selsta: git clone github.com/monero-project/monero.git && cd monero && git submodule init && git submodule sync && git submodule update && git checkout v0.17.2.0 && git submodule update # this fails, only --recursive on the update or the clone would fix it

lru: you have to do ?git submodule sync`

sorry, typo

I did, see above :-)

you have to do `git submodule sync`after you checkout a branch / tag.

not before

ok, let me try

yep, order matters, thanks

`git submodule sync` and `git submodule update --init --force --recursive` usually solve all submodule issues

I see the .gitmodules file changed between master and v0.17.2.0, which makes sync make sense now

yep, we changed a submodule to upstream

<selsta> sometimes the window doesn't close; other times, if I leave it a few seconds, it "closes" but continues to run in the background

or if I click the 'x' button twice. I'm not sure how this user would have had an instance running unknowningly without something similar happening, but perhaps

Lyza: It COULD be related to this, since you're talking about closing the app:

This and the linked PRs.

I don't know how much into C++ you are, but destructors are being called when an application exits.

And Undefined Behavior earns its name.

<NikolaIgnatovNed> "stopwatch" <- Nice try.

mj-xmr[m]: it's a GUI issue

"<mj-xmr[m]> I don't know how much into C++ you are" just enough to be dangerous :D

but yeah it's a GUI issue tho

^ proposal on deprecating integrated addresses due to their bottlenecking of batching withdrawals and other reasons, comments would be appreciated.

Triptych embedded data could be used to store up to 3 of those per ring.

Isn't it only per tx?

+1 for anything which makes the network more homogenous

IMO the major reason would be UX, the concept of integrated address is difficult to explain to new users

especially when they send to an integrated address and the base address appears somewhere in the wallet and they think they sent to the wrong address

<selsta> "especially when they send to..." <- The UX for this is *awful* as a Ledger user.

The Ledger displays the non-integrated address, so it never matches.

I would love to see IAs deprecated/removed in favor of subaddresses once and for all.

As said in the github proposal, it probably makes sense to embed a "subaddress only" plan in with whatever the eventual "big ring upgrade" ends up beinh

As that will require big upgrades across the ecosystem regardless

I just finished making a payment processor for my website that uses integrated addresses, because I read that integrated addresses were the more performant option when tracking many payments as once.

It was also easier to implement, because I didn't need to synchronize any counters between threads to keep track of which subaddresses are in use at any point in time.

Do the downsides of integrated addresses outweigh these advantages?

carrington: Seth your comments on the issue would be appreciated, otherwise it'll get lost in the noise.

* carrington: Seth your comments on the github issue would be appreciated, otherwise it'll get lost in the noise.

Yes, sorry, will comment there!

> <@busyboredom:monero.social> I just finished making a payment processor for my website that uses integrated addresses, because I read that integrated addresses were the more performant option when tracking many payments as once.... (full message at libera.ems.host/_matrix/media/r0/do…6b6898aad446bc338d36216e64be02f22e6)

Fair enough, I'm happy to switch my project to subaddresses if that really is the recommended route now :)

Or if it ends up being the recommended route

It is the recommended route, and has been since it appeared on mainnet.

Sounds good, subaddresses it is then. Side note then since it seems its info might be outdated, monerodocs is where I got my info when deciding whether to use integrated addresses or subaddresses: monerodocs.org/public-address/integrated-address

is this normal: Exception: boost::wrapexcept<boost::bad_weak_ptr>

found in the logs