When using base64, if your input size is not a multiple of 3, you are guaranteed to have at least one symbol, the = used for padding.
On Sat, Nov 12, 2016, 5:45 PM Kenny Evitt <[email protected]> wrote: > You're right, and you provided me with a nice little (fun) project > yesterday. I asked a question on the statistics Stack Exchange site > <http://stats.stackexchange.com/questions/245610/not-generating-a-valid-password> > about how likely it is for those commands to fail to generate a password of > sufficient length. While writing up the question I also tested it. In the > 12,000 times I ran the commands only *one* execution failed to generate a > password of less than 32 characters (really, 130 characters). The one > failure was actually the 'worst' possible failure – the password was empty! > But given a success rate of 11,999 out of 12,000, the commands seem even > more likely to be good enough for a lot of cases. > > Regarding the larger discussion, I think options or settings to specify an > external password generator, both as a default and as an argument for > generating a specific password, would be both useful and keeping with the > relative compactness of Pass's features. > > On Fri, Nov 11, 2016 at 4:41 AM, Kevin Cox <[email protected]> wrote: > > On 11/11/16 09:26, Henrik Christian Grove wrote: > > Den 10-11-2016 kl. 16:14 skrev Kevin Cox: > > > On Nov 10, 2016 14:43, "Kenny Evitt" <[email protected] > <mailto:[email protected]>> wrote: > > I then edited "test.txt", inserted a newline between the `4` and > `5`, and saved the file. I reran the same command as before: > > > Head be default takes 10 lines. So you'll need a couple more newlines. > > By default, yes, but he explicit specifies '-c 10' overriding the > default, and tr will have removed the newlines before head gets them. > Can you actually demonstrate the bug you claim to be there? (I'm not > sure no kind of nastiness can hide in those pipes) > > .Henrik > > PS. The lack of a way to choose passwords without/with only certain > characters is one of the reasons I still haven't switched to pass (using > pwsafe), even though I like the project. > _______________________________________________ > Password-Store mailing list > [email protected] > http://lists.zx2c4.com/mailman/listinfo/password-store > > > I'm talking about the first head, not the second. This command gives you > 12345 when you expected 12345678. > > echo '12345 > > > > > > > > > > 67890' | head | tr -dc '0-9 | head -c 8 && echo > > A fixed command looks like this (drop the unnecessary head) > > tr -dc 'A-Za-z0-9!"#$%&'\''()*+,-./:;<=>?@[\]^_`{|}~' </dev/random | head > -c 32 && echo > > > _______________________________________________ > Password-Store mailing list > [email protected] > http://lists.zx2c4.com/mailman/listinfo/password-store > > > _______________________________________________ > Password-Store mailing list > [email protected] > http://lists.zx2c4.com/mailman/listinfo/password-store >
_______________________________________________ Password-Store mailing list [email protected] http://lists.zx2c4.com/mailman/listinfo/password-store
