🇬🇧 / 🇫🇷 Need some help with WeasyPrint? You’re in the right place to ask :)
In HTML previews, these tags work, but it seems like the PDF images are always bigger
HTML Preview ^
PDF ^
With:
img { max-width: 100%; max-height: 250px; width: auto; height: auto;}
Oh, I think I know what’s happening
Is the image in a table?
You can use a smaller
max-height if you think that the second photo is too tall.
a customer of mine is interested in sending out PDFs with some basic forms (text box, checkboxes) from my application. we don't use weasyprint in that part of the application for HTML->PDF conversion yet, but there are some long term plans in that direction, pending support, of course. so the question is: would this support materialize with throwing some money at the problem and what kind of money would that be? :)
The project use headless chrominium + skia patch to render any html page to svg
There is some additionnal information in the first issue on the github project.
If you are seeing slow generation of reports or significantly larger output files make sure you are running 57.1 and not 57.0. I haven't seen a reference in the changelog to this (but I could easily have missed it) but we had experienced both massively longer generation time and file sizes that resolved after moving to 57.1 from 57.0. (12 page report with 22 images and an SVG logo normally generates in ~20 seconds and file size about 3.5MB was taking ~3 minutes and 20 seconds and was generating a file about 32MB)
just wondering if there is a way for weasyprint to output an image file thats just fit the html raw size, without limiting it with print page canvas (e.g. A4)?
@AlexanderNeilson is right, you can change the page size in the
@page block. But if you’re looking for a dynamic page size, it’s not implemented (you can take a look at Kozea/WeasyPrint#193)
Hello! I'm trying to generate a variant pdf/a-2b with weasyprint 57.1 and I cannot solve one final validation failure (I validate the pdf with verapdf). Fault description is "Except for annotation dictionaries whose Subtype value is Popup, all annotation dictionaries shall contain the F key." and the errorMessage is "A dictionary of Link annotation does not contain F key"... Any ideas of what I can do to fix this?
Could you please share the HTML document (you can send a mail to contact@courtbouillon.org if you don’t want it to be public)
24 replies
we are currently investigating an issue where some reports we generate with ~45 images in them totaling between about 15 and 18 MB on disk sometimes produce a final PDF of ~6MB and sometimes ~77MB
I haven't yet been able to confirm this is always a common factor but so far I can only see that one file's original extension was JPG and the others was JPEG however I do not expect that to cause a difference in behaviour as it appears weasyprint's image functions decode data from a URI as bytes.io from the string so the original file extension is potentially not even known
So while this was correct here
These ones further down showed the original taken dimensions of the image before it was resized. My working theory is that Pillow internally makes the larger "image plane" and then when the smaller image is read in a sort of "crop box" operates around it and therefore lots of space is reserved inside for what turns out to be a very small image
Could you please share an example in a bug report so that we can investigate and fix this problem?
11 replies
katex isn't working
I think the fonts are not working