For example https://commons.wikimedia.org/wiki/File:Molen_-_Veenoord_-_20239028_-_RCE.jpg contains a link to https://tools.wmflabs.org/heritage/api/api.php?action=images&imcountry=nl&imid=33786&format=html&props=img_name , but that doesn't contain any images. Judging from the last update at https://commons.wikimedia.org/wiki/Commons:Monuments_database/Indexed_images/Statistics I think the whole process is broken.
Description
Description
Related Objects
Related Objects
Event Timeline
Comment Actions
Yeah, the process does not work :-( I have actually spent quite some time recently extending that script (gathering stats on geolocated images, etc.) but it does not go through : the database connection drops (even with a ping=True)
Comment Actions
Plan of attack: load the SQL results into memory, and process that data structure, instead of holding open the database connection.
That data structure should be memory-efficient.