How to manually upgrade the port and make the patch.

Status
Not open for further replies.
I'd like to upgrade the new port, but this handbook:
http://www.freebsd.org/doc/en_US.ISO8859-1/books/porters-handbook/
is long, unfriendly and keeps people away from doing this.
I'm strictly interested how to upgrade the version, not by reading how does it work from scratch (I don't have time for this).

I just want to upgrade my port from:
mysql-workbench-oss-5.2.1.tar.gz
into:
mysql-workbench-gpl-5.2.29-src.tar.gz

I've downloaded the package, done the sums:
Code:
make makesum
or using md5, sha256 shell commands.
I just don't like reading tons of pages, I need specified instruction, I'd like to see guide with real commands step by step for my specified case (which is simple version upgrade).
I'm not interested how to build Makefile (which I have), usage of different variables, SDL, TCL/TK, QT4 and the other stuff.
I want change only the numbers and make a diff.
I'm following OpenBSD guide, which is far easier,
http://www.undeadly.org/cgi?action=article&sid=20080318060000
But still some commands are different.
Is there any alternative, unofficial guide step by step to upgrade manually my port?
I just want upgrade my software to work (currently it's crashing every few minutes).
 
Maybe that's why most of the versions in ports are so old and not maintained, because nobody wants to read this and people simply don't have time, because of work?
All I want is only simple instructions instead of description how to write and configure Makefile.
Anything like link to another post, already answered thread, some wiki.
Basically I see that there is a lack of nice wiki's, and other users basically they don't care, because it's open-source.
I just want to make it easier, not only for me, but for others by creating new upgrade into port.
If you think differently, I could just compile untar, compile and install the software and give a damn. Or just install Windows as a Desktop, I'm sure that it will work.
 
kenorb said:
Maybe that's why most of the versions in ports are so old and not maintained, because nobody wants to read this and people simply don't have time, because of work?

I'd like to see you try and maintain 22000+ ports.
 
SirDice said:
I'd like to see you try and maintain 22000+ ports.

With tke proper wikis and some funds, why not.
Could be automatized for most of the QA tasks.
Simply cron tasks which checking the new version of softwares from specified feeds or source. When there is a new version, run virtual machine inside which scripts changing the ports files (sums, filenames, etc.), testing the ports by compiling, installing and run some tests if it does work. If not, post the issue with compiling error and let's community together solve the problem and report the bug or do some patches.
This is how open-source should work.

Example with Drupal:
http://qa.drupal.org/
All the patches for core system and selected contributed code are tested with SimpleTest against latest HEAD in the network of VMs.
Easy deployment of a PIFR test machine

Other thing, trust me, none of the Drupal wiki or documentation is so long as this, and the things are even more complicated. Are just separated/divided for the specific use-cases.

Maybe that's why Drupal is used as open source by most of the government organizations such as WhiteHouse, most of the UK government organizations (NS, DoC), etc.
Because it's easy enough documented?
It doesn't matter that's web solution, not system. It's open-source which works in the same way.
2001 and 1993 - 10 years difference and a very big difference between user community and organisation. Difference is that they follow the technology and how the world is changing.
Drupal vs FreeBSD
Drupal - 1680000 results
FreeBSD - 2150000 results
Let's repeat this test after 1 year:)

People have no problem creating their own documentation or in easy way be part of community (everybody has access to edit wiki's, so they are never so old), they don't have to send PR for that reason.
People have access to their own repositories so they can contribute their own code (or ports) in very easy way.

If this is open-source, how I can edit/update the wiki's, how I can create new one, how I can apply for access to svn repository, how I can be part of community? Forum, on which people pasting the links to the old handbooks, which nobody wants to read it?
I just see dead people, nothing more.
 
  • look at the commit logs
  • figure out what constitutes an update
  • do the same
  • test
  • submit the patch into GNATS
As you can guess the shortest way to contribute (to any project) is to start from the end.

I for one was too lazy to read Porter's Handbook and ended up reading the code instead. After a while a storm of PRs ensued. ;)
 
eye said:
I for one was too lazy to read Porter's Handbook and ended up reading the code instead. After a while a storm of PRs ensued. ;)

I'm not lazy, I'm a programmer and it's easier and faster for me to read the code source, backtraces or raw command lines (with short descriptions what does it do), than the long handbook where English is not my first language. Not everybody must to read the books to achieve something, I hate it, especially if 80% of it is off-topic to what I'm looking for. In this case I just can't focus and find the right thing, my eyes are just jumping around and looking for interesting chapter for my needs. If after 15 minutes I can't find it (searching of this is a waste of time, not the reading the proper found section), I'm searching for easier alternatives in Google. Then if not found, I'm posting the thread here. And I'm sure that other people has the same problems with the long handbooks (but they don't share their opinions).
Some people have to tolerate that. And fixed schemes with one main handbook is not the solution.
 
kenorb said:
I'm not lazy, I'm a programmer and it's easier and faster for me to read the code source, backtraces or raw command lines (with short descriptions what does it do), than the long handbook where English is not my first language. Not everybody must to read the books to achieve something, I hate it, especially if 80% of it is off-topic to what I'm looking for. In this case I just can't focus and find the right thing, my eyes are just jumping around and looking for interesting chapter for my needs. If after 15 minutes I can't find it (searching of this is a waste of time, not the reading the proper found section), I'm searching for easier alternatives in Google. Then if not found, I'm posting the thread here. And I'm sure that other people has the same problems with the long handbooks (but they don't share their opinions).
Some people have to tolerate that. And fixed schemes with one main handbook is not the solution.

Like some fellow programmer once said: "Shut-up and hack"
 
If there's anybody that has problems reading documentation it'll be me. That's one of the downsides of having AD/HD. But I did read it, learned from it, tried it, made mistakes, read it again, tried some more etc. It took a lot of effort but in the end it pays off. Don't try to cut corners, it won't work.

Famous quote: Give a man a fish and you feed him for a day. Teach a man to fish and you feed him for a lifetime.
 
kenorb said:
I'm not lazy, I'm a programmer and it's easier and faster for me to read the code source, backtraces or raw command lines (with short descriptions what does it do), than the long handbook where English is not my first language.

The docs are available in languages besides English.

Not everybody must to read the books to achieve something, I hate it,

Not a FreeBSD problem.

especially if 80% of it is off-topic to what I'm looking for. In this case I just can't focus and find the right thing, my eyes are just jumping around and looking for interesting chapter for my needs. If after 15 minutes I can't find it (searching of this is a waste of time, not the reading the proper found section), I'm searching for easier alternatives in Google.

Your rush to get going is actually making things slower. Learn how to do things correctly on FreeBSD, then go fast. The other way around is a fast way to unstable, broken systems, as has been already demonstrated.

Then if not found, I'm posting the thread here. And I'm sure that other people has the same problems with the long handbooks (but they don't share their opinions).
Some people have to tolerate that. And fixed schemes with one main handbook is not the solution.

Everyone appreciates brevity, so please provide patches to make the Handbook shorter. Leave only the information on what must be done to get things working correctly.
 
kenorb said:
FreeBSD problems are open-source community problems.
If I've the problem, it's a FreeBSD problem also.

Not everybody must to read the books to achieve something, I hate it,

Many experts have worked many hours to describe how to do things in FreeBSD the easiest, quickest, and most effective way.

The problem is that you hate to read it. That problem is yours. It's not something FreeBSD can change. You are the only one who can do anything about it. The easy fix would be to find some other operating system, presumably without documentation to read. The harder fix would be to realize that skipping documentation takes much longer and makes the process much more difficult, because you've rejected all that free expert advice.

There's a third choice, which would be to try to change FreeBSD to fit a preconception. That probably isn't going to happen, but you can try. It'll take a long time. By the end you'll be an expert on FreeBSD, and know everything that's in the documentation. Just like you would have learned by reading it to begin with, only it'll take a lot longer.
 
Status
Not open for further replies.
Back
Top