputs SQL insert statments.

First, place your CSV file in the folder.

Usage

mkdir input
echo '1, john@example.com' >> input/input.csv

cd input
docker run -v $(pwd):/input registry.gitlab.com/tomfordweb/csv-dumper:latest \
    -f=/input/input.csv  \
    -c=key,email:string  \
    -t=users 

This will create a bind mount in the current directory and move the files to the /input directory on the container. Any directory is okay as long as it is not where the app is installed (do not use the /app folder!).

Arguments

--input,-i

The input CSV data to process. It is important that this not contain any headers.

--columns, -c

The columns of the CSV file, you can also provide data types for the columns if they are known.

Providing datatypes should be done as much as possible and will provide a much better experience when working with large CSV files or ones with lots of bad data.

Each column should be seperated by a comma , Each datatype for the column is seperated by a colon :

Example:

--columns foo:str,bar:int,baz:float,fizz,buzz

The above argument will tell pandas that the column foo contains strings, the column bar contains integers, the column baz contains floats, and fizz and buzz are unknown.

Pandas will still atttempt to infer the datatype if you do not provide one, however it consumes more memory and time.

You can read more about dtypes here

--table, -t

The SQL table to dump into

--delimiter,-d

The delimter of the CSV, defaults to tab