In building up a complicated mock type, there are often cases where we want to test "the same thing, but with one key different."

In this case, we want to test a valid agora tx, which we call BASE_AGORA_TX. All of the "default" keys here, I am also prefixing with "BASE_"

This is not any kind of official convention, it's just something I started doing for chronik-client tests and have continued here.

That way, if we want to test "agora tx that is kind of like the base one but a little different," we can do

{...BASE_AGORA_TX, someKey: 'some value that is not the base value'}

This is the logic I am using in composing the tests. So, generally, there may be alt values ... but usually they are on-off and do not get their own variable.

It's a little more complicated for these tests because we can't just store an agora tx as an object, since it includes special types. We have to recreate the Script for each test. This is why the tests here are all built in this file instead of pulled from vectors.

For now, the function is expected to overwrite existing Map entries. myListings and offeredListings are never cached, they are state variables that depend on the current lokadId tx history for agora txs.

They are re-calculated every time the NFT page is loaded (or every time the tx history changes, see NFT market diff).

Potential optimization to cache these but would be handled in a later diff. Tricky to handle though. Lokad Id tx history won't change, but an "old" lokad id tx can change -- i.e. it is unspent, but becomes spent. So you can't just cache the lokad history. Need to check something on every page load or have a chronik plugin that indexes only the unspent offers.

Probably not. This is pretty technical dev documentation. Only people who would want or need to know are devs.

For devs, this is already documented (well enough) by the integration tests in ecash-agora.

Cashtab users don't need to worry about this.

😐