[Mini announcement] Server update addressing duplicate image names
11 views
Skip to first unread message
Stephen Chan
Jul 14, 2025, 4:14:39 PMJul 14
to CoralNet Users
Hi all,
I've made a server update which changes how duplicate image names are detected.
In many cases, CoralNet would previously catch situations where you were about to have two images with the same name in your source. For example, there was a check when you were about to upload images, and there was a check in the Edit Metadata grid page. However, a few situations were overlooked. For example:
- It would check for dupes on the image-upload preview step, but not when it was time to actually upload.
- It didn't check on the single-image 'Edit Image Details' page.
And when these situations resulted in dupes sneaking through, that would sometimes lead to mysterious errors later on, such as when uploading CSV metadata.
While I was addressing these situations, I also made dupe checking case-insensitive. Previously it was case-sensitive, so if you had IMG_0001.jpg and IMG_0001.JPG, those would have been accepted as two different images. I changed to case-insensitive for a couple of reasons:
- Case-sensitive is probably confusing for most people, especially since Mac and Windows (to my knowledge, at least) do not allow two filenames in the same folder that differ only in case.
- Case-insensitive is probably more intuitive for the "Image name contains" field on Browse Images, as well as alphabetical sorting of image names. It is good to have all of these name-casing behaviors consistent (e.g. all case-insensitive), for site performance reasons.
Of course, since dupe name checking wasn't so watertight before, there was the matter of what to do with existing dupes. I added suffixes to each set of dupe names to ensure that, as of now, all image names within a source are guaranteed case-insensitively unique. So, the filenames in the above example would become something like IMG_0001__dupe-name-01.jpg and IMG_0001__dupe-name-02.JPG.
About 2% of sources contained at least one dupe name. To check if that includes your source, go to Browse Images, type "dupe-name" (without the quotes) in the "Image name contains" field, hit Search, and see if you get any search results. Then, you may choose what to do (if anything) with the results you see.
For example, if you indeed find that you accidentally uploaded the same image twice, you may want to delete one of the copies. Or, if you have two completely different images that happened to share a name, you may want to give new names to those images for your own organizational purposes.
I've made a server update which changes how duplicate image names are detected.
In many cases, CoralNet would previously catch situations where you were about to have two images with the same name in your source. For example, there was a check when you were about to upload images, and there was a check in the Edit Metadata grid page. However, a few situations were overlooked. For example:
- It would check for dupes on the image-upload preview step, but not when it was time to actually upload.
- It didn't check on the single-image 'Edit Image Details' page.
And when these situations resulted in dupes sneaking through, that would sometimes lead to mysterious errors later on, such as when uploading CSV metadata.
While I was addressing these situations, I also made dupe checking case-insensitive. Previously it was case-sensitive, so if you had IMG_0001.jpg and IMG_0001.JPG, those would have been accepted as two different images. I changed to case-insensitive for a couple of reasons:
- Case-sensitive is probably confusing for most people, especially since Mac and Windows (to my knowledge, at least) do not allow two filenames in the same folder that differ only in case.
- Case-insensitive is probably more intuitive for the "Image name contains" field on Browse Images, as well as alphabetical sorting of image names. It is good to have all of these name-casing behaviors consistent (e.g. all case-insensitive), for site performance reasons.
Of course, since dupe name checking wasn't so watertight before, there was the matter of what to do with existing dupes. I added suffixes to each set of dupe names to ensure that, as of now, all image names within a source are guaranteed case-insensitively unique. So, the filenames in the above example would become something like IMG_0001__dupe-name-01.jpg and IMG_0001__dupe-name-02.JPG.
About 2% of sources contained at least one dupe name. To check if that includes your source, go to Browse Images, type "dupe-name" (without the quotes) in the "Image name contains" field, hit Search, and see if you get any search results. Then, you may choose what to do (if anything) with the results you see.
For example, if you indeed find that you accidentally uploaded the same image twice, you may want to delete one of the copies. Or, if you have two completely different images that happened to share a name, you may want to give new names to those images for your own organizational purposes.
Reply all
Reply to author
Forward
0 new messages