fillable PDFs

got it to work by using a Raspberry Pi as the middle system. It was already running as a Node.js/Express print server to print labels from Ninox, so I added another endpoint for filling out PDF files. Ninox makes a request to the raspberry pi, the pi make the request to pdfOtter, PDFOtter responds with the PDF file which the raspberry pi places in it's webserver directory. The raspberry pi responds to Ninox with the url to the PDF.

I needed to add a delay between the response from the raspberry pi because the PDF takes a few seconds to build and download. I wasn't sure what to do, so I used a dialog() as a delay.

This is what the ninox button code looks like, minus the stuff to create the URL:

let response := http("GET", url, {}, null);
if response.error then
alert(text(response.error))
else
dialog("Need a Delay!", "Are you Ready?", ["yes"]);
Text := text(response.result);
importFile(myID, text(response.result))
end

This is what the Express endpoint looks like on the raspberry pi side:

// Express route for incoming requests for a list of all inputs
app.get('/fill', function (req, res) {

exec('curl -X POST https://www.pdfotter.com/api/v1/pdf_templates/tem_c214E5Y8Ycc3pM/fill \
-u test_YOUR-API-KEY: \
-d "data[nax]"='+ req.query.nax +' \
-d "data[apx]"='+ req.query.apx +' \
-d "data[rx]"='+ req.query.rx +' \
-d "data[name]"='+ req.query.name +' \
-d "data[phone]"='+ req.query.phone +' \
-d "data[dob]"='+ req.query.dob +'\
-d "data[homeAddress]"='+ req.query.homeAddress +' \
-d "data[homeState]"='+ req.query.homeState +' \
-d "data[homeCity]"='+ req.query.homeCity +' \
-d "data[homeZip]"='+ req.query.homeZip +' \
-d "data[mailAddress]"='+ req.query.mailAddress +' \
-d "data[mailState]"='+ req.query.mailState +' \
-d "data[mailCity]"='+ req.query.City +' \
-d "data[mailZip]"='+ req.query.mailZip +'\
-d "data[emailAddress]"='+ req.query.emailAddress +' \
-d "data[ssn]"='+ req.query.ssn +' \
-d "data[nfName]"='+ req.query.nfName +'\
-d "data[nfx]"='+ req.query.nfx +' \
-d "data[hName]"='+ req.query.hName +' \
-d "data[hx]"='+ req.query.hx +' \
-d "data[cFee]"='+ req.query.cFee +' \
-d "data[pgx]"='+ req.query.pgx +' \
-d "data[nop]"='+ req.query.nop +' \
-d "data[cbcFee]"='+ req.query.cbcFee +' \
-d "data[family1]"='+ req.query.family1 +' \
-d "data[family2]"='+ req.query.family2 +' \
-d "data[totalFee]"='+ req.query.totalFee +'\
-d "data[clStreet]"='+ req.query.clStreet +'\
-d "data[clState]"='+ req.query.clState +' \
-d "data[clCity]"='+ req.query.clCity +' \
-d "data[clCounty]"='+ req.query.clCounty +' \
-d "data[clZip]"='+ req.query.clZip +' \
-d "data[indoorx]"='+ req.query.indoorx +' \
-d "data[indoorDescription]"='+ req.query.indoorDescription +' \
-d "data[outdoorx]"='+ req.query.outdoorx +' \
-d "data[outdoorDescription]"='+ req.query.outdoorDescription +'\
-d "data[clpoName]"='+ req.query.clpoName +'\
-d "data[clpoAddress]"='+ req.query.clpoAddress +' \
-d "data[clpoZip]"='+ req.query.clpoZip +' \
-d "data[clpoState]"='+ req.query.clpoCity +' \
-d "data[clpoCounty]"='+ req.query.clpoCounty+'\
-d "data[clpoCity]"='+ req.query.clpoCity +' \
-d "data[cgName]"='+ req.query.cgName +' \
-d "data[date]"='+ req.query.date +'\
> /var/www/pdf/result.pdf', function (err,stdout,stderr){});

console.log('printed http://192.168.1.31/pdf/result.pdf');
res.status(200).send('http://192.168.1.31/pdf/result.pdf' );
});

oh, this only with the ninox apps when they are on the LAN with the raspi pi.

Thanks. I changed some things around on the Node side so the response doesn't come until the curl command is funished. It seems to be working pretty well now.

I changed 

var child =require('child_process').exec(‘MY_LONG_CURL_COMMAND')
child.stdout.pipe(process.stdout)
child.on('exit',function(){

console.log('Finished curl');
res.status(200).send('http://192.168.1.31/pdf/result.pdf' );

})

This is the initial tutorial I based my raspberry pi Node.js stuff off of:

http://www.robert-drummond.com/2013/05/08/how-to-build-a-restful-web-api-on-a-raspberry-pi-in-javascript-2/

I found a less complicated way to do what I wanted to do. I realized Ninox can treat a PDF as an image file.

I split the PDF into single pages with PDFSAM, then create a print layout in Ninox that uses the PDF as a static image for the background.

If I need to I merge the single PDF pages back together I use PDFSAM again.