Fix handling UTF-8 characters in FpdfOutput and TcPdfOutput

[KamilPesek]

Fixes #131

[sprain]

Thank you!

The style guide states:

The following sans-serif fonts are permitted:
– Arial
– Frutiger
– Helvetica
– Liberation Sans

Therefore I don't think we should allow the font to be as flexible as this pr makes it.
Maybe we could instead provide a utf8-compatible font which is fixed within this library. What do you think?

[KamilPesek]

I have to say I am lost. I tried to played with FpdfOutput but with no success. I tried some test code like

( I tried to add custom Helvetica and Frutiger generated from ttf from here http://www.fpdf.org/makefont/make.php)

Font has been added, font is also set (I can see it's used) but it has still problem with ěčřšžýýáíé

EDIT: Ok, so I was on false lead. Problem was in this library. I added convertText() which I believe could fix it.

[KamilPesek]

@sprain how should I proceed?

[sprain]

I gave this PR a try.

I used the name Pia-Maria Mareč with FpdfOutput and got Pia-Maria Mareè in the pdf. So I doubt this is actually a fix for the issue.

[KamilPesek]

Hm, thinking what could be the difference:

I am on Windows 10, CRLF

$fpdf = new \Fpdf\Fpdf('P', 'mm', 'A4');
$fpdf->AddPage();

$output = new FpdfOutput($qrBill, 'de', $fpdf);
$output
    ->setPrintable(false)
    ->getPaymentPart();

@$qrPdfTempPath = tempnam('tmp', 'test.pdf');
$fpdf->Output($qrPdfTempPath, 'F');

header('Content-type: application/pdf');
header('Content-Disposition: inline; filename="' . $qrPdfTempPath . '"');
header('Content-Transfer-Encoding: binary');
header('Accept-Ranges: bytes');
echo file_get_contents($qrPdfTempPath);

[sprain]

Looking around it seems that utf8 and FPDF is basically a big pain. Because of stuff like this I use html and convert it with PrinceXML into pdfs in my projects. It just works :)

Maybe one could try tFPDF instead?
https://github.com/Setasign/tFPDF

Also, maybe @Franco2911 has an idea, who provided the original FpdfOutput for this library?

[Franco2911]

Hello,
I use utf8_decode() to use utf8 characters.
My textes cames from database.

a little example maybe can help.

$this->Cell(0,0, utf8_decode($address->getName()));

[KamilPesek]

I use utf8_decode()

That's what I had to replace to make it work. The question is why it works differently for me. I also tried it on our Linux server and it also works there.

[sprain]

Looks like we're kinda stuck here 🤔
New ideas are welcome.

[Franco2911]

I made many and many tests before a good result.

Your entire flow must be UTF8 compliant.

You have to use utf8 decode on each rows(cell and multicell)

[sprain]

Closing this due to missing activity. Feel free to re-open if there is a need.