Dan Langille – Page 19 – FreshPorts News

Commits to multiple branches

Leave a Comment / Branches / Dan Langille

Tonight I started doing some database structure changes for branches. I thought: Hopefully nobody commits to two branches at once. Yes. They do. Oh oh. mat gave me a few examples: http://docs.freebsd.org/cgi/getmsg.cgi?fetch=758626+0+current/svn-src-all http://docs.freebsd.org/cgi/getmsg.cgi?fetch=747218+0+current/svn-src-all http://docs.freebsd.org/cgi/getmsg.cgi?fetch=735782+0+current/svn-src-all Look at those URLs and you’ll see: Changes in other areas also in this revision. The code already ignores everything after

Commits to multiple branches Read More »

What branches means to FreshPorts

Leave a Comment / Branches, Development / Dan Langille

What do you want out of FreshPorts? You want to know: What is available What version Any notes in UPGRADING Any vulnerabilities You want to know about this for HEAD and for RELENG_9_1_0. You want to know that on HEAD, sysutils/bacula-server is version 5.2.12_3, but on head, it’s 5.2.12. That’s important. That’s commit history. FreshPorts

What branches means to FreshPorts Read More »

Session variables! Almost. Not quite.

2 Comments / Branches, Development, PostgreSQL / Dan Langille

Yesterday, I wrote about some great progress on getting FreshPorts to work with multiple repositories to cater for the branches of the FreeBSD ports tree. At the end of that post, I talked briefly about wanting global settings or variables for my stored procedures. Shortly after posting that, I found a solution in the form

Session variables! Almost. Not quite. Read More »

Multiple repos – good progress

1 Comment / Development, PostgreSQL / Dan Langille

Today I managed to make quite a bit of progress trying to incorporate multiple ports trees into the commit processing. The diff is about 985 lines long. It’s not as complex as it sounds. I just have to pass $CommitBranch around to various functions. The result: I have the commits going into the database, but

Multiple repos – good progress Read More »

Multiple branches, multiple repos

Leave a Comment / Branches, Development / Dan Langille

I just realized (although, deep in the recesses of my innocent mind, I suspect I knew) that FreshPorts will require one repo for each branch. The email on the svn ports branches mailing list includes multiple branches. A quick glance around the archives finds me these branches: RELENG_9_1_0 RELENG_9_2_0 2014Q1 I know there will be

Multiple branches, multiple repos Read More »

Adding code, removing more code to reduce duplication

Leave a Comment / Branches, Development, PostgreSQL / Dan Langille

In a recent post, I said: That second patch uses the same string you saw above in the previous patch. Yes, it would be better to have added that string to just one place. I welcome patches for that change. Today I will show you that code. With help from Matthias Andree on IRC, I

Adding code, removing more code to reduce duplication Read More »

Multiple ports trees for FreshPorts

2 Comments / Branches, Development / Dan Langille

As mentioned in my previous post, if FreshPorts is to display commits from every branch, it is going to need one check-out of each branch of the ports tree. Tonight I started looking into that and hit a snag in an unexpected area: name length. I do most of my development work from within jails.

Multiple ports trees for FreshPorts Read More »

Processing the first port commit from svn-ports-branches

Leave a Comment / Branches, Development / Dan Langille

Over the past few weeks, I’ve made some progress towards supporting the branches for FreshPorts. Today, I started making the first code changes and I wanted to share that with you as I went along. Finding the messages FreshPorts processes commits by parsing the messages from the mailing list. I think it might be feasible

Processing the first port commit from svn-ports-branches Read More »

Fixing up a referential integrity issue

1 Comment / PostgreSQL / Dan Langille

Earlier today, I wrote about a fix which broke search. At the end of that post, I mentioned a few things which needed to be done to fix up the broken relationships. clean up the database add a foreign key, on delete set null trigger to find the right value when set null Let’s get

Fixing up a referential integrity issue Read More »

Bad SQL corrupts search results

Leave a Comment / Bug fixes, PostgreSQL / Dan Langille

Bernard emailed me, and Tim Stoddard posted in the forum about search not finding what it should find. I quickly agreed; it was wrong. Fortunately, it was a small fix. See also my follow-up post where I fixed the broken data. The problem The problem was searching for ports such as bird didn’t include that

Bad SQL corrupts search results Read More »

Author name: Dan Langille