#!/usr/bin/env python3

# Convert HTML pages to PDF.

# Copyright 2017 by Akkana Peck -- share and enjoy under the GPL v2 or later.

# If PdfFileMerger fails,

# To convert all the output files into a single multi-page PDF, use this:

# gs -dNOPAUSE -sDEVICE=pdfwrite -sOUTPUTFILE=multipage.pdf -dBATCH file*.pdf

import sys

import urllib.request

from PyQt5.QtCore import QUrl, QSizeF, QMarginsF

from PyQt5.QtWidgets import QApplication

from PyQt5.QtWebEngineWidgets import QWebEnginePage, QWebEngineProfile

from PyQt5.QtGui import QPageLayout, QPageSize

from PyPDF2 import PdfFileMerger

from io import BytesIO

class HeadlessPagePrinter():

"""Load a single page and convert it to PDF.

When the PDF is ready, call the pdf_ready() callback.

"""

def __init__(self, app, pdf_ready, x_res=1024):

self.profile = QWebEngineProfile()

self.webpage = QWebEnginePage(self.profile, app)

self.webpage.loadFinished.connect(self.load_finished)

self.x_res = x_res

self.pdf_ready = pdf_ready

def load_url(self, url):

"""pdf_ready is a callback called with a QString

of the PDF bytes 'printed'.

"""

# Sadly, QWebEnginePage has no way to apply custom stylesheets

# or otherwise change style.

# So we have to intercept the load and insert our print style

# directly into the page, rather than just loading a URL.

# self.webpage.load(QUrl(url))

# Use urllib because it handles file:// requests, unlike requests.

r = urllib.request.urlopen(url)

html = r.read().decode(r.headers.get_content_charset() or 'utf-8')

html = html.replace('</head>', '''

<style type="text.css">

@media print {

body {

width: %dpx;

height: 768px;

position: absolute; left: 0px; top: 0px;

border: 0; margin: 0; padding: 0;

overflow: hidden;

}

#points li { visibility: visible !important; }

#notes, #nextdiv { display: none; }

}

</style>

</head>''' % self.x_res)

print(html)

self.webpage.setHtml(html, QUrl('file:///home/akkana/Docs/Preso/moon-peec/'))

def set_html(self, html):

self.webpage.setHtml(html)

def load_finished(self, ok):

print("Load finished ... trying to print")

# contentsSize is always 0x0. Not clear if it's supposed to be useful.

# print("Contents size is %f x %f" %

# (self.webpage.contentsSize().width(),

# self.webpage.contentsSize().height()))

margins = QMarginsF(0, 0, 0, 0)

# With QSizeF(271, 203.3), it makes a 1024x768 PDF; but it

# prints the whole 1366 web page so it will fit in a 1024x768 box,

# even if the content extends outside.

# But an @media print CSS rule to keep everything at 1024 fixes that.

# With QSizeF(362, 203.3) it can print a full 1366x768.

if self.x_res > 1024:

xmm = 362

else:

xmm = 271

ymm = 203.3

print("Page size of", xmm, ymm)

layout = QPageLayout(QPageSize(QSizeF(xmm, ymm),

QPageSize.Millimeter),

QPageLayout.Portrait, margins,

QPageLayout.Millimeter, margins)

self.webpage.printToPdf(self.pdf_ready, layout)

class SlidePrinter:

def __init__(self, app, outfile, x_res=1024):

self.renderer = HeadlessPagePrinter(app, self.pdf_ready, x_res=x_res)

# PdfFileMerger always fails with "Multiple definitions in dictionary"

# unless you add strict=False.

# https://github.com/mstamy2/PyPDF2/issues/244

self.merger = PdfFileMerger(strict=False)

self.outfile = outfile

def add_pages(self, urllist):

self.urllist = urllist

self.urlindex = 0

self.next_page()

def next_page(self):

if self.urlindex >= len(self.urllist):

print("All done")

sys.exit(0)

print("\n=== Rendering url", self.urlindex, ":",

self.urllist[self.urlindex])

self.renderer.load_url(self.urllist[self.urlindex])

def pdf_ready(self, pdf_bytes):

print("finished rendering to", len(pdf_bytes), "bytes")

if pdf_bytes:

filename = "/tmp/page%d.pdf" % self.urlindex

with open(filename, "wb") as fp:

fp.write(pdf_bytes)

fp.close()

bytes_fp = BytesIO(pdf_bytes.data())

self.merger.append(bytes_fp)

else:

print("No output for", self.urllist[self.urlindex-1])

self.urlindex += 1

if self.urlindex >= len(self.urllist):

print("\nTrying to save the merged file")

with open(self.outfile, "wb") as fp:

self.merger.write(fp)

print("Wrote", self.outfile)

sys.exit(0)

else:

self.next_page()

if __name__ == '__main__':

if len(sys.argv) <= 1:

print("what pages?")

sys.exit(1)

# -1024 or -1366

if len(sys.argv) > 2 and sys.argv[1].startswith('-'):

try:

x_res = int(sys.argv[1][1:])

args = sys.argv[2:]

except:

print("Can't parse flag", sys.argv[1])

sys.exit(1)

else:

x_res = 1024

args = sys.argv[1:]

if len(args) < 1:

print("Usage: %s [-1024|-1366] file.html file.html ..."

% os.basename(sys.argv[0]))

sys.exit(1)

app = QApplication(sys.argv)

slideprinter = SlidePrinter(app, "/tmp/outfile.pdf", x_res=x_res)

slideprinter.add_pages(args)

app.exec_()