DZone Snippets: url code

Extract the filename from a URL

jrobertson (James Robertson) — Sun, 20 Apr 2008 15:24:54 GMT

This ECMAScript extracts the filename from a URL. eg. the filename 'index4.svg' would be extracted from the document.URL with the value 'http://rorbuilder.info/r/whiteboardqueue/index4.svg'.



var regexp = /(\w|[-.])+$/

str = document.URL

a = regexp.exec(str)

alert(a[0])

Reference: Regular Expressions: Methods - Doc JavaScript [webreference.com]

Ruby: Escape, Unescape, Encode, Decode, HTML, XML, URI, URL

sikelianos (Zeke Sikelianos) — Tue, 26 Feb 2008 23:23:08 GMT

This example will show you how to escape and un-escape a value to be included in a URI and within HTML.



require 'cgi'



# escape

name = "ruby?"

value = "yes"

url = "http://example.com/?" + CGI.escape(name) + '=' + CGI.escape(value) + "&var=T"

# url: http://example.com/?ruby%3F=yes&var=T

html = %(example)

# html: example



# unescape

name_encoded = html.match(/http:([^"]+)/)[0]

# name_encoded: http://example.com/?ruby%3F=yes&var=T

href = CGI.unescapeHTML(name_encoded)

# href: http://example.com/?ruby%3F=yes&var=T

query = href.match(/\?(.*)$/)[1]

# query: ruby%3F=yes&var=T

pairs = query.split('&')

# pairs: ["ruby%3F=yes", "var=T"]

name, value = pairs[0].split('=').map{|v| CGI.unescape(v)}

# name, value: ["ruby?", "yes"]

Generate short URLs using Ruby

jrobertson (James Robertson) — Thu, 14 Feb 2008 05:00:18 GMT

This Ruby code generates shortened URLs. It has the added feature that it will not allow words to be created accidentally eg. swear words.



#!/usr/bin/ruby



class ShortUrl

  def initialize()

    @chars = ('a'..'z').to_a + ('1'..'9').to_a + ('A'..'Z').to_a

    @array_size = @chars.size

    @h = Hash.new

    @chars.each {|c| @h[c] = '0'}

    vowels = %w(a e i o u A E I O U 4 3 1 0)

    vowels.each {|v| @h[v] = '1'}

    nums = %w(2 5 6 7 8 9)

    nums.each {|n| @h[n] = '2'}

    @count = 0

    @a = Array.new(7, -1)

    @k = 0

  end 



  def iterate_chars(array_size)

    (0..array_size).each {|i| 

      increment_index(@k)

      convert_to_chars()

    }

  end

  

  def convert_to_chars()

    buffer = ''

    @a.each {|i|

      buffer << @chars[i] if i >= 0

    }

    a = buffer.reverse.scan(/./)

    k = a.length  

    if  (k > 1)  

      if ((@h[a[k-2]] + @h[a[k-1]]) != '10')

        puts buffer.reverse 

      end

    else

      puts buffer.reverse

    end

  end



  def get_short_url(count)

    if count >  @array_size

      new_count = count - @array_size

      iterate_chars(@array_size)

      get_short_url(new_count)

    else

      iterate_chars(count)

    end

  end

  

  def increment_a(i)

    if @a[i] < @array_size - 1

      @a[i] = @a[i] + 1

      return i 

    else

      @a[i] = 0

      return i += 1

    end 

  end



  def increment_index(k)

    old_k = k

    k = increment_a(k)

    

    if k != old_k

      increment_index(k)

    else

      k = 0

    end

    k

  end



end



if __FILE__ == $0

  su = ShortUrl.new  

  su.get_short_url(222761)

end

Note: You will most likely see the " `convert_to_chars': stack level too deep (SystemStackError)" error message after around 253950 iterations, I'm presuming this is expected due to the memory constraints of the language. However 253k short urls is enough for linking my website to itself.

spaces in URL

kelyar (Kelyarsky Evgeniy) — Sun, 13 Jan 2008 15:45:12 GMT

This is a workaround to get spaces in URL as pluses not %20 like it is in Rails2.



module ActionController::Routing

  class DynamicSegment

    def interpolation_chunk

      "\#{CGI.escape(#{local_name}.to_s)}"

    end



    def match_extraction(next_capture)

      default_value = default ? default.inspect : nil

      %[

        value = if (m = match[#{next_capture}])

          CGI.unescape(m)

        else

          #{default_value}

        end

        params[:#{key}] = value if value

      ]

    end

  end

end

Using Apache with RewriteMap and a text file

jrobertson (James Robertson) — Thu, 25 Oct 2007 22:51:14 GMT

This code will read a text file using Apache then upon a user request, look for a url containing /l/ eg. http://mysite.com/l/digg and rewrite the URL in full eg. http://mysite.com/gwd/feed/digg.html . The example file is shown here for completeness.

RewriteMap uses a text file as a convenient alternative for the administrator when declaring many rewrite rules. Code based on the article from ONLamp.com - A Day in the Life of #Apache http://urltea.com/1vwb





scotsman /gwd/feed/scotsman.html

digg /gwd/feed/digg.html







RewriteEngine on

RewriteMap links txt:/var/www/localhost/links.txt

RewriteRule ^/l/(.*) ${links:$1|http://mysite.com/} [R]

Note: The [R] at the end of RewriteRule means redirect, to have a clean url simply remove that switch.

*update 15-Feb-08 *
Restart Apache rather reloading the module when switching redirection on or off for a RewriteRule. The Apache version 2.2.6 (Unix) didn't pick up my settings correctly when I tried /etc/init.d/apache2 reload, instead I needed to use /etc/init.d/apache2 restart.

A simple mod_rewrite example

jrobertson (James Robertson) — Wed, 24 Oct 2007 09:11:02 GMT

Rewrite a long url as a short easy to remember url with mod_rewrite. Place this code in httpd.conf or the virtual hosts file. Original code from sitepoint - mod_rewrite: A Beginner's Guide to URL Rewriting http://urltea.com/1v6d .





RewriteEngine on

RewriteRule ^/shortcut$ /complicated/and/way/too/long/url/here

Create strings for SEO-friendly URLs

labuschin (Martin Labuschin) — Thu, 27 Sep 2007 11:53:58 GMT

this method returns string, which is perfect for SEO-friendly URLs.

features:
- converts every improper character to a hyphen
- returns a lowercase string

I POSTED AN EVEN BETTER METHOD ON MY BLOG: http://tinyurl.com/2vurbq



def create_callname (title)

	title.downcase.gsub(/[^a-z0-9]+/i, '-')

end

Rails URL Validation

jnewland (Jesse Newland) — Wed, 12 Sep 2007 14:03:00 GMT

No regexes, allows URLs with ports or IPs. Inspiration from here



  validates_each :href, :on => :create do |record, attr, value|

    begin

      uri = URI.parse(value)

      if uri.class != URI::HTTP

        record.errors.add(attr, 'Only HTTP protocol addresses can be used')

      end

    rescue URI::InvalidURIError

      record.errors.add(attr, 'The format of the url is not valid.')

    end

  end

Reading an xml file from a URL and saving it locally

jrobertson (James Robertson) — Sat, 08 Sep 2007 14:20:04 GMT

// description of your code here
This code reads an xml file, modifys an element and saves the file locally.



require 'net/http'

require 'rexml/document'



url = 'http://www.example.com/journal080907.xml'



element_name = ARGV[0] # basic_category

xpath = ARGV[1] # eg.'entries/entry'

file = ARGV[2] # eg. 'journal080907.xml'

element_value = ARGV[3] # 'ruby'



file = File.new(file,'w')

# get the XML data as a string

xml_data = Net::HTTP.get_response(URI.parse(url)).body



# extract event information

doc = REXML::Document.new(xml_data)

docx = doc



node = docx.elements[xpath]

element = node.elements[element_name]

puts element.text

element.text = 'ruby'

file.puts docx

ISAPI URL Rewrite

mornlee (mornlee) — Fri, 29 Jun 2007 08:01:32 GMT

// description of your code here



#include 

#include 



#define MAX_URL_LEN 4096



BOOL WINAPI GetFilterVersion(HTTP_FILTER_VERSION * pVer)

{

      pVer->dwFlags = (SF_NOTIFY_SECURE_PORT | SF_NOTIFY_NONSECURE_PORT | SF_NOTIFY_PREPROC_HEADERS | SF_NOTIFY_ORDER_HIGH);

      pVer->dwFilterVersion = HTTP_FILTER_REVISION;

      strcpy_s(pVer->lpszFilterDesc,9,"Blah blah blah");

      return TRUE;

}



void ReMapURLs(CHAR* pUrl,HTTP_FILTER_CONTEXT* pfc,PHTTP_FILTER_PREPROC_HEADERS pHeaders)

{

      if (pUrl[0] != '/') return;



      CHAR *iUrl = 0;

      BOOL doSet = FALSE;



      char *sOldUrls[] = { "/test/", "/TEST/" };

      char *sNewUrls[] = { "/go/", "/GO/" };



      for (int i=0; i<2; i++)

      {

            if (iUrl = strstr(pUrl,sOldUrls[i]))

            {

                  doSet = TRUE;

                  memcpy(iUrl,sNewUrls[i],strlen(sNewUrls[i]));

            }

      }



      if (doSet) pHeaders->SetHeader(pfc, "url", pUrl);

}



DWORD WINAPI HttpFilterProc(HTTP_FILTER_CONTEXT *pfc,DWORD NotificationType,VOID * pvData)

{

      PHTTP_FILTER_PREPROC_HEADERS pHeaders;

      DWORD cUrlOrig = MAX_URL_LEN;

      DWORD cUrl = cUrlOrig;

      CHAR rgUrl[MAX_URL_LEN];

      CHAR *pUrl;

      BOOL result;



      switch ( NotificationType )

      {

            case SF_NOTIFY_PREPROC_HEADERS:

                  pHeaders = (PHTTP_FILTER_PREPROC_HEADERS) pvData;

                  result = pHeaders->GetHeader(pfc, "url", rgUrl, &cUrl);



                  if (!result && cUrl > cUrlOrig)

                  {

                        pUrl = (CHAR*)LocalAlloc(0, cUrl);

                        result = pHeaders->GetHeader(pfc, "url", pUrl, &cUrl);

                        if (!result)

                        {

                              LocalFree(pUrl);

                              break;

                        }

                        ReMapURLs(pUrl, pfc, pHeaders);

                        LocalFree(pUrl);

                  }

                  else

                        ReMapURLs(rgUrl, pfc, pHeaders);

                  break;

            default:

                  break;

      }

      return SF_STATUS_REQ_NEXT_NOTIFICATION;

}