Extracting the actual in-text title from a PDF - pdf

There seems to be a lot of questions about extracting a title from a PDF (using its metadata). However, the large majority of the titles do not seem to exist in the metadata. I found this out when using pypdf .
Is there anyway to actually retrieve the in text title from a pdf? I tried to export to a text file then search but there is no consistent formatting. Is there any way to export the pdf to a document with its formatting, then check for a font size >= 14 ?

This is a very good question. Applications that create PDFs don't seem to do anything useful with the available metadata fields.
Take pdflatex as an example: even when one sets the \title{...} and \author{...} in the preamble, this information is not reflected in the metadata. After a quick search, the solution appears to be to introduce a block in the preamble which is read only by pdflatex [1]:
\pdfinfo
{
/Title{...}
/Author{...}
...
}
...which is then placed in the the relevant metadata fields of the PDF. It is strange that this is necessary, though.
I cannot speak for word processors like Word or Writer. One presumes such metadata fields have to be set manually by the user.
Perhaps a heuristic approach is the only way you can approach your problem if your PDFs are not generated by you. [2] seems like it does something similar to what you want, but I guess it depends how well published the PDFs are -- this tool seems to be scientific-paper oriented.
I hope that is at least some help.
[1] http://wlug.org.nz/PdfLatexNotes
[2] http://www.molspaces.com/d_cb2bib-metadata.php

Related

Understanding the PDF DOM

I am writing an application that has to read and interpret data stored in some PDF files. The reading part is done but I am only able to get a dump of all the words on a page and not the format of the words. What I mean is that if I have to extract a table, I am getting the numbers in the table but not the markup which defines the table.
Further, there is some formatting used which displays a few of these numbers within parentheses (meaning that those numbers are negative) but the parentheses themselves are not part of the text. Hence, I am not able to distinguish between positive and negative numbers present in the PDF table!
How do you get the PDF markup along with the text? Is a PDF similar in structure to an XML with tags used to markup tables etc.? If not, then, is there a resource which describes the salient features of the PDF DOM?
I am using VBA and the Acrobat library (AcroExch etc.)
There is no such thing as "PDF markup" in the sense of HTML etc. A table in PDF cannot be distinguished from line art, other than by using OCR, which can be error-prone if the layout is complex. It is simply drawn using geometrical shapes, like in a vector-based graphics program.
"Is a PDF similar in structure to an XML with tags used to markup tables etc.?"
No, not at all.
And there is no such thing as a 'DOM' either. Google for a file named *PDF32000_2008.pdf*. The current PDF specification for v1.7 (ISO spec) is that file. You should be able to locate it on the Adobe website.
As omz stated, text inside PDF does not really have a structure. You can take a look on the specification here. However, for some very specific files, there is something called PDF Tags, or PDF Marked Content, which is fairly new, and it aims to give PDF documents some kind of structure. If you target this kind of files specifically, you might be able to achieve something. Take a look on chapter 10 (Document Interchange) of the Adobe's specification for further details.
Maybe what you want to achieve can be done with less effort and faster by using TET, the Text Extraction Toolkit made by the fine folks from pdflib.com ( http://www.pdflib.com/products/tet/ ) ??
AFAIR, the TET has some (limited) support for table detection as well....

Creating ODT and PDF files as end result

I've been working on an app to create various document formats for a while now, and I've had limited success.
Ideally, I'd like to dynamically create a fairly simple ODT/PDF/DOC file. I've been focusing my efforts on ODT, because it is editable, and open enough that there are several tools which will convert it to any of the other formats I need.
The problem is that the ODT XML files are NOT simple, and there aren't any good-quality API's I could find (especially in python). So far, I've had the most success creating a template ODT file, and then manipulating the DOM in python as needed. This is ok generally, but is quickly becoming inadequate and requires too much tweaking every single time I need to alter one of the templates.
The requirements are:
1) Produce a simple document that will have lists, paragraphs, and the ability to draw simple graphics on the page (boxes, circles, etc...)
2) The ability to specify page size, and the different formats should generally print the exact same output when sent to a printer
My questions:
1) Are there any other ways I can produce ODT/PDF/DOC files?
2) Would LaTeX be acceptable? I've never really used it, does anyone have experience converting LaTeX files into other formats?
3) Would it be possible to use HTML? There are a lot of converters online. Technically you can specify dimensions in mm/cm, etc..., but I am worried that the printed output will differ between browsers/converters....
Any other ideas?
have you tried pandoc? i've been using it with good success for the conversion of different formats into each other. why try to invent the wheel twice?
I suppose to be successful, you'd have to define how you want to input everything. Why don't you just use openoffice? it will save to ODT (duh...), PDF, and HTML (though it's not clean HTML, it's actually quite ugly).
In my recent experience, I've had success going from latex -> xhtml via LaTeXML (i had to compile from source). LaTeX is seeming more and more like a terminal format. It's great for PDF, but once you need some flexibility, it kind of fails. I should also note that there is no latex -> dvi in my workflow, so I can't comment on things like tex4ht that reads out of a dvi file (I have too many graphics that don't work with DVI to switch them now).
Shortly I'll be moving everything into docbook 4.5-- i like the docbook-utils package which supports latex, html, and i even saw a converter to ODT. But docbook is super-heavy on the markup, which is annoying, but it will provide me with the flexibility i need going forward.
Since you're using python, have you just considered using ReStructured Text?
I've also really enjoyed publishing from emacs' orgmode, which is a super light weight markup that goes into a bunch of different formats.

Is there a reliable way to determine if a PDF was generated from a Powerpoint file?

Like the title says. Reason I ask is that we're converting PDFs to formatted ASCII text (using pdftotext) and only want to display the ones that look reasonably sane.
PPT files tend to have text over images, diagonal text and others things that don't translate to ASCII very well, so we'd like to filter them out if we can.
The creating application of a PDF is listed in its XMP metadata. You can see this quite easily in Acrobat 9 (and I believe earlier): go to File > Properties, click Additional Metadata..., then go to Advanced and it's listed under both XMP Core Properties and PDF Properties:
xmp:CreatorTool: Microsoft PowerPoint
pdf:Creator: Microsoft PowerPoint
I'm guessing you want to find this programatically, so you'll need to find a library to read this metadata that works with your language. Here is a list of some XMP tools.
Short answer:
No, I don't think so.
Long answer:
No, I don't think so, because there are may ways to convert a PowerPoint file to pdf, for example Adobe Acrobat and PDFCreator and many many others. It's up to the converters to embed specific information in the PDF file, even if you find a way to detect PowerPoint-source pdf from one convert, the same method may not work for another.
Even longer answer:
No, I don't think so, because of the reasons described in the "long answer". And I don't think detecting the source of the PDF is the best approach to the problem you are trying to solve. Not just PowerPoint produces overlapped text and images. I think it's much better to detect the actual layout of the PDF file. If there are overlay of image and text, then you do some filtering or pre-processing to cater for that.
Your reasoning is very arbitrary - there are surely plenty of PPT files without the features you describe, and plenty of PDF files with them, that were generated from another source.
In theory a better method would just be to detect when these "unwanted" situations occur. However, even though the PDF format is partly open (only for reading, apparently, so it's not truly an open format), extracting complex data like that would be incredibly difficult.
All PDFs can have this problem regardless of their source. Most desktop publishing suites are capable of outputting PDF and are often sold boasting their high quality and flashier PDF presentations ...
A "saner" method would be to use a PDF parser, ITextSharp, or pdfNet...etc, Using the library of your choice, find all image rectangles, and all text rectangles, SORT THE RECTANGLES, and then see if there is substantial overlap of text and image rects -- ignoring image to image overlaps. If so, reject the page and/or document.
That won't be perfect, but at least it's going to catch many PDFs that aren't sane, regardless of source. Other heuristics to add would include color analysis. (i.e. are the colors in the overlapping region sufficiently different to allow "sane" results?)
Best of luck to you
It might put its name in the creator or producer info, but I don't have a copy to check this theory with.
In general, it is not an easy task to programmatically determine (reliably) where a file came from or how it was generated based on its contents. After all, a file is just a collection of bits.
Unless you have a lot of resources to expend building the heuristics to determine whether a file looks "reasonably sane" according to your needs, I would consider this a task for human beings.
some converter from ppt to pdf preserve creator in comments at begin of pdf.
I think that PDF's generated from most applications seem to be the same. It may have some meta-data that you can read from the file...

How to extract data from a PDF file while keeping track of its structure?

My objective is to extract the text and images from a PDF file while parsing its structure. The scope for parsing the structure is not exhaustive; I only need to be able to identify headings and paragraphs.
I have tried a few of different things, but I did not get very far in any of them:
Convert PDF to text. It does not work for me as I lose images and the structure of the document.
Convert PDF to HTML. I found a few tools that helped me with this, and the best one so far is pdftohtml. The tool is really good presentation wise, but I haven't been able to successfully parse the HTML.
Convert PDF to XML. Same as above.
Anyone has any suggestions on how to tackle this problem?
There is essentially not an easy cut-and-paste solution because PDF isn't really very interested in structure. There are many other answers on this site that will tell you things in much more detail, but this one should give you the main points:
If identifying text structure in PDF documents is so difficult, how do PDF readers do it so well?
If you want to do this in PDF itself (where you would have the majority of control over the process), you'll have to loop over all text on pages and identify headers by looking at their text properties (fonts used, size relative to the other text on the page, etc...).
On top of that you'll also have to identify paragraphs by looking at the positioning of text fragments, white space on the page, closeness of certain letters, words and lines... PDF by itself doesn't even have a concept for a "word", let alone "lines" or "paragraphs".
To complicate things even more, the way text is drawn on the page (and thus the order in which it appears in the PDF file itself) doesn't even have to be the proper reading order (or what us humans would consider to be proper reading order).
PDF parsing for headers and its sub contents are really very difficult (It doesn't mean its impossible ) as PDF comes in various formats. But I recently encountered with tool named GROBID which can helps in this scenario. I know it's not perfect but if we provide proper training it can accomplish our goals.
Grobid available as a opensource on github.
https://github.com/kermitt2/grobid
You may do use the following approach like this with iTextSharp or other open source libraries:
Read PDF file with with iTextSharp or similar open source tools and collect all text objects into an array (or convert PDF to HTML using the tool like pdftohtml and then parse HTML)
Sort all text objects by coordinates so you will have them all together
Then iterate through objects and check the distance between them to see if 2 or more objects can be merged into one paragraph or not
Or you may use the commercial tool like ByteScout PDF Extractor SDK that is capable of doing exactly this:
extract text and images along with analyzing the layout of the text
XML or CSV where text objects are merged or splitted into paragraphs inside a virtual layout grid
access objects via special API that makes it possible to address each object via its "virtual" row and column index disregarding how it is stored inside the original PDF.
Disclaimer: I am affiliated with ByteScout
PDF files can be parsed with tabula-py, or tabula-java.
I made a full tutorial on how to use tabula-py on this article. You can tabula in a web-browser too as long as you have installed Java.
Unless its is Marked Content, PDF does not have a structure.... You have to 'guess' it which is what the various tools are doing. There is a good blog post explaining the issues at http://blog.idrsolutions.com/2010/09/the-easy-way-to-discover-if-a-pdf-file-contains-structured-content/
As mentioned in the answers above, PDF's aren't very easy to parse. However, if you have certain additional information regarding the text that you want to parse, you can pull it off.
If your headings are positioned at specific parts of the page, you can parse the PDF file and sort the parsed output by coordinates.
If you have prior knowledge of the spacing between headings and paragraphs, you could also leverage this information to parse the file.
PDFBox is a PDF parsing tool that you can use for extracting text and images on top of which you can define your custom rules for parsing.
However, for parsing PDFs you need to have some prior knowledge of the general format of the PDF file. You can check out the following blogpost Document parsing for more information regarding document parsing.
Disclaimer:I was involved in writing the blogpost.
iText api:
PdfReader pr=new PdfReader("C:\test.pdf");
References:
PDFReader

Create destinations for all bookmarks in a PDF file with iText API

I'd like to write some (java) code that takes a PDF document, and creates named destinations from all of the bookmarks. I think the iText API is the easiest way of doing this, but I have never used the API before.
How would you go about writing this sort of code with the iText API? Can iText do the parsing needed to manipulate existing PDFs by itself? The kind of manipulations I am thinking of are:
Open,
Find bookmarks,
Create destinations,
Save,
Close.
Or is there a different API that would be better?
Followup: I submitted a patch to iText a few months ago (it has now been accepted and is part of HEAD) that adds text parsing capabilities to iText. PdfBox (mentioned below) has (had?) problems with reading newer PDFs that use xref streams instead of the older xref table format.
Another library that is very good at parsing existing PDF files is PdfBox It can also be used for modifying an existing PDF. FYI - this is the text parser that Lucene uses.
I will also mention that iText does have the ability to parse a PDF file, it's just not great at parsing the text content on each page. If you are looking at accessing the PDF higher level constructs (Dictionaries, etc...) that are used for storing bookmarks, etc... and you don't mind getting your hands a little dirty with reading the PDF spec, you can absolutely do what you are asking about (we do it quite a bit ourselves).
The PDF Spec is big, but readable for the most part, and you don't have to worry about the bulk of it (which is geared towards actual page content and rendering) if all you are trying to do is extract bookmarks.
I'll just warn you up front that you may be disappointed with this. iText isn't really intended to be used as a parser. It's really more for creating entirely new PDF documents, but you can take a whack at it.
To start, using iText, you won't be able to modify the existing PDF document. What you can do, though, is to make a copy with the additional features that you want. (If somebody else knows better, please let me know, this drives me crazy.)
What you will want to do is create a PdfReader object from an input stream on your source file. Then create a PdfCopy object (which is just an extended PdfWriter that makes getting data from an existing source more convenient) for your destination.
As far as I can tell, the bookmarks cannot be obtained from iText at all. Another library may be needed. I think jpedal may have the ability to extract them (it can get them as an XML document, which you may then have to parse to get what you want.) However you get them, you can then add them to a java.util.List, and set that list as outline on the PDFCopy. The bookmarks themselves are just HashMaps with a particular set of keys. I'm not sure what all of the values are, but they include "Title", "Action" (which seems to be where you'd specify that this is a named destination, though I don't know what that value would be), and "URI" (which is used if this is an external link -- I suspect that this would specify the name of the named destination that you're linking to). Again, the specifics are hard to find.
Then iterate over the pages of the reader, importing each page to the PdfCopy. this page may help you.
Sorry I'm not more helpful to you. Good luck.
P.S. If anybody else knows of a better tool that's either (L)GPL or BSD licensed, I'd love to hear about it.