---
id: "csvdetective"
category: "detector"
url: "http://registry.apicrafter.io/tool/csvdetective"
---

# csv-detective (`csvdetective`)

Python package by the data.gouv.fr team that detects what the columns of a tabular file (CSV, Excel, ODS, Parquet) contain. For each column it combines checks of the values (regular expressions, casting, comparison with official lists) with hints from the header. Version 0.13.0 has 55 formats: generic ones such as dates, emails, URLs, UUIDs and WGS 84 or Lambert-93 coordinates, and many French ones such as SIREN, SIRET, INSEE commune and department codes, postal codes, UAI school codes, RNA association numbers, NAF activity codes, FANTOIR street codes and RNB building identifiers; several of them use the frformat library. udata-hydra runs it on the tabular files published on data.gouv.fr. Formats without a registry datatype (int, float, json, binary, canton names, socio-professional categories and historical association numbers) are not listed. Released under the MIT license.

Website: <https://github.com/datagouv/csv-detective>

## Supported datatypes (41)

- `address`
- `boolean`
- `country`
- `countrycode_alpha2`
- `countrycode_alpha3`
- `countrycode_num`
- `date`
- `datetime`
- `dayofweek`
- `easting`
- `email`
- `epcicode`
- `fr_commune`
- `fr_department`
- `fr_department_code`
- `fr_naf`
- `fr_region`
- `fr_region_code`
- `fr_rivoli`
- `fr_rna`
- `fr_rnb_id`
- `fr_uai`
- `frpostcode`
- `gender`
- `geojson`
- `geopoint`
- `inseecode`
- `latitude`
- `longitude`
- `money`
- `mongodbid`
- `month`
- `northing`
- `percentage`
- `phone`
- `siren`
- `siret`
- `url`
- `username`
- `uuid`
- `year`