Well, here’s the thing. Inside all those spring jars, in the META-INF directory you’ll see a file

and another file

Everyone of the spring jars, spring-core and spring-context and spring-aop and so on contains a directory META-INF which contains spring.handlers and spring.schemas files. Most or all of them also have the particular schema file contained as well — this lets you have your spring contexts validated even if you’re running offline.

Here’s what the spring.schemas file looks like if you open up spring-aop-3.0.2.RELEASE.jar and have a look:

Notice you do NOT see anything about spring-context or spring-util or spring-task … and so on.

Now the maven-shade-plugin basically expands the jars and rewrites them into the work directory before jarring everything back up again. This normally works fine when you don’t have naming and pathing COLLISIONS, like you do with these schema files. So whichever spring jar was processed last by maven-shade will have its META-INF/* contents written into the final jar you’re creating, overwriting any other content with the same name.

So if you’re using a spring context file that looks something like this:

…you will get the complaint mentioned in the title of this blog entry. Spring can’t find the namespace handler for “util”, because it is not mentioned in the META-INF/spring.schemas that result in the final product.

Here’s the solution – you need to take advantage of the example in the maven-shade-plugin page and do this in your configuration section:

The maven-shade-plugin will append the contents of all META-INF/spring.schemas and META-INF/spring.handlers files it finds during the expansion.

Here’s a quote from the GNU gettext utilities page:

After preparing the sources, the programmer creates a PO template file. This section explains how to use xgettext for this purpose.

xgettext creates a file named domainname.po. You should then rename it to domainname.pot. (Why doesn’t xgettext create it under the name domainname.pot right away? The answer is: for historical reasons. When xgettext was specified, the distinction between a PO file and PO file template was fuzzy, and the suffix ‘.pot’ wasn’t in use at that time.)

So that’s it. It’s the same file, with a different name. Just thought I’d mention that, considering how much time I pissed off trying to find the answer!

[edit]
It looks like WordPress sees this a little differently. the POT file, the “PO” Template, is the raw output from the xgettext, untranslated. the PO file is the text-editable and edited pot file with the translations. the MO file is built with msgfmt (among others) and is what is actually used by php’s gettext.

GNU ‘gettext’ Utilities

These docs for gettext are C-oriented, but the associated tools such as xgettext are easily used by PHP, so let’s not get too excited about the C.
And I’m going to assume you have installed GNU’s gettext on your system.  I’m pretty sure it arrives by default with your ubuntu or other linux installation.  Mac and Windows you’re going to have to find for yourself

Edited:  download GetText for Windows here

There’s also a lot of emacs crap in this doc, which I’m also going to ignore.  We’re going to use tools like poedit, gtranslator for what we need … coming posts.

on my ubuntu 9.10 box the following command installs the php-gettext stuff in /usr/share/php/php-gettext

and includes these three files:

• gettext.inc
• gettext.php
• streams.php

I’m not going to deal with the internals of these files here in this post.

What I do is copy these into my working project somewhere, say in a subdirectory named “library”, “lib”, “i18n” … etc. I’m going to call it ./library/i18n just for this.

.po, .pot, .mo files – what are these?

http://www.gnu.org/software/gettext/manual/gettext.html#Files

That’s about all for now, sorry I couldn’t be more helpful in this post.  This is just the setup stuff I’ve done for my new entry into this world of Internationalisation (i18n), Localisation (l10n), and Native Language Support (NLS) — see Concepts.

I used WP-SyntaxHighlighter, I tried some others, but the simplest and most expedient plugin I found for highlighting code in my blogs comes from Dmytro Shteflyuk’s Code Colorer.
Like most of them it’s based on the Geshi Generic Syntax Highlighter … much easier syntax.

I recommend this one for bloggers who wish to include code in their posts.

One thing I have noticed – almost all of these plugin authors do NOT use classes in their design.  Their .php files are full of functions, most of the time written without the

or

How do they expect to avoid naming collisions?  If they have to have a function by a certain name which probably exist in some other plugin developer’s file somewhere, then even if they use this !function_exists business they’ll end up getting the other author’s behavior for that function.

Why don’t these people just use classes! I know I’m coming at this from the java world where you have to use classes for everything, and there’s packaging to avoid naming collisions, but still – this seems like an obvious question to me. Do the modern php programmers have no experience with a real class-oriented language?

Here’s a silly example I’m doing. I have a plugin I’m developing to allow Zlatina’s web administrators to deal with their database of students and contacts. It’s called Bcc DBManager.
I created a subdirectory wp-content/plugins/bcc-dbmanager, with a file named bcc-dbmanager.php … duh. I want to hook into the activate_${plugin} (see this link for an explanation of that). Here’s a partial …

How hard is that? I certainly don’t know much about plugin development at this point. But I would have to describe the use of functions rather than class instances and methods as Code Smell.

[Edit] Since I wrote this post I found this link on Nerdaphernalia: Use Classes in your WordPress plugins to increase code portability and reduce name conflicts

and you can actually type in http://localhost/index.php and get your phpinfo page coming out.  But it seems nowhere in the docs can you find out how to get apache2 to recognize index.php as the first page to deliver.

So, I found out where, finally.  In your WAMP environment there’s only the http.conf file, you look for the DirectoryIndex, and move index.php or insert “index.php” to the front of the line.  In this ubuntu 9.10 LAMP setup, where the hell is the file?

/etc/apache2/mods-available/dir.conf

That’s where it is.  open the file, it looks like this after your normal lamp setup in this environment

Notice the “index.php” somewhere in the DirectoryIndex list … move that to the front of the list as in

WordPress Plugin Development: Beginner’s Guide

I also purchased this eBook, for €19.99

Now remember, you want to be developing on localhost if you can.  You can always upload your “finished” plugin to your site wherever it is.  I have a windows box for www.hagelnx.com, I have a linux box for developing my plugins.  Do NOT waste your time trying to develop on windows if you can help it.  The apache2-php-mysql .. LAMP setup for ubuntu, just as an example, works really well with apt-get installing all the modules you need.  All the config files are updated without your having to edit them, both php ini files and apache2 ini files.

I installed a new dvd r/wr drive on my windows box … or rather a friend of ours Dimitar (Mitko) installed it for me   It is good to know people who know hardware, I could not survive as a software developer on this planet without people like that.  It’s good to know that there are people like Mitko here in Bulgaria.