This is the mail archive of the
mailing list for the Cygwin project.
- From: Achim Gratz <Stromeko at nexgo dot de>
- To: cygwin-apps at cygwin dot com
- Date: Mon, 03 Aug 2015 21:07:16 +0200
- Subject: Re: setup
- Authentication-results: sourceware.org; auth=none
- References: <87r3nsw0nb dot fsf at Rainer dot invalid> <20150729125127 dot GA8328 at calimero dot vinschen dot de> <87y4hv0zwv dot fsf at Rainer dot invalid> <87vbcwfdvw dot fsf at Rainer dot invalid>
Achim Gratz writes:
> I've just found out that MD5 checksums get parsed OK, but wind up as
> solid zeros in the package database in setup. I have no idea how this
> is even possible, since the exact same code works just fine for SHA512
> checksums. I didn't check yet if this actually works in the release
> version of setup. An extra pair of eyeballs would be appreciated. I
> don't see how my changes could be responsible and I was sure I'd checked
> it beforeâ :-(
Found it, I feel stupid nowâ
> Also, I want to clean up the parser code a bit and fix a very old
> formatting bug found while debugging the problem with the MD5 checksums.
> Another thing: it seems that the SHA512 digest is checked for existence
> by checking that the first byte is non-zero. Is that actually a
> guarantee that the SHA2 standard makes or is it just very unlikely to
> encounter 0x00 as the first byte?
If the SHA512 checksum is all zeros in the package DB it doesn't get
checked, which is actually the reason I didn't catch the error earlier.
Have to fix that also.
+<[Q+ Matrix-12 WAVE#46+305 Neuron microQkb Andromeda XTk Blofeld]>+
SD adaptations for Waldorf Q V3.00R3 and Q+ V3.54R2: