Christopher Faylor | 31 Jul 23:40 2014

Withdrawing from the project

As I mentioned in the cygwin mailing list: I'm withdrawing from the

I'm sending a separate note here with information about the cygwin-apps
part of this decision.

The upset perl script lives in ~cygwin/setup.  This is also a git
repository.  When you push to it, it updates the active version of upset
so be careful.  The companion Makefile which runs upset is also in this

Scripts for keeping cygwin alive:

  Keeps the mirror list alive, creates the mirrors web page, and mirrors
  list file.  Check cygwin-admin crontab file for how it is run.  Someone
  should change the crontab to have it send email to them.  It used to
  notify me and I temporarily just changed it to sourcemaster@...
  Adds a new user when fed "SSH key for upload access" mail.
  I've been running this script as root.  I don't remember why
  now.  Presumably it can be run as the just-created cygwin-admin
  user.  If this actually does need root you'll need to coordinate with

If I missed something look in /sourceware/infra/bin or in the
cygwin-admin crontab.

I'd be happy to receive well wishes or answer questions about my
decision in private email (me period cgf period cx) but please don't
(Continue reading)

Tony Kelman | 26 Jul 10:00 2014

[ITA] p7zip


I noticed that all of Chuck Wilson's packages are now listed as orphaned. 
I'm particularly interested in getting one of them, p7zip, uploaded to the 
64-bit distribution. I sent a few messages to the main list in March 
discussing what I needed to change in order to get the existing cygport to 
build in 64 bit Cygwin. I'm willing to adopt the package, if someone can 
guide me through the rest of the process.

Let me know if I'm doing this right. Package files are uploaded here:

I put the source up in a GitHub repository so you can see the two small 
tweaks I had to make:
One required repackaging the upstream source tarball to fix line endings on 
a man file (so a patch would apply correctly), the other was to switch to a 
different makefile fragment to use a C version of a CRC routine instead of 
an architecture-dependent assembly version. There's a 64 bit assembly 
version of the same routine included in the source, but I could never get it 
to work reliably. I reported the problem upstream and never got a response:

Let me know if any comments or anything else I need to do.


P.S: For fun and to see if it would work, I used the AppVeyor CI service to 
build this. You can see the log here 
(Continue reading)

Yaakov Selkowitz | 22 Jul 18:12 2014

man-db: cache handling


There are a few issues with the current man-db setup:

1) the cache is not strictly necessary for operation of man(1);

2) mandb -c is quite slow with a large number of manpages (~20min on my

3) generating the cache only during man-db postinstall doesn't help for
manpages updated or added between man-db releases.

Therefore, we need to discuss further how to best handle the man-db
cache.  In the meantime, to help avoid user complaints, could you ship a
new release without the postinstall/preremove?


Achim Gratz | 18 Jul 06:39 2014

[ATTN maintainer] mined

The mined directory for both architectures contain temporary files that
apparently sftp has left there (they have the suffix SftpXFR.<PID> I
think).  These should be removed, but maybe the transfer script that
moves the files from the upload area could be extended to ignore or
delete such files so they never get distributed to the mirrors.


+<[Q+ Matrix-12 WAVE#46+305 Neuron microQkb Andromeda XTk Blofeld]>+

Factory and User Sound Singles for Waldorf rackAttack:

Christian Franke | 14 Jul 19:55 2014

ruby-rake-10.0.4-1: minor bug in rake.1.gz

Uncompressing the rake man page results in an error message:

$ cygcheck -f /usr/share/man/man1/rake.1.gz

$ gzip -t /usr/share/man/man1/rake.1.gz
gzip: /usr/share/man/man1/rake.1.gz: unexpected end of file

This does not affect the "man rake" command. The uncompressed data is 

There is only a bogus '\n' at the end of the .gz file:

$ od -t x1 rake.1.gz
0002520 bf df c8 48 f4 e4 0a 00 00 0a
.......... ----CRC---- ---SIZE----

$ head -c -1 rake.1.gz | gzip -t

Probably an upstream issue. The rake.1.gz file is also included in the 
source package.


Dave Kilroy | 9 Jul 22:08 2014

Missing enscript dependency

Hi all,

Just tried to use enscript for the first time in a while, and noticed 
that I no longer have an lpr...

Given that `enscript foo.txt` prints to the default printer, it seems 
that enscript is missing a dependency on cygutils-extra.



Andrew Schulman | 8 Jul 03:14 2014

cygport upload command?

Using cygport, I think that packaging has become quite easy now.  At least,
once the cygport script is built and working, updating a package to a new
release is as easy as updating the version number in the cygport script,
then running 'cygport ... download all'.

Except for one thing: the upload step.  Maintainers still have to go
through the procedure at  Although it's
easier than the old manual upload process, it's still a little tedious and

Yaakov, would you consider adding an 'upload' command to cygport, that
would handle the uploading details?  That would take away the last bit of
manual work in a routine package update.

If you don't have the time or interest to add an upload command yourself,
would you consider a patch?  I've spent some time looking through cygport
code, and I might have time to give it a try, or maybe someone else would.


Mark Hessling | 3 Jul 23:58 2014

SSH key for upload access

Name: Mark Hessling
Package: regina-rexx
Comment: "2048-bit RSA, converted by mark <at> Windows7-64bit from OpenSSH"

Jason Tishler | 1 Jul 02:57 2014

Re: [SECURITY] python, python3


On Tue, Jun 24, 2014 at 11:38:35AM -0500, Yaakov Selkowitz wrote:
> On 2014-06-24 09:42, Yaakov Selkowitz wrote:
> >A security vulnerability (CVE-2014-4616) has been announced in python
> >and python3 builtin _json modules.  Patches have been committed
> >upstream; could we get the latest 2.7.7 and 3.2.5 point releases with
> >these patches added?
> >
> >
> >
> Your previous message stated that you were set up for 64bit
> building. Any chance you could do this soon?

I will be AFK for most of July, so I won't be able to look into this
until August.


Achim Gratz | 29 Jun 20:14 2014

[ATTN maintainer] rcs

I've developed a patch for the infamous and two-year-old rcs bug that
destroys data, based on the analysis by Peter Wagemans and an earlier
demonstration script by Don Hatch.  The patch has been sent upstream,
but there was no reaction for the last week:

I have compiled test packages for both architectures including this

# The first line sets up shell variable wget, to be used by all later lines
# simply paste into the shell from here
--8<---------------cut here---------------start------------->8---
wget="wget -rxnH"
--8<---------------cut here---------------end--------------->8---

# The first line sets up shell variable wget, to be used by all later lines
# simply paste into the shell from here
--8<---------------cut here---------------start------------->8---
wget="wget -rxnH"
(Continue reading)

DJ Sylvester | 26 Jun 22:56 2014

tmux-1.9a2 ctrl+b not working

I searched Google.
I looked at sourceforge tickets
I've read this Cygwin tmux announcement:
I've reinstalled tmux with cygwin-x86 installer.

I've had no luck.

ctrl+b "command" is not working for me.

I'm running a default mintty as installed by cygwin-x86 on Win-7 with a
few aliases defined, nothing else. My.tmux.conf file is empty.

To reproduce:
1. start tmux
2. ctrl+b % (outputs 5u)
3. ctrl+b " (no outputs, does nothing)
4. ctrl+b c (no outputs, does nothing)

No commands are working. Not sure what to try next.