Check the file type of the thumbnail before passing it to Photon

[krysal]

Fixes

Fixes #2067 by @krysal

Description

To prevent known errors from Photon due to unsupported file types, this PR adds the following check before making the call to Photon:

1. Get the thumbnail extension from the URL. We can't assume it'd be the same of the main image (in the case of images/), so try to guess it first from the URL. I believe it is worth trying this route, since this is where most of the cases will fall.
2. If can't guess the extension from the URL then make a HEAD request to get the content-type of the file, caching the result
3. If the media type is not supported by Photon then raise an exception: 415 Unsupported Media Type.

Testing Instructions

If you want to manually check, try replacing the first row (after the header) of the sample_data/sample_image.csv file with the following:

0e3315c5-3328-4a99-80ab-567ac32f685f,2022-12-21 17:29:54.000000+00,2022-12-21 17:29:54.000000+00,provider_api,wikimedia,wikimedia,,https://www.flickr.com/photos/54633257@N04/51745822704,https://upload.wikimedia.org/wikipedia/commons/9/92/Flower_sewing_pattern_outline.svg,,433,1024,97150,by-sa,2.0,Jooja,https://commons.wikimedia.org/wiki/User:Jooja,Flower sewing pattern outline,"{""views"": ""5991"", ""pub_date"": ""1639443671"", ""date_taken"": ""2021-07-20 18:19:25"", ""description"": ""dressed"", ""license_url"": ""https://creativecommons.org/licenses/by-sa/3.0/"", ""raw_license_url"": null}",,f,2021-12-15 20:49:19.976193+00,f,svg,illustration

The run just api/recreate and visit http://localhost:50280/v1/images/0e3315c5-3328-4a99-80ab-567ac32f685f/thumb/

You should see the new response message 'Unsupported media type "svg" in request.'

Checklist

• My pull request has a descriptive title (not a vague title likeUpdate index.md).
• My pull request targets the default branch of the repository (main) or a parent feature branch.
• My commit messages follow best practices.
• My code follows the established code style of the repository.
• I added or updated tests for the changes I made (if applicable).
• I added or updated documentation (if applicable).
• I tried running the project locally and verified that there are no visible errors.
• I ran the DAG documentation generator (if applicable).

Developer Certificate of Origin

Developer Certificate of Origin

Developer Certificate of Origin
Version 1.1

Copyright (C) 2004, 2006 The Linux Foundation and its contributors.
1 Letterman Drive
Suite D4700
San Francisco, CA, 94129

Everyone is permitted to copy and distribute verbatim copies of this
license document, but changing it is not allowed.


Developer's Certificate of Origin 1.1

By making a contribution to this project, I certify that:

(a) The contribution was created in whole or in part by me and I
    have the right to submit it under the open source license
    indicated in the file; or

(b) The contribution is based upon previous work that, to the best
    of my knowledge, is covered under an appropriate open source
    license and I have the right under that license to submit that
    work with modifications, whether created in whole or in part
    by me, under the same open source license (unless I am
    permitted to submit under a different license), as indicated
    in the file; or

(c) The contribution was provided directly to me by some other
    person who certified (a), (b) or (c) and I have not modified
    it.

(d) I understand and agree that this project and the contribution
    are public and that a record of the contribution (including all
    personal information I submit with it, including my sign-off) is
    maintained indefinitely and may be redistributed consistent with
    this project or the open source license(s) involved.

[AetherUnbound]

The code and tests look good, great call on adding caching at various levels! I have a few suggestions/questions but nothing blocking 🚀 🖼️

[stacimc]

LGTM and tested well! I appreciated the pro-active work to reduce Sentry messages 🚀

[AetherUnbound]

Oh, one other thing I was curious about! I know we do the dead link filtering with using async to parallelize requests:

openverse/api/api/utils/check_dead_links/__init__.py

Line 46 in 098925a

async def _make_head_requests(urls: list[str]) -> list[tuple[str, int]]:

I think that's because they're all happening as part of one request, whereas queries for the thumbnail are each their own request, right? That to say, do we need to try and parallelize this or is it okay to make the HEAD request as part of the request since all thumbnail requests will be issued separately, at the same time?

[krysal]

do we need to try and parallelize this or is it okay to make the HEAD request as part of the request since all thumbnail requests will be issued separately, at the same time?
– @AetherUnbound

I'm not sure I fully understand the question 🤔 These are all individual request so I don't see how a group of thumbnails requests can be parallelized in this point. We could create another endpoint that receives a group of media IDs to return their thumbnails, and there they can be parallelized, but here it might be a simple request of the link returned from /<media>/<uuid>/ view.

---

Thank you all for the suggestions!

[AetherUnbound]

Got it, just got myself mixed up and had to talk through it to understand - thanks @krysal!