Alan Grimes [ARCHIVE] on Nostr: 📅 Original date posted:2011-07-12 🗒️ Summary of this message: The writer is ...
📅 Original date posted:2011-07-12
🗒️ Summary of this message: The writer is facing design problems while writing a bitcoin daemon and suggests refactoring the codebase to make it easier to audit and manage.
📝 Original message:Yeah, I'm starting to run into real design problems in my attempt to
write a bitcoin daemon, namely I'm trying to figure out how to manage
connections and peers and how to interpret and coordinate messages from
peers.
While spying on the old code, I noticed one major problem that could be
fixed quite easily. That is, the 1 class-per .h/.cpp rule is completely
ignored in main.h/cpp and net.h/cpp If all of the classes in the project
were re-factored to their own files, it would be much easier to audit
what was interacting with what. I think this work could be done within
16 programmer hours.
The old codebase doesn't build on my system but I'd be willing to rough
it in anyway.
I think this should be done in parallel to efforts, such as mine, to
re-implement major functionality.
--
E T F
N H E
D E D
Powers are not rights.
Published at
2023-06-07 02:04:31Event JSON
{
"id": "a112fb58fb10f345511b07733b06275b7b5a59f34f177f3ede30c01f0e235cab",
"pubkey": "1ff44081309d1630c10d6c98e652a908562fdfc06fd3def22ca78c6c1aa9cee0",
"created_at": 1686103471,
"kind": 1,
"tags": [
[
"e",
"4eaf1439aa67825cabdea581be3956c38ea4a62226b60d5367bace64dca7945b",
"",
"root"
],
[
"e",
"6d79f723e042e61df5f2ac80a2c7f07123ac5488a342c2d8e64dc97e8b18b96b",
"",
"reply"
],
[
"p",
"6ac6a519b554d8ff726a301e3daec0b489f443793778feccc6ea7a536f7354f1"
]
],
"content": "📅 Original date posted:2011-07-12\n🗒️ Summary of this message: The writer is facing design problems while writing a bitcoin daemon and suggests refactoring the codebase to make it easier to audit and manage.\n📝 Original message:Yeah, I'm starting to run into real design problems in my attempt to \nwrite a bitcoin daemon, namely I'm trying to figure out how to manage \nconnections and peers and how to interpret and coordinate messages from \npeers.\n\nWhile spying on the old code, I noticed one major problem that could be \nfixed quite easily. That is, the 1 class-per .h/.cpp rule is completely \nignored in main.h/cpp and net.h/cpp If all of the classes in the project \nwere re-factored to their own files, it would be much easier to audit \nwhat was interacting with what. I think this work could be done within \n16 programmer hours.\n\nThe old codebase doesn't build on my system but I'd be willing to rough \nit in anyway.\n\nI think this should be done in parallel to efforts, such as mine, to \nre-implement major functionality.\n\n-- \nE T F\nN H E\nD E D\n\nPowers are not rights.",
"sig": "a60850a34ac55c83a6beb9f44fe243d728b9fba7a56a213666e6d2136c959ad10cd77b000b1004daa26dbe244cbf64176c62a942014db13dc5da640810f8aaa8"
}