Commons:Batch uploading/Wikipedia Loves Art

From Wikimedia Commons, the free media repository
Jump to: navigation, search

Wikipedia Loves Art[edit]

Upload all the photos from the Wikipedia Loves Art photo pool on Flickr that have the machine tag "caption=yes".

This process actually involves several phases (each with its own script):

  1. The first script uses the Flickr API to examine all the photos in the pool, and copy all the relevant tags, comments, file info, and metadata into a local mySQL database. I utilized the phpFlickr class for most of this.
  2. The second script downloads all the appropriate Flickr photos from the pool at the highest resolution available.
  3. The third script attempts to create a unique human-readable filename for each photo by examining the tags and comments (too complicated to describe here).
  4. A fourth script lets humans review the filename suggestions and change them as appropriate. See Wikipedia:Wikipedia Loves Art/Filename review contest
  5. Once the filenames are approved, a fifth script renames all the files locally.
  6. The files are then batch uploaded to Commons with templates filled in according to the data stored in the mySQL database. This script is partially based on Eloquence's file upload script.
Assigned to Progress Bot name
Kaldari (talk) Phase 6 (partially complete) File Upload Bot (Kaldari)


Just a thought for future events: as the participants made two images of each object (one with an index card, one of the image for articles), it might be worth uploading both under the same filename. (First the one with the index card obviously). This could help identification later. As once in a while there are questions about the colors of paintings, it might be worth taking the first picture with an index card and a color chart. -- User:Docu at 07:22, 23 September 2009 (UTC)