'[Errno 13] Permission denied' - Jupyter Labs on AWS SageMaker

Question

我在AWS SageMaker上使用Jupyter实验室实例。

内核：conda_mxnet_latest_p37。

url_lib包含一些错误的url，我会对此进行异常处理。

['15', '259', '26', '58', 'https://imagepool.1und1-drillisch.de/v2/download/nachhaltigkeitsbericht/1&1Drillisch_Sustainability_Report_EN_2018.pdf', 'https://imagepool.1und1-drillisch.de//v2/download/nachhaltigkeitsbericht/2018-04-06_1und1-Drillisch_Sustainability_Report_eng.pdf', '6', 'http://youxin.37.com/uploads/file/1556248045.pdf', '80', 'https://multimedia.3m.com/mws/media/1691941O/2019-sustainability-report.PDF', 'https://s3-us-west-2.amazonaws.com/ungc-production/attachments/cop_2020/483648/original/GPIC_Sustainability_Report_2020__-_40_Years_of_Sustainable_Success.pdf?1583154650', 'https://drive.google.com/open?id=1_dnBcfXWjexy9QoWRhOk_3gnOkWfYRCw', 'http://aepsustainability.com/performance/docs/2020AEPGRIReport.pdf']  # sample

然而，那些正在工作的URL会抛出以下错误：

[Errno 13] Permission denied: '/data'

我没有打开目录，也没有文件，因为我没有下载它们。

我在终端运行时运气不佳：

sh-4.2$ chmod 777 data
sh-4.2$ chmod 777 data/
sh-4.2$ chmod 777 data/gri
sh-4.2$ chmod 777 data/gri/

代码：

import pandas as pd
import opendatasets as od
import urllib
import zipfile
import os
csr_df = pd.read_excel('data/Company Sustainability Reports.xlsx', index_col=None)
url_list = csr_df['Report PDF Address'].tolist()
for url in url_list:
try:
download = od.download(url, '/data/gri/')
filename = url.rsplit('/', 1)[1]
path_extract = 'data/gri/' + filename
with zipfile.ZipFile('data/gri/' + filename + '.zip', 'r') as zip_ref:
zip_ref.extractall(path_extract)
os.remove(path_extract + 'readme.txt')
filenames = os.listdir(path_extract)
scans = []
for f in filenames:
with Image.open(path_extract + f) as img:
matrix = np.array(img)
scans.append(matrix)
# shutil.rmtree(path_extract)
os.remove(path_extract[:-1] + '.zip')
except (urllib.error.URLError, IOError, RuntimeError) as e:
print('Download PDFs', e)

输出：

Download PDFs list index out of range
Download PDFs list index out of range
Download PDFs list index out of range
Download PDFs list index out of range
Download PDFs <urlopen error [SSL: CERTIFICATE_VERIFY_FAILED] certificate verify failed: Hostname mismatch, certificate is not valid for 'imagepool.1und1-drillisch.de'. (_ssl.c:1091)>
Download PDFs <urlopen error [SSL: CERTIFICATE_VERIFY_FAILED] certificate verify failed: Hostname mismatch, certificate is not valid for 'imagepool.1und1-drillisch.de'. (_ssl.c:1091)>
Download PDFs list index out of range
Download PDFs [Errno 13] Permission denied: '/data'
...

如果还有什么我需要澄清的，请告诉我。

Answer 1

download有一个正斜杠/作为保存目录的第一个字符(第二个参数(。我删除了这个：

download = od.download(url, 'data/gri/')

输出：

...
Downloading http://youxin.37.com/uploads/file/1556248045.pdf to data/gri/1556248045.pdf
450560it [00:02, 207848.59it/s]
...