290 entries in 0.635s
lobbes: ascii_field: I can confirm the same behavior on my graphical winblowz browsers (Firefox, Chrome). On Firefox, I get additional message of 'MrTAHIXE
.sig.part could not be saved, because the source file could not be read.'
ben_vulpes: armored and clearsigned body + patch + patch
.sig, right?
assbot: Logged on 04-09-2015 23:54:25; asciilifeform: mircea_popescu: here's how i do it: 1) clearsigned (asciiarmoured) body. 2) attachment 3) signature (asciiarmoured, detached) of attachment, having the filename attach.foo
.sig for every attach.foo
mircea_popescu: the "with clearsigned body and
.sig signature" version is exact replica of the single email i successfully sent to date, back in the days of the V patch.
mircea_popescu: ok, for the record : sent it with no body, and with a clearsigned body ; the latter with .asc armored signature and with
.sig binary signature. still nothing there.
mod6: so if we do <patchname>.<wotGuy>
.sig then will fail
mod6: these sigs that you posted won't verify out-of-the box, not that i've tried. but the detached sig must match the file name with the exception of the trailing
.sig on the end.
mod6: asciilifeform: I see you attached the sigs for the vpatch files, and included the tarballs in your orchestra+breath-of-life (UPDATED) email. I think it'd like to attach the raw .vpatch files as well as the
.sigs for completeness. Anything against this other than it's a bit redundant since you already posted the tarballs of the .vpatches?
mod6: <+asciilifeform> mod6: i recommend to follow the format nameofpatch.vpatch.mod6
.sig << cool, sounds good.
jurov: mircea_popescu: only clearsigned text is mandatory, there can be none or any number of further attachments(not necessarily only patches), each with detached attachment.name.ext
.sig ascii_field: mircea_popescu: third, if the second exists - a detached
.sig for it
hanbot: 3197e61aff68e988c611b6ef4c4d7092645804a800112435089da3af1a9ef2bea8fb6198de3f0ab17483e601cc1ce1a0e0afd4132607bdc7f6ef5d08953e1f83 ak47_05243673e45fc4c4ce30467f8ffc003deec1d184.sh
.sig mod6: `gpg --verify stator.tar.gz
.sig`
mod6: then `mv stator_7447d6ad798179d04e9d277acb72799b3c7d0eae.tar.gz stator.tar.gz` and `mv stator_72424e6da0f81aea5ab09165c377fd8b7418983f.tar.gz
.sig stator.tar.gz
.sig`
mod6: once pulled down do this `sha1sum stator_7447d6ad798179d04e9d277acb72799b3c7d0eae.tar.gz stator_72424e6da0f81aea5ab09165c377fd8b7418983f.tar.gz
.sig` and ensure that the output hashes match what's embedded in the file name.
punkman: well if it doesn't reject files that don't match
.sig, I could just put them in a tar.gz
punkman: also works if I download
.sig and --verify it
jurov: if you sent lenna_full.jpg and lenna_full.jpg
.sig it should add them to list, too
jurov: mats as the
.sig file verifies, the hash must be therein
mod6: so after review you just expect or perhaps exepect to just see a
.sig file attached alone without the patch itself?
jurov: then why are all .tar.gz
.sig?
danielpbarron: the patch
.sig is by ascii's key, so it shouldn't matter what it hashes to or what btc-dev says
mod6: had 2 attachments, a .patch file and a
.sig file
decimation: ~/bitcoin-v009/build/chicken/bitcoin-asciilifeform.1.patch
.sig mod6: hmm, still, should be ok. i just re-ran on my side and got: >> Signature verified for bitcoin-asciilifeform.1.patch
.sig mod6: so you're saying that this doesn't work: gpg --logger-file alert-snip
.sig.log --homedir .gnupg
mod6: - --verify chicken/bitcoin-asciilifeform.1.patch
.sig mod6: gpg --logger-file chicken
.sig.log --homedir .gnupg \
jurov: it will make blablabla...hexporn.patch
.sig ben_vulpes: i'd really like punkbot to take in the raw doc and
.sig files. but...whatever.
mod6: so in this case, ben
.sig & mod6
.sig, each yields a PGP armored message, that's just a signed file that can be verified.
mod6: each party will want to name ``out
.sig'' something different obv.
ben_vulpes: if each party runs gpg --output out
.sig --sign --armor thing-to-sign.txt, you can run gpg --decrypt out
.sig on the blob
ben_vulpes: jurov: why bother checking body? instead maybe just check .patch
.sig for l2 sig, and compare that to the file.
ben_vulpes: mircea_popescu, asciilifeform: re sig stack is the implication that a signs patch, b signs patch and a
.sig, c patch and b
.sig...
jurov: and whoever want to sign it, sends patch-author-id
.sig too
jurov: thus ... for every attachment named patch-author-id attach also patch-author-id
.sig punkman: Azelphur: try this: git config --global user
.signingkey FFFFFFFF
mjr_: because it would be awesome to be able to contribute to a business that you are invested in..
.signing up helps your actual investment