Amazon-Web services without fopen () - is that possible

[bernard_hinault]

good day ,

well i have some troubles

want to run a AWS product feed and - unfortunatley my server is limited - it is configured with allow_url_fopen to off! that means:

i may have troubles to process requests - since (in my humble opinion) Amazon web services only function with the fopen-option: is this correct. Does any AWs (even in the new standard ecs 4 ) make usage of the fopen - function?

i have tried to get the thing going with a look at several solutions.

the amazon product feed of mrRat see http://www.mrrat.com

the php-store here: http://www.accessories.me.uk/forum-2.htm?

the amazonModule of Edgar Diaz - see http://www.preciogasolina.com -

the standalone scripts of this guy here http://www.chipdir.nl/amazon/ Jaap's Free Amazon PHP Scripts This page describes PHP scripts that can interface with the XML version of Amazon's Web Services (AWS). It also contains links to more information about PHP and AWS and to sites with other scripts. see http://www.chipdir.nl/amazon/

well - nothing runs on my restricted server - i run PHP ver 4.3.11 and the allow_url_fopen is set to off. My questin is : how can i go round this issue - and ho can i get Aamazon product feeds running.-

to say it clear; i do not need many many features - i only want to show up a handy number of books

My need is to create a site with only fifty ItemIds: thats all! besides those books i do not need any further service from amazon - /(as of today!!!)

i need a solution that makes this going - i need a solution that can do this; and can perform a request - without the need of changing the server settings; does any body konw how to do this.

thanks in advance
bernard hinault

[bubblenut]

I'm not sure of how the allow_url_fopen option restricts the server (it may well prevent sockets as well) but have you thought of using sockets? It would be a little more work but you could easily write yourself a wrapper class to (or even just a function if your needs aren't to complex) hide all the messy stuff. I first port of call I would say is [man]fsockopen[/man] and just test to see if your server allows it.

PS. remember, http requests are on port 80 😉

[bernard_hinault]

hi bubblenut,

thx for the reply. this is a interesing starting point.

Originally posted by bubblenut
I'm not sure of how the allow_url_fopen option restricts the server (it may well prevent sockets as well) but have you thought of using sockets? It would be a little more work but you could easily write yourself a wrapper class to (or even just a function if your needs aren't to complex) hide all the messy stuff. I first port of call I would say is [man]fsockopen[/man] and just test to see if your server allows it.

PS. remember, http requests are on port 80 😉

well if i take for instance this

http://www.chipdir.nl/amazon/amazonli.php.txt

<?php

//this is a simple stand-alone PHP example script
//for the lite version of the new Amazon Web Services interface
//20020816..29/Jaap van Ganswijk <[e-mail]>, released under the GNU GPL
//20041207/JvG, added the correct handling of German Umlauts and other unusual characters

//you normally would like to change the following two variables
//(but the script should also work as it is)
$search ='machiavelli'; //replace this with your own search term, you can give upto 3 terms separated by '%20' like 'cat%20dog', a '+' should also work instad of %20
//$search ='title:"civil+engineering"'; //replace this with your own search term, you can give upto 3 terms separated by '%20' like 'cat%20dog', a '+' should also work instad of %20
$associates_id='chipdir'; //replace this with your own associate id
//get one from http://associates.amazon.com/exec/panama/associates/apply

//--You don't have to change anything after this line, to start earning money.--

$dev_token='D2WMCOIPS9D14E'; //replace this with your own developer token if you like
//get one from http://associates.amazon.com/exec/panama/associates/join/developer/application.html
//you don't have to change this token unless you change the code a lot it seems:
//http://forums.prospero.com/n/mb/message.asp?webtag=am-assocdevxml&msg=378.1&ctx=0

$mode='books'; //change this if you want to offer something else as books
//other modes are:
//mode=baby (Baby)
//mode=books (Books)
//mode=classical (Classical Music)
//mode=dvd (DVD)
//mode=electronics (Electronics)
//mode=garden (Outdoor Living)
//mode=kitchen (Kitchen & Housewares)
//mode=magazines (Magazines)
//mode=music (Popular Music)
//mode=pc-hardware (Computers)
//mode=photo (Camera & Photo)
//mode=software (Software)
//mode=tools (Tools & Hardware)
//mode=toys (Toys & Games)
//mode=vhs (Video)
//mode=videogames (Computer & Video Games)

search parameters:

$type ='lite'; //don't use this script for the 'heavy' XML!
$show_xml=false; //change false to true if you want to see the raw XML data (for debugging)

$f_rich =true; //this will convert the URL so you'll get 15% on qualifying books instead of 5%.
//change it to false when you prefer to get only 5% ;-)
//see http://forums.prospero.com/n/mb/message.asp?webtag=am-assosdev&msg=6.1&ctx=0
//I'm not sure if this method will also work for non-books, but I guess so

if (!$page) {
$page=1;
}

$file="onca/xml3?".
"dev-t=$dev_token".'&'.
"f=xml".'&'.
"KeywordSearch=$search".'&'.
"mode=$mode".'&'.
"page=$page".'&'.
// "PowerSearch=$search".'&'.
"t=$associates_id".'&'.
"type=$type".'&'.
"v=1";
//Please note that:
//- /onca/xml3 is the pathname of the URL
//- ? denotes the spot where the arguments start
//- that all arguments consist of 'argument_name=value' and that I have put them in alphabetical order
//- that '&' seperates the arguments
//you can also add:
//- page=1 for result entries 1..10
//- page=2 for result entries 11..20
//- etc.
//Each query result will contain only up to 10 entries. If you want to show more than 10 results, you will have to show several pages with results until you get a result with 0..9 entries, but when it's 0 you'll get an error message instead.
//To search for ASIN's, replace the KeywordSearch line by AsinSearch=<ASIN>,<another ASIN> upto 30 for lite and 10 for heavy searches.
//To search for UPC's use instead: UpcSearch=<upc>,<upc>.
//You can also use:
//- AuthorSearch =.. (only for mode=books)
//- ArtistSearch =.. (only for mode=music,classical)
//- ActorSearch =.. (only for mode=dvd,vhs,video)
//- DirectorSearch =.. (only for mode=dvd,vhs,video)
//- ManufacturerSearch=.. (only for mode=electronics,kitchen,videogames,software,photo,pc-hardware)
//- SimilaratiesSearch=ASIN (no mode)

//Amazon lists (in the lite version) per book a record like:
// <Details url="http://www.amazon.com/exec/obidos/redirect?tag=chipdir%26creative=D2WMCOIPS9D14E%26camp=2025%26link_code=xm2%26path=ASIN/039480001X">
// <Asin>039480001X</Asin>
// <ProductName>The Cat in the Hat</ProductName>
// <Catalog>Book</Catalog>
// <Authors>
// <Author>Seuss</Author>
// <Author>Theodor Seuss Geisel</Author>
// <Author>Dr. Seuss</Author>
// </Authors>
// <ReleaseDate>June, 1957</ReleaseDate>
// <Manufacturer>Random House (Merchandising)</Manufacturer>
// <ImageUrlSmall>http://images.amazon.com/images/P/039480001X.01.THUMBZZZ.jpg</ImageUrlSmall>
// <ImageUrlMedium>http://images.amazon.com/images/P/039480001X.01.MZZZZZZZ.jpg</ImageUrlMedium>
// <ImageUrlLarge>http://images.amazon.com/images/P/039480001X.01.LZZZZZZZ.jpg</ImageUrlLarge>
// <ListPrice>$8.99</ListPrice>
// <OurPrice>$8.99</OurPrice>
// <UsedPrice>$0.50</UsedPrice>
// </Details>
//
//in case there was an error or nothing was found Amazon will return something like:
//<ProductInfo><ErrorMsg>There are no exact matches for the search.</ErrorMsg></ProductInfo>

//global variables:
$totalresults='unknown';

//get the data from amazon
function getAmazon($file) {
global $host;
global $show_xml;
global $totalresults;

if ($hf=fopen("http://xml.amazon.com/$file",'r')) { //open a file from Amazon
if ($show_xml) { //print the raw XML instead of parsing it into an array?
print "<pre>\n";
}
while ($line=fgets($hf,10000)) { //handle each line of XML
$line=utf8_decode($line); //convert the UTF8 characters, 20041207/JvG added
if ($show_xml) { //print the raw XML instead of parsing it into an array?
print htmlentities(rtrim($line))."\n";
}
if (ereg('HTTP/[0-9].[0-9] ([13-9].)$',$line,$Atmp)) {
if (!$show_xml) print "HTTP error: ".$Atmp[1]."<br>\n";
}
else if (eregi('<ProductInfo><ErrorMsg>(.)</ErrorMsg></ProductInfo>',$line,$Atmp)) {
if (!$show_xml) print "Amazon search error: ".$Atmp[1]."<br>\n";
}
else if (eregi('<TotalResults>(.)</TotalResults>',$line,$Atmp)) {
$totalresults=$Atmp[1];
}
else if (eregi('<details url="(.)">',$line,$Atmp)) {
$E=$Atmp[1];
}
else if (eregi("<(.)>(.)</.*>",$line,$Ares)) {
$tag =$Ares[1];
$data=$Ares[2];
if ($tag=='Author') { //there can be multiple authors so put in sub array
$E[$tag][]=$data;
}
else {
$E[$tag]=$data;
}
}
else if (eregi("</details>",$line)) {
$A[]=$E;
$E=array(); //empty array again
}
}
if ($show_xml) { //print the raw XML instead of parsing it into an array?
print "</pre>\n";
}
}
else die("Can't open socket for xml.amazon.com.<br>\n");
return $A;
}

//print the array
function printAmazon($A) {
global $associates_id;
global $f_rich; //do you want 15% or 5%?
global $totalresults;


....and see for the rest here this

[url]http://www.chipdir.nl/amazon/amazonli.php.txt

and here we have the issue

if ($hf=fopen("http://xml.amazon.com/$file",'r')) { //open a file from Amazon

• you mean that i should wrap this with using sockets? - i am not a programmer - but if i would give it a try
  how it would be done in this case. Thanks for a hint.

ps. Guss that this means a great learing curve for me;-) if you would help me a bit i would be very thankful

looking forward to hear from you

bernard hinault 🙂

[MarkR]

using fopen is the proper way of doing HTTP connections in PHP now.

So simply get allow_url_fopen enabled. This can be done on a per virtual host basis in your Apache config (with php_admin_flag or something)

Cheers

Mark

[bernard_hinault]

hello MarkR, hello, bubblenut

many thanks for the input - i am happy to hear from you - you input is appreciate here - since i allmost gone mad

Originally posted by MarkR using fopen is the proper way of doing HTTP connections in PHP now.

So simply get allow_url_fopen enabled. This can be done on a per virtual host basis in your Apache config (with php_admin_flag or something) Cheers Mark

Mark - the thing the former poster mentioned is interesting - he wanted me to do some thing with fsockopen - well to that point i have something found out. many thanks for the input.

is that a way to work round the issues? at least it Sounds possible. I searched Go Ogle and found this which sounds like a good way to implement what you are talking about.[/quote]

MarkR and bubblenut, that sound very very interesting. Again many thanks for the response.- well that semms to be agreat idea -

the thread here: http://7sheaven.pingouin.ca/forum/showthread.php?t=13

please have a look at it

hmm as i am no programmer - can we get this to work - and can we get a work around with that - so that amazon feeds were possible and doable with a server-configuration that is as restrict as mine -. Guess that many many user have server that are so badly configured

hmmm - how can we get the ball rolling - can you help me with a quick and dirty way to patch the script or at least the eaysy one - here - the scripts of this guy here

http://www.chipdir.nl/amazon/

look - there are a vew very handy scripts that can be runnded with ease - but all of thoes make usage of fopen. Perhaps we can hack some of them.

what do you mean Mark and bubblenut, what do you think
look forward to your ideas and suggestions.

with regards

bernard hinault 🙂

here i Found a way to get round fopen issue:

http://7sheaven.pingouin.ca/forum/showthread.php?t=13

theese guys had the same problem and were able to get a work around so there is now rss feed possible:

to share the code with you

/ This function is a replacement the file(URL) function which can not be used with the PHP configuration settings "allow_url_fopen = off"
// This is a generic function, which calls a URL and returns the contents of the URL as a string.
// This function was originally posted by edwin at bitstorm dot org on PHP.net ([url]http://de.php.net/manual/en/function.fsockopen.php[/url])
function fetchURL($url) {
    $url_parsed = parse_url($url);
    $host = $url_parsed["host"];
    $port = $url_parsed["port"];
    if ($port==0)
        $port = 80;
    $path = $url_parsed["path"];
    if ($url_parsed["query"] != "")
        $path .= "?".$url_parsed["query"];

$out = "GET $path HTTP/1.0\r\nHost: $host\r\n\r\n";

$fp = fsockopen($host, $port, $errno, $errstr, 30);

fwrite($fp, $out);
$body = false;
while (!feof($fp)) {
    $s = fgets($fp, 1024);
    if ( $body )
        $in .= $s;
    if ( $s == "\r\n" )
        $body = true;
    }

fclose($fp);

return $in;
} 

hmm as i am no programmer - can we get this to work - and can we get a work around with that - so that amazon feeds were possible and doable with a server-configuration that is as restrict as mine - Guess that many many user have server that are so badly configured

what do you mean. what do you think regards