search

culibraries / ir-scholar

CU Scholar - Institutional Repository Hyrax
0 stars 0 forks source link

Files other than "primary" data file for all pre-migration data sets with multiple files have been turned into HTML files #63

Open mbstacy opened 3 years ago

mbstacy commented 3 years ago

See example: https://scholar.colorado.edu/concern/datasets/wm117p69n. For this data set and apparently all other pre-migration data sets with multiple files, the "primary" data file is accessible as it should be, but all other data files (e.g., CSV version and readme in this case) have been turned into the same HTML file. We need to restore the correct files for all of these data sets.

sabetiv commented 1 year ago

Data was migrated from bepress to current Scholar setup (Samvera/Fedora/Solr). The original bepress data files are stored on S3 cubl-ir-bepress bucket. The uploaded and processed data files are on S3 cubl-ir bucket under original and processed keys. Filenames between bepress and uploaded/processed were at times changed (removal of white space, sometimes shorten, dash-vs-underscore). One method of matching the bepress files to the uploaded/process files to check on differences is via the "context_key" in the inventory list.