mirror of
https://github.com/ilri/csv-metadata-quality.git
synced 2025-02-28 12:38:45 +01:00
README.md: Add note about Pandas
This commit is contained in:
parent
cf6c01caaf
commit
a8a41d60b6
@ -1,14 +1,14 @@
|
||||
# CSV Metadata Quality [](https://builds.sr.ht/~alanorth/csv-metadata-quality?)
|
||||
A simple but opinionated metadata quality checker and fixer designed to work with CSVs in the DSpace ecosystem. Supports multi-value fields using the standard DSpace value separator ("||"). Despite the name it does support reading Excel files.
|
||||
|
||||
Written and tested using Python 3.7.
|
||||
Written and tested using Python 3.7. CSV and Excel support comes from the [Pandas](https://pandas.pydata.org/) library.
|
||||
|
||||
## Checks
|
||||
Supports checking the validity of the following metadata elements:
|
||||
|
||||
- ISSN
|
||||
- ISBN
|
||||
- Multi-value separators
|
||||
- Multi-value separators ("||")
|
||||
- Dates
|
||||
|
||||
## Fixes
|
||||
|
Loading…
x
Reference in New Issue
Block a user