On Tue, 28 Aug 2018 11:41:58 -0400 Brian Callahan wrote: > Hi Charlène -- > > On 08/28/18 11:19, Charlene Wendling wrote: > > Hi! > > > > Peter Selinger mailed me about the test failure, and decided that > > the support for '-' as a getopt shortoption will be dropped for the > > 1.12 release. He sent me a patch that i've tested and everything is > > fine. > > > > I have 2 solutions to deal with this: > > > > 1. Update the Makefile with proper commenting. > > 2. Include the one-liner patch, and drop the included getopt > > configure option. > > > > I took the decision to go for 1. and am sending an updated diff. > > Tell me if you want to go with 2., i'll comply. > > > > Charlène. > > > > > > On Sat, 25 Aug 2018 04:14:57 +0200 > > Charlene Wendling wrote: > > > >> Hi ports, > >> > >> Here is an update for ccrypt, a tool that encrypts and decrypts AES > >> files and streams. > >> > >> What's new upstream (only the most interesting bits for the port, > >> full changelog there [1]): > >> > >> - Many compilation warnings/errors fixes, --with-included-getopt > >> added to configure options > >> - Add stdin "-" argument support, for cat (-c) and unixmode (-u) > >> options. > >> - New Estonian translation > >> > >> What's new in the port: > >> > >> - Updated DESCR: we don't ship the emacs flavor anymore since > >> revision 1.15 [2] of the Makefile. > >> - PLIST wasn't up to date, so this update introduces French and > >> German translations from ccrypt v1.8, in addition to the Estonian > >> one. > >> - Added --with-included-getopt to CONFIGURE_ARGS because tests > >> fail if we use our own getopt, due to a different handling of '-' > >> as an argument. Thanks to ccrypt's author Peter Selinger for > >> pointers on this issue. > >> > >> Testing (proot/amd64 + i386): > >> > >> - No other port depends on this one > >> - 'make test' is OK. It skips the crypt(3) test because we're not > >> using the included replacement libcrypt. > >> - Tested the "-" argument support for -c and -u options, to check > >> if the getopt issue above was really solved, it is. > >> - No regressions found in my daily use of ccencrypt, ccdecrypt and > >> ccat. > >> > >> Comments/feedbacks are welcome! > >> > >> Charlène. > >> > >> [1] http://ccrypt.sourceforge.net/ChangeLog > >> [2] > >> https://cvsweb.openbsd.org/cgi-bin/cvsweb/ports/security/ccrypt/Makefile.diff?r2=1.22&r1=1.15&f=u > >> > >> PS: By the way, thanks for your comments on the discount update :)
Hi Brian! > I think your approach is fine. > Not a condition for committing the update but your mail makes it seem > like you're a regular user of this so I would invite you to consider > becoming MAINTAINER for it. Yes, i'm a *very* regular user of it. As long as you're fine with the fact that my C skills are low (i can send you the discount library test case if you want to witness this), i'm going for it, as upstream is not only very reactive, but willing to help. I'm sending an updated diff that also fixes one of the MASTER_SITES, i've totally missed it! Charlène. > ~Brian >
master_sites.log
Description: Binary data
ccrypt.diff
Description: Binary data
