David really hit it on the head here. Because this is really a big FAQ, a couple months ago I took a lot of time to research Windows and Linux solutions for creating PDFs, particularly from MV. I was very surprised and disappointed to find a triangle of mutually exclusive usage factors: Ease of Use, Low Cost, and Quality. The closer you get to one or two of those, the further you move from the third. This is especially true in Linux.
The biggest problems, as David points out, are in pagination and fonting. Decent business documents don't get overly creative with fonts, but you still need control over your page breaks. Also, common reports require careful placement of text on the page. To handle this, I've looked at various HTML to PDF solutions. The most common one in use is HtmlDoc which is opensource freeware but the supported version has a cost. This is very good software, but it does have its flaws related to the above. It is limited to HTML 3.2 so formatting must be done with tables, no CSS or HTML 4.0 with DHTML syntax. It has some issues even with v3.2 tables, so you can't just get your web designer to provide document templates for you. All of this said, I have do have clients using HtmlDoc freeware in Linux, we generate HTML files from the application and have a script that executes HtmlDoc to convert to PDF. They then use NebulaMail (plug) to mail the PDF as required. But because of the issues with HtmlDoc I am hesitent to recommend it to clients that tell me they need high quality PDF. My latest adventure into this area has been to send complex HTML 4.0 with nice formatting, fonts, etc into MS Word. Using the MS Word DOM which is very familiar with HTML and XML structures, I tweek the document for pagination via scripting (eliminating the line creep situation that David describes), then save the document as PostScript. I then execute GhostScript which is a perfect tool for converting PostScript to PDF. All of this is free and it works very well. While the environment is not elegant in the least this does create a solution right in the middle of that triangle, or if you will, it actually puts the pointer in all three corners. Because of the lack of elegance in a Linux environment, I'm now re-evaluating my personal sense of quality and may settle on using HtmlDoc for all Linux-based sites. However, for a site that wants the quality, wants versatility in the PDF version, and especially where it's required to have index, links, security, author data, or other PDF-specific hooks, I'll continue to use MS Word and PostScript as an intermediary format to define the PDF document. Note also that PDF itself is only a text document internally. Once you convert to PDF you can actually modify the document, though unless you have a clue about the code it's probably best to not get into there. I only mention this because some developers might want to create a nice PDF with Adobe Acrobat or some other tool, and then use that as a template, simply populating it with data from their MV app. Not a universal solution, but an option nonetheless. Yes, there are a lot of free PDF programs out there, but regardless of the cost I've found that they all need to be evaluated carefully - some free ones are excellent and some that cost hundreds of dollars are pathetic. If you prefer to off-load all of this hassle with scripts, installing weird packages, and hunting for mods, etc, there are some for-fee offerings in our market that might help. Names on request. Tony Nebula R&D [EMAIL PROTECTED] David Wolverton dwolvert-at-flash.net |U2UG| wrote: > The problem with software995 (and others) is the lack of > control for paging and fonts sizing -- I have docs that > work perfectly going to PCL, landscape, line printer > font, but none of the 'free' pdf creators will honor the > spacing required, or give you a way to really accurately > control it. The PCL commands sent via our application are > seen as extra 'data' by the PDF products as they expect > to be fed WINDOWS Printer insructions... > > So - if it's formatted for use via PCL, you may or may > not get what you want - we ended up with 'line creep' on > most all of the 'free' pdf products - or with the > rightmost data cut off on really wide/full page utilized > items! > > You'll have to try several potentially before you find > one that will work for you if you're trying to format > data to match paper forms - if you just need a raw 'data > dump' to PDF, then most any will perform - do a google on > Free PDF - there's a lot out there! > > David W. > ------- > u2-users mailing list > [email protected] > To unsubscribe please visit http://listserver.u2ug.org/ ------- u2-users mailing list [email protected] To unsubscribe please visit http://listserver.u2ug.org/
