mirror/gradio
2
0
Fork 0
mirror of https://github.com/gradio-app/gradio.git synced 2025-03-01 11:45:36 +08:00
Code Issues Packages Projects Releases Wiki Activity

Fix loading private Spaces (#4430)

Browse Source 
* replace api key with hf_token

* removed extra slash

* changelog

* update notebooks

* lint

* changelog

* changelog

* revert demo
This commit is contained in:
Abubakar Abid 2023-06-07 13:55:05 -05:00 committed by GitHub
parent 866b57b93c
commit d82dfb886d
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
10 changed files with 51 additions and 47 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
CHANGELOG.md
View File

@ -9,6 +9,8 @@
- Remove target="_blank" override on anchor tags with internal targets by [@hannahblair](https://github.com/hannahblair) in [PR 4405](https://github.com/gradio-app/gradio/pull/4405)
- Fixed bug where `gr.File(file_count='multiple')` could not be cached as output by [@freddyaboulton](https://github.com/freddyaboulton) in [PR 4421](https://github.com/gradio-app/gradio/pull/4421)
- Restricts the domains that can be proxied via `/proxy` route by [@abidlabs](https://github.com/abidlabs) in [PR 4406](https://github.com/gradio-app/gradio/pull/4406).
- Fixes bug where `/proxy` route was being incorrectly constructed by the frontend by [@abidlabs](https://github.com/abidlabs) in [PR 4430](https://github.com/gradio-app/gradio/pull/4430).
- Fix z-index of status component by [@hannahblair](https://github.com/hannahblair) in [PR 4429](https://github.com/gradio-app/gradio/pull/4429)
- Fix video rendering in Safari by [@aliabid94](https://github.com/aliabid94) in [PR 4433](https://github.com/gradio-app/gradio/pull/4433).
## Other Changes:
@ -27,7 +29,7 @@ No changes to highlight.
## Bug Fixes:
- Fix bug for get_continuous_fn by [@dkjshk](https://github.com/dkjshk) in [PR 4434](https://github.com/gradio-app/gradio/pull/4434)
- Allow `every` to work with generators by [@dkjshk](https://github.com/dkjshk) in [PR 4434](https://github.com/gradio-app/gradio/pull/4434)
- Fix z-index of status component by [@hannahblair](https://github.com/hannahblair) in [PR 4429](https://github.com/gradio-app/gradio/pull/4429)
- Allow gradio to work offline, by [@aliabid94](https://github.com/aliabid94) in [PR 4398](https://github.com/gradio-app/gradio/pull/4398).
- Fixed `validate_url` to check for 403 errors and use a GET request in place of a HEAD by [@alvindaiyan](https://github.com/alvindaiyan) in [PR 4388](https://github.com/gradio-app/gradio/pull/4388).

2
client/js/src/client.ts
View File

@ -745,7 +745,7 @@ function normalise_file(
if (!root_url) {
file.data = root + "/file=" + file.name;
} else {
file.data = "/proxy=" + root_url + "/file=" + file.name;
file.data = "/proxy=" + root_url + "file=" + file.name;
}
}
return file;

2
demo/autocomplete/run.ipynb
View File

@ -1 +1 @@
{"cells": [{"cell_type": "markdown", "id": 302934307671667531413257853548643485645, "metadata": {}, "source": ["# Gradio Demo: autocomplete\n", "### This text generation demo works like autocomplete. There's only one textbox and it's used for both the input and the output. The demo loads the model as an interface, and uses that interface as an API. It then uses blocks to create the UI. All of this is done in less than 10 lines of code.\n", " "]}, {"cell_type": "code", "execution_count": null, "id": 272996653310673477252411125948039410165, "metadata": {}, "outputs": [], "source": ["!pip install -q gradio "]}, {"cell_type": "code", "execution_count": null, "id": 288918539441861185822528903084949547379, "metadata": {}, "outputs": [], "source": ["import gradio as gr\n", "import os\n", "\n", "# save your HF API token from https:/hf.co/settings/tokens as an env variable to avoid rate limiting\n", "auth_token = os.getenv(\"auth_token\")\n", "\n", "# load a model from https://hf.co/models as an interface, then use it as an api \n", "# you can remove the api_key parameter if you don't care about rate limiting. \n", "api = gr.load(\"huggingface/EleutherAI/gpt-j-6B\", api_key=auth_token)\n", "\n", "def complete_with_gpt(text):\n", " return text[:-50] + api(text[-50:])\n", "\n", "with gr.Blocks() as demo:\n", " textbox = gr.Textbox(placeholder=\"Type here...\", lines=4)\n", " btn = gr.Button(\"Autocomplete\")\n", " \n", " # define what will run when the button is clicked, here the textbox is used as both an input and an output\n", " btn.click(fn=complete_with_gpt, inputs=textbox, outputs=textbox, queue=False)\n", "\n", "demo.launch()"]}], "metadata": {}, "nbformat": 4, "nbformat_minor": 5}
{"cells": [{"cell_type": "markdown", "id": 302934307671667531413257853548643485645, "metadata": {}, "source": ["# Gradio Demo: autocomplete\n", "### This text generation demo works like autocomplete. There's only one textbox and it's used for both the input and the output. The demo loads the model as an interface, and uses that interface as an API. It then uses blocks to create the UI. All of this is done in less than 10 lines of code.\n", " "]}, {"cell_type": "code", "execution_count": null, "id": 272996653310673477252411125948039410165, "metadata": {}, "outputs": [], "source": ["!pip install -q gradio "]}, {"cell_type": "code", "execution_count": null, "id": 288918539441861185822528903084949547379, "metadata": {}, "outputs": [], "source": ["import gradio as gr\n", "import os\n", "\n", "# save your HF API token from https:/hf.co/settings/tokens as an env variable to avoid rate limiting\n", "auth_token = os.getenv(\"auth_token\")\n", "\n", "# load a model from https://hf.co/models as an interface, then use it as an api \n", "# you can remove the api_key parameter if you don't care about rate limiting. \n", "api = gr.load(\"huggingface/EleutherAI/gpt-j-6B\", hf_token=auth_token)\n", "\n", "def complete_with_gpt(text):\n", " return text[:-50] + api(text[-50:])\n", "\n", "with gr.Blocks() as demo:\n", " textbox = gr.Textbox(placeholder=\"Type here...\", lines=4)\n", " btn = gr.Button(\"Autocomplete\")\n", " \n", " # define what will run when the button is clicked, here the textbox is used as both an input and an output\n", " btn.click(fn=complete_with_gpt, inputs=textbox, outputs=textbox, queue=False)\n", "\n", "demo.launch()"]}], "metadata": {}, "nbformat": 4, "nbformat_minor": 5}

2
demo/autocomplete/run.py
View File

@ -6,7 +6,7 @@ auth_token = os.getenv("auth_token")
# load a model from https://hf.co/models as an interface, then use it as an api
# you can remove the api_key parameter if you don't care about rate limiting.
api = gr.load("huggingface/EleutherAI/gpt-j-6B", api_key=auth_token)
api = gr.load("huggingface/EleutherAI/gpt-j-6B", hf_token=auth_token)
def complete_with_gpt(text):
return text[:-50] + api(text[-50:])

2
demo/automatic-speech-recognition/run.ipynb
View File

@ -1 +1 @@
{"cells": [{"cell_type": "markdown", "id": 302934307671667531413257853548643485645, "metadata": {}, "source": ["# Gradio Demo: automatic-speech-recognition\n", "### Automatic speech recognition English. Record from your microphone and the app will transcribe the audio.\n", " "]}, {"cell_type": "code", "execution_count": null, "id": 272996653310673477252411125948039410165, "metadata": {}, "outputs": [], "source": ["!pip install -q gradio "]}, {"cell_type": "code", "execution_count": null, "id": 288918539441861185822528903084949547379, "metadata": {}, "outputs": [], "source": ["import gradio as gr\n", "import os\n", "\n", "# save your HF API token from https:/hf.co/settings/tokens as an env variable to avoid rate limiting\n", "auth_token = os.getenv(\"auth_token\")\n", "\n", "# automatically load the interface from a HF model \n", "# you can remove the api_key parameter if you don't care about rate limiting. \n", "demo = gr.load(\n", " \"huggingface/facebook/wav2vec2-base-960h\",\n", " title=\"Speech-to-text\",\n", " inputs=\"mic\",\n", " description=\"Let me try to guess what you're saying!\",\n", " api_key=auth_token\n", ")\n", "\n", "demo.launch()\n"]}], "metadata": {}, "nbformat": 4, "nbformat_minor": 5}
{"cells": [{"cell_type": "markdown", "id": 302934307671667531413257853548643485645, "metadata": {}, "source": ["# Gradio Demo: automatic-speech-recognition\n", "### Automatic speech recognition English. Record from your microphone and the app will transcribe the audio.\n", " "]}, {"cell_type": "code", "execution_count": null, "id": 272996653310673477252411125948039410165, "metadata": {}, "outputs": [], "source": ["!pip install -q gradio "]}, {"cell_type": "code", "execution_count": null, "id": 288918539441861185822528903084949547379, "metadata": {}, "outputs": [], "source": ["import gradio as gr\n", "import os\n", "\n", "# save your HF API token from https:/hf.co/settings/tokens as an env variable to avoid rate limiting\n", "auth_token = os.getenv(\"auth_token\")\n", "\n", "# automatically load the interface from a HF model \n", "# you can remove the api_key parameter if you don't care about rate limiting. \n", "demo = gr.load(\n", " \"huggingface/facebook/wav2vec2-base-960h\",\n", " title=\"Speech-to-text\",\n", " inputs=\"mic\",\n", " description=\"Let me try to guess what you're saying!\",\n", " hf_token=auth_token\n", ")\n", "\n", "demo.launch()\n"]}], "metadata": {}, "nbformat": 4, "nbformat_minor": 5}

2
demo/automatic-speech-recognition/run.py
View File

@ -11,7 +11,7 @@ demo = gr.load(
title="Speech-to-text",
inputs="mic",
description="Let me try to guess what you're saying!",
api_key=auth_token
hf_token=auth_token
)
demo.launch()

36
gradio/external.py
View File

@ -66,14 +66,14 @@ def load(
)
hf_token = api_key
return load_blocks_from_repo(
name=name, src=src, api_key=hf_token, alias=alias, **kwargs
name=name, src=src, hf_token=hf_token, alias=alias, **kwargs
)
def load_blocks_from_repo(
name: str,
src: str | None = None,
api_key: str | None = None,
hf_token: str | None = None,
alias: str | None = None,
**kwargs,
) -> Blocks:
@ -97,14 +97,14 @@ def load_blocks_from_repo(
src.lower() in factory_methods
), f"parameter: src must be one of {factory_methods.keys()}"
if api_key is not None:
if Context.hf_token is not None and Context.hf_token != api_key:
if hf_token is not None:
if Context.hf_token is not None and Context.hf_token != hf_token:
warnings.warn(
"""You are loading a model/Space with a different access token than the one you used to load a previous model/Space. This is not recommended, as it may cause unexpected behavior."""
)
Context.hf_token = api_key
Context.hf_token = hf_token
blocks: gradio.Blocks = factory_methods[src](name, api_key, alias, **kwargs)
blocks: gradio.Blocks = factory_methods[src](name, hf_token, alias, **kwargs)
return blocks
@ -134,12 +134,12 @@ def chatbot_postprocess(response):
return chatbot_value, response_json
def from_model(model_name: str, api_key: str | None, alias: str | None, **kwargs):
def from_model(model_name: str, hf_token: str | None, alias: str | None, **kwargs):
model_url = f"https://huggingface.co/{model_name}"
api_url = f"https://api-inference.huggingface.co/models/{model_name}"
print(f"Fetching model from: {model_url}")
headers = {"Authorization": f"Bearer {api_key}"} if api_key is not None else {}
headers = {"Authorization": f"Bearer {hf_token}"} if hf_token is not None else {}
# Checking if model exists, and if so, it gets the pipeline
response = requests.request("GET", api_url, headers=headers)
@ -439,15 +439,15 @@ def from_model(model_name: str, api_key: str | None, alias: str | None, **kwargs
def from_spaces(
space_name: str, api_key: str | None, alias: str | None, **kwargs
space_name: str, hf_token: str | None, alias: str | None, **kwargs
) -> Blocks:
space_url = f"https://huggingface.co/spaces/{space_name}"
print(f"Fetching Space from: {space_url}")
headers = {}
if api_key is not None:
headers["Authorization"] = f"Bearer {api_key}"
if hf_token is not None:
headers["Authorization"] = f"Bearer {hf_token}"
iframe_url = (
requests.get(
@ -473,7 +473,7 @@ def from_spaces(
raise ValueError(f"Could not load the Space: {space_name}") from ae
if "allow_flagging" in config: # Create an Interface for Gradio 2.x Spaces
return from_spaces_interface(
space_name, config, alias, api_key, iframe_url, **kwargs
space_name, config, alias, hf_token, iframe_url, **kwargs
)
else: # Create a Blocks for Gradio 3.x Spaces
if kwargs:
@ -483,11 +483,11 @@ def from_spaces(
"Blocks or Interface locally. You may find this Guide helpful: "
"https://gradio.app/using_blocks_like_functions/"
)
return from_spaces_blocks(space=space_name, api_key=api_key)
return from_spaces_blocks(space=space_name, hf_token=hf_token)
def from_spaces_blocks(space: str, api_key: str | None) -> Blocks:
client = Client(space, hf_token=api_key)
def from_spaces_blocks(space: str, hf_token: str | None) -> Blocks:
client = Client(space, hf_token=hf_token)
predict_fns = [endpoint._predict_resolve for endpoint in client.endpoints]
return gradio.Blocks.from_config(client.config, predict_fns, client.src)
@ -496,15 +496,15 @@ def from_spaces_interface(
model_name: str,
config: dict,
alias: str | None,
api_key: str | None,
hf_token: str | None,
iframe_url: str,
**kwargs,
) -> Interface:
config = streamline_spaces_interface(config)
api_url = f"{iframe_url}/api/predict/"
headers = {"Content-Type": "application/json"}
if api_key is not None:
headers["Authorization"] = f"Bearer {api_key}"
if hf_token is not None:
headers["Authorization"] = f"Bearer {hf_token}"
# The function should call the API with preprocessed data
def fn(*data):

2
js/app/src/components/Dataset/Dataset.svelte
View File

@ -23,7 +23,7 @@
}>();
let samples_dir: string = root_url
? "proxy=" + root_url + "/file="
? "proxy=" + root_url + "file="
: root + "/file=";
let page = 0;
$: gallery = components.length < 2;

2
js/upload/src/utils.ts
View File

@ -45,7 +45,7 @@ export function normalise_file(
if (root_url == null) {
file.data = root + "/file=" + file.name;
} else {
file.data = "/proxy=" + root_url + "/file=" + file.name;
file.data = "/proxy=" + root_url + "file=" + file.name;
}
}
return file;

44
test/test_external.py
View File

@ -68,7 +68,7 @@ class TestLoadInterface:
def test_summarization(self):
model_type = "summarization"
interface = gr.load(
"models/facebook/bart-large-cnn", api_key=None, alias=model_type
"models/facebook/bart-large-cnn", hf_token=None, alias=model_type
)
assert interface.__name__ == model_type
assert isinstance(interface.input_components[0], gr.Textbox)
@ -77,7 +77,7 @@ class TestLoadInterface:
def test_translation(self):
model_type = "translation"
interface = gr.load(
"models/facebook/bart-large-cnn", api_key=None, alias=model_type
"models/facebook/bart-large-cnn", hf_token=None, alias=model_type
)
assert interface.__name__ == model_type
assert isinstance(interface.input_components[0], gr.Textbox)
@ -86,7 +86,7 @@ class TestLoadInterface:
def test_text2text_generation(self):
model_type = "text2text-generation"
interface = gr.load(
"models/sshleifer/tiny-mbart", api_key=None, alias=model_type
"models/sshleifer/tiny-mbart", hf_token=None, alias=model_type
)
assert interface.__name__ == model_type
assert isinstance(interface.input_components[0], gr.Textbox)
@ -96,7 +96,7 @@ class TestLoadInterface:
model_type = "text-classification"
interface = gr.load(
"models/distilbert-base-uncased-finetuned-sst-2-english",
api_key=None,
hf_token=None,
alias=model_type,
)
assert interface.__name__ == model_type
@ -105,7 +105,7 @@ class TestLoadInterface:
def test_fill_mask(self):
model_type = "fill-mask"
interface = gr.load("models/bert-base-uncased", api_key=None, alias=model_type)
interface = gr.load("models/bert-base-uncased", hf_token=None, alias=model_type)
assert interface.__name__ == model_type
assert isinstance(interface.input_components[0], gr.Textbox)
assert isinstance(interface.output_components[0], gr.Label)
@ -113,7 +113,7 @@ class TestLoadInterface:
def test_zero_shot_classification(self):
model_type = "zero-shot-classification"
interface = gr.load(
"models/facebook/bart-large-mnli", api_key=None, alias=model_type
"models/facebook/bart-large-mnli", hf_token=None, alias=model_type
)
assert interface.__name__ == model_type
assert isinstance(interface.input_components[0], gr.Textbox)
@ -124,7 +124,7 @@ class TestLoadInterface:
def test_automatic_speech_recognition(self):
model_type = "automatic-speech-recognition"
interface = gr.load(
"models/facebook/wav2vec2-base-960h", api_key=None, alias=model_type
"models/facebook/wav2vec2-base-960h", hf_token=None, alias=model_type
)
assert interface.__name__ == model_type
assert isinstance(interface.input_components[0], gr.Audio)
@ -133,7 +133,7 @@ class TestLoadInterface:
def test_image_classification(self):
model_type = "image-classification"
interface = gr.load(
"models/google/vit-base-patch16-224", api_key=None, alias=model_type
"models/google/vit-base-patch16-224", hf_token=None, alias=model_type
)
assert interface.__name__ == model_type
assert isinstance(interface.input_components[0], gr.Image)
@ -143,7 +143,7 @@ class TestLoadInterface:
model_type = "feature-extraction"
interface = gr.load(
"models/sentence-transformers/distilbert-base-nli-mean-tokens",
api_key=None,
hf_token=None,
alias=model_type,
)
assert interface.__name__ == model_type
@ -154,7 +154,7 @@ class TestLoadInterface:
model_type = "text-to-speech"
interface = gr.load(
"models/julien-c/ljspeech_tts_train_tacotron2_raw_phn_tacotron_g2p_en_no_space_train",
api_key=None,
hf_token=None,
alias=model_type,
)
assert interface.__name__ == model_type
@ -165,7 +165,7 @@ class TestLoadInterface:
model_type = "text-to-speech"
interface = gr.load(
"models/julien-c/ljspeech_tts_train_tacotron2_raw_phn_tacotron_g2p_en_no_space_train",
api_key=None,
hf_token=None,
alias=model_type,
)
assert interface.__name__ == model_type
@ -175,7 +175,7 @@ class TestLoadInterface:
def test_text_to_image(self):
model_type = "text-to-image"
interface = gr.load(
"models/osanseviero/BigGAN-deep-128", api_key=None, alias=model_type
"models/osanseviero/BigGAN-deep-128", hf_token=None, alias=model_type
)
assert interface.__name__ == model_type
assert isinstance(interface.input_components[0], gr.Textbox)
@ -288,8 +288,10 @@ class TestLoadInterface:
pass
def test_private_space(self):
api_key = "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" # Intentionally revealing this key for testing purposes
io = gr.load("spaces/gradio-tests/not-actually-private-space", api_key=api_key)
hf_token = "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" # Intentionally revealing this key for testing purposes
io = gr.load(
"spaces/gradio-tests/not-actually-private-space", hf_token=hf_token
)
try:
output = io("abc")
assert output == "abc"
@ -298,9 +300,9 @@ class TestLoadInterface:
pass
def test_private_space_audio(self):
api_key = "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" # Intentionally revealing this key for testing purposes
hf_token = "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" # Intentionally revealing this key for testing purposes
io = gr.load(
"spaces/gradio-tests/not-actually-private-space-audio", api_key=api_key
"spaces/gradio-tests/not-actually-private-space-audio", hf_token=hf_token
)
try:
output = io(media_data.BASE64_AUDIO["name"])
@ -309,18 +311,18 @@ class TestLoadInterface:
pass
def test_multiple_spaces_one_private(self):
api_key = "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" # Intentionally revealing this key for testing purposes
hf_token = "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" # Intentionally revealing this key for testing purposes
with gr.Blocks():
gr.load("spaces/gradio-tests/not-actually-private-space", api_key=api_key)
gr.load("spaces/gradio-tests/not-actually-private-space", hf_token=hf_token)
gr.load(
"spaces/gradio/test-loading-examples",
)
assert Context.hf_token == api_key
assert Context.hf_token == hf_token
def test_loading_files_via_proxy_works(self):
api_key = "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" # Intentionally revealing this key for testing purposes
hf_token = "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" # Intentionally revealing this key for testing purposes
io = gr.load(
"spaces/gradio-tests/test-loading-examples-private", hf_token=api_key
"spaces/gradio-tests/test-loading-examples-private", hf_token=hf_token
)
assert io.theme.name == "default"
app, _, _ = io.launch(prevent_thread_lock=True)