@jeremybmerrill can you take a look at this one while you're at it? :)

@jazzido I'm getting empty responses on a PDF that should work (and do work off of master) from this.

For several different selections that work on master, I get this response for each, regardless of the extraction method I choose: [{"extraction_method":"","top":0.0,"left":0.0,"width":0.0,"height":0.0,"data":[[{"top":0.0,"left":0.0,"width":0.0,"height":0.0,"text":""}]], "spec_index": 0}]. The params to the request look right (and do work right, as I said, on master).

Does any of that ring a bell as to what could be happening? (It's possible, I guess, that my local environment is screwed up...?)

@jazzido: Was upgrading my JRuby to test this when I figured I might as well do it for Travis too :)

The only change besides using PDFBox for rendering, was upgrading tabula-java to 1.0.0.

That means that your extraction should also fail in 1.0.0. Can you try reproducing that with 0.9.2 and 1.0.0?

Thanks?

Interesting, looks like it: java -jar lib/jars/tabula-1.0.0-jar-with-dependencies.jar -a 193.163,34.425,333.158,365.67 ~/Downloads/cs-en-us-pbms.pdf does also return an empty string.

Then it's a bug. Can you send me that PDF?

http://www1.nyc.gov/assets/nypd/downloads/pdf/crime_statistics/cs-en-us-pbms.pdf

Thanks. Just added tabulapdf/tabula-java#171

Tested locally with tabulapdf/tabula-java#173, seems to work fine. Will update this branch with the new version of tabula-java as soon as I cut a release to Maven.

@jeremybmerrill @mtigas I went ahead and closed #418 on this PR. Besides removing the ugly fat jar from the repo, it'll force us to depend only on released versions of tabula-java.

There's an additional build step, I updated the README accordingly.

Let me know what you think.

I think it's looking pretty good for merging to master. Any outstanding issues?

BTW, I vote for bumping up to 1.2.0.

Yes, a beta first?.

Let me get a fix for the button feedback thing. Will be done by EOD.

And I'll aim to fix the anchors on the page thumbnails before we release 1.2.0 for real.

Yes, a beta first?.

Sounds good.

Let me get a fix for the button feedback thing. Will be done by EOD.

👍🏼

And I'll aim to fix the anchors on the page thumbnails before we release 1.2.0 for real.

Wouldn't a simple e.eventPropagation() fix that?

And I'll aim to fix the anchors on the page thumbnails before we release 1.2.0 for real.
Wouldn't a simple e.eventPropagation() fix that?

How so? I think the problem is that there's no JavaScript involved in that navigation, it's just the browser's built-in behavior for anchors. But with the new-ish <base> tag, href="#" is interpreted as navigating to the "base" URL.

Oh, gotcha. It's been a while since I touched the frontend code :)

Okay, I added the feedback.

Now I'm being annoying :) Is there a way to make the template buttons wrap nicely?

Man you must have a tiny screen. But yeah, I should figure out how to do that.

Got any layout ideas? I was thinking about making the autodetected tables show up as if they were a template (because they, kind of, are).

Maybe I'll right-align the Preview & Export Extract Data button, since it's the main button you'd use. Hmm...

Got any layout ideas?

File name on the first line, buttons below?

Should I go ahead and merge to master?

@jazzido if you're cool with releasing a beta/merging to master with these two bugs, sure. (the buttons wrapping funny on smaller screens; the thumbnail clicking being broken)

I don't think either's a blocker, though I'll try to fix 'em soon.

OK, I'm going to go ahead and merge to master. Thanks for fixing #712 !