Google Base Feeder Support Thread [OLD]

[tj1]

I have three websites, this mod worked on two of them, but on one of them Google gave the following error.

Error Bad data Line #6
We found a mismatched XML tag. - help

This is what line 6 has for this file:

<description>Sells fine solid gold jewelry.</description>r-jewelry.com/images/large/gld/r3248_LRG.jpg</g:image_link>

There are obviously broken tags on this line. I wasn't able to figure out why this happened. This was with v1.6.0 and 1.6.1.
I just tried v1.6.2 and I no longer have the problem reported above; however, now the file generated is missing more than half the items.

I'll wait and see if anyone else has this issue or if it could be an issue on my server.

BTW, do you have a link where I can download the last version that created tab delimited files. Old links in this thread don't seem to work. I'd like to take a look at it and see if I can integrate some of its features to EasyPopulate.

[numinix]

You can find older versions in the Zen Cart Downloads section of this website.

You still haven't said what is different about the missing products. Perhaps they were inactive, some custom product type, etc.

The Google Base Feeder only supports the 5 product types found in Zen Cart by default. To add additional customized product types, you would add the following line to the SQL query for your feed type:

OR p.products_type = # 

Where # is the product_type id.

[tj1]

Thanks, I found and downloaded the older versions.

There's is nothing different about the missing products. They're all general products.

Actually, after opening the file generated I'm seeing it's missing the the closing tags as well, so this might indicate a time out, correct?

[numinix]

Yes, it's timing out then.

[TonyBenedetti]

Some of our product names include a percent sign -- e.g., "100% Kona Coffee (1 Pound Bag)". That percent sign is important to our "branding" ... no blend of beans grown other than on the Kona Coast of the Big Island of Hawaii may use the "100% Kona Coffee" appellation.

But the percent sign (which shows up just fine on our Zen Cart pages) disappears from the XML file which we send to Google Base. BTW, the same "disappearing act" is true of the Windows Live Product Search file. We aren't using the Yahoo feeder, but I would guess that similar code is used there as well.

I tried to look at the code but I must admit that my eyes glazed over when I got to the "sanitizer" function.

Numinix, is there a chance that you could look at the Google / Live (and Yahoo!) code to see if the percent signs could be preserved in the output? Or, perhaps, is there a workaround that I should use in the original ZC product entries?

We really need those little percent sign devils to show up in the product search results.

TIA

[TonyBenedetti]

I forgot to add the details in my above posting -- we are using:
Google Base Feeder V1.6.2
Zen Cart V1.3.7.1
PHP Version 4.4.7 (we will be moving to PHP5 next month)

BTW, I tried to use HTML entity (37) in the product entry but ZC nicely turned it into a percent sign.

[CascianoLtd]

I am using Zen Cart 1.3.8 and google_base_froogle_feeder_1-7-4. I would like to replace the description with my meta description instead of the product description. Is that something that can be easily changed.

I can usually find the code and if it's a simple substitution I think I should be able to do it.

Thanks,
Ken

[TonyBenedetti]

Some of our product names include a percent sign -- e.g., "100% Kona Coffee (1 Pound Bag)". That percent sign is important to our "branding" ... no blend of beans grown other than on the Kona Coast of the Big Island of Hawaii may use the "100% Kona Coffee" appellation.
...
...
...

I sharpened my sword, girded my loins and dove into the code -- boy, do I hate (fear?) regular expressions.

In any case, the solution was pretty simple. I just added a percent sign to the list of acceptable characters in line 439 (in the zen_xml_sanitizer function of the googlefroogle.php script). Like so:

[FONT="Courier New"]$str = eregi_replace("[^[:alnum:][:space:].,!()'-_/+=?%]", "", $str);[/FONT]

Hopefully, that simple change won't come back to bite me.

Numinix, assuming that this little mod is benign --- could you sneak the change into the next update of each of your feeders?

[TonyBenedetti]

Running the Feeder from a CRON
FWIW here's a "heads up" ... the Google Base feeder uses three input values but if you literally follow the instructions in the readme file you might get bitten. I couldn't figure out why only the first of the three parameters was being recognized when I tried using cURL to run the googlefroogle.php script.

The readme suggests something like:

0 5 * * * curl -s http://your_http_catalog/googlefroogle.php?feed=yes&upload=yes&type=products

But the URL passed to cURL should be quoted like so:

0 5 * * * curl -s 'http://your_http_catalog/googlefroogle.php?feed=yes&upload=yes&type=products'
... or ...
0 5 * * * curl -s "http://your_http_catalog/googlefroogle.php?feed=yes&upload=yes&type=products"

Quoting the URL worked for me with cURL and I would assume would also work for GET and wget (although I haven't tested with those programs).

This from the cURL FAQ:

4.2 Why do I get problems when I use & or % in the URL?
In general unix shells, the & letter is treated special and when used, it
runs the specified command in the background. To safely send the & as a part
of a URL, you should quote the entire URL by using single (') or double (")
quotes around it.

An example that would invoke a remote CGI that uses &-letters could be:

curl 'http://www.altavista.com/cgi-bin/query?text=yes&q=curl'

In Windows, the standard DOS shell treats the %-letter specially and you
need to use TWO %-letters for each single one you want to use in the URL.

Also note that if you want the literal %-letter to be part of the data you
pass in a POST using -d/--data you must encode it as '%25' (which then also
needs the %-letter doubled on Windows machines).