When to save URL images in a database vs when the file-system is enough

if u want to use your app as api also then url i think is best

@fsdolphin I save just the filenames of uploaded files in my database. That way, I’m not tied to one particular storage mechanism. I can also move images between folders and change it one place, instead of having to upload multiple rows across many tables in a database.

@martinbean Is that generally the process you follow, regardless of what type of app you are building?

I have done what @martinbean has done, where just the filename of the image is in the database. But I took it further, where I also had the target filesystem as a column, as well as the reference key for that target filesystem.

So, for instance, if I start app development with local storage, with the intent of moving to Amazon S3 later, i have set 'filesystem' as local, then the 'reference' is the relative path to the image.

I have also written automated migrations where the files are moved from local to S3, and just update the 'filesystem' and 'reference' after the copy is complete. That way, the platform did not have an outage, and users never noticed.

@fsdolphin Pretty much, and have done for years. Has worked well up to now. As I say, means I can store the file in any location and at any path, so long as the filename stays the same.

@martinbean One last question, does it matter where do I put my images, in the storage folder or the public folder? Is there a common practice?

u should use storage folder and make symbolic link to map storage folder to public, it will secure your storage

@vipin93 How does doing a symbolic link to the storage folder make files in there any more “secure”?

@fsdolphin If storing files on the same filesystem as your application, then common practice is to save the original uploads in the storage, and then serve generated thumbnails from somewhere under the public directory.