Closed robertmin1 closed 1 month ago
One observation I can make is that we must transform the invoice into a paid status after sending it
The cli
currently lacks a "pay invoice"
command. My plan is to focus on implementing for GUI exclusively, unless circumstances change.
The other thing we should probably do is, make the "Check name availability..." button check whether a name_new is already in the wallet for that name (by doing the same check that the name_firstupdate command does), and if so, the UI should change to a new UI that only does name_firstupdate (as opposed to name_autoregister) -Jeremy
The
cli
currently lacks a"pay invoice"
command. My plan is to focus on implementing for GUI exclusively, unless circumstances change.
I think the name_new
RPC command will work fine for this, if you pass the destination
, commitment
, and amount
params (can you confirm this?). It won't be able to accept these params in URI format, but that's an upstream limitation and isn't our problem.
The other thing we should probably do is, make the "Check name availability..." button check whether a name_new is already in the wallet for that name (by doing the same check that the name_firstupdate command does), and if so, the UI should change to a new UI that only does name_firstupdate (as opposed to name_autoregister) -Jeremy
Is this the check I should be looking at
For the UI, I plan to open show_configure_name
with a custom is_new
parameter, such as firstupdate.
Then using this parameter to modify the UI and its functionality here.
@JeremyRand
The
cli
currently lacks a"pay invoice"
command. My plan is to focus on implementing for GUI exclusively, unless circumstances change.I think the
name_new
RPC command will work fine for this, if you pass thedestination
,commitment
, andamount
params (can you confirm this?). It won't be able to accept these params in URI format, but that's an upstream limitation and isn't our problem.
Yes, that should work, do we plug it into payto
or create an independent command?
The
cli
currently lacks a"pay invoice"
command. My plan is to focus on implementing for GUI exclusively, unless circumstances change.I think the
name_new
RPC command will work fine for this, if you pass thedestination
,commitment
, andamount
params (can you confirm this?). It won't be able to accept these params in URI format, but that's an upstream limitation and isn't our problem.Yes, that should work, do we plug it into
payto
or create an independent command?
Neither, AFAICT this should already work out of the box if you call name_new
on the console with those params present. So there's nothing left to do CLI-wise unless I've missed something.
The other thing we should probably do is, make the "Check name availability..." button check whether a name_new is already in the wallet for that name (by doing the same check that the name_firstupdate command does), and if so, the UI should change to a new UI that only does name_firstupdate (as opposed to name_autoregister) -Jeremy
Is this the check I should be looking at
Yes. Feel free to factor that out into a separate function.
For the UI, I plan to open
show_configure_name
with a customis_new
parameter, such asfirstupdate.
Then using this parameter to modify the UI and its functionality here.
Yes that sounds fine; we already have an autoregister and update code path there, adding a 3rd firstupdate codepath should be straightforward.
Alright!
The other thing we should probably do is, make the "Check name availability..." button check whether a name_new is already in the wallet for that name (by doing the same check that the name_firstupdate command does), and if so, the UI should change to a new UI that only does name_firstupdate (as opposed to name_autoregister) -Jeremy
Is this the check I should be looking at
Yes. Feel free to factor that out into a separate function.
For the UI, I plan to open
show_configure_name
with a customis_new
parameter, such asfirstupdate.
Then using this parameter to modify the UI and its functionality here.Yes that sounds fine; we already have an autoregister and update code path there, adding a 3rd firstupdate codepath should be straightforward.
If we make a separate the function, the password needs to be passed here This would mean prompting the user to enter the password for every check name availability?
If we make a separate the function, the password needs to be passed here This would mean prompting the user to enter the password for every check name availability?
Good point, I hadn't thought of that. OK, in that case, how about you instead edit the name_autoregister
function, so that it attempts to call name_firstupdate
by itself, and only does the name_new
if name_firstupdate
returns an error due to a missing input? This also improves UX for the CLI/RPC interface.
This works well, definitely easier to implement
Code review passes for f784cae4b4f8f3c4f4aee2c593603b5f9bb5da61 other than the things I noted above. Will test shortly.
Testing successful for f784cae4b4f8f3c4f4aee2c593603b5f9bb5da61 other than the things I noted above.
ACK a24ce487d331873cba24bfbee77de1ad4aafd8f9
Continuation of #363