This script will read an exported UpLead csv of leads and grab all the relevant columns for the Customer Illuminated Airtable template.
$ ./translate_uplead_to_ci.py -h
usage: translate_uplead_to_ci.py [-h] -i INPUT -o OUTPUT [-r ROOT]
optional arguments:
-h, --help show this help message and exit
-i INPUT, --input INPUT
input filename
-o OUTPUT, --output OUTPUT
output filename
-r ROOT, --root ROOT root filepath to read / write the files from
It has a dependency on pandas, so feel free to pull the docker image to run the script.
The docker image will look for the input file (-i
) in the /data/
directory in its container, so mount a folder with your exported leads to /data
. The processed file (filename denoted by -o
) will also be written into that mounted directory.
Usage:
$ ls
UpLead_05-09-2019.csv
$ docker pull 10forward/uplead-ci-airtable
$ docker run -v $(pwd):/data 10forward/uplead-ci-airtable -i UpLead_05-09-2019.csv -o airtable-ci.csv
exported translated file to /airtable-ci.csv
$ ls -l
airtable-ci.csv
UpLead_05-09-2019.csv