Download files from subfolders doesn't work

[setoff]

Tried the same code:

from kaggle.api.kaggle_api_extended import KaggleApi

api = KaggleApi()
api.authenticate()

api.competition_download_file(
        <competition-name>, 
        'dataset/dataset_files/filename.ext' 
        path='local_path'
    )

Version 1.6.17 downloads file successfully.

Version 1.7.4.2 returns this error:

404 Client Error: Not Found for url: https://www.kaggle.com/api/v1/competitions/data/download/<competion-name>/dataset/dataset_files/filename.ext

[stevemessick]

@setoff I'm having trouble reproducing this. Can you give me an example using the command-line tool that works with 1.6.17? Thanks.

[github-actions]

Without additional information we're not able to resolve this issue, so it will be closed at this time. You're still free to add more info and respond to any questions above, though. We'll reopen the case if you do. Thanks for your contribution!

[setoff]

Sorry for the late reply.
You may use this one:
kaggle competitions download --file kaggle-environments-0.1.4/kaggle_environments/static/player.html -c connectx

With Kaggle API 1.6.17 downloads file successfully.
With Kaggle API 1.7.4.2:

404 Client Error: Not Found for url: https://www.kaggle.com/api/v1/competitions/data/download/connectx/kaggle-environments-0.1.4/kaggle_environments/static/player.html

[setoff]

@stevemessick added CLI to download file from one of the open competitions https://www.kaggle.com/competitions/connectx/overview

You need to join competition first and accept rules before running command I added in message above.

[stevemessick]

@setoff Thanks! I'll look into this.

[stevemessick]

This should be fixed in the latest release, last week.

[MelonLord8]

The issue still persists now:

k_api.dataset_download_file(
                dataset="allen-institute-for-ai/CORD-19-research-challenge",
                file_name='Kaggle/target_tables/0_table_formats_and_column_definitions/column_definitions.docx',
                path="cache/",
                force=False
            )

(with all necessary imports)

returns this error:

Traceback (most recent call last):
  File "d:\StockDataPrep\main.py", line 29, in <module>
    k_api.dataset_download_file(
  File "D:\StockDataPrep\.venv\lib\site-packages\kaggle\api\kaggle_api_extended.py", line 1617, in dataset_download_file
    response = kaggle.datasets.dataset_api_client.download_dataset(request)
  File "D:\StockDataPrep\.venv\lib\site-packages\kagglesdk\datasets\services\dataset_api_service.py", line 80, in download_dataset
    return self._client.call("datasets.DatasetApiService", "ApiDownloadDataset", request, HttpRedirect)
  File "D:\StockDataPrep\.venv\lib\site-packages\kagglesdk\kaggle_http_client.py", line 126, in call
    response = self._prepare_response(response_type, http_response)
  File "D:\StockDataPrep\.venv\lib\site-packages\kagglesdk\kaggle_http_client.py", line 191, in _prepare_response
    http_response.raise_for_status()
  File "D:\StockDataPrep\.venv\lib\site-packages\requests\models.py", line 1026, in raise_for_status
    raise HTTPError(http_error_msg, response=self)
requests.exceptions.HTTPError: 404 Client Error: Not Found for url: https://www.kaggle.com/api/v1/datasets/download/allen-institute-for-ai/CORD-19-research-challenge/Kaggle/target_tables/0_table_formats_and_column_definitions/column_definitions.docx?filename=Kaggle%2Ftarget_tables%2F0_table_formats_and_column_definitions%2Fcolumn_definitions.docx&raw=false

[HardRockDude]

Confirming this issue on Kaggle API 1.7.4.5.

I have a private dataset with the following structure:

.../<user>/<dataset>/dat_feat/dat_feat/*.npy
.../<user>/<dataset>/dat_lab/dat_lab/*.npy

with *.npy meaning several npy files in that last folder.
Now, I'd like to download all npy files only from dat_lab.
This is my Google Colab example:

from google.colab import files
files.upload()  # select kaggle.json

!pip install -q kaggle
!mkdir -p ~/.config/kaggle
!mv kaggle.json ~/.config/kaggle/kaggle.json
!chmod 600 ~/.config/kaggle/kaggle.json

import kaggle
from kaggle.api.kaggle_api_extended import KaggleApi
api = KaggleApi()
api.authenticate()

dataset = "<user>/<dataset>"
files = api.dataset_list_files(dataset).files
for f in files:  # first file is dat_lab/dat_lab/abc.npy
    if f.name.startswith("dat_lab/"):
        api.dataset_download_file(dataset, f.name, path="/content")

This immediately throws the error:
HTTPError: 404 Client Error: Not Found for url: https://www.kaggle.com/api/v1/datasets/download/<user>/<dataset>/dat_lab/dat_lab/abc.npy?filename=dat_lab%2Fdat_lab%2Fabc.npy&raw=false

(Also I'm aware that api.dataset_list_files only returns the first 20 files. I have a solution for that, but this is not the issue here)