Update graphql query hash used for stories

[Ristovski]

This updated graphql query hash returns all fields for tappable_objects. These fields were missing/partial when using the old query hash.
This allows to get the full json metadata of a story. Fixes #871

[aandergr]

That's a nice query hash, thanks!

Story-related tests still pass, and all attributes of the yielded Story and StoryItems objects are still accessible. It seems the data returned using this query hash contains all the (relevant) fields, and it includes more-complete tappable_objects information.

[Deleora]

So now is it possible to download the actuals posts shared in stories?

[aandergr]

So now is it possible to download the actuals posts shared in stories?

Yes, there is a field tappable_objects in the JSON that is now returned for each StoryItem. So with the JSON saved, something like the following will give you an id and a shortcode of a post shared in a story:

$ xzcat 2021-01-30_05-57-43_UTC.json.xz | jq .node.tappable_objects[0].media

or download it:

$ instaloader -- -`xzcat 2021-01-30_05-57-43_UTC.json.xz | jq -r .node.tappable_objects[0].media.shortcode`

However, the tappable objects are not yet exposed in the programmatic interface of StoryItem.

[Deleora]

Ohh, that explains why I wasn't seeing any such posts getting downloaded. Anyway, Thanks for one more cool feature!!

So one question, can I download such posts using the uncompressed JSON files? If so how? because I have only uncompressed JSON files.
And I didn't understand the purpose of having those backticks(`) in the command. Is it mandatory? [I have Windows 10]

[aandergr]

The backticks do what is called Command substitution, allowing the output of a command to replace the command name.

I think in Powershell and with the JSON files not compressed, the following command should do the job

instaloader -- -$(jq -r '.node.tappable_objects[0].media.shortcode' '2021-01-30_05-57-43_UTC.json')

(it requires jq to be installed)