ABI Type encoding support #238
Conversation
If the But, what I think should actually happen is that |
There was a problem hiding this comment.
Good work, I really like the use of an abstract base class for Type and the encode and decode methods on each type.
My comments are mainly about implementation and some minor interface improvements.
Additionally, could you add docstring comments to the encode and decode methods of each subclass? The reason being each subclass accepts and returns different types for these methods, so it would be good if we can document them all.
* Add initial support for ABI types * Finish up abi types and values * Add doc strings * Refactor and finish encoding functions * More cleanup * Fix minor bugs and add number decoding * Bump version to 1.8.0 and add changelog * Add change to changelog * Add decoding for arrays and dynamic types * Add more unit tests * Minor change from PR comment * Another small PR comment change * Split up files and remove circular imports * Address PR comments * Have arrays accept bytes and add docstrings * Minor clean up * Fix error messages for negative uint * Remove unnecessary imports * Refactor out abi_types since it is implicit by the class * Tuples don't need static lengths... * Address PR comments 1 * Fix head encoding placeholders * Fix the tuple docstring * Formatting fixes Co-authored-by: Brice Rising <[email protected]>
This PR adds ABI encoding support for the Python SDK:
abisupport for types, e.g.from algosdk.abi import UintTypeCloses #228