“L’Italia è una Repubblica democratica fondata sul lavoro. La sovranità appartiene al popolo, che la esercita nelle forme e nei limiti della Costituzione”.
Silvio, la Costituzione è stata fatta così per difenderci da persone come te.
code snippets
“L’Italia è una Repubblica democratica fondata sul lavoro. La sovranità appartiene al popolo, che la esercita nelle forme e nei limiti della Costituzione”.
Silvio, la Costituzione è stata fatta così per difenderci da persone come te.
Normally hudson will kill all the processes that have been created during the build tasks, this will ensure that after the build the system will be left in a consistent state.
However, sometimes, you may need to keep those tasks running, for example we use Hudson to automatically perform nightly installations on our test servers, in those cases you want that hudson will not kill the processes that have been started during the installation (i.e. jboss, tomcat).
A web search got me this page:
http://www.suryasuravarapu.com/2010/07/spawning-a-process-from-hudson.html
The page explain how to setup the project to prevent hudson from killing started process, basically it says:
set the environment variable BUILD_ID to something like 'dontKillMe' in the
process that should stay alive.Hudson looks for that environment variable when cleaning up stray processes.
However, if you are new to hudson, it does not say how to achieve that, well, you can use the setenv plugin, to install it you have to simply go to the “manage hudson” page where you will find the “manage plugins” shortcut, in the plugins page (in the “available” column) you can find the SetEnv plugin, just mark it and then click on “install” at the end of the page.
At the end of the plugin installation you can restart hudson and the Environment Variables field will appear when configuring the project.
Happy CI.
Installing hudson is pretty easy, you can download the war from http://java.net/projects/hudson/downloads/directory/war, for example you can download it directly from the server:
wget http://java.net/projects/hudson/downloads/download/war%252Fhudson-1.395.war
Then you can create a script like this that will start it:
#!/bin/bash
dir=/home/polopoly/hudson<
nohup java -jar $dir/hudson.war --httpPort=18080 --ajp13Port=18009 >$dir/hudson.log 2>&1 &;
Now you are free to create your tasks.
I’ve just replaced the old FCKEditor with CKEditor, in this page you can find some really helpful hints to achieve better performance: http://ckeditor.com/blog/CKEditor_Loading_performance_details
As always, measure your performance to understand if the changes you have done are really worth it.
Have a look at http://mickeyben.com/2009/12/30/using-nginx-as-a-load-balancer.html to see how to use Nginx as a load balancer.
upstream yoursite {
server yoursite1.yoursite.com;
server yoursite2.yoursite.com;
}
server {
server_name www.yoursite.com;
location / {
proxy_pass http://yoursite;
}
}The configuration is really simple, additionally you can use the ip_hash directive to forward all the request of the same client to the same server:
upstream yoursite {
ip_hash;
server yoursite1.yoursite.com;
server yoursite2.yoursite.com;
}Suppose your web page in the CMS is already using one version of jquery and you want to add a new HTML element with a javascript library that needs a more recent version of jquery.
You may think to use the google api like:
<script>
google.load("jquery", "1.3.2");
</script>
and then use the noConflict option like:
<script>
function OnLoad(){
jq13 = jQuery.noConflict(true);
}
google.load("jquery", "1.3.2");
google.setOnLoadCallback(OnLoad);
</script>
Unfortunately this will not work correctly in IE…so you may want to use the technique to dynamically load a javascript (see http://themnd.blogspot.com/2009/12/dynamically-load-javascript-on-webpage.html)
<script>
function OnLoad(){
jq13 = jQuery.noConflict(true);
}
function loadJQuery() {
loadJS('http://ajax.googleapis.com/ajax/libs/jquery/1.3.2/jquery.min.js', function() {
OnLoad();
});
}
jQuery(document).ready(function() {
setTimeout(function() {
loadJQuery();
}, 2000);
});
</script>
at this point you can load the needed library in the OnLoad function before setting the noConflict option. Please note that we are still using the google cdn to fetch the jquery library. The OnLoad function would be something like:
function OnLoad(){
loadJS(plugin_url, function() {
jq13 = jQuery.noConflict(true);
// now you can call the plugin methods here:
jq13.XXXX()
});
}
When you work with a CMS sometimes you may need to dynamically add a javascript library on the fly.
The easiest thing to do is:
var headID = document.getElementsByTagName("head")[0];
var newScript = document.createElement('script');
newScript.type = 'text/javascript';
newScript.src = url;
headID.appendChild(newScript);
But what if you want to be notified when the javascript has been loaded? You can add the onload parameter:
newScript.onload = function(){
func();
};
but you will find out that this will not work in IE, so what you really need to use is:
function loadJS(url, func) {
var headID = document.getElementsByTagName("head")[0];
var newScript = document.createElement('script');
newScript.type = 'text/javascript';
if (newScript.readyState) { //IE
newScript.onreadystatechange = function() {
if (newScript.readyState == "loaded" || newScript.readyState == "complete") {
newScript.onreadystatechange = null;
func();
}
};
} else { //Others
newScript.onload = function(){
func();
};
}
newScript.src = url;
headID.appendChild(newScript);
}
We will rely on the onreadystatechange event that IE will fire, unfortunately that event is, in reality, the progress of the loading and sometimes it reach the “complete” state, sometimes only the “loaded” state, so if we really want to be sure that our callback will be called we have to get the first one and then cancel the event.
You can learn more at: http://www.nczonline.net/blog/2009/06/23/loading-javascript-without-blocking/
In a similar way you can also dynamically load CSS:
function loadCss(url) {
var headID = document.getElementsByTagName("head")[0];
var cssNode = document.createElement('link');
cssNode.type = 'text/css';
cssNode.rel = 'stylesheet';
cssNode.href = url;
cssNode.media = 'screen';
headID.appendChild(cssNode);
}
This post is mostly for who have to use an HTML template generated by an external Web designer and making it “dynamic”, that means we are handed the html page and we have to create the template that will be used by a web application to render content using that templates like polopoly (www.atex.com) but it is also true for other systems like joomla.
1. do not use cellpadding and cellspacing in tables, set them to 0 and eventually use CSS to control spacing, firefox, chrome and internet explorer interprets them differently.
2. do not use SPACER tag since internet explorer will not understood it, use a transparent 1 pixel gif with an img tag instead.
3. do not comment a css line with // since internet explorer will read it anyway, comment using /* and */ if you really need to.
4. if you start doing evaluate the layout with internet explore it would be more easy to adjust it for firefox and chrome (if any adjustment needs to be made).
5. always start with IE6 if you need to target it since it is the poorer browser that probably will need more care.
In a java program when you want to execute an external program you will normally write something like this:
Runtime rt = Runtime.getRuntime();
Process proc = rt.exec(cmd);
proc.waitFor();
Where cmd is the command line needed to be executed.
However, if you run the same piece of code from a tomcat servlet it may just hung up until you kill it. The problem is documented in the documentation of the exec method, you need to consume the inputStream and the outputStream, so a quick way of doing it is to start other two threads that will consume those stream.
First create the thread class that will process the stream:
class StreamGobbler extends Thread
{
InputStream is;
String type;
StreamGobbler(InputStream is, String type)
{
this.is = is;
this.type = type;
}
public void run()
{
try {
InputStreamReader isr = new InputStreamReader(is);
BufferedReader br = new BufferedReader(isr);
String line=null;
while ( (line = br.readLine()) != null) {
System.out.println(type + ">" + line);
}
} catch (IOException ioe) {
ioe.printStackTrace();
}
}
}
Now we can modify the previous code to handle InputStream and ErrorStream:
Runtime rt = Runtime.getRuntime();
Process proc = rt.exec(cmd);
StreamGobbler errorGlobber = new StreamGobbler(proc.getErrorStream(), "CMDTOOL-E");
StreamGobbler outputGlobber = new StreamGobbler(proc.getInputStream(), "CMDTOOL-O");
errorGlobber.start();
outputGlobber.start();
proc.waitFor();
et voilà the code will execute just fine in a servlet.
The idea and code has been taken from: http://www.javaworld.com/javaworld/jw-12-2000/jw-1229-traps.html?page=1
If you use apache to forward request to a local running tomcat, you may encounter this error message in the error.log:
[error] (13)Permission denied: proxy: HTTP: attempt to connect to 127.0.0.1:8080 (*) failed
If you have SELinux running you have to check if it allows local connections, by running:
getsebool -a | grep httpd
you have to check that the following is written:
httpd_can_network_connect –> on
if it is off, you have to enable it:
setsebool -P httpd_can_network_connect=1
and restart your apache.
This tip has been taken from: http://swapyourbooks.com/welcome/knowledgebase/misc/selinux-allow-httpd-tomcat-proxy.html
With Visual Studio 2003 you can use member template functions, for example:
class Foo {
template <class T>
};
that you can call in this way:
void main() {
int a = 10;
Foo obj;
obj.bar(a); }
Make sure that when you declare the bar method you include also the method definition otherwise you will end-up with a linker error.
I had a soap server built in python with SOAPpy and I wanted to use it from an ASP.Net application using the visual studio 2005 built in web services tool. I encountered several problems:
For the WSDL problem you have just to define a function like:
def wsdl(self):
wsdl_text = ''
f = open('service.wsdl', 'r')
for l in f.readlines():
wsdl_text += l f.close()
return wsdl_text for the namespece part, suppose you ns is 'http://localhost/MyServer/2007/09/Services/MyService/01', you server will be:
class MySoapServer(SOAPServer):
def __init__(self, ip = "localhost", port = 8080):
from SOAPpy import NS
# visual studio want's the soap envelope defined as 's'
NS.ENV_T = "s"
self.ns = 'http://localhost/MyServer/2007/09/Services/MyService/01'
SOAPServer.__init__(self, (ip, port))
self.registerFunction('wsdl')
self.registerFunction('soap_myFunction', 'myFunction', namespace = self.ns)
In this way Visual Studio will be able to correctly call your exposed methods. Later we will see how to let Visual Studio correctly understand your results.
I had problem with a copy of a K2Vector, basically I had
an interface's method like:
class fooInterface {
virtual void setVector(const K2Vector<fooObject>& src) { fVec =
src; };
protected:
K2Vector<fooObject> fVec;
}
The linker's error I had was like "no operator= was
defined for K2Vector<fooObject>"...I had the K2Vector.h included but it
didn't solved my problem. It was not the first copy between vectors in the
plugin so I was a bit shooked by this error, after tries I include
"K2Vector.tpp" and that solved the problem. Really strange.
If you look at
the bottom of K2Vector.h you can see that it include "K2Vector.tpp".
Anyway, that's solved my problem and another lesson learned.
Class
{
kSampleCmdBoss,
kInvalidClass,
{
IID_ICOMMAND, kSampleCmdImpl,
IID_ISTRINGDATA, kStringDataImpl,
IID_ISAMPLENAMEDATA, kStringDataImpl,
}
},
resource ImplementationAlias(1)
{
{
kSampleNameDataImpl, kStringDataImpl,
}
};
Class
{
kSampleCmdBoss,
kInvalidClass,
{
IID_ICOMMAND, kSampleCmdImpl,
IID_ISTRINGDATA, kStringDataImpl,
IID_ISAMPLENAMEDATA, kSampleNameDataImpl,
}
},
Just look at the InDesign knowledge base (http://support.adobe.com/devsup/devsup.nsf/docs/50352.htm).
Hint: use the stream Luke.