# -*- coding: utf-8 -*-
# Copyright (C) 2004-2008 Tristan Seligmann and Jonathan Jacobs
# Copyright (C) 2012-2014 Bastian Kleineidam
# Copyright (C) 2015-2020 Tobias Gruetzmacher
import os
import time
from urllib.parse import quote as url_quote
import codecs
import json
import imagesize
from . import rss, util, configuration
from .output import out
# Maximum width or height to display an image in exported pages.
# Note that only the displayed size is adjusted, not the image itself.
MaxImageSize = (800, 800)
class EventHandler(object):
"""Base class for writing events to files. The currently defined events are
start(), comicDownloaded() and end()."""
def __init__(self, basepath, baseurl, allowdownscale):
"""Initialize base path and url."""
self.basepath = basepath
self.baseurl = baseurl or self.getBaseUrl()
self.allowdownscale = allowdownscale
def getBaseUrl(self):
'''Return a file: URL that probably points to the basedir.
This is used as a halfway sane default when the base URL is not
provided; not perfect, but should work in most cases.'''
components = util.splitpath(os.path.abspath(self.basepath))
url = '/'.join([url_quote(component, '') for component in components])
return 'file:///' + url + '/'
def getUrlFromFilename(self, filename):
"""Construct URL from filename."""
components = util.splitpath(util.getRelativePath(self.basepath, filename))
url = '/'.join([url_quote(component, '') for component in components])
return self.baseurl + url
def start(self):
"""Emit a start event. Should be overridden in subclass."""
pass
def comicDownloaded(self, comic, filename):
"""Emit a comic downloaded event. Should be overridden in subclass.
Parameters are:
comic: The ComicImage class calling this event
filename: The target filename
"""
pass
def comicPageLink(self, scraper, url, prevUrl):
"""Emit an event to inform the handler about links between comic pages.
Should be overridden in subclass. Parameters are:
scraper: The Scraper class calling this event
url: The current page url
prevUrl: The previous page url
"""
pass
def end(self):
"""Emit an end event. Should be overridden in subclass."""
pass
class RSSEventHandler(EventHandler):
"""Output in RSS format."""
name = 'rss'
def getFilename(self):
"""Return RSS filename."""
return os.path.abspath(os.path.join(self.basepath, 'dailydose.rss'))
def start(self):
"""Log start event."""
today = time.time()
yesterday = today - 86400
today = time.localtime(today)
yesterday = time.localtime(yesterday)
link = configuration.Url
self.rssfn = self.getFilename()
if os.path.exists(self.rssfn):
self.newfile = False
self.rss = rss.parseFeed(self.rssfn, yesterday)
else:
self.newfile = True
self.rss = rss.Feed('Daily Dosage', link,
'Comics for %s' % time.strftime('%Y/%m/%d', today))
def comicDownloaded(self, comic, filename):
"""Write RSS entry for downloaded comic."""
imageUrl = self.getUrlFromFilename(filename)
size = None
if self.allowdownscale:
size = getDimensionForImage(filename, MaxImageSize)
title = '%s - %s' % (comic.scraper.name, os.path.basename(filename))
pageUrl = comic.referrer
description = ''
if comic.text:
description += '
%s' % comic.text
description += '
View Comic Online' % pageUrl
args = (
title,
imageUrl,
description,
util.rfc822date(time.time())
)
if self.newfile:
self.newfile = False
self.rss.addItem(*args)
else:
self.rss.addItem(*args, append=False)
def end(self):
"""Write RSS data to file."""
self.rss.write(self.rssfn)
def getDimensionForImage(filename, maxsize):
"""Return scaled image size in (width, height) format.
The scaling preserves the aspect ratio."""
try:
origsize = imagesize.get(filename)
except Exception as e:
out.warn("Could not get image size of {}: {}".format(os.path.basename(filename), e))
return None
width, height = origsize
if width > maxsize[0]:
height = max(round(height * maxsize[0] / width), 1)
width = round(maxsize[0])
if height > maxsize[1]:
width = max(round(width * maxsize[1] / height), 1)
height = round(maxsize[1])
if width < origsize[0] or height < origsize[1]:
out.info("Downscaled display size from %s to %s" % (origsize, (width, height)))
return (width, height)
class HtmlEventHandler(EventHandler):
"""Output in HTML format."""
name = 'html'
encoding = 'utf-8'
def fnFromDate(self, date):
"""Get filename from date."""
fn = time.strftime('comics-%Y%m%d', date)
fn = os.path.join(self.basepath, 'html', fn + ".html")
fn = os.path.abspath(fn)
return fn
def addNavLinks(self):
if self.yesterdayUrl:
self.html.write(u'Previous Day | ' % self.yesterdayUrl)
self.html.write(u'Next Day\n' % self.tomorrowUrl)
def start(self):
"""Start HTML output."""
today = time.time()
yesterday = today - 86400
tomorrow = today + 86400
today = time.localtime(today)
yesterday = time.localtime(yesterday)
tomorrow = time.localtime(tomorrow)
fn = self.fnFromDate(today)
if os.path.exists(fn):
out.warn('HTML output file %r already exists' % fn)
out.warn('the page link of previous run will skip this file')
out.warn('try to generate HTML output only once per day')
fn = util.getNonexistingFile(fn)
d = os.path.dirname(fn)
if not os.path.isdir(d):
os.makedirs(d)
try:
fn_yesterday = self.fnFromDate(yesterday)
fn_yesterday = util.getExistingFile(fn_yesterday)
self.yesterdayUrl = self.getUrlFromFilename(fn_yesterday)
except ValueError:
self.yesterdayUrl = None
self.tomorrowUrl = self.getUrlFromFilename(self.fnFromDate(tomorrow))
self.html = codecs.open(fn, 'w', self.encoding)
self.html.write(u'''