We have an off-the-shelf windows application which has several Thousand licensed copies in the field.
For our next version, we would like to add a feature which relies on reading (not writing) several user_timelines and displaying individual tweets (respecting the https://dev.twitter.com/terms/display-requirements). The user may or may not already be following the timeline.
I’m having some problems finding an explanation on how to use the REST API in this case.
If I understand correctly, then using the application-only authentication (https://dev.twitter.com/docs/auth/application-only-auth) because
- if we hardcode our application key, we’ll exhaust the applicaiton limit fast because there are instances of the application running on thousands of computers.
- I don’t think the idea is that users need to create/register “their own application” and then type in the key/secret in our program.
So how does the twitter integration work in this scenario ? None of the ones on https://dev.twitter.com/docs/auth/obtaining-access-tokens seem to be an Obvious fit, or I’m just missing something.
Do I (the developer of the software) still need to create an application on twitter?
Our app runs on Windows, is written in C++, uses MFC, has a GUI.
I can launch a separate browser and provide (ShellExecute a URL), but I can’t then get feedback on what is happening in the browser.
I can make a dialog with an embedded html control (IWebBrowser2) and get some return info from there, but I don’t think that’s a really good idea, some users may (correctly) assume it’s fake.
I can make webservice calls
What are the steps needed to get a user to log into twitter and get our app to make API calls with the user’s credentials ?