Add initial prototype supporting the files API

[MarkDaoust]

WIP

When this is done, the code below should work

pip install -U git+https://github.com/MarkDaoust/generative-ai-python@files

pip install -U git+https://github.com/google-generativeai-python

import google.generativeai as genai

genai.configure(api_key="...")

f = genai.upload_file(path="image.png")

m = genai.GenerativeModel(model_name="models/gemini-1.5-pro")
response = m.generate_content(["hello? Pleasecdescribe this file:", f])

print(response.text)

[TYMichaelChen]

remove: &labels=GOOGLE_INTERNAL. Also, should we always point to generativelanguage.googleapis.com here?

[MarkDaoust]

&labels=GOOGLE_INTERNAL

Done.

always point to

If they override the endpoint shouldn't we respect that? I can't really test this code without that.

[Qxxxx]

Hi @MarkDaoust, thanks for this api. That would be helpful. Though I meet the following error:

Traceback (most recent call last):
  File "/Users/qiguang/gemini_init/test.py", line 5, in <module>
    f = genai.upload_file(path="IMG_0521.PNG")
        ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/Users/qiguang/miniconda3/envs/pytorch/lib/python3.11/site-packages/google/generativeai/files.py", line 49, in upload_file
    response = client.create_file(
               ^^^^^^^^^^^^^^^^^^^
  File "/Users/qiguang/miniconda3/envs/pytorch/lib/python3.11/site-packages/google/generativeai/client.py", line 75, in create_file
    result = request.execute()
             ^^^^^^^^^^^^^^^^^
  File "/Users/xxxx/miniconda3/envs/pytorch/lib/python3.11/site-packages/googleapiclient/_helpers.py", line 130, in positional_wrapper
    return wrapped(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^^^^
  File "/Users/qiguang/miniconda3/envs/pytorch/lib/python3.11/site-packages/googleapiclient/http.py", line 938, in execute
    raise HttpError(resp, content, uri=self.uri)
googleapiclient.errors.HttpError: <HttpError 501 when requesting https://generativelanguage.googleapis.com/upload/v1beta/files?key=AIzaSyA8mR0GDyiha1hN1j0llZ7VoRWlWy6fQEA&alt=json&uploadType=multipart returned "Endpoint not implemented.". Details: "Endpoint not implemented.">

Also, does it means it support type: application/pdf now? Though it could by uploading pdf to aistudio, but api wise I don't know how to do it. That would be nice if you could help me with this issue! Thanks in advance!

[TYMichaelChen]

Hi @Qxxxx - The API is currently in private preview, so that error is expected.

We will not support PDFs via the API. I would recommend taking a look at LlamaIndex for pdf parsing on the client side

[Qxxxx]

Thanks a lot, this repo is very helpful!