--- Log opened Mon Mar 17 00:00:47 2025
05:21 -!- HumanG33k [~HumanG33k@82.66.65.160] has quit [Ping timeout: 265 seconds]
05:21 -!- b10c [~quassel@user/b10c] has quit [Quit: https://quassel-irc.org - Chat comfortably. Anywhere.]
05:22 -!- HumanG33k [~HumanG33k@82.66.65.160] has joined #secp256k1
05:23 -!- b10c [~quassel@static.33.106.217.95.clients.your-server.de] has joined #secp256k1
05:23 -!- b10c [~quassel@user/b10c] has changed host
05:47 -!- jamesob15665 [~jamesob@pool-108-44-244-6.clppva.fios.verizon.net] has joined #secp256k1
06:07 -!- jon_atack [~jonatack@user/jonatack] has joined #secp256k1
06:09 -!- jonatack [~jonatack@user/jonatack] has quit [Ping timeout: 260 seconds]
07:10 < josie> regarding function argument names, when passing a pointer to a fix sized array, in some places the size is added to the var_name (e.g., unsigned char* var_name32). in the silent payments module, we've ended up using a mix of adding the size suffix and not adding it. is that a convention we want to follow? iirc, there was some discussion awhile back on whether or not the size of the array could
07:10 < josie> be specified directly (e.g., unsigned char *(var_name[32])), but i cant remember what the outcome of that discussion was 
10:26 < roconnor> something like #define SECP256K1_ARRAY(typ, len, nam) typ[static len] nam
10:27 < roconnor> with other definitions for systems that don't support type[static len] syntax.
10:27 < roconnor> But the open tasks was to double check that everything would be ABI compatible.
11:47 < sipa> IIRC *if* that approach is acceptable (because ABI compatibility), we can introduce it everywhere in the library without it being a breaking change
11:47 < sipa> in which case it can be done once, orthogonally to the introduction of new features (like silementpayments) that would use it
11:50 < sipa> but if it's not acceptable to do that, then perhaps it's preferable that new APIs use pointer-to-array-of-known-size (which would be an API break, so not something we can just do without hard break)
13:29 < nickler> see also here for the discussion (should probably be an issue): https://github.com/bitcoin-core/secp256k1/pull/1479#discussion_r1846865805
--- Log closed Tue Mar 18 00:00:52 2025