[thehigheredcio]

I already have a single Products page that displays all products using the [wp_eStore_all_products_stylish:end] shortcode.

It seems that since the Merchant feed already has the Product page URL, adding it in eStore as the product URL would cause customers to loop back if they clicked on it.

It seems that for a customer experience a landing page may be best.

[thehigheredcio]

I should also note that for some future enhancements I can see how other Google recommended attribute might be nice to have in the eStore anyway such as Brand and Category (UPC or other). For now the Add-on is perfect for me since I have only one category and brand.

[thehigheredcio]

I think the store page as a default would be a problem.

For other reasons I have often wondered if I could generate a link to a single product listing without having to create a product page. This might be a a dynamic page that is built on the fly invoking a product number much like the use of shortcodes but for external links and only for the single product.

For now I guess I need a couple of new pages unless you might have an idea.

[thehigheredcio]

Upgraded to v1.1

Invalid URL Errors corrected! Thanks for the quick fix.

Type 2 Products with Product Page URL defined in eStore loaded in Google Merchant without error.

Type 1 Products that do not have a Product Page URL (no page exists) fail to load with a “Missing Attribute Error” for the product Link attribute. Here isthe feed xmlfor one example:

<item>

<title><![CDATA[Business Impact Analysis (BIA) Questionnaire]]></title>

<g:id>The Higher Ed CIO-10</g:id>

<description><![CDATA[Business Impact Analysis (BIA) Questionnaire]]></description>

<g:image_link>http://thehigheredcio.com/wp-content/uploads/2012/06/Business-Impact-Analysis-BIA.png</g:image_link&gt; <g:condition>new</g:condition> <g:availability>in stock</g:availability>

<g:price>0.99 </g:price>

<g:brand>The Higher Ed CIO</g:brand>

I can create pages for these products but that could become problematic and not very elegant.

[thehigheredcio]

After fixing other errors and resubmitting the Base Feed I am left with only URL Errors as noted above other than the warnings for the attributes not supported in the eStore product attributes.

In looking over the URL Errors again, it appears they all have a space between the <link> operator and the href=…. even the Type 1 products. By comparison, the Image Link Attribute does not have the space in it.

Hope that helps

[thehigheredcio]

WZP – Everything is working great since the changes. Fast downloads, no empty files or errors and I am very comfortable with the security workaround.

[thehigheredcio]

Thanks for that T&T Link (BTW maybe it could have an S3 Tag added for searchability).

I don’t know what is happening behind the scenes on the Get Request and I know very little about S3 but it appears eStore is using a Signed URL method because it allows for it to be Time Limited. The trade off though it sacrifices securityeven if the credentials are ‘public’ which is contrary to S3 security best practices. It likely also means they are passing in clear text from eStore to AWS as part of the URI Get Request.

If you do go back in to tweak the S3 Integration to strip out the credentials in the URL it would be nice if you could add support for using a Virtual Host name from a field in the S3 Config of eStore (like My Domain) or the Bucket Name.

Thanks again for the help – on Sunday no less!

[thehigheredcio]

I set up S3 and migrated my files over and tested them all with no errors and much better download performance speeds. I have an S3 security concern I will open in a new post and revert back until this issue is resolved.

[thehigheredcio]

I made the change to cURL and cleared my logs.

I do have some failures from US based users but I did not look up what state they are in. It does appear that at least half, perhaps more than half (~75%) are international users. UK, Canada, Russia, India, Austria, Australia, and several from South Africa.

If I do move over to using Amazon S3, is there a preferred Download Method for S3?

[thehigheredcio]

Bingo. Thx Again

PS – Can you flush the log data?

[thehigheredcio]

I am going to leave the page live but reset the pricing just in case.

[thehigheredcio]

Perfect. This solves so much. And, I appreciate you writing up the How To.

Resolved

[thehigheredcio]

I am just getting started with eStore and understand the need to keep the code lightweight, but I have old download stats from my previous method and have begun revising products, their descriptions and their landing pages and so far I have to track these changes to see if it helps in a separate location.

This will become a more useful capability once I start selling products.

Thanks for considering it.

[thehigheredcio]

I will give Analytics a try. So far I am using the same approach for all free downloads which is the just a squeeze form using the Free Download Ajax shortcode as shown on this page: http://wp.me/p1RYyA-ZA

Thanks for considering the suggestion.

[thehigheredcio]

That was it. Thanks!

It appears earlier products which were pdf files also had spaces in the filenames but work fine. On closer look it appears as though eStore auto-inserted hyphens in those names where there spaces were or they got in there through the copy/paste method – not sure. But hyphens were not inserted into the latest files I set up.

Guess I can’t be lazy anymore in naming my files – thanks for the resolution.

[Author]

Posts