Uploading Large Numbers of Images

What is the recommended way to do a bulk upload of thousands of images, each of about 1 MB?
The images will need sorting into albums, and Titles/Descriptions adding. This will be done by a team of people, so I'd prefer to upload and then do the sorting etc. within Zenphoto rather than offline (but I may be wrong).
I see that there is a 64MB limit for an upload, although FTP is also mentioned (but I know little about it).

Tags:

Comments

  • acrylian Administrator, Developer

    see that there is a 64MB limit for an upload,

    This is a limit of your server, not Zenphoto. The best to upload really that many images is FTP directly to the albums folder. Zenphoto is file system based so uploading folders there will make them albums.

    I would not recommend to upload that many images all at once because of processing image sizes and generating the image cache: https://www.zenphoto.org/news/caching/

    This will likely be overloading your server's resources:
    https://www.zenphoto.org/news/problems-with-albums-and-images/

  • If it is possible to name them ahead of time so that you can just sort by name, it would probably go better.

    It might be very time consuming to sort one the pictures are in ZP.

    Also.... you might look at adding info to exif and then using that to sort. you might even be able to automate it better.

  • I feel it is easier and quicker to sort relevant images into an offline folder prior to uploading to an album. However, I have a related issue in that the information I want to add to the database for each image is stored in a spreadsheet row. Is it possible to do some sort of bulk transfer of data rather than having to copy and paste one cell at a time?

  • acrylian Administrator, Developer
    edited May 2023

    You mean you want data like descriptions from an offline spreadsheet to be stored in the database?

    It is technically possible to import from a CSV file for example but since nothing for this exists you would have to code that first… That's not exactly a simple solution.

    The best and recommended way is to store the data within each images meta data (Exif/IPTC) itself already as Zenphoto then would import that automatically on discovery. To do that from a spreadsheet is out of Zenphoto's domain and I have no idea if there might be tools for this.

  • The Title, Description, Tags, etc. for each image are input to cells in a spreadsheet by a group of people who are not very technically literate. They can manage this, but would probably not be able to handle metadata, so I believe the spreadsheet has to be the initial input document.
    Perhaps it is outside the scope of Zenphoto, but I feel it would be a useful enhancement for crowdsourced projects involving a large number of images (I have around 80,000) if input from spreadsheet was possible.
    Perhaps it is possible to have an intermediate stage by transferring data in bulk from a spreadsheet into metadata?

  • ctdlg Member

    Hello Nick,

    You could learn how to use online converter tools.
    Search "Convert Excel to Insert SQL" on the web,
    Export your current database using phpmyadmin,
    Study the zp_images structure.
    And try. You will fail, fail, but then probably succeed !

    Use a localhost server for this kind of work.
    You can also ask questions on specialized sql forums.
    Because perhaps someone already did what you want to do.

  • acrylian Administrator, Developer

    What ctdlg suggest may be an options. But be aware that Zenphoto is file system based so you need to upload images first and then do such database updates.

    Perhaps it is possible to have an intermediate stage by transferring data in bulk from a spreadsheet into metadata?

    That is something you have to do offline. Zenphoto cannot write meta data to images itself.

    Another way would be to find a way to export this meta data to xmp sidecar files. Using the xmp plugin Zenphoto could read the meta data from these files.

  • acrylian Administrator, Developer
    edited May 2023
Sign In or Register to comment.