Importing 11-page PDF into Scrivener document

I conducted an online survey for my dissertation and now I need to put the questionnaire in the appendix. The only way I can export my questions from the survey server is as a PDF that’s formatted as a survey with boxes and lines and stuff so it can be printed out and filled in by hand. The PDF is 11 pages long.

I’d like Scrivener to fill 11 pages entirely (I need header and footer to be available, though) with this PDF. How do I do this?

I imported the PDF, but if I drag it into a document, it is turned into a Scrivener link (I think). I can convert it to JPGs or something else before importing, but what resolution do I choose for the page to be filled (with header and footer staying available)? Do I have to split the file into 11 separate files?

So many questions! :open_mouth:

Please, somebody!

Maybe somebody can just tell me what the resolution of the writeable (ugly word) are in a RTF/DOC is? Then I’ll convert every page to a JPG of the right resolution and do it that way.

You can not import a PDF into a Scrivener document and when you drag it you indeed get a Scrivener link.

Converting the PDF into an image might do the trick but I don’t think it’s wise to do so with lots of images like in this case.

Keep in mind that Scrivener is made for writing and not for the final output if you need more than just a manuscript with basic formatting. If so, then a word processor, LaTeX or maybe InDesign would be the application for further processing of your Scrivener generated manuscript, including the addition of the survey PDF.

If your Scrivener manuscript fits all your needs except for the mising PDF appendix you could simply compile your manuscript (sans survey) to PDF and add the survey PDF to your manuscript PDF in Preview or any other PDF processing program.

If your PDF mostly contains of text you could convert it to RTF, with this free service of Devon Technologies maybe:

devon-technologies.com/files … ervice.zip

“Boxes and lines and stuff” does sound like a more complex PDF but maybe a conversion service like this can come handy in other cases.

Lots of ifs, but maye some of this can help you. The one part of your posting I did not understand is what you mean with Scrivener “filling” you PDF—it does contain all survey data already, doesn’t it?

As suavito says, there’s not really a great way of doing this. You can actually import a PDF document into the text of a Scrivener document, but there is a bug that currently disallows you from doing so by dragging a PDF from the binder (this is fixed for 2.0.3), which will be out later this week with a bit of luck). You can drag in a PDF file from the Finder, though.

The problem then, however, is that:

a) The PDF file will get converted to a JPG or PNG the next time you open Scrivener, because of the way Scrivener saves the text files internally, thus resulting in a loss of quality and pages. This has been improved for 2.0.3, though, and so will no longer apply.

b) Even once (a) is fixed in 2.0.3, the OS X text system just plonks multi-page PDF files into the text in a little control box with its own scroll bar. This isn’t surprising as there is no really good way of taking a multi-page PDF file and splitting its pages up and making them tally with the target application.

So there are two potential solutions, neither great:

  1. Cut the PDF file up into individual pages, cropping them so that only necessary information is included, then wait for 2.0.3 and drop each PDF page into a text document (they can all go in the same document, or in different text documents). This way each page will get printed separately rather than in a scroll box. (You need to wait for 2.0.3 to avoid the quality loss mentioned above.)

  2. Recreate the PDF file in Scrivener - obviously this is the most work and probably not worth it if you can get away with (1). Unfortunately there isn’t any great way of copying and pasting the contents of PDF files, as you’ll know if you’ve ever tried to copy the contents of a PDF from Preview and paste it into a word processor.

Sorry it’s not better news.

All the best,
Keith

Thanks to both of you.

The ‘filling’ part is irrelevant, really. The survey has been conducted online and all questions and responses are stored exclusively in an online database. For my dissertation I simply need an empty version of the survey that I can put in the appendix so my examiners can have a look at the questions I was asking. I could type all the questions into Scrivener by hand, but if you’ve ever done a survey, you will know that the formatting is a b***h. So I was hoping I could export the questions from the database into Scrivener somehow. The only way I can access the questions, however, is via a PDF export. Hence my problem.

Compiling to PDF and then inserting the survey as a PDF would screw up the page numbers, but I could compile to RTF and do it in Word. I’ll try the Devon service. Thanks for the link! I also just tried an online pdf-to-rtf website, but didn’t get the rtf yet. I was about to play around with these suggestions and then ping comes another reply notification from Keith!

So I guess I’ll be waiting for 2.0.3. Which is really coming in the nick of time since I have to hand in my bound final seminar document in 2 weeks! (And the final final dissertation in 2 months.)

Yeah, recreating is not an option. I’m running out of time. Cropping and cutting up sounds OK. But will the PDF be treated as a JPG internally? In that case, or actually generally, what maximum resolution should I use for JPGs? How wide and high is a page pixel-wise?

I appreciate you guys taking the time to answer (and Keith, the layout/characters not appearing bug hasn’t popped up yet, but I’m being observant).

Cheers!

In 2.0.3 the PDF will still be stored and treated internally as PDF data - the problem with the current version is that it gets converted internally as JPG or PNG, which is why you should wait for 2.0.3 before inserting the cropped pages.

All the best,
Keith

Cool, looking forward to it.

This sounds very promising:

I can’t tell you if it works but I am very interested if it does. So if you give LibreOffice a try please share your experience.

Would like to, but unfortunately I can’t really take time off at the moment to try a new office suite. After I hand in my diss maybe. Although hopefully I won’t be touching a keyboard for a long while then. :wink:

It’s better but still doesn’t seem to work for me.

I installed 2.0.4, cropped the PDF, split it up into 11 separate PDFs (both tasks done in Acrobat X), and inserted those pieces into one Scrivener document.

Scrivener does insert the actual PDF now, so that’s good. But it’s bigger than the printable area in the exported RTF and I can’t resize it, which means it gets cropped in the compiled RTF. On top of that, the quality is rather mediocre.

Original PDF at 300%:

Compiled RTF document in Word:

Am I doing something wrong?
The only solution I see at the moment is to export the PDF to JPG/PNG/TIFF and insert it as images. But scaling in Scrivener to the width of my eventual Word page means the images are only 420 pixels wide! That’s really low quality! Printing on a laser printer at 600dpi means I could in theory print an image nearly 3,500 pixels wide on one page! Does that mean I would be better off compiling the document without the images and inserting them later on in Word?

@suavito: I tried the Devon plugin, but it couldn’t cope with the formatting of the PDF. Might come in handy on future occasions, though.

Could you not compile your document, save it as a PDF, and then directly add the PDF survey pages where appropriate? That way, you do not need to worry about image conversions or DPI. The point of a PDF is to have a print-ready file, so why try and fight it?

Alternatively, do the PDFs contain text data? If so, you could just copy the text. If it’s pixel data, then run it through an OCR program. But I really think the first solution is the simplest, unless you have more exacting requirements of which I am unaware.

Yes, that would be the simplest solution, but I would run into trouble with page numbers. The finished dissertation needs to be one long printable document with a header and a footer. :frowning:

OCR won’t help me much. The PDF has quite elaborate formatting (it’s a survey with lots of tick boxes and lines and text fields). :frowning:

Yes.
And use Word 2008 or 2011, not earlier, because earlier versions do funny things to inserted pdfs.

The rendering quality is all down to the OS X text system, so I have no control over that. Regarding the other issues:

  1. PDF images cannot be resized - PDF is a very different format to JPG, PNG etc, and resizing part or all of PDF page is a complicated business, so you need to do that before you import it into Scrivener.

  2. As I mentioned earlier, the RTF format does not support embedded PDF. Scrivener now uses some custom RTF codes to store PDF data inside the RTF files in the .scriv package. When you export to PDF, though, the embedded PDF files will still get converted to PNG or JPG. There’s no way around this - Word has no way of reading Scrivener’s custom RTF codes that I created to store the PDFs in Scrivener.

The only change 2.0.4 has made is that embedded PDF files now retain their quality in Scrivener, and when printing directly from Scrivener. This is now back to the same standards as 1.x in this regard. The only advantage of this is really to those who wish to print directly from Scrivener or who are using MMD to write scientific files and generate LaTeX with embedded PDFs.

Unfortunately, I don’t think you mentioned that your objective was to get back into Word again, or I would have clarified this earlier.

All the best,
Keith

No worries Keith. As long as I get done what I need to get done all is good. But sorry about wasting your time a bit.
Also, I’m only going back to Word, because I know it. I wouldn’t mind experimenting with MMD or LaTeX, but I simply don’ t have the time anymore. First deadline is in 6 days.

I always thought you couldn’t insert PDFs into word documents. Now I played around and found out you could do it via drag&drop. But I can’t find an option in the menu (Word 2011) and I was told to not use drag&drop with images because then they are stored differently from a proper menu command and you can run into trouble down the line (with printing, etc.). Is there a menu command there hidden somewhere? Does it really matter?

Thanks for mentioning this nicka!

You haven’t wasted my time at all - this discussion actually prompted me to come up with the solution I found to embedding PDF documents in Scrivener for 2.0.4.

In Word, you can import a PDF by going to Insert > Photo > Picture From File… Presumably .docx now supports embedded PDFs (in fact, having just tested and decompiled a test .docx file, this does seem to be the case, although the PDF file is converted internally to an .emf file, which embeds the PDF itself, and of course you can only import one page of a PDF file). The problem remains, of course, that the RTF format doesn’t support PDF data, and Scrivener’s .docx exporter, which is just the standard OS X one provided by Apple and used in TextEdit too, doesn’t support images at all…

All the best,
Keith

Good. So the first thing I should probably do after compiling to RTF is to save as DOCX, so I get the full PDFembeddednessglory?

Yes, save in Word as doc or docx, NOT rtf, or your pdfs will all be rasterised. If you insert a pdf, and save as doc or docx, then the pdf’s vector information is preserved (from Word 2008 – previously this only worked with ps or eps).

You can even print the doc to pdf (i.e. produce a pdf using the standard Print dialogue in Word) and the result will have proper vector images and text (assuming that your original pdf did). Try printing to pdf then zoom in on that pdf in Preview. The text that was in the pdf you embedded in the Word file should stay sharp.

Awesome! Thanks guys.

Good luck with the final stages of the dissertation.

By the way, I know it’s no good at this stage to think of switching word processor, but the next time you write a dissertation, you might consider Mellel. It really shines with long documents with inserted figures, lots of footnotes etc., and has excellent integration with Bookends for references. (No connection with the company, except as a user and beta tester.)