zeehaven's Introduction

zeehaven

a tiny tool to convert zeeschuimer ndjson files into csv format.

web interface available at: https://publicdatalab.github.io/zeehaven/

if ndjson file is collected from instagram or twitter the tool structures the csv file accordingly, otherwise it outputs the whole ndjson file as a csv (which may contain many columns).

if you have suggestions or encounter problems you can add an issue.

developed by iain. themed by jonathan.

zeehaven's People

Contributors

Stargazers

Watchers

zeehaven's Issues

Data processing step needed

For zeehaven to play a role equivalent to 4CAT when it comes to accessing zeeschuimer data in csv format, there is a step that is implemented in 4CAT and that would need to be implemented in zeehaven too to map the ndjson categories into readable and useful categories for the csv user.

Here are the scripts that 4CAT uses to convert zeeschuimer ndjson files into csvs:
instagram: https://github.com/digitalmethodsinitiative/4cat/blob/master/datasources/instagram/search_instagram.py
tiktok: https://github.com/digitalmethodsinitiative/4cat/blob/master/datasources/tiktok/search_tiktok.py

The 4CAT scripts for processing the other zeehaven data sources can be found here: https://github.com/digitalmethodsinitiative/4cat/tree/master/datasources

Would it be possible to implement these in zeehaven too?

add data processing for tiktok

as per this issue: #1

Is the tool down?

Currently nothing happens when dropping a file into the window. A couple of days ago the csv file was immediately returned.

Locations from tweets not displaying

Observed behaviour:
When I convert tweets to csv using the tool, all the entries in the "place_name" column are empty. If I check the original ndjson, I can see that the location data is present (data.core.user_results.result.legacy.location).

Expected behaviour:
Locations are displayed in the zeehaven csv output.

Recommend Projects