On Tue, Feb 19, 2013 at 7:10 AM, Martin Weissenboeck <[email protected]> wrote:
> Hi Mariano,
>
> thank you very much for your detailed answer.
>
> I know, you are not the creator of the library, but I think you do a very
> important job. Maybe my question could help others and me to use this
> program.
>
> I have played a little bit more with the examples.
> http://pyfpdf.googlecode.com/files/web2py.app.fpdf_20130218.w2p
>
> The error messages have gone - thank you!
>
> But I have some more questions / comments. I have made some changes and here
> are the result. By the way: I am using the last web2py version from trunk.
>
> (1) in def get_me_a_pdf():
> line 24:
>
> text = 'blä€ß Χανιά' * 10000
>
> or
>
> text = u'blä€ß Χανιά' * 10000
>
>
> The result: the “ά” is not shown
> That is interesting, because the accent is important in the Greek language
> and it could change the meaning of a word..“ά” is replaced by a black
> square. Maybe the reportlab font does not know the Unicode character U+03AC?
Sorry, I cannot help you with reportlab, it is a complex software piece.
In fact, that's why I don't use reportlab and decided to go with pyfpdf :-)
> (2) in def get_me_a_pyfpdf():
> line 48:
> text = "blä€ß Χανιά" * 10000
> Result: “blä€ß Χανιά”
>
> line 48:
>
> text = u"blä€ß Χανιά" * 10000
>
>
> Error:
>
> <type 'exceptions.UnicodeEncodeError'> 'latin-1' codec can't encode
> character u'\u20ac' in position 3: ordinal not in range(256)
>
>
> Ok, I see utf-8 or unicode characters are not allowed.
Yes, in pyfpdf, if you don't use a unicode font, non utf8 characters
are not allowed.
> (3) in def get_me_a_unicode_pyfpdf():
>
> line 70:
>
> text = u"blä€ß Χανιά" * 10000
>
>
> Perfect, the pdf-file shows “blä€ß Χανιά”
Please note that for some languages / characters, you will need
special fonts (for example, chinese, thai, etc.)
> (4) in def listing():
>
> line 142:
> response.title = "web2py sample listing with äö€ß Χανιά"
>
> listing.html is ok
> listing.pdf:
>
> “web2py sample listing with äö€ß Χανιά”
>
>
> I think there is a problem. It seems, that HTML-texts with no-latin-1
> characters (like “äö€ß” or “Χανιά") or HTML-equivants (like “α”) do
> not work.
Yes, you need to set a unicode font like in get_me_a_unicode_pyfpdf()
to get non-latin1 support.
It's a bit tricky, you could do the following:
# add the font to the pdf document:
pdf.add_font('DejaVu', '', os.path.join(request.folder, "private",
"font", 'DejaVuSans.ttf'), uni=True)
# create the basic html parser:
h2p = HTML2FPDF(pdf, image_map)
# update the supported font list:
h2p.font_list = list(h2p.font_list) + ['DejaVu']
# convert basic html to pdf:
h2p.feed(text)
Then, in your html text, you have to specify the font tag:
<font face="DejaVu"> .... </font>
> (5) def invoice():
>
> line 290
> f["company_name"] = u"Sample Company Χανιά "
> Greek letters disappeared: “Sample Company”
>
> line 299
> f['customer_name'] = u"Sample Client äöü߀"
>
> “€” disappered: “Sample Client äöüß”
Idem to previous issues, you need to add a explicit ttf font with
unicode support
Change the font face in import_csv so it is used in pdf_element
instead the buil-in latin1 fonts:
v['pdf_template_id'] = template_id
v['font'] = 'DejaVu' # add this!
db.pdf_element.insert(**v)
Then, in invoice() set the font:
f = Template(format="A4",
elements = elements,
title="Sample Invoice", author="Sample Company",
subject="Sample Customer", keywords="Electronic TAX Invoice")
# add this to set the font:
f.pdf.add_font('DejaVu', '', os.path.join(request.folder,
"private", "font", 'DejaVuSans.ttf'), uni=True)
detail = "Lorem ipsum dolor sit amet, consectetur. " * 5
>
> (6) Some typos:
>
> line 90:
>
> "example to generate a pdf based on basic html (use http:/.../report.pdf
> download the pdf)"
>
>
> line 218:
>
> "example to generate an invoice based on a generic pdf template"
>
Thanks, could you send me the final file fixed, or fill an issue?
(some text are truncated in the email)
>
> (7) From the example on http://code.google.com/p/pyfpdf/wiki/Unicode
>
> Arabic: مرحبا العالم
>
> Hebrew: שלום עולם
>
>
> right-to-left scripts are printed in the wrong direction:
>
> שלום עולם is written as םלוע םולש
>
> مرحبا العالم is written as ملاعلا ابحرم
>
With reportlab how are they printed?
As far as I know, pyfpdf outputs the same pdf as the php versions.
> I speak neither Arabic nor Hebrew, but it seems wrong to me.
Me neither, but if you need to change the direction of the text, you could see:
http://stackoverflow.com/questions/12233949/changing-page-direction-to-right-to-left-in-fpdf
The advantage of pyfpdf is that we can reuse a lot of scripts made for
the php versions ;-)
Best Regards,
Mariano Reingart
http://www.sistemasagiles.com.ar
http://reingart.blogspot.com
--
---
You received this message because you are subscribed to the Google Groups
"web2py-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to [email protected].
For more options, visit https://groups.google.com/groups/opt_out.
