Well, AWS works

i got the code for the amazon API off a devshed article a while ago, but i've had problems with it since the start... it takes like 15-20 seconds for any page with the API on it to load and i can't seem to figure out why.. i've broken up the code into different parts and can't really figure out what's causing the delay

sample page is here:
http://www.lyricallysquared.com/amaz...n_artist=Finch

and some of the code i used is here (this is what i think is most likely causing the problem anyway)

PHP Code:

---

// include class
include("nusoap.php");

// create a instance of the SOAP client object
// remember that this script is the client,
// accessing the web service provided by Amazon.com
$soapclient = new
soapclient("http://soap.amazon.com/schemas3/AmazonWebServices.wsdl",
true);

// create a proxy so that WSDL methods can be accessed directly
$proxy = $soapclient->getProxy();

// invoke the method 
$result = $proxy->BrowseNodeSearchRequest($params); 


---

Google has indexed more of my tools site (and kissitmakeitbetter.com). My tools site now gets 3-4k page views a day and yesterday I sold 36 items through AWS. 15 or so of them were less than $5 little bandaid things but still.

Quote:

---

Originally posted by Chris
Google has indexed more of my tools site (and kissitmakeitbetter.com). My tools site now gets 3-4k page views a day and yesterday I sold 36 items through AWS. 15 or so of them were less than $5 little bandaid things but still.

---

Okay, you've got about 19,000 pages in Google, you have got to tell me how you did that.

My site doesn't seem to get spidered fast at all. Every now and then 50 more pages get into Google, but it's not happening as fast as it seems for you. What's the secret? :(

It's just not the same for all sites. The time varies. I've had some AWS sites listed within a week (completely) and some that still don't have any more than the first page listed. Just keep submitting... it's all you can do.

I have a dedicated server so maybe it responds faster than what you have. SE friendly urls help of course as well.

how many sales are u guys averaging from amazon a day ?

0 :d

3

2

Though, I'm still getting seriously low traffic. Like, 150 - 200 uniques a day for each site (camcorders and lcd are getting none, because they're not indexed yet).

Quote:

---

Originally posted by Chris
I have a dedicated server so maybe it responds faster than what you have. SE friendly urls help of course as well.

---

I suppose it's because I don't have a dedicated server then, because I do have SE friendly urls (e.g http://www.aspit.net/faq/miscellaneo...ofessional.asp)

I guess I'll just keep submitting my sites, and the main subpages, and wait for Google to finally index ALL my pages.

Quote:

---

Originally posted by chromate
Though, I'm still getting seriously low traffic.

---

Same here. If only Google Japan would propogate to www.google.com :D

0 - Site not launched yet and I haven't finnished AWS.

So you're getting 0 sales from a site that doesn't exist yet? Thanks for letting us know Incka ;) lol

In addition to my URLs being SE friendly they're short.

domain.com/item/asin

domain.com/market/node

That might be helping.

Yeah, I always prefer sites with www.site.com/category/article rather than www.site.com/category?article=articlename

I've made 35 sales so far this quarter... so about 1 per day. But it only kicked off around the second week of January.

What is your AWS site url Mark?

Quote:

---

Originally posted by Chris
In addition to my URLs being SE friendly they're short.

---

Chris, I noticed that about your URLs. I was surprised that you were not putting any potential keywords in the URLs. On my site I include the item title in the URL, but it certainly makes the URL longer.

What's better... short URLs, or keywords in URLs?

Interesting point. All my kitchen sites use short url's like Chris's. My other two sites (camcorders and LCD Monitors) both use longer keyword rich URLs.

All the kitchen sites got deep crawled pretty quickly. The other two sites (using the longer URLs) are still waiting for a deep crawl.

Wonder if the URL's have something to do with it?

Having said that Mike's site uses the short URL's and is still waiting for a deep crawl. So we're probably reading too much into it.

I'm not adverse to putting a couple keywords in the url, like on the main part of this site. However amazon titles are probably 10 words on average and that makes the url extremely long. Too long for my tastes.

Quote:

---

Originally posted by cameron
What is your AWS site url Mark?

---

I actually have two - together, the sites have made 58 sales this year (items sold, that is).

www.buymetal.net is the one doing best.
www.mygroovywedding.com/shop/ was my first (basic) attempt at AWS using Mr Rat's perl script.

Buy Metal uses the same script, but more trimmed down as far as items spiderabobble.

Oh, and on the 29th someone ordered 10 coffee mills LOL

I'm still trying to get my product pages to work like /product/asin

I can do it for none amazon pages, but no amazon ones...

Quote:

---

Originally posted by incka
I'm still trying to get my product pages to work like /product/asin

I can do it for none amazon pages, but no amazon ones...

---

Which method are you using?

Quote:

---

Originally posted by Chris
I'm not adverse to putting a couple keywords in the url, like on the main part of this site. However amazon titles are probably 10 words on average and that makes the url extremely long. Too long for my tastes.

---

Yep. I agree. A lot of the words used in the title are just additional features of the product and don't really contribute much. If you look on my URL's I've just included the first 4 words of the product title. That way it normally includes the make and model of the product, plus two other words for good measure.

I figure that a lot of other AWS stores will be using the product title for their own html page title in a detailed product page. So I've actually just used the first 4 words on these page titles too. Hopefully, it will increase the keyword density of the important keywords that people will use in their search terms and give a slight edge over the competition that are using the whole product title and wasting the density over inconsequential words. Of course the danger of just using the first 4 words is titles that just seem to stop abruptly and therefore people will be less likely to click on them from the search engine. The majority of the time they make sense though. It's too early to tell how this is going to work out, because my sites that do it haven't even been deep crawled yet.

Im using Dan's script with .htaccess and

$var_array = explode("/",$PATH_INFO);
$ASIN = $var_array[2];

Try this code:

PHP Code:

---

$var_array = explode("/",$REQUEST_URI);

$var = $var_array[2];
$Asin = $var; 


---

Remember variable names are case sensitive so if you need a variable named "Asin" then "ASIN" will not suffice.

Chris, I'm wondering if you're able to share a slice of your code?

I have nodes stored in an sql DB, in a table which is simply:

Code:

---

ID      NODE
6      67207

---

as an example.

In my code, I have:

PHP Code:

---

$var_array = explode("/",$REQUEST_URI);

$sql = "SELECT node FROM cat WHERE id=$var_array[3]";
$result = mysql_query($sql);

if ($result == 0) { $cat = 67207; } else { $cat = $result; } 


---

And then $cat is called when asked for the node number later on.

But it only produces an error with AWS - obviously the node is not being carried through... Any ideas?

Mark, I think there's a few things missing from your code. Try:

PHP Code:

---

$var_array = explode("/",$REQUEST_URI); 

$sql = "SELECT node FROM cat WHERE id='$var_array[3]'"; 
$result = mysql_query($sql); 

if ($result == 0) { $cat = 67207; } else { $cat = mysql_result($result, 0, 'node'); } 


---

I'm not certain, but that's just what I would do out of habit. Try echoing your $sql var and then run it through mysql manually. Also try echoing $result to make sure $cat is being set to something meaningful.

That seems to have done the job, mate :)

Why was my code not working?

All I need now is Chris's node spider script to do all this for me automatically ;) Any chance of that, Chris? :)

Depending on how you have your field types set in your database, you needed the quotes in your sql to make it valid.

You can't always access $result directly either.

www.buymetal.net/ - PHP-powered AWS now up :) Have to wonder how to power the main page without just having it static or redirected... will have to work on that I think.

Grrr.. I still can't get my code to work...

Here is my SQL structure. The pid is what I use to access the categories. Amazonid is obvious. Parentid is the parent category's amazon id. textid is an optional text identifier to use instead of pid (I'm not using it). Name is the name. mode is the mode (mode is sometimes necessary).

CREATE TABLE blank_categories (
pid mediumint(9) NOT NULL auto_increment,
amazonid int(12) NOT NULL default '0',
parentid int(12) NOT NULL default '0',
textid varchar(70) NOT NULL default '',
name varchar(70) NOT NULL default '',
mode varchar(20) NOT NULL default '',
PRIMARY KEY (pid),
UNIQUE KEY amazonid (amazonid,parentid),
KEY textid (textid),
KEY parentid (parentid),
KEY mode (mode)
) TYPE=MyISAM;

Here is the code I use to parse browsenodes.com files into my db:

Code:

---

$quotelist = file("$url");

$db = mysql_connect("localhost", "user", "pw");
if (!$db) {  echo( "<P>Unable to connect to the " . "database server at this time.</P>" );  exit();}

mysql_select_db("dbname", $db);

function addquote($element){
        $quote_array = explode("|",$element);
        $name = $quote_array[0];
        $parentid = $quote_array[1];
        $amazonid = $quote_array[2];
        $mode = $quote_array[3];

global $db;

$result1 = mysql_query("INSERT blank_categories SET name = '$name', parentid = '$parentid', amazonid = '$amazonid', mode = '$mode'", $db);
if (!$result1) {  echo("<P>Error performing query: " . mysql_error() . "</P>"); 



}

}

array_walk($quotelist, 'addquote');

?>

---

Its modified from a script I made to do much the same thing with quotes for my literature site's quote db so that explains the naming conventions. Access it like http://www.example.com/script.php?ur...enodesfile.txt

When getting the browsenodes.com files use | as a seperator and make sure it includes mode.

Oh My God.

You'll never guess how many items I sold through AWS yesterday.



200 of these: http://www.kissitmakeitbetter.com/item/B0000532MK/

I'll only make say $30 off of it, but damn. That is alot of items. I don't know why but people are always buying the health type stuff in bulk.

Whats great about that is that it will help power me up to the next tier.

Now Amazon only has 4 left in stock.

lol. That helps :)

I only have 6 items left needed to get to my next tier (5%)

I missed out on getting into the scaling tier thing this quarter - damn me and my laziness!

CHRIS! Thanks very much for posting all that - extremely generous of you!

BTW, how long did you have the site open for spidering before you announced it to us here?

And would URLs like: www.domain.com/cat/1 be less spiderable that www.domain.com/cat/1/ ?