zestsoftware / collective.sendaspdf Goto Github PK

Hello,

wkhtmltopdf has moved the downloads provider from SourceForge to gna.org. As mentioned in the linked message, all earlier downloads have been removed from SourceForge and you are strongly advised to migrate to the new URLs. The following files in your repository will need to be updated:

• buildout.d/osx.cfg
• buildout.d/base.cfg
• .travis.yml

It is recommended to either subscribe to the mailing list or Twitter feed to receive updates on new releases and further changes.

Thanks,
wkhtmltopdf maintainer

This has worked fine for a long time as far as I know. But today downloading an https page as pdf misses all styles. Somehow the styles (plus presumably any images) cannot be loaded.

It may be due to this stackoverflow question: http://stackoverflow.com/questions/26529863/how-do-i-configure-wkhtmltopdf-to-request-https-pages-by-url-not-ip-on-windows-8

Possibly a recent openssl update on the server causes this, but that is a guess.

It does not matter whether wkhtmltopdf itself is running on the server that hosts the https website, or locally on my laptop. This gets executed:

parts/wkhtmltopdf/wkhtmltopdf --disable-javascript --toc-header-text Inhoudsopgave --margin-right 10 --margin-left 10 --margin-bottom 10 --margin-top 10 --print-media-type file:///tempfile.html /tmp/tempfile.pdf

Any https links in tempfile.html cannot be loaded. When I replace https with http, it works just fine:

cat tempfile.html | sed "s/https/http/g" > newfile.html

I tried just now with running wkhtmltopdf on https://google.com, but that looks fine. So there may be something wrongly configured on the server.

Workaround that I did on the server to fix it: I edited transforms/wk.py and in the html_to_pdf function I added a line source = source.replace('https://', 'http://'). (Yes, I did that by editing the collective.sendaspdf distribution on the server. I am bad.)

There's any motivation about the reason that the JavaScript source use:

• compression="none"
• cookable="False"

?

I'm trying to run collective.sendaspdf on a server with wkhtmltopdf 0.12.1 installed.

This fails because of the option "--toc-header-text", whose syntax has apparently changed. The exact same setup works fine on another server with wkhtmltopdf 0.9.9

I'm able to get it to work by simply dropping 'toc-header-text' from the variable 'valued_options' in transforms/wk.py. However, this doesn't seem very elegant and given this error string:

Please check that you use a version compatible with your OS and the version is 0.9.

should collective.sendaspdf be considered "locked" to wkhtml 0.9.*? Are there plans to support later versions like 0.12+ at some point?

When installing collective.sendaspdf at the Plone products page, jquery.pyproxy is also listed, but is not installed.

collective.sendaspdf version 2.9

Thanks.

This works:

(Pdb) extract_from_url('http://.../pgTimePrint/OpenBareForm?ref=562a&to_date=2013-12-31', context_url)
('OpenBareForm', {'ref': '562a', 'to_date': '2013-12-31'})

This does not:

(Pdb) extract_from_url('http://.../pgTimePrint/OpenBareForm?ref=562a&to_date=2013-12-31 00:00', context_url)
(None, None)

In the second case, the to_date parameter includes a time (00:00). The parsing chokes on the :.
It also chokes on urlencoded parameters, e.g.

(Pdb) extract_from_url('http://.../pgTimePrint/OpenBareForm?ref=562a&to_date=2013-12-31+00%3A00', context_url)
(None, None)

Perhaps extract_from_url could be changed to use urlparse.parse_qs instead of regex-parsing the URL.