Free Amazon book store, need testers.

Want to help debug the next free book store? Got a free MySQL DB?

We need your help!

So far I have the back end written. It pulls from the amazon XML feed and populates the DB with available books.

To test you'd have to...

• Backup your site and DB's so nothing goes wrong!
• Have an amazon associate ID.
• Have an amason Developers Token.
• Create a DB and DB user or use an existing DB. (TABLE prefixes are used and configurable.)
• Run 2 queries to setup the two tables.
• Edit the import script.
• Upload the import script.
• Run the import script.
• Report errors or suggestions.
• Report on inspection of DB records.

All of the above is easy to do in phpmyadmin.
Running the script is as simple as an FTP upload.

While you play with the back end, I'll be coding the front end since the back end works for me. That don't mean it will work for you perfectly however, this is testing.

Thanks

 

How will you make this script different from Mr. Rat's Amazon Products Feed?

 

I would like a script that all in php, I freaking hate perl, becasue I suck at it, and there always seems to be some host issue cgi scripts

 

Well I'm going to post the code Mr. Will, so you can see for yourself. However since I've never seen that code, I can't imagine it's the same.

Ferret, it is 100% php.

 

why limit it to just books?

 

It's my starting point. I want the content pages I thought I got with FREEBIE and didn't. So I'm doing books for now.

I intend to go further, but for now I'm cutting my teeth with amazon. I think I may have made the server mad once or twice, but that's not a problem now, got the timing down and they feed us as fast as we pull. Over 800 books pulled in under 10 minutes. (I didn't actually time it)

 

I will like to test run this script. I have played with several amazon web scripts, some easy to configure, some very difficult, so far I haven't found what I am looking for

email me at kash88888 (at) hotmail.com

 

OK guys, this is going to be stupid simple IMO.

The output will be a table that can be put into about any style. Basically you'll put the top half of you sites style in a header file and the bottom half in a footer file. Break it in half where it can recieve a table. I am trying to code it so it looks ok wide and narrow.

The front end is not started yet, so suggestions are welcome at this time.

The code will be released here. FREEBS - Free Book Store Software Look at the bottom, it has it's own new forum.

I should have the back end code posted later today. For now, I'm off to the park.

Cheers, regards, and thanks.

 

if you are serious about this being free, you ought to open it to sourceforge and have them host it for you. You'll get good feedback from developers there.

 

That's a cool idea, I was thinking CVS myself too. Forgot SF, thanks.

For now my biggest issue is populating the DB with books and all the infomation possible on a book in one XML query. I have not gotten the XML to feed me all the data that can appear on the actual amazon page yet. I do not want to pull and parse the actual amazon HTML item sales page to get the data.

The code is basically done and working. The timing is good and amazon feeds as fast as we request. With the one second rule the app is friendly to amazon and your server.

All books are updated on the pull and dated so you can show the price accurate as of date. Any books returned as out of stock or not available and such are removed from the DB too if in there and never added if they are unavail the first time we find them. The update is fast too. I can update with 1500 returns from the XML query in under 10 minutes! This is a huge benefit for that weekly maintenance required by amazon you will run with this back end app of ours.

I am trying to plan if the book page is accepable without related book links and or editorial reviews. Those are the two items that the XML don't feed me yet with my current query.

Related links are fed as ASIN numbers, I could pull the item XML data for the related items and create the links though. This adds an XML query for each related asin though during the populate/update proceedure I'm referring to as the back end.

For the editorial reviews, well I could live without um, but I want them to come to me in the XML. I'm off to figure out what to add to my XML query if it's possible at all.

What I need now is one or two private beta testers to run the back end and shake it down. What is does is populate a DB. Running it will show you how many pages you will get of books which can be dynamicly called and sorted.

Anyone that wants to help test it, please PM me. The only requirement is confidentiality, the code is not public yet.

Thanks

 

Progress report...

Using the amazon SOAP class we can do some powerful stuff...

Using the keyword motorcycle yields 4335 search results from amazon.

The SOAP pull for the search results to populate the DB takes about 4 minutes.

Of the 4335 there are 859 usable books.

To do a full weekly maintenance of the DB after the XML pull takes under 5 minutes.

There are no editorial reviews pulled since most of them are copyrighted, pulling them from the HTML page would be illegal in most cases and that's not something we are going to do. If anyone knows different, holler.

So anyway, the back end is done and done using the amazon api. It's under 500 lines of code and is complient to the weekly rule in the amazon tos.

The amazon soap calls and the MySql database are blazingly fast!

The store front end should be easy for all of you.

EDIT:
Opps, We have to resolve the similar links.

Humm, maybe I'll do it on the front so it's always current. The back end will always get updated in the weeklys with the asin numbers for similars.

Yeah, that's it.

 

OK the official beta release 1 of FREEBS - Free Book Store is posted.

This is the back end only, but it's the meat of the application.

Please feel free to test and review the code.

FREEBS - Free Book Store Support

 

I'd be happy to test (thought I may need a litte help) or help provide hosting to the project.

 

Any help would be appreciated, we do have hosting covered though.

If you can create a database using phpmyadmin, run the query file from phpmyadmin, edit the config.php file and upload the required files, you can test the app.

It's working pretty good from where I sit. Tomorrow I'll do the main index page, I hope.

 

I think I may have made the server mad once or twice, but that's not a problem now, got the timing down and they feed us as fast as we pull. Over 800 books pulled in under 10 minutes. (I didn't actually time it)
Amazon's Terms of Service are quite explicit that no one IP address can take more than one call a second. Just so's you know.

 

Well thanks for that tip, however the SOAP API fixed that issue for us.

I threw away all timing code, no need for it. I replaced it with the amazon API that uses SOAP and it does the timing transparenlty. Nice stuff.

Since we can only currently pull 250 pages or 2500 results, we can do a monthly/weekly database update for our cached info in about 7 minutes.

Pricing is live and to the moment on the purchase pages (SOAP call) and cached on the info page that links to the purchase page.

Here's the working model. FREEBS - Free Book Store

There's a link around here for the backend demo too, you can time it.