Selfhosted

I had rolled my own PDF generator with an entire layout engine before I found this. I was disappointed but immediately threw it all away in favor of that, because it's so much easier. It's just soooo damn much easier to use than keeping adding features to a custom engine. It handles CSS print stylesheets pretty well too, so you get to do all your page breaks and everything and the output is pretty clean and usable.

I think it does generate mildly heavier/complex PDFs but at least it looks great.

PHP PDF libraries 10 years ago were awful.

Oh, 1st of all, I have to make a small correction, I use an odt as base document.

You run this in the directory with the odt file and a temp directory t

cd t
unzip ../${document_template}.odt
sed -i "s/ReplaceThis/${var1}/g" content.xml
sed -i "s/ReplaceThisToo/${var2}/g" content.xml
zip -r ../${document_target}.odt .
cd ../
libreoffice --headless --convert-to pdf:writer_pdf_Export ${document_target}.odt

The file content.xml is the actual text of the document, the rest is just the dressing (formatting). What you need to do is make sure the text you want to replace (ReplaceThis in the example) is unique apart from the spots where you want the text to be replaced. Check the content.xml to be sure, but be warned, the content.xml is a 2 line file with line 1 being the xml header. (it's a terrible layout of the xml) Use an file editor you know well to find the words you want to replace. Also, be warned, correcting small typos in the odt results in garbage in the content.xml. You need to replace the whole word as 1 action. (Editing history has it's uses, but is an issue here)

Instead of converting it to file you can print it as well. (DDG result)

libreoffice --headless --print-to-file --printer-name the_printer_name --outdir /tmp ${document_target}.odt
lpr -U username /tmp/filename.ps -P the_printer_name
rm /tmp/filename.ps

Ok, will post it tomorrow.

Oh, 1st of all, I have to make a small correction, I use an odt as base document.

You run this in the directory with the odt file and a temp directory t

cd t
unzip ../${document_template}.odt
sed -i "s/ReplaceThis/${var1}/g" content.xml
sed -i "s/ReplaceThisToo/${var2}/g" content.xml
zip -r ../${document_target}.odt .
cd ../
libreoffice --headless --convert-to pdf:writer_pdf_Export ${document_target}.odt

The file content.xml is the actual text of the document, the rest is just the dressing (formatting). What you need to do is make sure the text you want to replace (ReplaceThis in the example) is unique apart from the spots where you want the text to be replaced. Check the content.xml to be sure, but be warned, the content.xml is a 2 line file with line 1 being the xml header. (it's a terrible layout of the xml) Use an file editor you know well to find the words you want to replace. Also, be warned, correcting small typos in the odt results in garbage in the content.xml. You need to replace the whole word as 1 action. (Editing history has it's uses, but is an issue here)

Instead of converting it to file you can print it as well. (DDG result)

libreoffice --headless --print-to-file --printer-name the_printer_name --outdir /tmp ${document_target}.odt
lpr -U username /tmp/filename.ps -P the_printer_name
rm /tmp/filename.ps