Simple page examiner: "Failed to extract URL from container *" despite correct URL in DtA

[Hypocee]

Since an initial lesson in JSON escapes, I've made dozens of successful page expanders. In the last few months, I've been having trouble with "Failed to extract URL" but the problems have all been complicated so I assumed I was writing bad regex. This case seems as simple as possible, and I can't make it work. Either I'm making a fool of myself again or something's wrong.

I want to patch some holes in my 8-Bit Theater run. Code:

{
    "type": "resolver",
    "ns": "downthemall.net",
    "author": "Hypocee",
    "prefix": "8-Bit Theater",
    "match": "nuklearpower.com",
    "static": true,
    "finder": "img src=\"(http://.\\.cdn\\.nuklearpower\\.com/.*?)\"",
    "builder": "{1}"
}

The regex is a little sloppy, yes, but it finds the right thing and only the right thing.

Used on pages like: http://www.nuklearpower.com/8-bit-theater-archive/?archive_year=2001

AC processes the container page properly. The target images appear in the DtA manager with their proper filenames, but all downloads fail with the "Failed to extract URL" error. If I get "Download Information" on any download in DtA and copy-paste that URL into Firefox, I go directly to the desired image. Save directory was created from the DtA filepicker, is empty and remains empty. DtA 2.0.8, AC 20010913 nightly WinXP Pro 32 SP3 FF 7.0.1

[nmaier]

Not a support forum, and really old issue, but the problem likely is a loop: Once something gets transformed, AC will try any other plugins, incl. the same one, on the result. What happens here is that the filter matches again and again. AC detects that and aborts.