Stuck on an issue?

Lightrun Answers was designed to reduce the constant googling that comes with debugging 3rd party libraries. It collects links to all the places you might be looking at while hunting down a tough bug.

And, if you’re still stuck at the end, we’re happy to hop on a call to see how we can help out.

Add Testing Infrastructure

See original GitHub issue

While it is easy to add a new site to scan, it is really hard to know if it is actually supported properly. Especially for the sites that do not have an explicit error code. For these sites, Sherlock relies on knowing the exact text of the informational page that the social media site presents. Which, even if it works today, could very well break tomorrow.

I am thinking that there would be a tests directory that would contain Python code that would run various tests. The main input to these tests would be a curated list of user names that are supported at a given social media site, and another list of user names that are not.

For Each Social Media Site A List Of User Names Supported Curated list of user names known to be supported.
For Each Social Media Site A List Of User Names Not Supported Curated list of user names known to not be supported.
Python Module That Runs Through List And Compares Results To Expected This would be the code that would run after changes have been made to the module. And, it would be run before a release was made.

Issue Analytics

State:
Created 5 years ago
Comments:12 (11 by maintainers)

Top GitHub Comments

1reaction

hoadlckcommented, Jan 22, 2019

Added some basic testing infrastructure with #127. I did this change without any of the restructuring that we have talked about… I am not sure exactly where that is at, and I wanted to get these tests out there so they could help people.

One thing that I do not like about the current way that the tests run is that the working directory gets littered with result files from all of the tests. I was going to change the main routine around so that the result file was optional, but I did not want to add churn before the restructuring is done.

0reactions

hoadlckcommented, Feb 10, 2019

With #169, that puts the test coverage at 100%!

Issues that exist with tests at the current time:

Some of the sites are randomly intermittent, and we get failures because Sherlock could not connect to them.
The tests hang up because some site is not responding.
Sites like AngelList which interpret the Sherlock queries coming from one IP address as some kind of bot activity.
Tor options are not tested.
Proxy options are not tested.

But, this is a good first pass. Any restructuring that we do in the future will have a complete test of site connections available.

There can be other issues added in the future to add more sophisticated options to the testing, but I think that this issue can be closed now.