Free open-source Java library to convert PDF -Image, PDF - HTML, PDF - extract images and text

Question

I need Java Library to perform following tasks 1) Convert Pdf pages to Image 2) Extract html text from PDF pages with there locations on the page 3) Extract images from PDF pages

I have already tried

PDFBox - it fails with error --unsupported/disabled operation: BDC and EMC
icePDF - it works for task 1) and 3) but again its paid.
PDFRenderer - it fails
BFO - its paid library but able to perform tasks 1) and 3)

Can anyone suggest better solution.

I suppose that is for Converting HTML to PDF, not vice versa — Yashpal Singla, Oct 30 '12 at 07:50
I'd be interested in such a Java library, too. At the moment we are using http://en.wikipedia.org/wiki/Poppler_(software) on Linux systems to do similar tasks. — Udo Klimaschewski, Oct 30 '12 at 08:06
@YashpalSingla itext can do this, i belive. For example - how to extract text from pdf: http://itextpdf.com/examples/iia.php?id=275 — user1516873, Oct 30 '12 at 08:45
and this may be helpful for extraction images http://stackoverflow.com/questions/6851385/save-tiff-ccittfaxdecode-from-pdf-page-using-itext-and-java — user1516873, Oct 30 '12 at 08:59
I have used pdfToHTML to get HTML from pdf in ubuntu and than i have used cutycapt to get image from it. Than i parsed that HTML to extract text with styling and locatoin using JSoup.Output was satisfactory to what i required. — Yashpal Singla, Oct 31 '12 at 13:18
"unsupported/disabled operation: BDC and EMC" is not an error, it is an INFO and is irrelevant to most PDF files. — Tilman Hausherr, Dec 18 '14 at 08:23
@user1516873, iText is paid product and its license for single developer costs above $ 2k. — deathrace, Dec 08 '15 at 06:44
@deathrace No, it has dual licensing, for open-source project you can use AGPL version http://itextpdf.com/AGPL — user1516873, Dec 08 '15 at 08:48

score 0 · Answer 1 · answered Oct 31 '12 at 13:08

0

Have you tried JOD Converter? It's a Java API to a self-booted Open Office Server.

To see whether it converts to/from the formats you want, just install Open Office, open a file, and try to "Save As" the format you need, to see if it's supported.

answered Oct 31 '12 at 13:08

Stewart

17,616
8
52
80

Thanks @Stewart, i have tried JOD Converter, but the output is not useful as html – Yashpal Singla Oct 31 '12 at 13:11

Yashpal Singla · Accepted Answer · 2013-01-08T07:59:35.133

0

I have followed following steps to solve the issue in Ubuntu Enviornment

Step 1) Used pdftohtml library to convert pdf to html

Step 2) Used Jsoup to extract text with styling and position from html in step 1)

Step 3) Used CutyCapt to generate snapshot of HTML (if required)

We can also use pdftoppm command to extract images directly from pdf

edited Jan 08 '13 at 07:59

answered Oct 31 '12 at 13:21

Yashpal Singla

1,924
4
21
38

Singha All of them can be accomplished by using pdfbox – Neeraj Nov 05 '12 at 12:03
Its not singha its Singla, anyhow thanks for your comment – Yashpal Singla Nov 06 '12 at 06:55
@Neeraj Can you provide me a link to an example of converting pdf to html using pdfbox. – Rachit Agrawal May 07 '13 at 09:17
@Rachit Agarwal: java -jar pdfbox-app-x.y.z.jar ExtractText -html http://neerajkarimpuzha.wordpress.com/2012/04/04/pdf-to-html-update/, http://pdfbox.apache.org/commandlineutilities/ExtractText.html – Neeraj May 08 '13 at 03:48
Thanks..but i was looking forward to do this from java code. Not invoke it through command line arguments – Rachit Agrawal May 08 '13 at 06:35

score -2 · Answer 3 · edited Jan 05 '15 at 22:24

-2

You can do all those things with PDFBox. But for getting the position there is no API. Download the latest PDFBox. Go to the following links to find your solutions.

Convert Pdf pages to Image
Extract images from PDF pages
Extract html text from PDF pages with there locations on the page is a little bit different. Using the API you will not get the position information. But you can get all the position information using PDFBox.

Please have a look at this link. There you can see getTextPos() function. getTextPos().getXPosition(), getTextPos().getYPosition() will give you X and Y coordinates.

edited Jan 05 '15 at 22:24

bummi

27,123
14
62
101

answered Nov 05 '12 at 07:48

Neeraj

1,612
7
29
47

1

i have tried PDFBox already, as you can see in my post, but it results in BDC and EMC error, if you can help me in resolving that, that would be great – Yashpal Singla Nov 06 '12 at 06:58
@singla : Please check the above links, and try it out. I have accomplished these with pdfbox. If you are getting errors, let me know – Neeraj Nov 06 '12 at 08:14
@Singla :converting to image and extract images can be done directly using api. Download pdfbox and please check the link.. – Neeraj Nov 06 '12 at 08:17

Free open-source Java library to convert PDF -Image, PDF - HTML, PDF - extract images and text

3 Answers3