Repository navigation
[Core] Skip loading\processing of EXIF\metadata when opening file #11
Description
Activity
JimBobSquarePants commented
on Oct 30, 2016 MemberAuthorMore actionsFrom @dlemstra on August 18, 2016 21:30
The EXIF won't be parsed until you try to get information from it. When the decoder reads the EXIF data it just reads a bunch of bytes and stores them in an array. The encoder will then just copy those bytes in the output image. I don't think this will have a noticeable effect on your performance.
@JimBobSquarePants Might be good to still have an option disable reading/writing metadata in the encoder/decoder though.
p.s. Waar gebruik je dit?
JimBobSquarePants commented
on Oct 30, 2016 MemberAuthorMore actionsYeah, we could do that as an overload I'm sure and pass the option through to the decoder.
JimBobSquarePants commented
on Oct 30, 2016 MemberAuthorMore actionsFrom @KLuuKer on August 19, 2016 7:46
@dlemstra our company makes a saas webshop and end users don't need to see EXIF data (if it exists in the first place). It also makes the file bigger if it's included, and with allot of users hitting the site it saves a ton of bandwidth (for all parties involved moving that data around).
p.s. hé kijk een nederlander :)
JimBobSquarePants commented
on Oct 30, 2016 MemberAuthorMore actionsFrom @dlemstra on August 19, 2016 8:10
If optimization of file size is an issue then you might also want to take a look at a feature that exists in my library (Magick.NET). It will require an extra dependency but it works on .NET Core on Windows. If you need to have it work on an other platform you will need to wait a while.
My library has a class that can be used to compress the size of a JPEG/PNG/GIF without loss of quality. I have created a tiny example for the lossless compression here: https://magick.codeplex.com/wikipage?title=Lossless%20compression&referringTitle=Documentation
JimBobSquarePants commented
on Oct 30, 2016 MemberAuthorMore actionsFrom @KLuuKer on August 19, 2016 8:28
Unfortunately i have to run on linux and windows so that's not a option.
I am also not a fan of using native dll's for security reasons.JimBobSquarePants commented
on Oct 30, 2016 MemberAuthorMore actionsFrom @dlemstra on August 19, 2016 8:42
I think we will make this 'NoExif' option an extra option that you can pass to the
SaveAsFormat()methods. And we will also need to add an option for when an image is read. Maybe we could also add LoadAsJpeg @JimBobSquarePants ?JimBobSquarePants commented
on Oct 30, 2016 MemberAuthorMore actionsFrom @KLuuKer on August 19, 2016 9:8
Maybe name it 'NoMetadata' (as its a more generic term), for when there will be supported for WebM\WebP and to skip loading of RIFF data
JimBobSquarePants commented
on Oct 30, 2016 MemberAuthorMore actionsFrom @dlemstra on August 19, 2016 9:16
The quotes were added to try to clarify that my name was just a suggestion 😄 . I personally think
IgnoreMetadatais a better name and yes it should be used by multiple coders. Planning to work on ImageMagick this weekend so you will need to wait a bit before this will be added.p.s. You could help us out and send us a pull request 😺 We can continue this conversation on gitter if you need more info.
JimBobSquarePants commented
on Oct 30, 2016 MemberAuthorMore actions- added a commit that references this issue
on Nov 27, 2019 - added a commit that references this issue
on Jan 18, 2020 - added a commit that references this issue
on Oct 9, 2026
From @KLuuKer on August 18, 2016 20:15
Great library!
Also good that you are supporting EXIF data but....
I would like to completely skip the processing of EXIF (and any other non relevant metadata) and also not include it in the final output.
Because I am processing\downscaling\compressing images for use on a public facing website and nobody needs to see the EXIF data anyway.
This question was bound to show up :)
Copied from original issue: JimBobSquarePants/ImageProcessor#458