Adding images requires file extensions

[tpaulshippy]

When I run this:

response = chat.ask "What kind of architecture is shown here?", with: { image: "https://example.com/eiffel_tower" }

if my URL does not end in .jpg I am unable to make it work. I was able to workaround this by saving the image to a temp file and naming it with the .jpg extension.

It seems like if I know the image type, I should be able to provide it to the library and skip the detect_from_path method entirely.

[crmne]

Unfortunately the APIs require a MIME type to work properly, and we're inferring that from the extension.

We could expose the a way to set the MIME type manually, wdyt?

[tpaulshippy]

Yeah that would be sufficient for my needs.

[crmne]

@tpaulshippy do you have an example URL for me? I'm implementing the tests

[tpaulshippy]

https://lh3.googleusercontent.com/zonAUYRq0164wHlEMjd8tZEhriCOBg1104qnQ26OSRXegbdY8vFlg2Oe_xEevmGvkGU59RaaGw=w640-h400-e365

[crmne]

@tpaulshippy just rewrote the whole content and attachment system to be a lot simpler and not needing to specify mime types manually: fb5e4e4

will work on the rails integration tomorrow

[tpaulshippy]

And I just got it working haha. It's ok - should be easy to switch. Thanks!

[crmne]

Pushed 53fca94