HBR processed the image imports from yesterday, and the results are good. I then went back to the second batch of proposed imports, consisting of 1,310 items in the "HER" set, and worked to refine my image-matching code. The result is that I have matching images for 732 of these items. I've also rewritten the HTML-generating code so that it automatically generates the image copy script to create a folder of images for import, with sane image names, and the CSV-generation code so that it creates the image import CSV as well as the item CSV files for upload. This seems quite robust now. I have 14 CSV files to import tomorrow, following which I'll run SQL to publish them, and then put the image stuff on the server, and ask HBR to run that. With luck we'll have a lot more stuff by Christmas.