Linux productivity tools for me

This is the stack/are the tools that keep me productive in my day to day programming experience on linux.

Well, first of all let me tell you, that this might not fit your needs , as I’m a full time java programmer so have OS choices open (still I won’t go to my java-related stuff here, to make infos general and possibly usable for non-java guys as well.).

Linux is my choice for quite some years already and this post won’t detail the reasons, rather will focus on tools/utilities that help me survive in this world.

Distro

Well, let’s start from the ground. I don’t have a clear winner here. In fact I run currently 3 different ones (didn’t count my router and phone here):

• Fedora - on my work laptop
• Xubuntu - on my home laptop
• Raspbian (Debian) - on my Raspberry Pi server

In fact the info present in this post won’t detail too much on the 3.rd one, as it’s a server that just runs. The only thing I need there is a ssh access and that’s pretty much it, so this post would be rather short for that one :)

The reasons for Fedora, and Ubuntu in my case are:

• stability (well, let’s say things don’t crash that often for me),
• packages for most of the SW I need available in repositories and
• the software packages are not at some archaic versions.

Window manager

Well, the nice thing is that most common window managers are available across all the popular distros. My choice here is Xfce.

Well, I don’t hate others, this just fits my needs best:

• no need for fancy effects for me as
• speed is rather my preference.
• still I need something that I would not fight with every day.

Project I’m keeping my eye on is: LXQt. Still waiting for a more stabilized release (0.8 was released recently, which says to be ready for production desktops, but I’d plan to give it a try with the next release as well as when I have some time to set it up).

Shell

I have 2 candidates here:

• bash and
• zsh.

bash I use for scripting. It’s been my choice for quite some time already and I don’t see a need for change any time soon. As I’m quite used to it and the most people/projects I share the scripts with are OK with that/have used it already.

zsh as my default shell. This is still the world I consider new to me (see my recent post on that).

Shell env sync

As I have multiple machines I’m working on, there are 2 essential projects in the area for me:

• oh-my-zsh as well as
• homeshick.

oh-my-zsh

Provides for my zsh:

• nicely/systematically structured plugin approach,
• autocompletition and
• all sorts of aliases.

To document my use case a bit here, these are the plugins I have present in my .zshrc:

plugins=(git mvn glassfish yum colored-man vagrant z common-aliases gradle homeshick vim-interaction powerline tmuxinator tmux)

Please note: Some are not available officially, and in the time of writing are just present as pull request from me to the project. Namely:

• homeshick and
• powerline.

Well, might sound like a self promotion, but feel free to check and provide feedback on these if you find them useful.

Homeshick

Lets me synchronize all my custom zsh plugins as well as .bashrc files over git repository.

The only requirements are:

• git and bash on all the clients
• and the git repo accessible from all my clients.

In my case, there are things I don’t want to expose to public, but have no problem to have it hosted in some private git repo. So bitbucket is my choice, as I have there private git repo for free for the purpose.

Shell sessions bootstrapping

There are 2 important projects for me here:

• tmux and
• tmuxinator.

Tmux and tmuxinator enable me to have just one yaml formatted file for bootstrapping my terminals.

I preffer it to having multiple tabs open in some GUI terminal, as it:

• bootstraps all my shells daily in a same/reproducible way,
• provides me with the nice way on organizing these and
• navigation between these works with keyboard only => switching between different tasks becomes after some time of usage just a routine.

So no more searching in countless cards/windows for particular task.

Shell productivity

I don’t want to list all the linux utils I use, as I guess I’d make list too long and too boring (even for me to write). So let’s just name some, that I consider worth it:

• ack - powerful grep replacement (I might document in the separate blog post),
• notify-send and
• all sorts of aliases and zsh/bash functions, that would be too many to list here, moreover might be too specific and useless for others.

Notify-send

I blogged about notify-send already.

I use it for all the app-server (in my case Glassfish) lifecycle management operations (start/stop/restart domain) as well as deployment stuff.

As these:

• take quite some time to finish and
• might end up with errors,

I let it myself updated with notification having:

• exit code
• chunk of the last couple lines from the log file.

This is great, as I can work in parallel and it gets my attention, once the job is done.

Desktop app launcher

Without having an option for fast startup of my favourite programs I’d waste my time searching the icons or menus.

As Xfce doesn’t provide me with the powerful one, I use: Synapse.

File manager

Well, there are times when I play with files in the command line, but sometimes it just fits better to use some UI for the purpose.

My choice is Krusader. Well, the most stuff I need is available, namely:

• file/folder manipulation,
• file folder comparison and
• file/folder contents search.

The only thing that bothers me is the fact that development is rather stalled in the project.

Still, viable alternative might be: Double commander, which seem to be even cross-platform and can use the total commander plugins (which used to be my choice on windows for the purpose).

Editor

I tried to live with Gedit and for simple note-taking it might be a good choice, however as I like to play with ruby these days, I tried to find something that would help more in the area.

(G)Vim

After searching I came to conclusion, that (G)Vim is quite popular in the Linux world.

Well, it’s worth to mention vim joke here, which expresses the feelings of many on this editor: Infinite Vim monkey theorem (see Infinite monkey theorem for explanation).

Still, I’ve seen:

• many people favouring it,
• it’s the first editor I remember to have available once logging in to my school linux account (OK no real argument, just sentiment),
• it seemed to provide countless plugins for all sorts of stuff and
• it’s incredibly powerful, as far as I’ve seen and learning it might pay back.

So I decided cca 2 years back to uninstall the editors I’ve been used to and force myself to use (G)Vim.

Great source of information was for me: vimcasts, which helped me a lot in the area.

Well, I plan to document my .vimrc setup in a separate post, as it could make this one way too long.

Files sync tool

Unison is my choice her. For more info on my use case, see this blog post.

Conclusion

Hope you find some inspiration here. And would be glad to hear from you guys about any I missed, but you could not live without.

Still, I can’t believe anyone read this far. As I guess I would not force myself to :).

Kindle for offline html docs reading

Summer is here, while laying on the beach, you can use the time to learn new programming language or get some overview on the framework of your interest. Possibly the things you had no time left for over the year.

I bought myself Kindle, my beach holiday is just a day away, so it’s time to fill the reader with content.

Goal

Get an online html docs to Kindle for offline reading.

Required software

• wget - download program
• calibre - ebook reader/converter/…

Process

1. find the docs you’re interested in (I’ll refer to <http://your-manual.com> as the url of the docs of my interest)
2. download it (I used wget to do so, inspired by stackoverflow answer, please note this is expecting all docs no one domain only):
   
wget --recursive --timestamping -l inf --no-remove-listing --page-requisites <http://your-manual.com>

3. start calibre
4. drag&drop the downloaded index (main) page to calibre
5. convert to mobi format (via right click on the book and choosing Convert Books -> Convert individually afterwards editing metadata and choosing mobi as the Output format)
6. connect your kindle (via usb)
7. send ebook to device - via right clicking the ebook and choosing: Send to device -> Send to main memory
8. done!

The valuable part

Well, the challenging part is for me always to actually digest the docs. So go ahead and read it (if it’s worth it). Self-discipline, I need to work on that, I guess.

Migrating from bash to zsh

These are my steps in migration from bash to zsh. I’m documenting these to keep the future reference for me (and possibly other readers as well).

Motivation?

Check out the Slideshare presentation: Why Zsh is Cooler than Your Shell

Current shell

to find out the current shell:

ps -p $$

it’s useful during migration to make sure the new shell is really in use.

Migration

Migration itself included (on my Fedora) reconfiguring:

• default shell
  (inspired by: http://unix.stackexchange.com/questions/39881/running-chsh-does-not-change-shell)

  chsh -s /bin/zsh

• konsole
  in Settings -> Edit Current Profile
  for: Shell type:

  /bin/zsh

• gnome-terminal
  no changes required, seems to go for the default shell.

• xterm
  no changes required, seems to go for the default shell.

• tmux
  (from: http://superuser.com/questions/253786/how-can-i-make-tmux-use-my-default-shell)
  in the: ~/.tmux.conf added:

  set-option -g default-shell /bin/zsh

oh-my-zsh

There exists users’ maintained collection of zsh customizations, that I found useful. Called: oh-my-zsh. (Please note the amazing Github’s - Star count. Once seeing that I didn’t doubt I’ll give it a try.)

For installation I just went for:

curl -L http://install.ohmyz.sh | sh

oh-my-zsh with powerline

I’m used to powerline everywhere ((g)vim/konsole/tmux/bash) already, so let’s keep it working in zsh as well.

I created file: ~/.oh-my-zsh/custom/powerline.zsh with contents:

. ~/.local/lib/python2.7/site-packages/powerline/bindings/zsh/powerline.zsh

Please note: You need to go for the actual location of powerline installation on your system.

and in the file: ~/.zshrc I had to comment out line:

# ZSH_THEME="robbyrussell"

otherwise oh-my-zsh theme seems to conflict with powerline one.

oh-my-zsh plugins

The real benefit of oh-my-zsh commes with plugins, so I went for those I find useful. Via changing the file: ~/.zshrc having:

plugins=(git mvn glassfish yum colored-man vagrant z common-aliases gradle)

oh-my-zsh custom stuff

As my ~/.bashrc held already quite some customizations, I migrated those to:

~/.oh-my-zsh/custom/*.zsh

Even this part was nice, as I could tidy up things a bit and omit unused stuff. Moreover I decided to create multiple *.zsh files (as these are auto-loaded) to achieve modularity.

Done

Honestly in the beginning, I had no clue how much I would enjoy my new shell.

Xfce: How to disable toolbar text in KDE apps

Recently I’ve experienced quite a funny situation. I use Krusader for file browsing that calls Kompare for file comparisons. However I was not able to copy changes from left to right (or vice versa) in the compared files, as I could not see all the toolbar buttons (these were too big). I didn’t even notice for quite some time, that there are others, not displayed on my screen.

But who needs text description in toolbars? Well, I don’t!

I remembered global setting from the Gnome world that it should be possible, so searched from the KDE equivalent.

Found it (in my Fedora + Xfce) in:

Xfce menu button -> Options (hope this one is right, as I have it as Einstellungen) -> KDE SystemOptions (for me KDE SystemEinstellungen) -> Style (Stil) -> Fine Tuning tab -> Toolbars section
where I just chose No Text in the both: Main toolbar text as well as in the Secondary toolbar text.

That did the job for me.

Apache Ant tasks for JMX access

I wanted to invoke JMX operations from the Ant tasks. However finding usable ant tasks library as well as the usage was rather tricky. So let me share my experience to make things easier for others.

Ant tasks for JMX operations

I decided to follow Tomcat documentation and used ant tasks distributed with tomcat.

Just for the record the usage is not restricted to Tomcat deployed JMX mBeans. For me it worked for java process accessible via JConsole via Remote connection.

Retrieving the library

1. As I wanted to get the latest version I used maven central repository “search by classname” feature and searched for: org.apache.catalina.ant.jmx.JMXAccessorTask (see the query)
2. afterwards I went for the tomcat 8 jar file (called tomcat-catalina-ant-8.0.8.jar)
3. and just copyied the latest available to my $ANT_HOME/lib dir.

Usage

I didn’t have a chance (or motivation?) to check all the tasks available, the full list of tasks available can be seen in the zipped file: org/apache/catalina/ant/jmx/antlib.xml, following were present for me:

<typedef
    name="open"
    classname="org.apache.catalina.ant.jmx.JMXAccessorTask" />
<typedef
    name="set"
    classname="org.apache.catalina.ant.jmx.JMXAccessorSetTask" />
<typedef
    name="get"
    classname="org.apache.catalina.ant.jmx.JMXAccessorGetTask" />
<typedef
    name="invoke"
    classname="org.apache.catalina.ant.jmx.JMXAccessorInvokeTask" />
<typedef
    name="query"
    classname="org.apache.catalina.ant.jmx.JMXAccessorQueryTask" />
<typedef
    name="create"
    classname="org.apache.catalina.ant.jmx.JMXAccessorCreateTask" />
<typedef
    name="unregister"
    classname="org.apache.catalina.ant.jmx.JMXAccessorUnregisterTask" />
<typedef
    name="equals"
    classname="org.apache.catalina.ant.jmx.JMXAccessorEqualsCondition" />
<typedef
    name="condition"
    classname="org.apache.catalina.ant.jmx.JMXAccessorCondition" />

out of these, I gave following a try:

org.apache.catalina.ant.jmx.JMXAccessorTask
org.apache.catalina.ant.jmx.JMXAccessorInvokeTask
org.apache.catalina.ant.jmx.JMXAccessorQueryTask

For the demonstration purposes I’m using Glassfish 4.0.

Example: Listing JMX MBeans

Let’s assume we want to retrieve the MBean by name (namely: java.lang:type=Memory). Please note username and password were not required for access (otherwise they should be specified via respective properties).

Noteworthy here is the resultproperty, which could hold array from which we could get a name. So having in ant build script:

<typedef 
    name="jmxQuery"
    classname="org.apache.catalina.ant.jmx.JMXAccessorQueryTask" />

<jmxQuery
    host="localhost"
    port="8686"
    echo="true"
    name="java.lang:type=Memory"
    resultproperty="memory" />

<echo>Retrieved MBeans count: ${memory.Length}</echo>
<echo>The 1.st one has name: ${memory.0.Name}</echo>

results for me in following output:

 [jmxQuery] memory.Length=1
 [jmxQuery] memory.0.Name=java.lang:type=Memory
     [echo] Retrieved MBeans count: 1
     [echo] The 1.st one has name: java.lang:type=Memory

Example: Invoking operation via JMX

Here is a 2 step approach required:

1. connect to remote server via JMX and afterwards
2. invoke the operation on the particular MBean.

For demonstration purposes, let’s assume we want to call garbage collection (via invoking operation: gc() on MBean named: java.lang:type=Memory)

Sample ant build file chunk does the job (please note ref property value that has to be the same across these 2 tasks):

<typedef 
    name="jmxOpen"
    classname="org.apache.catalina.ant.jmx.JMXAccessorTask" />

<typedef 
    name="jmxInvoke"
    classname="org.apache.catalina.ant.jmx.JMXAccessorInvokeTask" />

<jmxOpen
    host="localhost"
    port="8686"
    ref="glassfish"
    failOnError="true" />

<jmxInvoke
    name="java.lang:type=Memory"
    operation="gc"
    echo="true"
    delimiter=" "
    failOnError="true"
    ref="glassfish" />

Further information

For more details, I recommend reading the official Tomcat documentation as well as Javadocs.

Using Genius TVGo DVB-T03 on Fedora

My other laptop was not available and the start of Football WM 2014 was just less than 30 minutes away. Let’s see if I can watch it on my laptop with Fedora (16) with my Genius TVGo DVB-T03 device.

The most of the steps were inspired by fedora wiki: https://fedoraproject.org/wiki/DavidTimms/DVB

DVB-T required package

I installed dvb-apps package via:

sudo yum install dvb-apps

Afterwards checked the file list in package via:

rpm -q --fileprovide dvb-apps

which contained also:

/usr/bin/scandvb

(please note we’ll need this one later).

Firmware

I plugged in my dvb-t usb and seen after running dmesg:

[11140.318728] usb 2-1.2: new high-speed USB device number 6 using ehci-pci
[11140.397936] usb 2-1.2: New USB device found, idVendor=0458, idProduct=4012
[11140.397946] usb 2-1.2: New USB device strings: Mfr=1, Product=2, SerialNumber=0
[11140.397952] usb 2-1.2: Product: DVB-T 2
[11140.397956] usb 2-1.2: Manufacturer: Afatech
[11140.557325] usbcore: registered new interface driver dvb_usb_af9015
[11140.557753] usb 2-1.2: dvb_usb_v2: found a 'Genius TVGo DVB-T03' in cold state
[11140.557900] usb 2-1.2: dvb_usb_v2: Did not find the firmware file 'dvb-usb-af9015.fw'. Please see linux/Documentation/dvb/ for more details on firmware-problems. Status -2
[11140.557905] usb 2-1.2: dvb_usb_v2: 'Genius TVGo DVB-T03' error while loading driver (-2)

so searched a bit and found + downloaded rpm from: http://pkgs.org/fedora-centos-rhel-opensuse-mandriva/olea/dvb-usb-af9015-fw-4.95.0-1.noarch.rpm.html

Afterwards installed the downloaded firmware via:

sudo rpm -ivh dvb-usb-af9015-fw-4.95.0-1.noarch.rpm

And finally unplugged and plugged to usb again to see:

[11796.948500] usb 2-1.1: new high-speed USB device number 7 using ehci-pci
[11797.038703] usb 2-1.1: New USB device found, idVendor=0458, idProduct=4012
[11797.038714] usb 2-1.1: New USB device strings: Mfr=1, Product=2, SerialNumber=0
[11797.038720] usb 2-1.1: Product: DVB-T 2
[11797.038725] usb 2-1.1: Manufacturer: Afatech
[11797.041521] usb 2-1.1: dvb_usb_v2: found a 'Genius TVGo DVB-T03' in cold state
[11797.042331] usb 2-1.1: dvb_usb_v2: downloading firmware from file 'dvb-usb-af9015.fw'
[11797.119176] usb 2-1.1: dvb_usb_v2: found a 'Genius TVGo DVB-T03' in warm state
[11797.542424] usb 2-1.1: dvb_usb_v2: will pass the complete MPEG2 transport stream to the software demuxer
[11797.542542] DVB: registering new adapter (Genius TVGo DVB-T03)
[11797.575204] i2c i2c-7: af9013: firmware version 4.95.0.0
[11797.578207] usb 2-1.1: DVB: registering adapter 0 frontend 0 (Afatech AF9013)...
[11797.589973] tda18271 7-00c0: creating new instance
[11797.597341] TDA18271HD/C1 detected @ 7-00c0
[11797.802656] Registered IR keymap rc-empty
[11797.802848] input: Genius TVGo DVB-T03 as /devices/pci0000:00/0000:00:1d.0/usb2/2-1/2-1.1/rc/rc0/input12
[11797.803814] rc0: Genius TVGo DVB-T03 as /devices/pci0000:00/0000:00:1d.0/usb2/2-1/2-1.1/rc/rc0
[11797.803825] usb 2-1.1: dvb_usb_v2: schedule remote query interval to 500 msecs
[11797.803832] usb 2-1.1: dvb_usb_v2: 'Genius TVGo DVB-T03' successfully initialized and connected

so firmware seems to work now.

Generating channels.conf

Following command did the job for me (as I’m located in Bayern):

scandvb /usr/share/dvb/dvb-t/de-Bayern  > channels.conf

Enjoying the broadcasts

There is no me-tv (that I’m used to it from Ubuntu world), still, there is a vlc, that does the job for me quite well. Just running:

vlc channels.conf

made it and I could watch the game!

Eclipse plugin for svn integration

I’ve found it tricky to install Eclipse plugin providing svn integration.

The problem is:

• which plugin to choose,
• which version of the plugin to choose as well as
• which packages to install.

I’m running latest Eclipse (Kepler SR2) and plugin of my choice is subclipse.

The Eclipse plugin update site has to be chosen based on the svn version installed on the machine (check it running svn --version):

• for version 1.8.x: http://subclipse.tigris.org/update_1.10.x and
• for version 1.7.x: http://subclipse.tigris.org/update_1.8.x.

Following packages were required for installation (for me):

• Subclipse (Required)
• SVNKit Client Adapter (Not Required)
• SVNKit Library

That’s it! As it covers my Eclipse/svn day-to-day workflow.

Markdown editor for blogger

Markdown is quite popular these days and I like it for it’s simplicity. So I’ve been searching for a way to integrate it in my blog writing workflow. In Blogger, it’s unfortunately not an option.

There I could go:

• either for WYSIWYG editor or
• manual HTML writing.

However WYSIWYG generates too much HTML I don’t need. Moreover some pre-publish manual HTML editing is a pain then. And the other option I consider it rather slow/unproductive.

So, I looked for other options, and found stackedit.io. Even this post is (and couple others as well) written in it.

stackedit

I consider Stackedit great as it:

• is Markdown editor,
• is open source (see github repo),
• has awesome UI (including live preview, key bindings, …),
• has integration with other popular services that I use anyway:
  
  • for storage (google drive, …) and
  • for publishing (blogger,…),
• has active community (just check the stars count and commit activity on the github repo) and
• has impressive feedback time (resolution on my issues/questions came in couple hours)

So my post writing/editing workflow goes like this:

1. write/edit post on stackedit.io,
2. sync it to google drive and
3. publish/republish it to blogger.

That’s it! No further in-blogger updates required!

Blogger specifics

Still there are some specifics in my workflow (to provide smooth blogger integration).

• using the Interpreted variables: title and tags via:

---
title: Markdown editor for blogger
tags: blogger markdown stackedit.io
---

• syntax highlighting via prettify (I use in my blog posts) on blogger works for me using: Github fenced code blocks

p6spy 2.0.0 is out!

After 8 years P6spy came to it’s next stable release!
You can get it here.

The last stable release happened (based on maven central repo) on 27-Dec-2005 (1.3 version). That is quite some time, so one would expect many things to happen in the meantime. Well the truth is that project was (half) dead for quite some time.

Changes

I can’t comment on full history since 1.3 release (since my interest in project started last summer), still I’ve noticed following:

• project hosting was moved from sourceforge to github,
• major part of the legacy code was refactored,
• Java 6/7 JDBC API support introduced,

• proxying via modified JDBC URLs only was implemented,
  So for for MySQL original url would be:

  jdbc:mysql://<hostname>:<port>/<database>

  the one proxied via p6spy would one:

  jdbc:p6spy:mysql://<hostname>:<port>/<database>

  without a need for any further configuration,

• XA Datasource support has been introduced,
• configuration via:
  
  • system/environment properties and
  • JMX properties
  • as an alternative to file configuration only
  • or even zero config use case supported,
• slf4j support (more flexible as previously used log4j),
• junit tests were migrated to junit 4 (well most of the old ones were failing anyway),
• Continuous integration using Travis was setup providing testing on popular:
  
  • DB systems (namely: Oracle, DB2, PostgreSQL, MySQL, H2, HSQLDB, SQLite, Firebird, and Derby), see build status on: travis-ci as well as
  • application servers (namely: Wildfly 8, JBoss 4.2, 5.1, 6.1, 7.1, Glassfish 3.1, 4.0, Jetty 7.6, 8.1, 9.1, Tomcat 6, 7, 8, Resin 4, Jonas 5.3 and Geronimo 2.1, 2.2), see build status on: travis-ci.

Full changelog

For the full changelog, see issues fixed in: 2.0.0-alpha1 as well as 2.0.0.

Postgres and Oracle compatibility with Hibernate

There are situations your JEE application needs to support Postgres and Oracle as a Database.
Hibernate should do the job here, however there are some specifics worth mentioning.
While enabling Postgres for application already running Oracle I came across following tricky parts:

• BLOBs support,
• CLOBs support,
• Oracle not knowing Boolean type (using Integer) instead and
• DUAL table.

These were the tricks I had to apply to make the @Entity classes running on both of these.

Please note I’ve used Postgres 9.3 with Hibernate 4.2.1.SP1.

BLOBs support

The problem with Postgres is that it offers 2 types of BLOB storage:

• bytea - data stored in table
• oid - table holds just identifier to data stored elsewhere

I guess in the most of the situations you can live with the bytea as well as I did. The other one as far as I’ve read is to be used for some huge data (in gigabytes) as it supports streams for IO operations.

Well, it sounds nice there is such a support, however using Hibernate in this case can make things quite problematic (due to need to use the specific annotations), especially if you try to achieve compatibility with Oracle.

To see the trouble here, see StackOverflow: proper hibernate annotation for byte[]

All- the combinations are described there:

annotation                   postgres     oracle      works on
-------------------------------------------------------------
byte[] + @Lob                oid          blob        oracle
byte[]                       bytea        raw(255)    postgresql
byte[] + @Type(PBA)          oid          blob        oracle
byte[] + @Type(BT)           bytea        blob        postgresql

where @Type(PBA) stands for: @Type(type="org.hibernate.type.PrimitiveByteArrayBlobType") and @Type(BT) stands for: @Type(type="org.hibernate.type.BinaryType").

These result in all sorts of Postgres errors, like:

ERROR: column “foo” is of type oid but expression is of type bytea

or

ERROR: column “foo” is of type bytea but expression is of type oid

Well, there seems to be a solution, still it includes patching of Hibernate library (something I see as the last option when playing with 3.rd party library).

There is also a reference to official blog post from the Hibernate guys on the topic: PostgreSQL and BLOBs. Still solution described in blog post seems not working for me and based on the comments, seems to be invalid for more people.

BLOBs solved

OK, so now the optimistic part.

After quite some debugging I ended up with the Entity definition like this :

@Lob
private byte[] foo;

Oracle has no trouble with that, moreover I had to customize the Postgres dialect in a way:

public class PostgreSQLDialectCustom extends PostgreSQL82Dialect {

    @Override
    public SqlTypeDescriptor remapSqlTypeDescriptor(SqlTypeDescriptor sqlTypeDescriptor) {
    if (sqlTypeDescriptor.getSqlType() == java.sql.Types.BLOB) {
      return BinaryTypeDescriptor.INSTANCE;
    }
    return super.remapSqlTypeDescriptor(sqlTypeDescriptor);
  }
}

That’s it! Quite simple right? That works for persisting to bytea typed columns in Postgres (as that fits my usecase).

CLOBs support

The errors in misconfiguration looked something like this:

org.postgresql.util.PSQLException: Bad value for type long : ...

So first I’ve found (on String LOBs on PostgreSQL with Hibernate 3.6) following solution:

@Lob
@Type(type = "org.hibernate.type.TextType")
private String foo;

Well, that works, but for Postgres only.

Then there was a suggestion (on StackOverflow: Postgres UTF-8 clobs with JDBC) from to go for:

@Lob
@Type(type="org.hibernate.type.StringClobType")
private String foo;

That pointed me the right direction (the funny part was that it was just a comment to some answers). It was quite close, but didn’t work for me in all cases, still resulted in errors in my tests.

CLOBs solved

The important was @deprecation javadocs in the org.hibernate.type.StringClobType that brought me to working one:

@Lob
@Type(type="org.hibernate.type.MaterializedClobType")
private String foo;

That works for both Postgres and Oracle, without any further hacking (on Hibernate side) needed.

Boolean type

Oracle knows no Boolean type and the trouble is that Postgres does. As there was also some plain SQL present, I ended up In Postgres with error:

ERROR: column “foo” is of type boolean but expression is of type integer

I decided to enable cast from Integer to Boolean in Postgres rather than fixing all the plain SQL places (in a way found in Forum: Automatically Casting From Integer to Boolean):

update pg_cast set castcontext = 'i' where oid in ( select c.oid from pg_cast c inner join pg_type src on src.oid = c.castsource inner join pg_type tgt on tgt.oid = c.casttarget where src.typname like 'int%' and tgt.typname like 'bool%');

Please note you should run the SQL update by user with provileges to update catalogs (probably not your postgres user used for DB connection from your application), as I’ve learned on Stackoverflow: Postgres - permission denied on updating pg_catalog.pg_cast.

DUAL table

There is one more specific in the Oracle I came across. If you have plain SQL, in Oracle there is DUAL table provied (see more info on Wikipedia on that) that might harm you in Postgres.

Still the solution is simple. In Postgres create a view that would fill the similar purpose. It can be created like this:

create or replace view dual as select 1;

Conclusion

Well that should be it. Enjoy your cross DB compatible JEE apps.

yeoman - testing with mocha and chai instead of jasmine

I've been playing around with yeoman these days. It seems like project worth a try for my free time single page application.

After the inital setup (using AngularJS generator) I decided to let my tests run with mocha + chai (with expect asertions) rather than jasmine (generated by default).

In fact it was super-simple. The diffs that matter:

diff --git a/yo/karma.conf.js b/yo/karma.conf.js
index bc0e168..385045f 100644
--- a/yo/karma.conf.js
+++ b/yo/karma.conf.js
@@ -7,7 +7,7 @@ module.exports = function(config) {
     basePath: '',
 
     // testing framework to use (jasmine/mocha/qunit/...)
-    frameworks: ['jasmine'],
+    frameworks: ['mocha', 'chai'],
 
     // list of files / patterns to load in the browser
     files: [
diff --git a/yo/package.json b/yo/package.json
index 3e3db14..9649a55 100644
--- a/yo/package.json
+++ b/yo/package.json
@@ -34,7 +34,8 @@
     "karma-html2js-preprocessor": "~0.1.0",
     "karma-firefox-launcher": "~0.1.3",
     "karma-script-launcher": "~0.1.0",
-    "karma-jasmine": "~0.1.5",
+    "karma-mocha": "~0.1.1",
+    "karma-chai": "~0.0.2",
     "karma-coffee-preprocessor": "~0.1.2",
     "requirejs": "~2.1.10",
     "karma-requirejs": "~0.2.1",
diff --git a/yo/test/spec/controllers/main.js b/yo/test/spec/controllers/main.js
index eb7cda2..f9becde 100644
--- a/yo/test/spec/controllers/main.js
+++ b/yo/test/spec/controllers/main.js
@@ -17,6 +17,6 @@ describe('Controller: MainCtrl', function () {
   }));
 
   it('should attach a list of awesomeThings to the scope', function () {
-    expect(scope.awesomeThings.length).toBe(3);
+    expect(scope.awesomeThings.length).to.equal(3);
   });
 });

Afterwards just running (to fetch missing packages):

npm install

and tests succeed:

grunt test