What's Wrong With This HTML Code?

[Gawdzilla Sama]

I find it amusing that code people are supposed to be orderly and concise and the HTML code gets generated in a total mush.

[Thinking Aloud]

I find it amusing that code people are supposed to be orderly and concise and the HTML code gets generated in a total mush.

If the HTML was "generated" in something like MS Word (and a lot of people did that) then you may as well start again.

[klr]

I find it amusing that code people are supposed to be orderly and concise and the HTML code gets generated in a total mush.

I think I see the flaw in your logic.

I find it amusing that code people are supposed to be orderly and concise and the HTML code gets generated in a total mush.

If the HTML was "generated" in something like MS Word (and a lot of people did that) then you may as well start again throw yourself under the next bus.

[Gawdzilla Sama]

I find it amusing that code people are supposed to be orderly and concise and the HTML code gets generated in a total mush.

If the HTML was "generated" in something like MS Word (and a lot of people did that) then you may as well start again.

I get it from FineReader, my OCR program.

[Gawdzilla Sama]

So what's a good strategy for converting a 700+ page hardcopy to HTML with some nice clean, neatly laid out code?

[Thinking Aloud]

So what's a good strategy for converting a 700+ page hardcopy to HTML with some nice clean, neatly laid out code?

A human, and a lot of time.

The Gutenberg Project does a similar job on old texts, and relies on many thousands of volunteers to do all the format-fixing and corrections of OCRed documents.

[klr]

So what's a good strategy for converting a 700+ page hardcopy to HTML with some nice clean, neatly laid out code?

A human, and a lot of time.

The Gutenberg Project does a similar job on old texts, and relies on many thousands of volunteers to do all the format-fixing and corrections of OCRed documents.

There is no "magic bullet" for this work, especially when it comes to proof-reading and correcting the OCRed text.

As for HTML'ing it ... I don't know. It's possible that there are tools out there that would do a lot of the basic work in terms of HTML generation, but you'd still be advised to check everything afterwards.

[Gawdzilla Sama]

So, invoke the OPU (Organic Processing Unit) and chug through it. That hasn't changed in lo these many years. (1990 was my first HTML experience.)

And I've donated to Gutenberg, they're on the same server as Hyperwar, at UNC.

[klr]

So, invoke the OPU (Organic Processing Unit) and chug through it. That hasn't changed in lo these many years. (1990 was my first HTML experience.)

And I've donated to Gutenberg, they're on the same server as Hyperwar, at UNC.

It's possible software like Calibre (for generating eBooks) can also assist in generating HTML from OCRed text. It might be able to "intelligently" detect paragraph breaks, headings, etc. If you're really, really, really lucky, it might even be able to build a usable TOC.

[Gawdzilla Sama]

So, invoke the OPU (Organic Processing Unit) and chug through it. That hasn't changed in lo these many years. (1990 was my first HTML experience.)

And I've donated to Gutenberg, they're on the same server as Hyperwar, at UNC.

It's possible software like Calibre (for generating eBooks) can also assist in generating HTML from OCRed text. It might be able to "intelligently" detect paragraph breaks, headings, etc. If you're really, really, really lucky, it might even be able to build a usable TOC.

I'm actually learning more HTML doing it the hard way.

The WarDamageReports are ready to upload. (Except the long ones, they'll be for next week's goodie list.)

[Pappa]

So what's a good strategy for converting a 700+ page hardcopy to HTML with some nice clean, neatly laid out code?

A human, and a lot of time.

The Gutenberg Project does a similar job on old texts, and relies on many thousands of volunteers to do all the format-fixing and corrections of OCRed documents.

I find judicious usage of regular expressions works 95% of the time, and it cuts down the time required by a couple of orders of magnitude (or more).

Sorry... I meant for fixing badly generated HTML, not OCR correction.

[Gawdzilla Sama]

So what's a good strategy for converting a 700+ page hardcopy to HTML with some nice clean, neatly laid out code?

A human, and a lot of time.

The Gutenberg Project does a similar job on old texts, and relies on many thousands of volunteers to do all the format-fixing and corrections of OCRed documents.

I find judicious usage of regular expressions works 95% of the time, and it cuts down the time required by a couple of orders of magnitude (or more).

For example?

Where do I start learning that?

[klr]

http://en.wikipedia.org/wiki/Regular_expression

http://www.regular-expressions.info/reference.html

http://www.weitz.de/regex-coach/

http://en.wikipedia.org/wiki/List_of_re ... n_software

[Gawdzilla Sama]

Got 'em. Thanks.

Now to go screw up some files.

[klr]

That list of "supporting software" is incomplete. Almost all programming/HTML editors should support regular expressions, either out of the box, or via plugins.

It also doesn't cover programming language support for regular expressions, which is very widespread. There could be situations where that would come in very handy indeed.