When I went looking for commands to convert to the new jail strategy, I failed to find two. I have remedied that oversight now. make showconfig This command extracts the configuration items from the port. Like all the other scripts, it is fairly straight forward. The following is from make-showconfig.sh: make master-sites-all This command extracts […]
I found a few more makes to convert to jails Read More »
The jail approach mentioned in previous articles has been implemented on the dev system. It took about 4 hours of coding this afternoon. The setup script I spent some time today gathering together the stuff I’d mentioned in the two previous posts on this topic. The result is a one-line command which will create a
dev.FreshPorts.org now using jail Read More »
Recently, I wrote about using a jail to get better results. Today, I’m altering the scripts so they make use of this jail. For the most part, this will involve the creation of scripts that will run in the jail and duplicate the functionality of the existing FreshPorts code. What needs to be changed? I’m
Altering scripts to use the jail Read More »
emmh wrote to mention that the vuxml posts on FreshPorts were out of date. I checked. Sure enough. They were. Checking the sites On the homepage, under Latest Vulnerabilities, the latest entry was dated Oct 17. And here it was, November. Something is wrong. The beta website showed Aug 14. And my dev website showed
vuxml updates not being posted Read More »
FreshPorts has always strived to be correct. Even when it’s difficult. One issue which always always proven difficult is ports that rely upon other ports. or cannot be installed with other ports. Case in point, sysutils/msktutil, which includes this message: IGNORE: cannot install: SASL support requested and openldap-client-2.4.20 is installed That’s not really appropriate on
Getting more accurate results… Read More »
FreshPorts processes commit emails, updates via svn, then extracts data via make -V. That’s the short of it. However, stuff goes wrong. And sometimes we can’t update. Or the extract fails. This leaves data in the database which does not reflect what is in the port. This query, which I was creating tonight, may help
What ports need to be refreshed? Read More »
Why can’t things be simple? The basic premise of FreshPorts is parse, an email, put it in the database. Sounds simple? Well, it is. Until you start dealing with the exceptions. Such as slave ports. Such as UTF-8…. I only just got used to dealing with base64-encoded messages when a real UTF-8 character shows up.
Why can’t things be simple? Read More »
I think I have a solution for base64 emails. I emailed Matthew Seaman on Wednesday about this issue. He put me onto two very useful Perl modules: Email::MIME mail/p5-Email-MIME Email::Simple mail/p5-Email-Simple/ Email::Simple::Header The solution is coded below: sub myGetMessage { my ($message); my ($encoding); while () { $message .= $_; } $encoding = myGetMessage_ContentTransferEncoding($message); print
Solution found for base64 emails Read More »
I have to say, FreshPorts would be a very simple piece of code if it wasn’t for all the exceptions to the rules. But then, that is true for most software. Case in point, I’ve seen three base64 encoded emails in the past week or so. Each one makes FreshPorts barf. It has no idea
base64 encoded emails Read More »
In the past few days, I have implemented the full pathnames solution on dev. It was first implemented on gelt, a server setup explicitly for developing full pathnames. I am so glad that I keep notes when making changes such as this. Sometimes it is days or weeks between testing out a new approach and
Full path names now on dev Read More »