In our project we use checkstyle as static analyzer (as maven plugin, I mean maven-checkstyle-plugin). I am new to this tool, and there is one thing I want to find out - what is the difference between failOnError and failOnViolation?
I think the plugin documentation is pretty clear, but the minimum expectation is going to be:
Do you want a log of the errors before failing the build? Then use failOnViolation
Or
Do you want to fail the build immediately, without waiting for errors to be logged? Then use failsOnError
Of course, if you set logViolationsToConsole to false (or on the CLI as -Dcheckstyle.console=false), then both act almost identically and can cause some confusion.
What is the best way to create a custom OpenShift cartridge?
Looking at documentation and examples, I am seeing a lot of old-school compiling from source installation of the component that the cartridge needs to run.
Some examples https://www.openshift.com/blogs/lightweight-http-serving-using-nginx-on-openshift https://github.com/boekkooi/openshift-diy-nginx-php/blob/master/.openshift/action_hooks/build_nginx https://github.com/razorinc/redis-openshift-example/blob/master/.openshift/action_hooks/build and a ton of others are compiling from source..
I need to create some custom cartridges on my project, but doing it this way feels wrong.
Is there any reason I cant use yum and puppet/augeas to do the building, instead of curl, make and sed?
Or is this the best practice? In that case, why are we doing this 2000 style?
I'll do my best to explain this the best way I can. Feel free to let me know If I need to explain anything in more detail.
I'm assuming you're creating a custom binary cartridge (ie. you're creating a language cartridge such as ruby, python, etc.). Since none of the nodes have that binary installed on the system the custom cartridge you're creating will need to provide that binary and its libraries.
When you install a package with yum its going to install items in several different directories (/etc, /usr/, /var, etc). Since you're creating cartridge that will be copied over to several nodes you'll need to package all these items in a way that can be copied over to a node and then be executed without having to install them to the system.
As for doc's, I would suggest taking a look at these:
https://www.openshift.com/developers/download-cartridges
https://www.openshift.com/blogs/new-openshift-cartridge-format-part-1
https://www.openshift.com/blogs/new-openshift-cartridge-format-part-2
I am trying to use clojure kORMa with compojure, the webapp is here nepleaks.
First I added kORMa dependency
[korma "0.3.0-RC4"]
lein deps works cool, then created src/hotel/conf/datasource.clj which looks as follows :
(ns hotel.conf.datasource)
(use [korma.db])
(use [korma.core])
)
(defdb db (mysql {:db "nepleaks"
:user "root"
:password "mysql55"}))
(defentity users)
lein ring server is neither throwing any exception nor creating the database entity.
Am I missing something? Do I need more configuration?
My understanding is that Korma does not create the database nor the database tables for you, so you need to create them beforehand manually.
In the spirit of an example being worth 1000 words, I would recommend checking out the luminus clojure web framework. See http://www.luminusweb.net/ This framework sets up a lot of really useful libraries with very basic example use, including korma, ragtime, ring, compojure and some other nice libs. The default templates show you lots of nice ways of using these libraries and how to integrate them. You can then pick the bits you want. All you need to do to get started is run
lein new luminus <your project> +mysql
The web site has good documentation and the code is pretty clean and easy to understand. Saves hours when wanting to get up and running with a new library you are still comming to terms with.
you need lein ragtime plugin manage the database.
Ok, I've asked a few related questions here and only ended up with more questions and I realized now it's because I don't have enough background info. So I'll make it more generic:
I need to make a simple web application. Static HTML/JQuery pages will send AJAX POST requests to some server side code, which will:
Read the POST variables passed in
Run some very simple logic
Hit a MySQL database for simple CRUD ops
Return a plain string of data to be consumed by the javascript on the page
I was assuming Ruby was a good choice for this as everyone is raving about how well it's designed, and I've been playing with it - not RoR, just Ruby for simple scripting tasks - and I kind of like it.
My question is, I'm hopelessly confused by the trillion helper libraries and frameworks out there. I don't know what these are and therefore if I need any/all of them: Rack, Sinatra, Camping, mod_ruby, FastCGI, etc.
Would it be easier to just learn PHP and us that? Or can I get away with just dropping my .rb files into the cgi-bin folder(I'm using Apache for hosting) and use the ruby cgi library to get my variables?
EDIT: As far as Rails, I'm just assuming that it's overkill for what I want but I might be wrong. I looked at it, and it seemed cool for generating data based web sites quickly, but that's not what I'm trying to do. I don't want any forms pages for the user. I don't want them entering data or viewing records. I don't even want to return any HTML. I just want a ruby script to sit on the server, get passed a few variables in a post request, and return a JSON string in response. I will need some basic cookie/session/state managment
This is a really easy thing to do in C# and ASP.NET with webservices, but it seems very confusing with the open source technologies.
You don't want to use any feature from a fully blown framework so don't use one. Less code = less bugs = less security nightmares.
CGI
CGI has some performance drawbacks in comparison to other methods, but is still (in my opinion) the simplest and easiest to use one. This is how you use the builtin cgi library:
require "cgi"
cgi= CGI.new
answer= evaluate(cgi.params)
cgi.out do
answer
end
rack
Another low tech easy to use variant would be rack. Rack is an abstraction layer which works for many webserver interfaces (cgi, fastcgi, webrick, …). It's simplicity can be compared to the one of only using cgi. Put the following into a file wich ends with .ru into your cgi directory.
#!/usr/bin/rackup
require "rack/request"
run (lambda do |env|
request= Rack::Request(env)
anwser= evaluate(request.params)
return [200, {}, answer]
end)
This does not seem very different from cgi, but it gives you much more possibilities. If youst execute this file on your local machine rackup will start the webrick webserver. This webserver will deliver the webpages you described in your .ru file.
Other interfaces
fast-cgi
fast-cgi works almost like CGI. The difference is, in CGI your script get's started for every request it has to work on. With fast-cgi, your script only starts once for all requests. There is a library available to write fast-cgi script in ruby.
mod_ruby
mod_ruby is a builtin ruby interpreter for apache. It works analog to mod_php in apache.
mongrel
mongrel is a standalone webserver for ruby applications. This is a simple hello world example with it.
require 'mongrel'
class SimpleHandler < Mongrel::HttpHandler
def process(request, response)
response.start(200) do |head,out|
head["Content-Type"] = "text/plain"
out.write("hello world!\n")
end
end
end
h = Mongrel::HttpServer.new("0.0.0.0", "3000")
h.register("/hello", SimpleHandler.new)
h.run.join
Mongrel is often used for rails and other ruby frameworks. Most people use an apache or something else on port 80. This webserver than distributes the requests to several mongrel servers running on other ports. I think this is totaly overkill for your needs.
phusion passenger
passenger is also called mod_rails or mod_rack. It is a module for apache and nginx to host rails and rack applications. According to their websites rails with passenger uses 1/3 less ram than rails alone. If you write your software for rack, you could make it a little faster by using passenger, instead of cgi or fast-cgi.
Use jQuery and PHP.
Both technologies are well documented, and you should be able to get an application up and running in a matter of hours. You sound like you know a thing or two - talking about CRUD operations and so on - so I won't bore you with examples. And as far as JSON goes, there are probably a million PHP libraries out there, for outputting JSON objects.
Sinatra is very simple to learn and use. It's also easy to deploy with the use of Phusion Passenger (which is like mod_php for ruby frameworks like Rails and Sinatra). Instructions here: http://blog.squarefour.net/2009/03/06/deploying-sinatra-on-passenger/
If you find that you need more than what Sinatra will give you, I recommend Rails. Setting up that with Passenger is even easier since hardly any configuration is required. (see modrails.com).
PHP is very easy to use because it's designed specifically for this sort of thing. Want to read POST variables? They're in $_POST. Want to query MySQL? mysql_query("SELECT `something` FROM `table`");. And if you ever need help, Google searches for "php what_you_need_to_do" almost always return results on php.net, which is very helpful.
And for what you're doing, you don't need any additional frameworks.
I am curious about what I perceive to be your resistance to trying Rails. You say that you want "to spend more time on the scripting itself and less on configuration", and yet you seem to dismiss Rails out of hand. Rails is all about convention over configuration. If you take the time to learn how Rails does things, you can get an incredible amount of functionality "for free" just by following the conventions of the framework.
If you want to make a simple web app, Rails is really a very painless and good way to start. You can use a sqlite database and not even mess with MySQL (won't scale, but for learning or simple apps it's fine). It's true that there are simpler frameworks, but since you seem new to web programming, I'd recommend that you start with what will get you the most support in terms of documentation and knowledgeable folks. Follow the old adage: Get it working first, then optimize later.
The only sticking point I can see is the Apache integration... The consensus on Rails deployment these days seems to be focused on using lightweight httpds in place of Apache. There is a mod_fcgid which seems to be the best way to do it with Apache (mod_ruby is deprecated, buggy and slow, last I read) if you can do custom mods. Or there's Phusion Passenger, which seems to be the latest and greatest way to do it. Running Rails in a standard CGI environment will yield awful performance (but that goes for any CGI framework, really) due to the overhead of executing the interpreter + framework for every request. You'll get much better performance if you go with something that keeps the interpreter + framework in memory.
I personally like Django. I had a problem with Ruby on Rails where I just got overwhelmed by everything when I just wanted to do something simple, which it sounds like you want to do (since you said ROR feels like overkill). The cool thing I found with Django is that if you WANT everything, then you can get everything by plugging it in...but if you want less then you just don't plug in that technology and it's that much more lightweight.
Take for example "views". Django, like ROR, uses MVC. But if you just want to return a string of data and don't need the view, then you don't need to plug in the view. But if later on you decide that it will be more organized in a view then you can easily plug it in at that time.
Here's their website: http://www.djangoproject.com/
I'm looking for some way to replace patterns in files with values, during build time.
E.g. a configuration file may look the same except that different machines requires different hostnames in some setting. In that case i want to have a template file, where hostname is replaced with ##hostname##
Then when building, I want to create separate versions of the file with the patter replaced with the correct value for each environment.
In ant you could use "replace", is there something similar in maven2? I know that I can run ant form maven, but if theres a maven plugin doing it, I'd prefer that one.
I found http://code.google.com/p/maven-replacer-plugin/
but it's very new...
Suggestions?
Thanks!
The resource plugin is what does filtering in Maven. IIRC it supports the ant syntax as well as the ${foo} maven syntax.
keep up with the maven-replacer-plugin, the project is still alive and well
Maven's resource filtering might help you out.
Why don't you use the Maven-ant plug-in?