I would like to install a complete new version of Trac alongside of our current version (0.11.7) and I am looking for ways to do this. After some research, it says to use python's virtualenv, but I am trying to find specific steps on how to accomplish this without interfering with our 0.11.7 version at all.
I am using Ubuntu as the OS. Any input including any possible pitfalls is appreciated.
Try virtualenvwrapper that makes using python-virtualenv a breeze.
The steps to create and use such a Python virtual environment are explained in the user documentation. These environments form the core setup of my own Trac plugin development. It allows to even use custom python versions, if you ever need that. I found the need to give each environment a self-explaining name and use it with different Trac environment directories matching the db version required by different Trac versions, i.e. virtualenv "trac-0.11_py2.4" with Trac env "sandbox_0.11", "trac-0.12_py2.6" with Trac env "sandbox_0.12", etc.
Related
recently I tried to install an apache webserver on an ubuntu machine. I discovered two ways to do that.
The first one is to install it like it is described at this source apache docs using ./configure, make, make install to the downloaded apache sources.
And the second way is to install it via apt-get update && apt-get install apache2.
I also noticed that when I run the installation via apt-get it seems that there is a different configuration of the apache for example the directory /etc/apache2/* is only available over this way of installation. So when I install it manually the directories sites-available, sites-enabled, ... are just missing.
Is there also a way to get these folders when running the manual installation?
Where do this differences come from?
This is nothing that can be summed up in a few sentences. You basically ask: how does software management work under Linux systems? In short:
using the apt utility on a Ubuntu based system you are using the systems package management system, that should nearly always be what you want. That way the system can take care to keep your apache installation up to date, you can remove software again without leaving artefacts, the software is guaranteed to work with the system libraries already installed in your system. Potential conflicts are resolved. You can be certain that the configuration matches your system.
using the build system is a more generic (archaic) way: you do not only install the software, you build it from scratch from the software sources prior to installing it. That is only possible for OpenSource software, obviously. This certainly allows for more flexibility. But you are responsible yourself for a whole lot of things, starting with first setting up a complete build system, then configuring the package, select what you actually want to build and last but not least you are yourself responsible to update software you installed that way. This rarely is a good idea, with two exceptions:
you absolutely cannot find a pre build package for your Linux distribution or
you want to make own modifications to the software itself
The difference in the folder layout of the apache configuration is a separate thing. There are two aspects that come into play here:
you can change that layout using the uncounted build options offered by the build system (namely the options you can hand over to the configure utility on the command line).
typically the configuration of complex software packages (like the apache http server with all it's modules) is broken up into
various sub folders so that you can keep an overview and
allows additional packages to drop their additional configuration in place (this is mostly useful for prebuild packages)
Long story short: in 99,8% of all cases you want to use prebuild packages prepared for your distribution. That is the power of the software management systems under Linux (that still have no comparable counterpart in other operating systems).
Trying to set up IntelliJ IDEA with the Lua plugin. This requires me to direct it to the location of a Lua SDK. All instructions I have been able to find for this are on a Windows system, pointing to a specific directory in Program Files. I am running a Linux system. I've built and installed Lua 5.3.4 (newest version at time of writing) from source, but as far as I can tell that doesn't include any equivalent to the SDK needed – it just drops a variety of necessary files in /usr/local/{bin,include,lib,man/man1} and creates two empty directories /usr/local/{lib,share}/lua/5.3.
Am I missing something conceptual? Do I need to get something else (the existence of which I have been unable to confirm)? Should I just give up on this and find another IDE? The plugin seems to come with two SDKs, Kahlua and LuaJ, but I don't know how well either of these match up to either standard Lua or LuaJit, which are what I would be using.
This is assuming you are using this Lua plugin for IntelliJ:
If you have installed Lua via DNF or yum simply guide IntelliJ to /urs/bin and it'll find it. (I assume this also works for other installers like apt or brew.)
It looks like the SDK is just looking for two files lua and luac which both live in /usr/bin.
Installing PHPUnit with composer globally seems more convenient to me for those two reasons:
1. Using it everywhere without needing an extra install.
2. Just running phpunitinstead of vendor/bin/phpunit (using an alias might solve this)
Are there any reasons why a local install might be the better choice? For example: using the exact same versions every time. (don't have a lot of experience with PHPUnit, so not sure if this really is an issue or not)
The big disadvantage of installing packages globally is that you might end up with different versions of PHPUnit between developers in your team (unless you are the only developer). This might cause some side effects.
If you install it locally using composer.json, then every developer in your team will have exactly the same version as you do for that specific application. Also, everybody will see when you change the version in composer.json.
If you don't like typing vendor/bin/phpunit, you can use Makefile (which is also in your project):
test:
vendor/bin/phpunit --configuration=test/Unit/phpunit.xml
then run it ...
make test
I like to install it via composer and the require-dev block, but another way that does come highly recommended is to download the phpunit.phar into the project, to use that.
Either way, you control exactly which version is being used (and when it's updated) - which is the most important part, as you can't so easily control what people have installed globally.
A script needs to support two servers, Fedora and Centos, with different versions of httpd (Apache). CFEngine is used to create the configuration file and it needs to put different sections to the file depending on which httpd version is installed in the system. How can I set a variable to true/false depending on which version is installed so that I can then generate the file appropriately? All the examples I could find only deal with installing, upgrading or removing packages.
You could use packagesmatching to get the version of httpd that's currently installed. You might combine it with ifelse, strcmp, regcmp so that you can set different variables and classes which you might use in your template.
I use CFEngine (3.9) on CentOS 7.3. This version sets a hard class centos. You can use this to create the different sections. (Is also sets a hard class redhat, because CentOS is a derived from RedHat). You can find the classes that are availabe on your system with:
cf-promises -v
I'm trying to build an embedded simple web browser for an embedded device and I've decided to use WebKit / WebKitGTK+. However, our device uses a Linux environment somewhat based on CentOS 5.8. I haven't been able to find any RPMS or mention of support for WebKit / WebKitGTK+ for CentOS 5.8 while doing several web searches.
Does anybody know if it's possible to build an older version of WebKitGTK+ such as 1.2.6-2.el6_0 which works well on CentOS 6.3? Are any RPMS available for CentOS 5.8?
The goal here is to be able to run a relatively current, at least 1.2.6 version of WebKitGTk on CentOS 5.8
Note: I was able to sort everything out. Just took a long time compiling all of the dependencies in the correct order with the correct options. I was able to get WebKitGTK 1.6.0 running on Centos 5.8.
You shouldn't have any problems building an old version of webkit if you can install the older versions of libraries that it requires.
If you have older or newer versions of GTK+ etc installed than the old version of webkit requires it may need quite a bit of porting to compile.
I'm not aware of any RPMs that meet your requirements
Depending on the compilation options you should be able to compile the dependencies in an isolated directory. With each library you typically use the --prefix option to specify the destination. Then when compiling something that depends on that library, you typically have an option to specify where to look for that library - something like --with-libraryname=/path/to/library. You want to check ./configure --help of each thing you're compiling to get the correct options.
It'll be quite a bit of work, but you should be able to compile everything you need into an isolated directory without replacing anything on the system. I would highly recommend you avoid doing this in root to ensure you have the right options.