You need to rename to description, not item.description.
In the Regex module you can use the rule
In [item.description] replace [(.*)] with [<img src="$1" />] (omit outer []s).

Thanks. I tried your suggestion. Unfortunately, the images are not showing in the output. I looked at the description source in output debug. The output is simply <img/>. The SRC= tag is missing.

Sorry. I've got a little ADD. I had not renamed to "description," when I posted 20 minutes ago. Now, the regex expression is behaving. Thanks.

I'm not sure about this one.
In the debugger and in the RSS output in Firefox I see what I would expect. In the Pipes view/run page I see no description. In my usual RSS reader I see a "Screenshot not found" image.

Thanks. I was hoping that you'd take another look. The screenshot not found image is a bungie.net 404 of sorts. First guess is that the problem is on their end and is probably deliberate. They really don't want you pulling the screenshots off their site. That's why their original RSS interface didn't contain a description.

I'm going to consult bungie.

New development:

I was playing around with the image containers, the HTML code generated by the pipe. If I put them in a simple html page, they execute with the same error, the bungie gremlin 404. Then, I get an explorer security bar prompt. You know, the one that says blocked active content...do you want to view...etc. I click yes and the image is served like normal.

Then, when I run the pipe, that image and only that image appears in the xml feed.

Yikes this is a pain!

Here's my pipe -- do I need to publish it to let you see it?

http://pipes.yahoo.com/pipes/pipe.info?_...

My feed draws from a google reader feed -- specifically from a portion of an item containing a tag (for filtering in an earlier pipe), separators (for simpler regex), a URL (image source), and text (image alt).

I'd like to construct html like this [<img src="URL" style="width: USERINPUT;" alt="TEXT">]

But when I regex to create the string, I can't seem to insert the [<img src="] without having the engine insert its own [>] before the URL.

Is there a simple escape mechanism to just build the html, then have the feed output exactly what I'm aiming for?

The above would be great. Even better would be this: [<a target="_blank" href="ITEMLINK"><img src="URL" style="width: USERINPUT;" alt="TEXT"></a>]

This seems like it should be easier, but I'm a regex noob -- Is there a simple fix here?

S.

You really should have created a new thread for your question, since it's only marginally connected to the original theme.

Anyway, I think a couple a regular expressions should achieve what you want. These rules should be in a Regex module, which is different from a String Regex module.
In [item.width] replace [^] with [connect your width input here]
In [item.gr:annotation.content.content] replace [(?s).*?href="([^"]+).*::\s*(.*)] with [<a href="${link}"><img src="$1" style="width:${width};" alt="$2"></a>]
Remove all the outer []s. If any spaces appear in the middle parts of the rule please remove them, since they were added by the board editor.