14:26:19 <br-m> <rbrunner7> With the seed command the CLI wallet gives out a legacy seed as 3 lines, like so:
14:26:20 <br-m> <rbrunner7> nestle husband lemon smelting aglow five ozone abort
14:26:20 <br-m> <rbrunner7> mobile sushi rising lopped amidst juvenile hairy evaluate
14:26:20 <br-m> <rbrunner7> rotate dabbing island hurried taken gearbox upload wedge nestle
14:26:44 <br-m> <rbrunner7> I think it's like that since seeds quite in general were introduced.
14:27:51 <br-m> <rbrunner7> What do people think: Does this formatting have value? I would like to switch to giving out the seed as one long string.
14:29:16 <br-m> <rbrunner7> Background is a problem with seed input when restoring: Before Polyseed, the CLI wallet allowed entering seeds on multiple lines. That job was easy: Just ask for line after line until all together there are at least 24 words.
14:29:41 <br-m> <rbrunner7> Now, where we have the possibility that it's a Polyseed, seems to me that's no longer possible.
14:30:21 <br-m> <rbrunner7> If a user enters 16 words on a first line, is this now a complete Polyseed, or only the first 16 words of a legacy seed?
14:31:00 <br-m> <rbrunner7> What I want to prevent: That you can't take the seed as output with the seed command and directly restore with that.
14:31:55 <br-m> <rbrunner7> Having to re-format 3 lines back into 1 would be a pain.
14:33:35 <br-m> <rbrunner7> It's of course quite a small problem, but it somehow bugs me, and I don't want to break a decade of seed output tradition just like that :)
16:34:16 <br-m> <ofrnxmr> They have different wordlists > <@rbrunner7> If a user enters 16 words on a first line, is this now a complete Polyseed, or only the first 16 words of a legacy seed?
16:34:47 <br-m> <ofrnxmr> So it should be possible to detect based on the words alone
17:00:41 <br-m> <sneedlewoods_xmr:matrix.org> IMO changing the formatting to a single line sounds good. > <@rbrunner7> What do people think: Does this formatting have value? I would like to switch to giving out the seed as one long string.
17:00:44 <br-m> <sneedlewoods_xmr:matrix.org> Relying on the wordlists sounds more complex and only works if there are no words overlapping AFAICT (did not check if there are any).
17:17:34 <br-m> <rbrunner7> Yes, that thought crossed my mind: Continuously check whether the words as of now would be a valid Polyseed, and if not stop, stop to ask for further output.
17:18:21 <br-m> <rbrunner7> But this seems to be a bit the wrong question for me. Question is "Is that 3 line formatting important, is it useful in some aspect, so we do want to keep it".
17:19:04 <br-m> <rbrunner7> Beside "It always was that" and "Never change a running system" I personally don't see any advantages. But other people may see it differently.
17:19:27 <br-m> <rbrunner7> *and if yes, stop to ask for further input
17:43:24 <moneromooo> AFAIK it was just to avoid clipping words neatly, not on random letters as the text hit the terminal width.
17:43:47 <moneromooo> AFAIK it was just to clip words neatly, not on random letters as the text hit the terminal width.
18:57:09 <br-m> <rbrunner7> I see. Yes, that's an argument to keep the multiple lines ...
19:16:27 <moneromooo> I guess you could add "..." when continued, which the parser would interpret as "wait" if present when re-read.
19:20:57 <br-m> <rbrunner7> Hmm, yes. But other apps like the GUI wallet would have to learn that as well, or they would probably tax "..." as invalid word.
19:27:51 <moneromooo> Hmm. Clever. "Invalid seed, taxing next restored wallet at 0.001%" :D
23:42:05 <br-m> <ofrnxmr> @sneedlewoods_xmr:matrix.org: Well.. you could just press "enter" after entering 16 words
23:47:04 <br-m> <ofrnxmr> Or it can ask you to choose a seed