from HTML to PDF

Anon

I have one ASP(or PHP) page to generate one HTML page with different length. I'd like to provide user with PDF file format downloading. I searched a lot of solutions, with little success. (My server is Windows 2000, IIS)

One easy solution is using Adobe FDF tools. But it need the fixed length fields and fix contents in each page. My Asp page will generate different sized memo fields which FDF won't work out.

Another one is HTMLDOC tools (www.easysw.com). It basically "print" the input file to an output file, which is very sufficient for my problem. But it requires an input file in HTML, and an output file to the web server. This won't work, because in this way I need to generate the HTML file (not existing in the server) from the asp and save it to some unique temp name, then run the shell exe, save the output to the file with another unique temp name. This will cause huge trouble of the server storage and security problems.

They did offer some sample in Perl to get the input from the memory stream, which I can't make it work in windows machine.

So my troubles are:
1. ASP sending different sized HTML stream to the browser. After user click some button on that page, I need find a way to get all the HTML source (in memory only) to feed to some COM or EXE.
2. Get the PDF output from the COM, and then sending the PDF to user. (Memory only).
3. User using PDF plug in to view the PDF stream, and discard it or save it using PDF reader button in the browser.
4. The contents of the HTML is dynamic, the size of the contents, or the size of each field (or section) will change. So only the PDF "print" approach will work. The generating PDF on the fly approach won't work, because it is hard to figure out when to close the current page and create a new pdf page. How many lines of html or text should be put in that page(taken consideration of different font and style of the text and different graphic images.).

All suggestions are greatly appreciated.
Thanks.

maxpup979

try PDFLib

phillip

hi bruce,

i am currently facing the same problem - creating complex pdfs out of part template and part dynamic content.
i was looking into using the pdf lib functions to create the pdfs, but also checked out HTMLDOC.

i would be curious to hear your experience and what you finally decided to do.

thanks a lot,
phil!

(mailed and posted)

Anon

Hi Phil:

I couldn't find the right solution to this "simple" task. We end up only provide the html report at this phase.

Becuase we are using COM+ and win 2000 server, I was thinking to develop a COM. By calling a COM function, that function can easily control the memory, file directory, etc. At least we can call HTMLDOC in that COM. The COM can be developed by VB.

Anyway, good luck to your project.

Cheers,

Bruce

phillip

hi bruce!

But the trouble for us is we don't know what is the size for the fields. [...]
yes, i understood that from you first post. you have memo fields that can be very different in size. the following elements have to adapt to the size of your memos.

We need a "Auto grow" function in PDF or FDF [...]
maybe you can base your memo field output on the function pdf_show_boxed(). that would be my idea. plus write a function that keeps track of the pixel position you are currently at and place all following elements accordingly.

pdf_show_boxed outputs text into a rectancle-shaped area. check the length of your string and from that dynamically determine the box-size you will need. someone already contributed a function for dynamic size boxes in the manual for pdf_show_boxed.

if you have text that leaps over a page, i'd size the textbox 'til the end of the page. the remaining text (that didn't fit into the box you specified) is pdf_show_boxed's return value! neat, ey? you can continue drawing boxes on the next page.

i'm not saying i have this working, but after some studying of php's pdf functions this is the way i'd go. the pdf's i had to create are less complicated so we'll probably use HTMLDOC. generate static HTML and let HTMLDOC do the conversion. i don't see a big security problem - when exec()'ing HTMLDOC we will not pass any PHP variables which can be manipulated externally.
or am i overlooking something? i'm not an expert on that. also i am not sure about the performance issues with shell calls.

i'd be very grateful to hear about your experiences if you try what i described above. as i said already, i think it will work, but can't guarantee since i didn't yet try it myself.

For the pdflib, I don't know what kind of layout functions are availabe.
i think there are all funcions available to layout a pdf page nicely -
drawing lines/arcs/paths/geometric shapes, outputting text in different fonts/styles/kerning, placing images etc.
however i think it'd be some work to get used to the pixel positioning and keep track of the positioning of the objects in relation to eachother. but i
believe you could write some custom functions that make your life a little easier.

kind regards,
phillip!

ps: i'm talking about PDFlib here. i read ClibPDF's functions are similar, but there might be details that work better for our problem.(?)

(mailed and posted)

Anon

Is there a PHP function to save on a file a HTML stream ?