decode encrypted root.App.main.context.dispatcher.stores

[Rogach]

WIP decryption code for issue #1246

However this doesn't fix the issue completely, since QuoteSummaryStore is not found in the decrypted data - probably the data is now stored in some other place.

[Rogach]

I was using an incorrect ticker value - with correct ticker value scraping seems to work without problems (for my limited use-cases).

[Rogach]

But there seems to be some small differences - for example, if previously yfinance.Ticker("^IXIC").info["regularMarketPrice"] returned the value directly, now it returns a dict {'raw': 10705.414, 'fmt': '10,705.41'}.

[havedill]

Confirming that i loaded this into my environment with pip install 'yfinance @ git+https://github.com/ranaroussi/yfinance.git@refs/pull/1253/head'

It is working, but I agree I had to define ['raw'] in a decent number of places.

[jesusignazio]

pycryptodome is now a dependency as well.

[Rogach]

pycryptodome is now a dependency as well

I'm not sure where to add it, so I need some help from the maintainers (maybe in setup.py? but what version boundary should I use?).

[Rogach]

Ah, so the "raw" troubles were due to me deleting bunch of code that was actually needed. Sorry about that.

[ValueRaider]

I've just pushed a change to fix returned data structure, now the tests are passing. No worries @Rogach

Minimum version should be 3.6.6 or later because older contain a AES decryption bug. I'll update setup.py etc, then give others a chance to question.

[ValueRaider]

One more thing @Rogach . This also needs to be backported to branch release/0.1 (utils.py get_json()), because branch main is intended for the next big release. I don't mind doing this, I just want to give you chance in case you want to be in commit log. You will still get full credit in PRs and release notes.

[ValueRaider]

Regarding cryptography vs pycryptodome, is there any strong argument to choose one over other? cryptography has the high-level interface but this solution requires the low-level; it has larger community but pycryptodome has been free of known CVEs for longest time (unless pycryptodome has more unknown CVEs from less eyes).
@fredrik-corneliusson

EDIT: I have a Discussion pending in pycryptodome, hopefully they can answer because I can't find a decent answer anywhere.

[galashour]

Many thanks guys, this is much appreciated.

Is a new 'fixed' release is expected soon, or should we manually update a specific file on our local setup?

[ValueRaider]

Wait for a fix, it will come soon. We just need to decide with crypto module to use.

[fredrik-corneliusson]

Regarding cryptography vs pycryptodome, is there any strong argument to choose one over other? cryptography has the high-level interface but this solution requires the low-level; it has larger community but pycryptodome has been free of known CVEs for longest time (unless pycryptodome has more unknown CVEs from less eyes). @fredrik-corneliusson

Sorry for causing confusion, I worked on this fix before I saw @Rogach already had PR.
I only made my version as I had issues installing the crypto lib I thought the code in the comment used (required compiling).
But now that @Rogach has a working PR and pycryptodome seems to install on windows without issues my branch can be scrapped if there is no other compelling reason to use cryptography lib.
I committed the code anyway in my repo but I was not planning to make a PR for it, forgot how good GitHub is at sniffing out commit messages in forks.

[ValueRaider]

I decided to put both in the code but defaulting to cryptography just so I can get a release out. Will tidy when I get a good answer.

[Rogach]

@ValueRaider Sorry, was away from computer. Thanks for backporting the code!