Refuses to grab images from some feeds | HTML elements with more than one css class not selected

Hi,

"it seems to be an issue with the way imagegrabber handles cron caching"

It is not a problem of cron caching but imagegrabber can not handle tags with multiple classes. This is the tag you want to select:

<div class="field field-type-filefield field-field-image">

but because of the design of imagegrabber, it doesn't recognize this with class 'field-field-image'. As I am not an regex expert, I am working on my current query:

$class = "foo";
$xpath->query("//*[@class = '".$class."']");

to select the elements with multiple classes. If you can help, it will be great.

You can see the results of the 'feed refresh' when I changed the class description to 'field field-type-filefield field-field-image' (You can not change this from the interface, as no spaces will be allowed in class name, I changed directly into the the database)

http://publicmind.in/drupal/feed-item/2283

the last 2 items do not have a image, whereas the third item has a malformed image URL (url with spaces is considered malformed by the script).

I will change it to 'by design' and will update it as I resolve the issue.

Thanks for reporting the issue with such detailed explanation.

Hi,

One more thing, clicking on 'grab images' will only download images for those feed-items which were not processed due to lack of execution time and not who failed to get an image or resulted in error.

Also, after a cron run, some feed-items might not have the images due to lack of execution time (which is not probable as cron execution time limit is 120 seconds for imagegrabber), so it is always safe to click on 'grab images' after a cron run. To automate this process at the next cron run is already in the to-do list.

Thanks,

The issue has been fixed, I will soon put up a patch as well as release the dev version updating it.

For a quick fix:

change line no. 651 in feedapi_imagegrabber.module :

$items = $xpath->query("//*[@class = '".$id_class_description."']");

with

$items = $xpath->query("//*[@class and contains(concat(' ',normalize-space(@class),' '),' $id_class_description ')]");

Thats all what needs to be done.

It is now fixed in dev version. Visit: http://www.publicmind.in/files/FIG_readme.html

FeedAPI ImageGrabber v1.9 has been released. The new release resolves this issue. Please do post your comments about the new release , I am always looking for your feedback.

I am unable to get this to work with the 1.9 or dev version, if i state no class or id then it will probably download the sites logo but if will not download the class "field-field-ff"