pdf2htmlEX Windows Version

pdf2htmlEX Â Win32 static version, compiled with mxe under Ubuntu.

In one word,Â pdf2htmlEX Â is a tool thatÂ Convert PDF to HTML without losing text or format, bases onÂ poppler,Â FontForge ~~, andÂ jQuery~~.

Modification

support windows temp directory, tested under windows 64-bit system(tried all the example offered by the author, all successful).
change the share folder name to data
can automatically find the data folder, by default, Â the data folder Â is under the same folder withÂ pdf2htmlEX.exe

Introduction

pdf2htmlEX renders PDF files in HTML, utilizing modern Web technologies. It aims to provide an accurate rendering, while keeping optimized for Web display.

pdf2htmlEX is best for text-based PDF files, for example scientific papers with complicated formulas and figures. Text, fonts and formats are nativelyÂ preserved in HTML such that you can still search and copy. The generated HTML file is static, with optional features powered by JavaScript.

Features

Precise and native text in HTML

Flexible Output

Moderate Size

More PDF stuffs that you love: links, outlines & printing

LICENSE

GPLv3 with additional terms Â for most parts, MIT License forÂ share/*(in Windows version, I change the name of share to data).

ChangeLog

Download

pdf2htmlEX-win32-0.14.6-upx-with-poppler-data.zip (pdf2htmlEx.exe is packed with UPX )

pdf2htmlEX-win32-0.14.6-with-poppler-data.zip

pdf2htmlEX-win32-0.13.6

pdf2htmlEX v0.12-win32-static

pdf2htmlEX v0.11 (bases on v0.11 release version)
pdf2htmlEX-v0.11-win32-static(without poppler encoding data)

pdf2htmlEX-v0.11-win32-static-with-poppler-data
~~pdf2htmlEX-v0.11-win32-static(without poppler encoding data)~~

pdf2htmlEX-v0.11-win32-static-with-poppler-data

~~pdf2htmlEX-v1.0-win32-static~~

~~pdf2htmlEX-v0.9-win32-static~~

P.S.

The author isÂ Lu Wang, Â the original project and source code isÂ pdf2htmlEX,

I just did some modification and compiled for windows,Â the modified source is hereÂ .

128 Responses to "pdf2htmlEX Windows Version"

admin says:

September 16, 2014 at 6:43 pm

I am not sure which version you use, I just tested with pdf2htmlEX-0.12(not release because of fontforge bug), it works fine, so I share with you, it has some wired debug info, just ignore it.
hellocity says:

December 1, 2014 at 4:27 pm

hi, I downloaded the version 0.12 in Windows 7, most PDF files converted fine, thanks!
But, some special Chinese PDF files, such as this one: http://www.bimlib.cn/App/Upload/doc/2014-11-17/5469b39bbe6c5.pdf, the output html file was unreadable in browser.
I find that all unreadable text has the same class named .ff3, which set the font-family:ff3. When I clear this style set, the text display normally.
Would you please give some suggestion to help me solve it?
admin says:

December 11, 2014 at 4:08 pm

try using option –embed-external-font 0
if still have issue, please open a issue on https://github.com/coolwanglu/pdf2htmlEX/issues
tester says:

February 10, 2015 at 10:42 pm

Hello,
I’m facing an issue.
If I open individual html page, it doesn’t show fonts correctly but if I open main html file, all looks fine.
Is it possible to load individual html pages in side iframe.
Also is it possible to scale the html page inside iframe, responsively?

Thank u!
Terry says:

February 12, 2015 at 12:10 am

I’m trying to use pdf2htmlEX on Windows 7 and am having some problems:

I downloaded https://soft.rubypdf.com/download/pdf2htmlex/pdf2htmlEX-0.12-win32-static-with-poppler-data.zip, unzipped it into a folder and put that folder-name in my PATH.

Now if I type the command

pdf2htmlEX myfile.pdf

it seems to run to completion but it gives the error:

“Error: Cannot open the manifest file”

Windows 7 also reports that “pdf2htmlEX.exe has stopped working”. In fact this message shows every time I run the program, even if I just show the help:

pdf2htmlEX –help

What am I missing?

Thanks,
Terry
admin says:

February 12, 2015 at 11:19 am

please try go to the folder with pdf2htmlEx.exe, or you need special –data-dir

maybe you can write a batch file pdf2htmlex1.bat with the following data
pdf2htmlEx.exe –data-dir the_data_dir_of_pdf2htmlex %*

then you can use pdf2htmlex1 myfile.pdf
admin says:

February 12, 2015 at 11:22 am

no idea what issues you met, maybe you should give me the output html.
tester says:

February 12, 2015 at 3:02 pm

Hello,

When I convert the PDF, it converts properly.
In output folder, if I open main html file (i.e. if my PDF is abc.pdf then html files are, abc.html, page_1.html, page_2.html etc.)
So if I open abc.html, all looks fine.
But if I open page_1.html or page_2.html, output is not proper. It show first images and below that text without formatting.
I checked that, main file has all the js and css files added to it.
Can it be possible to add this code in all the files and run files individually?

I need to create a single responsive container to hold these files.
If I have converted multiple files, then is it possible to load all those files from single main file?

Thank you!
Terry says:

February 13, 2015 at 1:43 am

Thanks for the help with the –data-dir switch.

However, Windows 7 is still giving me the message
â€œpdf2htmlEX.exe has stopped workingâ€

It seems to have created the html file, but the ‘crash’ makes it hard to use in a script.

I can reproduce the problem by going to the folder with pdf2htmlEX.exe in it and simply typing:
> pdf2htmlEX.exe
or
> pdf2htmlEX.exe -v

I’m running Windows7 Pro, x64

Thanks,
Terry
Terry says:

February 20, 2015 at 12:38 am

Hi. I have an example (bug?) where the html output of pdf2htmlEX renders correctly in IE and FireFox, but not in Chrome. In Chrome, it sometimes writes lines over top of each other.

You can see an example in my dropbox at:
https://www.dropbox.com/sh/n4e2ztlbara45ha/AAB4BOvaMImgZInZf9mGzi26a?dl=0

Look in the HTML file, pages marked vii, 22, 34, 36, 49, 59, etc.

pdf2htmlEX is version 0.12
the –version switch reports:
poppler 0.26.3
libfontforge 20140516
cairo 1.12.14

Is there a way around this?

Thanks,
Terry
Helidium says:

April 8, 2015 at 7:43 pm

Yes, it seems this is a bug in webkit rendering engine. The ‘incoming’ branch has a fix, but not the master branch. Is it possible to compile the incoming branch for webkit browsers to display text properly.

Thank you,
Marko
jens says:

June 1, 2015 at 3:57 am

Hi, is there a chance to get a new version? The last release is v0.13.6. It might solve a problem I have. The programm is just crashing without an error message. I have a pdf with 132 pages. If I convert the first 131 pages or the last 3 everything is ok. If I convert all or the last 4 pages pdf2html crashes. I also have artefacts if I convert the last 2 pages which I don’t have, if I convert the last 3 pages. If you want to test this you find the pdf here:
http://liedtke.it/pdf2htmlEX/AM.pdf

Is there a description how I can build a new windows version myself?
zjzengdongyang says:

June 7, 2015 at 9:43 am

Could you update it?
v0.13.6
2015.04.29

* Use micro versions
* Do not support Poppler < 0.25.0 any more
* ENABLE_SVG is enabled by default
* Improved DrawingTracer
* Workarounds for chrome/webkit
Joey says:

June 25, 2015 at 3:35 am

I’m using 0.13.6 (have also tried 0.12 static w/ popplar) and I’m getting an APPCRASH (details at bottom). The PDF in question does not appear to be any different that the other 500+ files I tried, but crashes every time. The command I’m using to launch is:

pdf2htmlEX –zoom 1.3 –dest-dir html try/15snii15_week15_2015.pdf

I do get a series of warning about certain “Lookup ‘mark’ Mark Positioning” having offsets bigger than 65535, but I get those warnings for all of the files I am processing, and the results appear fine.

Is there a parameter or something else that I am missing?

Here are the details from the Windows crash popup:

Problem Event Name: APPCRASH
Application Name: pdf2htmlEX.exe
Application Version: 0.0.0.0
Application Timestamp: 5583e887
Fault Module Name: pdf2htmlEX.exe
Fault Module Version: 0.0.0.0
Fault Module Timestamp: 5583e887
Exception Code: c0000005
Exception Offset: 002baca5
OS Version: 6.3.9600.2.0.0.272.7
Locale ID: 1033
Additional Information 1: 5861
Additional Information 2: 5861822e1919d7c014bbb064c64908b2
Additional Information 3: a10f
Additional Information 4: a10ff7d2bb2516fdc753f9c34fc3b069
admin says:

June 25, 2015 at 9:20 am

please feel free to offer the PDF sample.
jens says:

July 12, 2015 at 12:34 am

Here are 2 similar PDF files which just stop. In an unix environment I get a perfect HTML file

http://liedtke.it/pdf2htmlEX/AM.pdf
http://liedtke.it/pdf2htmlEX/AM2.pdf
admin says:

July 16, 2015 at 11:43 am

works fine, maybe you need poppler-data
samchild says:

July 30, 2015 at 9:23 am

Hello,
does it not support the the absolute path with the input PDF file ?

Example:
C:\Users\hxc\Desktop\download\pdf2htmlEX-win32-0.14.6-with-poppler-data\pdf2htmlEX.exe “C:\Users\hxc\Desktop\download\pdf reference.pdf” –page-filename %d.html -f1 -l2 –split-pages 1 –embed cfijo –dest-dir “C:\Users\hxc\Desktop\download\pdf2htmlEX-win32-0.14.6-with-poppler-data\t5”

and will get error:
Error: Cannot open C:\Users\hxc\Desktop\download\pdf2htmlEX-win32-0.14.6-with-poppler-data\t5/C:\Users\hxc\Desktop\download\pdf reference.css for writing

I tried using 0.14.6 and all the other version in windows 7(32 bit), it’s the same error.
admin says:

July 30, 2015 at 9:44 am

It is a bug, will fix soon, but now, you can replace “C:\Users\hxc\Desktop\download\pdf reference.pdf” with “C:/Users/hxc/Desktop/download/pdf reference.pdf”
admin says:

July 30, 2015 at 1:03 pm

fixed, please download it again.
samchild says:

July 30, 2015 at 1:46 pm

Hi, thanks for your help.

and i got another issue:
it says:
Preprocessing: 2/2
Internal Error: Attempt to output 2147483647 into a 16-bit field. It will be truncated and the file may not be useful.

Internal Error: Lookup sub table, ‘mark’ Mark Positioning in Arabic lookup 1 subtable in ‘mark’ Mark Positioning in Arabic lookup 1, contains no data.
Internal Error: Lookup sub table, ‘mark’ Mark Positioning in Arabic lookup 2 subtable in ‘mark’ Mark Positioning in Arabic lookup 2, contains no data.
…..
…..
Internal Error: File Offset wrong for ttf table (name-data), -1 expected 570
Cannot save font to C:/Users/hxc/Desktop/download/pdf2htmlEX-win32-0.14.6-with-p
oppler-data/t5-tmp/pdf2htmlEX-a05900/__tmp_font1.ttf

Is this my PDF file wrong?
admin says:

July 30, 2015 at 1:55 pm

please offer the PDF to test.
samchild says:

July 31, 2015 at 9:34 am

I figured it out, the input pdf cannot be located in the C driver. I don’t know why.
The sample pdf can be converted in the D driver ,but not in the C driver.
All the parameter is the same except the input pdf file.
lee says:

December 9, 2015 at 12:23 pm

workaround for font names in bad encodings.
asd_u2 says:

December 20, 2015 at 10:37 pm

1-update version with new poppler-0.39
2- arabic text copy not allow after convert and cant edit the text
nagnath says:

January 11, 2016 at 8:50 pm

i installed both fontforge & poppler in mxe but still its showing package not found.. 🙁
Richard says:

February 1, 2016 at 8:20 am

How does this integrate with Fontforge? Does one have to install the windows version on the PC or something? Or does it come with the downloaded zip?
admin says:

February 1, 2016 at 8:58 am

this pdf2htmlEX is static version, no need download fontforge
ramakrishna says:

February 5, 2016 at 11:06 pm

When I am trying to convert a pdf this is the error I am getting. Could you let me know what is wrong with it.

pdf2htmlEX.exe –fi
t-width 1024 –embed CFIJo –clean-tmp 1 –process-outline 0 –dest-dir HTMLOutP
ut “test.pdf”
Preprocessing: 30/30
Lookup ‘mark’ Mark Positioning in Latin lookup 4 has an
offset bigger than 65535 bytes. This means
FontForge must use an extension lookup to output it.
Not all applications support extension lookups.

This is the test pdf that I was trying to convert.

http://www.2shared.com/document/B1vADMwM/test.html
Richard says:

February 21, 2016 at 10:38 am

Hm, so I built an MXE environment on an ubuntu virtual machine on vmware; can’t seem to get it to work though. How does one settle the dependencies—I don’t think the cross environment has all the dependencies needed to build pdf2htmlex right?
admin says:

February 21, 2016 at 3:53 pm

it still need fontforge, openjpeg, and recompile poppler again(depends on openjpeg)
Richard says:

February 21, 2016 at 8:34 pm

You wouldn’t happen to have the makefiles for those, would you? Because I’m getting a bunch of errors when trying to use the cmake inside mxe. Right now I’ve got an error saying “Error: no SVG support found in Cairo”
Richard says:

February 22, 2016 at 6:24 am

Actually, I think the better question would be what’s the easiest way to install a package INTO mxe (for cross-compiling use, satisfying dependencies you know). I know how to cross-compile I believe, that part I think is simple, but I’m a little confused on installing extra stuff into the cross environment (making the cross environment “better/”more compatible). How do I do it?
admin says:

February 22, 2016 at 10:37 pm

there is no an easy way to install a package into mxe, I check dependencies in a very dirty and stupid way(check dependencies from homepage). I still have no idea how to easily provide a better solution to compile it.
John Shotsky says:

March 7, 2016 at 9:03 pm

New use here.
It works pretty well, but messes up at least one page. I’ve tried numerous options, and no options, but page 49/50 gets messed up each time. Look for the phrase 1 cup balsamic vinegar, and you will see it is in the wrong place. Win 10-64. Most of this document seems to be ok – but it puts data in the wrong place. It also seems to put recipe titles in the wrong location, at the end instead of at the top. Also, it would be good to have an option to remove headers, footers, page numbers, etc. When I go from pdf to an ebook (epub), the ‘old’ page numbers are meaningless.
PDF file is shared below.
https://drive.google.com/open?id=0B-XDz04UkUaBU2FkQmRVOUMzUHc
Thanks!
John Shotsky says:

March 8, 2016 at 7:11 pm

Another problem that I’ve discovered, but don’t know how to solve. It seems that when opened in a browser, only images of the pages are seen. But opening the same file in Sigil, an ebook editor, shows BOTH the image of each page, AND the html, between the images. This may be due to the options I am using, or not using, but it is not clear what each option actually means. For example, exactly what is ‘fallback’? Is there a glossary somewhere that explains what these options actually mean, as opposed to simply using the name and defining it’s default? I need to be able to figure out why I would need a given option. FWIW, all I really want is a single html file, with nothing embedded, and images stored in an \images folder beneath the output folder. But half the images I see are actually images of all the pages. I don’t want any of those, I just want the actual images. Why is it providing an image of each page? (This is a different file from my previous shared file.)
Thanks!
pudding says:

April 14, 2016 at 8:19 am

adminä½ å¥½
æˆ‘ä¸‹è½½çš„ç‰ˆæœ¬0.12 ç‰ˆæœ¬0.14.6 è¿è¡ŒçŽ¯å¢ƒæ˜¯win7 64ä½
åœ¨cmdä¸æ‰§è¡Œ
D:\pdf2htmlEX-win32-0.14.6-upx-with-poppler-data\pdf2htmlEX.exe D:\pdf2htmlEX-win32-0.14.6-upx-with-poppler-data\test\test_output\1-page.pdf

cmdçª—å£å‡ºçŽ°
Preprocessingï¼š1/1
Workingï¼š1/1
è¿™æ˜¯æˆåŠŸäº†ï¼Œè¿˜æ˜¯æœ‰ä»€ä¹ˆé—®é¢˜
æˆ‘æ²¡æœ‰æ‰¾åˆ°ç”Ÿæˆçš„html
admin says:

April 14, 2016 at 1:36 pm

æ‰§è¡ŒæˆåŠŸäº†ï¼Œ å‘½ä»¤è¡Œå½“å‰ç›®å½•ä¸å¯»æ‰¾
pudding says:

April 14, 2016 at 5:32 pm

åœ¨cmdä¸æ‰§è¡Œ
D:\pdf2htmlEX-win32-0.14.6-upx-with-poppler-data\pdf2htmlEX.exe D:\pdf2htmlEX-win32-0.14.6-upx-with-poppler-data\test\test_output\1-page.pdf
cmdçª—å£å‡ºçŽ°
Preprocessingï¼š1/1
Workingï¼š1/1
è¿™æ˜¯æˆåŠŸäº†ï¼Œè¿˜æ˜¯æœ‰ä»€ä¹ˆé—®é¢˜
æˆ‘æ²¡æœ‰æ‰¾åˆ°ç”Ÿæˆçš„html

ç”Ÿæˆçš„htmlæ–‡ä»¶åœ¨å“ªé‡Œå•Šï¼Œæˆ‘æ²¡æ‰¾åˆ°ï¼Ÿï¼Ÿï¼Ÿï¼Ÿ
è°¢è°¢è€å¸ˆ
yuriko says:

June 21, 2016 at 6:37 pm

pdf2htmlEX-0.12-win32-static-with-poppler-data\pdf2htmlEX.exe D:/1/1-page
.pdf
I/O Error: Couldn’t open ‘nameToUnicode’ file ‘D:\tywork\2016\13-\\swfhtml5\pdf2htmlEX-0.12-win32-static-with-poppler-data\share\poppler\nameToUnicode\B
ulgarian’
I/O Error: Couldn’t open ‘nameToUnicode’ file ‘D:\tywork\2016\13-\\swfhtml5\pdf2htmlEX-0.12-win32-static-with-poppler-data\share\poppler\nameToUnicode\G
reek’
I/O Error: Couldn’t open ‘nameToUnicode’ file ‘D:\tywork\2016\13-\\swfhtml5\pdf2htmlEX-0.12-win32-static-with-poppler-data\share\poppler\nameToUnicode\T
hai’
Preprocessing: 1/1
Working: 1/1
Error: Cannot open the manifest file

è¿™ä¸ªé”™è¯¯ Error: Cannot open the manifest file æ€Žä¹ˆåŠžï¼Ÿæ€Žä¹ˆè§£å†³ï¼Ÿ
admin says:

June 21, 2016 at 10:02 pm

please run pdf2htmlex under the path where it locals or set it to path
travis young says:

July 12, 2016 at 7:31 pm

Could you please post how the program should be run? I cannot get 46 working on PATH, in the native directory, or, you know, how I want to actually run it 😉
Procyon82 says:

August 5, 2016 at 10:49 pm

How do you use pdf2htmlEX on multiple files or on a folder that contains pdf files?

I am able to convert single files just fine, but I obviously don’t want to run a command 100 times for 100 files.

I couldn’t find anything in the documentation and something like ‘*.pdf’ doesn’t work.

Thanks!
vytautas says:

August 17, 2016 at 8:12 am

hi i can’t run this program bunch of lines does not run for me please help i need this
Joe Grovel says:

September 5, 2016 at 5:12 pm

Hey, would you mind recompiling this with a recent version of poppler, cairo and the other libraries?
I’m getting much better results on my Macbook (brew-version) and suppose it’s because the libraries are much fresher. I’m really not able to compile this myself – this is far beyond my abilities. My entire workflow is bound to a PC and the switch to my private notebook for this one step means we cannot really deploy this solution.
B. Tran says:

September 5, 2016 at 9:06 pm

Hi,
When I tried to download ‘pdf2htmlEX-win32-0.14.6-upx-with-poppler-data.zip’, my windows 10 complained the zip file contained and stop me from downloading (Chrome).

I was able downloaded ‘pdf2htmlEX-win32-0.13.6’, but when I execute it I got this error message:

Error: Cannot open the manifest file

Any idea,

Thanks
B. Tran says:

September 5, 2016 at 9:08 pm

I mean the zip file contained Malware.
admin says:

September 5, 2016 at 9:56 pm

Have you tried https://soft.rubypdf.com/download/pdf2htmlex/pdf2htmlEX-win32-0.14.6-with-poppler-data.zip, without compressed by UPX(maybe a mistake from Win10).
admin says:

September 5, 2016 at 10:00 pm

sorry, I am some busy recently, and you can try to compile it with the help of mxe.cc.
Thomas says:

October 9, 2016 at 12:47 am

Thank you for this software, the conversions are very clean compared to others.

I am having a problem with some Japanese files though, all from Shimano. None of the text in the table gets converted, the output HTML shows a blank table:
http://fservice.shimano.co.jp/part/pdf/03238.pdf

I tried several versions of the converter, including the latest version. Also tried many different settings but didn’t have any luck.

Also, Microsoft Security Essentials reports that version 0.14.6 has the “Trojan:Win32/Spursint.A!cl” malware in the following two files:
pdf2htmlEX.js
pdf2htmlEX.min.js

This seems likely to be a false positive but perhaps something can be done?

Modification

Introduction

Features

LICENSE

ChangeLog

Download

Leave a Reply