Ah, see if I did that though, Hellblazer (2016) would just have the description "Volume 2."

http://comicvine.gamespot.com/the-hellblazer/4050-93381/

Yeah, that's exactly what happened. For me, this died down after a couple hours, but I also haven't run through a big batch of series. What sucks here is that none of the other search engines have a decent site search function, so the key is to stay as friendly with Google as possible.

I'm wondering if yours might've been exacerbated by the repeated use of .com from a .nl region, or if there's just a threshold of X weird searches (with a ton of inurl and intitle arguments) in a Y period that flags things.

What's funny is, this is actually always providing the description for the first issue, and not the series. Comicvine rarely, if ever, has a description field for the series itself. What's funny is the series descriptions on Comixology are frequently also just the descriptions from the first issue.

So I'm wondering if it's actually throwing an error, rather than not finding anything. I may need to mask the initial Google search better, since too many requests that look automated can trigger it to send you captcha images.

When I tried it, it found an ID and downloaded the image fine. I'll try to work in a better catch for Google errors. If you're in a non-US country, it might be worth changing google.com to your local Google (lines 68, 70 and 78).

If this sorts it out, I'll move that part to a variable at the top.

Just uploaded a new version with mogrify stuff in it.

Since I wrote this to be something you could run through some sort of automation, I figured it made more sense to use a setting in PageBuilder.sh instead of a command argument.

Under apikey there's now coverDownload and coverCreate, with download defaulting to true and create defaulting to false (with a note that you need imageMagick installed to use it). So if you only wanted to use a modified cover.jpg, you could set download to false and create to true. It should also download cover.jpg if it's missing.

If both are true, it'll try to download first, and create if it can't find a match on comixology.

Few times have I disliked Slackware as much as in getting ImageMagick installed... no package manager meant manually installing every dependency as it errored and asked for them.

Mogrify is a fantastic addition to this, though I'd make one change:

Rather than

cp "$(dirname "$1")/cover.jpg" "$(dirname "$1")/folder.jpg"

I would use

convert "$(dirname "$1")/cover.jpg" -resize 640 "$(dirname "$1")/folder.jpg"

I used A-Force (2016) as my test case.

Yours:

Mine:

It's more likely to include bits of title/label, but it looks less cut off image-wise.

Or maybe switch to -gravity north:

Though this wouldn't always get the title, since there are those rare cases where it's on the bottom.

Really, they're all significantly better than nothing. :) It's just down to personal preference at this point.

Just uploaded a version that fixes the ampersand issue. That was a tough fix to find.

The fuzzy search sorta goes against what I'm aiming for, so if/when I incorporate any of the mogrify stuff it'll be a fallback if there's no exact comixology match.

When it gets to that point, a google bypass would be a good idea. But at this point I haven't even looked at any of that.

All the more confusing, as I use it exclusively with Chrome...

I kept going, tightened up the google search to watch for the exact string in the page title, checks for year first and then falls back to without. New copy's on my dropbox.

The tough part was mimicking the request headers so Amazon/Cloudfare didn't flag my repeated attempts.

But this seems like pretty close to 100%. :)

I'll play around with the image stuff next, very cool stuff!