RussoMA
asked on
pdf to html?
Running coldfusion mx7, plus php extensions installed.
What I want to do is dynamically create an html document from a pdf file. I am curious if there is a public API for this type of conversion. I know that google employs this technology in Gmail for their "view as html" links. I have a stand alone copy of the pdftohtml opensource project that utilizes ghostwriter, but i want to eliminate that step. I have > 250 ever changing pdf files that i want to be able to display in my site upon request in the stead of using a embedded pdf (there is a reason, the print button points to a print version of the pdf, whereas the 'web' version has more content [links, reference materials, etc] that i want to just display as html and not give the user the ability to print.
So the question is, are there any server-side API's that can be used for conversion of pdf's to html on-demand? examples of use?
most people want to go the other way from html -> pdf (and i love the cfdocument feature for this)... i just want to go the opposite way.
What I want to do is dynamically create an html document from a pdf file. I am curious if there is a public API for this type of conversion. I know that google employs this technology in Gmail for their "view as html" links. I have a stand alone copy of the pdftohtml opensource project that utilizes ghostwriter, but i want to eliminate that step. I have > 250 ever changing pdf files that i want to be able to display in my site upon request in the stead of using a embedded pdf (there is a reason, the print button points to a print version of the pdf, whereas the 'web' version has more content [links, reference materials, etc] that i want to just display as html and not give the user the ability to print.
So the question is, are there any server-side API's that can be used for conversion of pdf's to html on-demand? examples of use?
most people want to go the other way from html -> pdf (and i love the cfdocument feature for this)... i just want to go the opposite way.
ASKER CERTIFIED SOLUTION
membership
This solution is only available to members.
To access this solution, you must be a member of Experts Exchange.
ASKER
i like the idea of using cfexecute, here's the problem i am having with it, i send this:
<cfexecute name="c:\pdf\pdftohtml.exe
the html file is created, but it does not invoke the ghostscript (http://sourceforge.net/projects/ghostscript) to create the image background.
pdftohtml 0.36 (http://pdftohtml.sourceforge.net/) uses ghostscript to create the image used for the background of the single page using the -c tag (complex, instead of text only).
it seems to not invoke that part of pdftohtml, because no png file is created.
is there something i need to do to declare the path of the ghostscript executable so that CF can call upon that while executing pdftohtml?
i logged the output and it looks the same as it does when it uses ghostscript successfullt (no error messages) just
Page-1
which it sends for each page that it converts.
suggestions?
<cfexecute name="c:\pdf\pdftohtml.exe
the html file is created, but it does not invoke the ghostscript (http://sourceforge.net/projects/ghostscript) to create the image background.
pdftohtml 0.36 (http://pdftohtml.sourceforge.net/) uses ghostscript to create the image used for the background of the single page using the -c tag (complex, instead of text only).
it seems to not invoke that part of pdftohtml, because no png file is created.
is there something i need to do to declare the path of the ghostscript executable so that CF can call upon that while executing pdftohtml?
i logged the output and it looks the same as it does when it uses ghostscript successfullt (no error messages) just
Page-1
which it sends for each page that it converts.
suggestions?
SOLUTION
membership
This solution is only available to members.
To access this solution, you must be a member of Experts Exchange.
ASKER
yes, i can run it from comand line and it creates a png file as the background image
Ok,
You will probably find it's to do with paths etc.
Try creating a batch file .bat that takes you can run from the command line with no paramaters and will work.
Then try running this from CFEXECUTE
the benefit of this, is you can change directories etc, within the batch file.
You will probably find it's to do with paths etc.
Try creating a batch file .bat that takes you can run from the command line with no paramaters and will work.
Then try running this from CFEXECUTE
the benefit of this, is you can change directories etc, within the batch file.
ASKER
ok, i have gotten it to work -
i had not formally installed the ghostscript on that machine (the command line question from plucka lead me down this road of realization), I had only copied the files to the directory on the webserver, and not had the install register whatever it is that now made it work. [black box that i dont care to dive into today]
but on some of the files, it doesnt show the png and instead shows a background image missing icon (IE is the browser used to test). if i refresh the page seconds later, the image is shown.
is there a way to make cfexecute pause before displaying the rest of the page? or is this a server issue that may not be recognizing the file creations fast enough to acknowledge the new image is indeed ready to be served on a cfinclude?
i had not formally installed the ghostscript on that machine (the command line question from plucka lead me down this road of realization), I had only copied the files to the directory on the webserver, and not had the install register whatever it is that now made it work. [black box that i dont care to dive into today]
but on some of the files, it doesnt show the png and instead shows a background image missing icon (IE is the browser used to test). if i refresh the page seconds later, the image is shown.
is there a way to make cfexecute pause before displaying the rest of the page? or is this a server issue that may not be recognizing the file creations fast enough to acknowledge the new image is indeed ready to be served on a cfinclude?
This will make coldfusion sleep for a bit
<cfset thread = CreateObject("java", "java.lang.Thread") />
<cfset thread.sleep(5000) />
put these two lines after the <CFEXECUTE and before the display
This is 5000 miliseconds, thus 5 seconds, so you can change it to whatever you like.
CF MX or later.
<cfset thread = CreateObject("java", "java.lang.Thread") />
<cfset thread.sleep(5000) />
put these two lines after the <CFEXECUTE and before the display
This is 5000 miliseconds, thus 5 seconds, so you can change it to whatever you like.
CF MX or later.
ASKER
that wraps this up, thanks to the help, i got my answer and i think this question will help out others that seek to convert pdf to html on the fly. i am splitting the points because sidfishes lead me to use cfexecute and plucka helped me to get it running as well as the pause (thanks)
http://www.adobe.com/products/acrobat/access_onlinetools.html
in combination with cfhttp
(you'd have to check whether adobe's licensing would allow this tho...)