CGI vars and their JSP equivilant

I never seem to remember the exact syntax for getting CGI variables in JSP. So, I put what I found here so I would not have to look again.

SERVER_NAME request.getServerName();
SERVER_SOFTWARE request.getServletContext().getServerInfo();
SERVER_PROTOCOL request.getProtocol();
SERVER_PORT request.getServerPort();
REQUEST_METHOD request.getMethod();
PATH_INFO request.getPathInfo();
PATH_TRANSLATED request.getPathTranslated();
SCRIPT_NAME request.getServletPath();
DOCUMENT_ROOT request.getRealPath("/");
QUERY_STRING request.getQueryString();
REMOTE_HOST request.getRemoteHost();
REMOTE_ADDR request.getRemoteAddr();
AUTH_TYPE request.getAuthType();
REMOTE_USER request.getRemoteUser();
CONTENT_TYPE request.getContentType();
CONTENT_LENGTH request.getContentLength();
HTTP_ACCEPT request.getHeader("Accept");
HTTP_USER_AGENT request.getHeader("User-Agent");
HTTP_REFERER request.getHeader("Referer");

Country codes for UPS

I'm not a big fan of the UPS site, so I try not to go there unless I have to - so I may be re-inventing the wheel here. I was wanting to do international shipping. While doing that, I found that UPS was kind enough to give a list of 2 Digit country codes with the full names inside their PDF. When you copy and paste, you get something like this:

Afghanistan AF
Åland Islands AX
Albania AL
Algeria DZ
American Samoa AS
Andorra AD
Angola AO
Anguilla AI
Antarctica AQ
Antigua and Barbuda AG
Argentina AR

....

This isn't the most helpful format - not really easy to parse with a bash script. I decided to continue my ruby education and create a short little Ruby script to turn this data into very basic database data. My needs were simple, so I created:

create table country (
 code char(2) not null primary key,
 country varchar(255) not null,
 enabled tinyint(1) default 0
);

I created a flat file with the data from the PDF - similar to above, but a full listing in a file and then wrote this ruby script:


#!/usr/bin/env ruby
  
# the file we want to create
writer = File.open("build-countries.sql", "w+")

# read in all the ups data and write it out in a format we want.
File.open("ups-country-codes.txt", "r") do |reader|
  while (!reader.eof)
   
    # read in a line
    line = reader.gets().chomp()

   
    # split into the 2 parts
    first = line.slice(0,line.length-3)
    second = line.slice(line.length-2,line.length )

    # ruby's string power makes formatting the output a snap
    out = "insert into country (country,code,enabled) values('#{first}','#{second}',0);"
    writer.puts(out);
  end
end

writer.close
#######################################

The result creates some nice SQL for me:

insert into country (country,code,enabled) values('Afghanistan','AF',0);
insert into country (country,code,enabled) values('Åland Islands','AX',0);
insert into country (country,code,enabled) values('Albania','AL',0);
insert into country (country,code,enabled) values('Algeria','DZ',0);
insert into country (country,code,enabled) values('American Samoa','AS',0);
insert into country (country,code,enabled) values('Andorra','AD',0);
insert into country (country,code,enabled) values('Angola','AO',0);
insert into country (country,code,enabled) values('Anguilla','AI',0);
insert into country (country,code,enabled) values('Antarctica','AQ',0);
insert into country (country,code,enabled) values('Antigua and Barbuda','AG',0);
insert into country (country,code,enabled) values('Argentina','AR',0);
......

Using wget to check if files exist

I had a set of xml files, each file contained a url in <url></url> and I need a way to check if the URL was valid. I came up with the following bash script


#!/bin/bash
for e in `find . -type f`; do

#strip the the junk before and after to get a clean url
url=`grep url $e|sed 's/.*<url>//'|sed 's/.<\/url>//'`

# i put wget output to a file - stdout and stderr to a temp file
wget -nv --spider $url > tmp.file 2>&1

# if it was good, then a grep count should return 0
# bad files will get added to a new delete script
# good files go to a good list - this way we can count the
# results with wc -l

if [ "`grep -c 404 tmp.file`" != "0" ]; then
  echo "rm -f $e" >> delete.lst
else
  echo "$e" >> good.lst
fi

Russ

Getting raw XML when testing Axis clients

I'm sure there's probably a really easy way to get the XML using code generated by Axis, but it wasn't so easy for me to find. I was using Axis 1.5 to do UPS's Tradablility Web Services. The web services are on https, so I didn't have a lot of luck with tcpmon to watch the data. The wsdl2java tool created a Stub and a unit test. I traced the unit test and found that the stub was making the web services calls. If you are looking, find your stub and look for where the code uses the org.apache.axis2.client.OperationClient. That code block should create a org.apache.axiom.soap.SOAPEnvelope. The SOAPEnvelope will have your XML

org.apache.axiom.soap.SOAPEnvelope env = ....

To get the XML:

String xml = env.getBody().toString();

After the web service call is made, you should see the code that gets the web service response e.g.:

org.apache.axis2.context.MessageContext _returnMessageContext = _operationClient.getMessageContext(org.apache.axis2.wsdl.WSDLConstants.MESSAGE_LABEL_IN_VALUE);

Then you can get the response XML:

org.apache.axiom.soap.SOAPEnvelope _returnEnv = _returnMessageContext.getEnvelope();

String xml = _returnEnv.getBody().toString();

I hope that makes sense.