Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

So you want to parse a PDF?

eliot-jones.com

Post reply on HN
Print view View original

101–110 of 236 posts

Re: So you want to parse a PDF?

#101

Post by nottorp »

throwaway4496 wrote: post #80
nottorp wrote: post #77

Earlier quoted context omitted.

How many different PDF generators have done those millions of PDFs tho? Because you're right if you're paid to evaluate all the formats with the Mark 1 eyeball and do a custom parser for each. It sounds like it's feasible for your application. If you want a generic solution that doesn't rely on a human spending a week figuring out that those 4 absolutely positioned text fields are the invoice number together (and in…

We process invoices from around the world, so more PDF generators than I care to count. It is hard a problem for sure, but the problem is the rendering , you can't escape that by rastering it, that is rendering. So it is absurd to pretend you can solve the rendering problem by rendering it into an image instead of a structured format. By rendering it into a raster, now you have 3 problems, parsing the PDF, rendering…

You are using the Mark 1 eyeball for each new type of invoice to figure out what field goes where, right?

Re: So you want to parse a PDF?

#102

Post by throwaway4496 »

jiveturkey wrote: post #62
diptanu wrote: post #15

Disclaimer - Founder of Tensorlake, we built a Document Parsing API for developers. This is exactly the reason why Computer Vision approaches for parsing PDFs works so well in the real world. Relying on metadata in files just doesn't scale across different source of PDFs. We convert PDFs to images, run a layout understanding model on them first, and then apply specialized models like text recognition and table recogn…

Doesn't rendering to an image require proper parsing of the PDF?

Yes, and don't for a second think this approach of rastering and OCR'ing is sane, let alone a reasonable choice. It is outright absurd.

Re: So you want to parse a PDF?

#103

Post by throwaway4496 »

nottorp wrote: post #101
throwaway4496 wrote: post #80

Earlier quoted context omitted.

We process invoices from around the world, so more PDF generators than I care to count. It is hard a problem for sure, but the problem is the rendering , you can't escape that by rastering it, that is rendering. So it is absurd to pretend you can solve the rendering problem by rendering it into an image instead of a structured format. By rendering it into a raster, now you have 3 problems, parsing the PDF, rendering…

You are using the Mark 1 eyeball for each new type of invoice to figure out what field goes where, right?

It is a bit more involved, we have a rule engine that is fine tuned over time and works on most of invoices, there is also an experimental AI based engine that we are running in parallel but the rule based Engine still wins on old invoices.

Re: So you want to parse a PDF?

#104

Post by viraptor »

sbrother wrote: post #52
BobbyTables2 wrote: post #45

Earlier quoted context omitted.

Kinda funny. Printing a PDF and scanning it for an email it would normally be worthy of major ridicule. But you’re basically doing that to parse it. I get it, have heard of others doing the same. Just seems damn frustrating that such is necessary. The world sure doesn’t parse HTML that way!

I've built document parsing pipelines for a few clients recently, and yeah this approach yields way superior results using what's currently available. Which is completely absurd, but here we are.

I've done only one pipeline trying parse actual PDF structure and the least surprising part of it is that some documents have top-to-bottom layout and others have bottom-to-top, flipped, with text flipped again to be readable. It only goes worse from there. Absurd is correct.

Re: So you want to parse a PDF?

#105

Post by lou1306 »

vander_elst wrote: post #81
diptanu wrote: post #15

Disclaimer - Founder of Tensorlake, we built a Document Parsing API for developers. This is exactly the reason why Computer Vision approaches for parsing PDFs works so well in the real world. Relying on metadata in files just doesn't scale across different source of PDFs. We convert PDFs to images, run a layout understanding model on them first, and then apply specialized models like text recognition and table recogn…

It might sound absurd, but on paper this should be the best way to approach the problem. My understanding is that PDFs are intended to produce an output that is consumed by humans and not by computers, the format seems to be focused on how to display some data so that a human can (hopefully) easily read them. Here it seems that we are using a technique that mimics the human approach, which would seem to make sense. I…

> the format seems to be focused on how to display some data so that a human can (hopefully) easily read them

It may seem so, but what it really focuses on is how to arrange stuff on a page that has to be printed. Literally everything else, from forms to hyperlinks, were later additions (and it shows, given the crater-size security holes they punched into the format)

Re: So you want to parse a PDF?

#106

Post by formerly_proven »

vander_elst wrote: post #81
diptanu wrote: post #15

Disclaimer - Founder of Tensorlake, we built a Document Parsing API for developers. This is exactly the reason why Computer Vision approaches for parsing PDFs works so well in the real world. Relying on metadata in files just doesn't scale across different source of PDFs. We convert PDFs to images, run a layout understanding model on them first, and then apply specialized models like text recognition and table recogn…

It might sound absurd, but on paper this should be the best way to approach the problem. My understanding is that PDFs are intended to produce an output that is consumed by humans and not by computers, the format seems to be focused on how to display some data so that a human can (hopefully) easily read them. Here it seems that we are using a technique that mimics the human approach, which would seem to make sense. I…

Yes, PDFs are primarily a way to describe print data. So to a certain extent the essence of PDF is a hybrid vector-raster image format. Sure, these days text is almost always encoded as or overlaid with actual machine-readable text, but this isn't really necessary and wasn't always done, especially for older PDFs. 15 years ago you couldn't copy (legible) text out of most PDFs made with latex.

Re: So you want to parse a PDF?

#107

Post by jeroenhd »

gcanyon wrote: post #57

The answer seems obvious to me: 1. PDFs support arbitrary attached/included metadata in whatever format you like. 2. So everything that produces PDFs should attach the same information in a machine-friendly format. 3. Then everyone who wants to "parse" the PDF can refer to the metadata instead. From a practical standpoint: my first name is Geoff. Half the resume parsers out there interpret my name as "Geo" and "ff" s…

There's a huge difference between parsing a PDF and parsing the contents of a PDF. Parsing PDF files is its own hell, but because PDFs are basically "stuff at a given position" and often not "well-formed text within boundary boxes", you have to guess what letters belong together if you want to parse the text as a word.

If you're interested in helping out the resume parsers, take a look at the accessibility tree. Not every PDF renderer generates accessible PDFs, but accessible PDFs can help shitty AI parsers get their names right.

As for the ff problem, that's probably the resume analyzer not being able to cope with non-ASCII text such as the ff ligature. You may be able to influence the PDF renderer not to generate ligatures like that (at the expense of often creating uglier text).

Re: So you want to parse a PDF?

#108

Post by retinaros »

diptanu wrote: post #15

Disclaimer - Founder of Tensorlake, we built a Document Parsing API for developers. This is exactly the reason why Computer Vision approaches for parsing PDFs works so well in the real world. Relying on metadata in files just doesn't scale across different source of PDFs. We convert PDFs to images, run a layout understanding model on them first, and then apply specialized models like text recognition and table recogn…

I do same but for document search. Colqwen + a VLM like claude.

Re: So you want to parse a PDF?

#109

Post by jeroenhd »

throwaway4496 wrote: post #23
Alex3917 wrote: post #19

Earlier quoted context omitted.

> This is exactly the reason why Computer Vision approaches for parsing PDFs works so well in the real world. One of the biggest benefits of PDFs though is that they can contain invisible data. E.g. the spec allows me to embed cryptographic proof that I've worked at the companies I claim to have worked at within my resume. But a vision-based approach obviously isn't going to be able to capture that.

Cryptographic proof of job experience? Please explain more. Sounds interesting.

In theory your (old) boss could sign part of your CV with a certificate obtained from any CA participating in Adobe's AATL programme. If you use the software right, you could have different ranges signed by different people/companies. Because only a small component gets signed, you'd need them to sign text saying "Jane Doe worked at X corp and did their job well" as a signed line like "software developer" can be yanked out and placed into other PDF documents (simplifying a little here).

I'm not sure if there's software out there to make that process easy, but the format allows for it. The format also allows for someone to produce and sign one version and someone else to adjust that version and sign the new changes.

Funnily enough, the PDF signature actually has a field to refer to a (picture of) a readable signature in the file, so software can jot down a scan of a signature that automatically inserts cryptographic proof.

In practice I've never seen PDFs signed with more than one signature. PDF readers from anyone but Adobe seem to completely ignore signatures unless you manually open the document properties, but Adobe Reader will show you a banner saying "document signed by XYZ" when you open a signed document.

Re: So you want to parse a PDF?

#110

Post by jeroenhd »

creatonez wrote: post #68
diptanu wrote: post #15

Disclaimer - Founder of Tensorlake, we built a Document Parsing API for developers. This is exactly the reason why Computer Vision approaches for parsing PDFs works so well in the real world. Relying on metadata in files just doesn't scale across different source of PDFs. We convert PDFs to images, run a layout understanding model on them first, and then apply specialized models like text recognition and table recogn…

While you're doing this, please also tell people to stop producing PDF files in the first place, so that eventually the number of new PDFs can drop to 0. There's no hope for the format ever since manager types decided that it is "a way to put paper in the computer" and not the publishing intermediate format it was actually supposed to be. A vague facsimile of digitization that should have never taken off the way it d…

PDFs serve their purpose well. Except for some niche open source Linux tools, they render the same way in every application you open them in, in practically every version of that application. Unlike document formats like docx/odf/tex/whatever files that reformat themselves depending on the mood of the computer on the day you open them. And unlike raw image files, you can actually comfortably zoom in and read the text.
Post reply on HN