requested changes from commit after issue #5

rushilsrivastava / image_search

Python Library to download images and metadata from popular search engines.

https://pypi.org/project/image-search/

MIT License

126 stars 34 forks source link

requested changes from commit after issue #5 #8

Open loopyd opened 6 years ago

loopyd commented 6 years ago

from issue #5:

includes the hash argument as an option with affecting the url argument's functionality (you can still use the program as before, but now with the addition of --hash yes if you wish
includes the error counter you voiced you liked, as a property of download_image
slight output cleanup - errors where being printed twice in your version (unnecessary), exception is now passed directly once to error() and printed optimally
no extra config file needed, this one directly uses properties of download_image

loopyd commented 6 years ago

FYI, your change to use io.open instead of open produces the same error i was intending to fix in python 2.7 with my structural workarounds. be advised you will have to patch this. i left it intact to show you why this occours. (check travis)

The correct method for 2.7, is:

        with io.open('dataset/logs/google/source.html', 'w', encoding='utf-8') as f:

The python 3 syntax works and should be left as it is.

Sorry there where two here, forgot the encoding specifier.

rushilsrivastava commented 6 years ago

Will start the review when build checks with Travis CI.

rushilsrivastava commented 6 years ago

Opening the file with utf-8 encoding should be enough, correct? Could you give me an example to test with?