Notes

Switched from Jekyll to Hugo

For my EasyBlogPosting project.. Testing out the Dropbox API.. 2020-12-09 trying this out Per here … pip install dropbox Went w/ that Getting Started page Through that page, I created a new test app for myself. Using the API explorer … I hit Get Token .. https://www.dropbox.com/oauth2/authorize?response_type=token&client_id=xxxxxxxxx&redirect_uri=https%3A%2F%2Fdropbox.github.io%2Fdropbox-api-v2-explorer%2F&state=file_requests_list!39%2B%2Bxxxxxxxx&token_access_type=online& That generated a token and I used https://dropbox.github.io/dropbox-api-v2-explorer/#files_list_folder and I was able to see actual folders in my account. That’s cool except I thought I had chosen something like ‘single folder access’ so I’m slightly confused why I can see the other folders. Maybe the generated token in the API explorer is separate and meant for testing. Hopefully it’s short lived. 2020-12-13 oh my special app folder was created w/ the acess token As soon as on my app page I hit “Generate access token” , my special App folder was created. import dropbox import os print("Initializing Dropbox API...") dbx = dropbox.Dropbox(os.getenv('DBX_ACCESS_TOKEN')) result = dbx.files_list_folder(path="") BadInputError: BadInputError('xxxxxxxxxxx', 'Error in call to API function "files/list_folder": Your app is not permitted to access this endpoint because it does not have the required scope \'files.metadata.read\'. The owner of the app can enable the scope for the app using the Permissions tab on the App Console.') Ah ok so I went to the Permissions tab, checked that ‘files.metadata.read’ box and tried again. Now seeing AuthError: AuthError('xxxxxxxxxxx', AuthError('missing_scope', TokenScopeError(required_scope='files.metadata.read'))) Ah according to stackoverflow , the scope cannot be retroactively granted to a token created before the scope was granted. Ok … trying to re-create a new token… Ok recreated. Retried. Worked! How to read? What’s the difference between the two files/download , files/export ( files/get_preview ) print("Initializing Dropbox API...") dbx = dropbox.Dropbox(os.getenv('DBX_ACCESS_TOKEN')) meta, response = dbx.files_download('/blarg.txt') # In [22]: response.text # Out[22]: 'foo blarg\n' Github interaction Hmm I was going down the path of building an AWS Batch container to commit to my git repo But hmmm I encountered that there may be a Github API which I can perhaps use??? And a nice tutorial Cool, so https://api.github.com/users/namoopsoo/repos , repos are here. However the oauth flow , looks kind of involved for a quick script But ah that’s right you can create a token for your account from here, Github settings . Ok cool I was able to perform a simple curl w/ the personal token I created for myself … ( per docs ) $ curl -v -H "Authorization: token ${GITHUB_TOKEN}" https://api.github.com/users/namoopsoo/repos ok but can your write using this API is the big question.. Hmm according to here you can commit. post /repos/{owner}/{repo}/git/commits But not seeing how to include the actual diffs in there. Maybe a blob? Ok according to this nice article , indeed the blob endpoint is for files! hmm step 1, get a reference.. like here curl -v -H "Authorization: token ${GITHUB_TOKEN}" https://api.github.com/repos/namoopsoo/namoopsoo.github.io/git/ref/heads/master Nice excellent, … ...