From 04b13e003d6af0de21e6c59e411ffee5b97b6134 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?J=C3=B6rg=20Frings-F=C3=BCrst?= Date: Sun, 1 Oct 2017 18:50:17 +0200 Subject: New upstream version 2.0.4 --- README.md | 89 ++++++++++------ config_template.py | 74 ++++++------- mwc.py | 298 +++++++++++++++-------------------------------------- mwcfeedserver.py | 31 +++--- mwctools.py | 239 ++++++++++++++++++++++++++++++++++++++++++ 5 files changed, 433 insertions(+), 298 deletions(-) mode change 100644 => 100755 config_template.py create mode 100755 mwctools.py diff --git a/README.md b/README.md index 8e78da6..69718e7 100644 --- a/README.md +++ b/README.md @@ -19,20 +19,23 @@ Some examples: sites = [ - {'shortname': 'mywebsite1', - 'uri': 'http://www.mywebsite1.com/info', - 'contentcss': 'div'}, - - {'shortname': 'mywebsite2', - 'uri': 'http://www.mywebsite2.com/info', - 'contentxpath': '//*[contains(concat(\' \', normalize-space(@class), \' \'), \' news-list-container \')]', - 'titlexpath': '//title'}, - - {'shortname': 'mywebsite3', - 'uri': 'http://www.mywebsite3.com/info', - 'type': 'text', - 'contentregex': 'Version\"\:\d*\.\d*', - 'user-agent': 'Mozilla/5.0 (X11; Fedora; Linux x86_64; rv:49.0) Gecko/20100101 Firefox/49.0'} + {'name': 'example-css', + 'parsers': [uri(uri='https://github.com/mtill', contenttype='html'), + css(contentcss='div') + ] + }, + + {'name': 'example-xpath', + 'parsers': [uri(uri='https://example-webpage.com/test', contenttype='html'), + xpath(contentxpath='//div[contains(concat(\' \', normalize-space(@class), \' \'), \' package-version-header \')]') + ] + }, + + {'name': 'my-script', + 'parsers': [command(command='/home/user/script.sh', contenttype='text'), + regex(contentregex='^.*$') + ] + } ] @@ -40,31 +43,55 @@ sites = [ * parameters: - * shortname - short name of the entry, used as an identifier when sending email notifications + * name + name of the entry, used as an identifier when sending email notifications + * receiver (optional) + Overrides global receiver specification. + + * parameters for the URL receiver: + * uri - URI of the website; If the scheme of the uri is 'cmd://', the string is interpreted as a command and the standard output (stdout) is parsed. - * type (optional; default: 'html') + URI of the website + * contenttype (optional; default: 'html') content type, e.g., 'xml'/'html'/'text'. - * contentxpath / titlexpath (optional) - XPath expression for the content/title sections to extract. If you prefer, you could use contentcss/titlecss instead. - * contentcss / titlecss (optional) - CSS expression for the content/title sections to extract. This is ignored if there is a corresponding XPath definition. - * contentregex / titleregex (optional) - Regular expression. If XPath/CSS selector is defined, the regular expression is applied afterwards. - * encoding (optional; default: 'utf-8') + * enc (optional; default: 'utf-8') Character encoding of the website, e.g., 'utf-8' or 'iso-8859-1'. - * splitregex (optional) - only works if type is set to 'text'; defines that content should be split to chunks based on the defined regex expression. - * receiver (optional) - Overrides global receiver specification. - * user-agent (optional) + * userAgent (optional) Defines the user agent string, e.g., - 'user-agent': 'Mozilla/5.0 (X11; Fedora; Linux x86_64; rv:49.0) Gecko/20100101 Firefox/49.0' + 'userAgent': 'Mozilla/5.0 (X11; Fedora; Linux x86_64; rv:49.0) Gecko/20100101 Firefox/49.0' * accept (optional) Defines the accept string, e.g., 'accept': 'text/html,application/xhtml+xml,application/xml;q=0.9,*/*;q=0.8' + * parameters for the Command receiver + + * command + the command + * contenttype (optional; default: 'text') + content type, e.g., 'xml'/'html'/'text'. + * enc (optional; default: 'utf-8') + Character encoding of the website, e.g., 'utf-8' or 'iso-8859-1'. + + * parameters for the XPath parser: + + * contentxpath + XPath expression for the content sections to extract + * titlexpath (optional) + XPath expression for the title sections to extract + + * parameters for the CSS parser: + + * contentcss + CSS expression for the content sections to extract + * titlecss (optional) + CSS expression for the title sections to extract + + * parameters for the RegEx parser: + + * contentregex + Regular expression for content parsing + * titleregex (optional) + Regular expression for title parsing * We collect some XPath/CSS snippets at this place: Snippet collection - please feel free to add your own definitions! diff --git a/config_template.py b/config_template.py old mode 100644 new mode 100755 index f394e52..02788bd --- a/config_template.py +++ b/config_template.py @@ -1,47 +1,49 @@ -import os.path +#!/usr/bin/env python3 +# -*- coding: utf-8 -*- -# Copyright: (2013-2014) Michael Till Beck +# Copyright: (2013-2017) Michael Till Beck # License: GPL-2.0+ -#We collect xpath snippets at this place: Snippet collection - please feel free to add your own definitions! + +# We collect xpath snippets at this place: +# Snippet collection +# Feel free to contribute! + + +from mwctools import URLReceiver as uri +from mwctools import CommandReceiver as command +from mwctools import XPathParser as xpath +from mwctools import CSSParser as css +from mwctools import RegExParser as regex +from mwctools import Content +from mwctools import Parser + sites = [ - {'shortname': 'mywebsite1', - 'uri': 'http://www.mywebsite1.com/info', - 'type': 'html', - 'titlexpath': '//h1', - 'contentxpath': '//div', - 'titleregex': '', - 'contentregex': '', - 'user-agent': 'Mozilla/5.0 (X11; Fedora; Linux x86_64; rv:49.0) Gecko/20100101 Firefox/49.0', - 'accept': 'text/html,application/xhtml+xml,application/xml;q=0.9,*/*;q=0.8' - 'encoding': 'utf-8'}, - - {'shortname': 'mywebsite2', - 'uri': 'http://www.mywebsite2.com/info', - 'type': 'html', - 'contentxpath': '//*[contains(concat(\' \', normalize-space(@class), \' \'), \' news-list-container \')]', - 'regex': '', - 'encoding': 'utf-8'}, - - {'shortname': 'mywebsite3', - 'uri': 'http://www.mywebsite3.com/info', - 'type': 'text', - 'contentxpath': '', - 'contentregex': 'Version\"\:\d*\.\d*', - 'encoding': 'utf-8'}, - - {'shortname': 'lscmd', - 'uri': 'cmd://ls -l /home/pi', - 'contentregex': '.*Desktop.*' - } + {'name': 'example-css', + 'parsers': [uri(uri='https://github.com/mtill', contenttype='html'), + css(contentcss='div') + ] + }, + + {'name': 'example-xpath', + 'parsers': [uri(uri='https://example-webpage.com/test', contenttype='html'), + xpath(contentxpath='//div[contains(concat(\' \', normalize-space(@class), \' \'), \' package-version-header \')]') + ] + }, + + {'name': 'my-script', + 'parsers': [command(command='/home/user/script.sh', contenttype='text'), + regex(contentregex='^.*$') + ] + } ] -subjectPostfix = 'A website has been updated!' +workingDirectory = '/path-to-data-dir/MailWebsiteChanges-data' -enableMailNotifications = True +enableMailNotifications = False maxMailsPerSession = -1 sender = 'me@mymail.com' smtphost = 'mysmtpprovider.com' @@ -51,9 +53,7 @@ smtpusername = sender smtppwd = 'mypassword' receiver = 'me2@mymail.com' -os.chdir('/var/cache/mwc') - -enableRSSFeed = True +enableRSSFeed = False rssfile = 'feed.xml' maxFeeds = 100 diff --git a/mwc.py b/mwc.py index 4df4799..6a48317 100755 --- a/mwc.py +++ b/mwc.py @@ -1,15 +1,11 @@ #!/usr/bin/env python3 # -*- coding: utf-8 -*- -# Copyright: (2013-2015) Michael Till Beck +# Copyright: (2013-2017) Michael Till Beck # License: GPL-2.0+ -import urllib.request, urllib.error, urllib.parse -import urllib.parse -from lxml import etree -from cssselect import GenericTranslator -import re import io +from lxml import etree import hashlib import smtplib @@ -22,8 +18,6 @@ import sys import getopt import traceback -import subprocess - import time from time import strftime import random @@ -32,164 +26,20 @@ import importlib config = None defaultEncoding = 'utf-8' -maxTitleLength = 150 # this is how an empty RSS feed looks like emptyfeed = """ MailWebsiteChanges Feed - https://github.com/Debianguru/MailWebsiteChanges + https://github.com/mtill/MailWebsiteChanges MailWebsiteChanges Feed """ -# Attributes in HTML files storing URI values. These values are automatically translated to absolute URIs. -uriAttributes = [['//img[@src]', 'src'], ['//a[@href]', 'href']] -cmdscheme = 'cmd://' - mailsession = None -# translates all relative URIs found in trees to absolute URIs -def toAbsoluteURIs(trees, baseuri): - for tree in trees: - if isinstance(tree, str): - continue - for uriAttribute in uriAttributes: - tags = tree.xpath(uriAttribute[0]) - for tag in tags: - if tag.attrib.get(uriAttribute[1]) != None: - if urllib.parse.urlparse(tag.attrib[uriAttribute[1]]).scheme == '': - tag.attrib[uriAttribute[1]] = urllib.parse.urljoin(baseuri, tag.attrib[uriAttribute[1]]) - - -def parseSite(site): - global defaultEncoding - file, content, titles, warning = None, None, None, None - - uri = site['uri'] - contenttype = site.get('type', 'html') - contentregex = site.get('contentregex', '') - titleregex = site.get('titleregex', '') - splitregex = site.get('splitregex', '') - enc = site.get('encoding', defaultEncoding) - - contentxpath = site.get('contentxpath', '') - if contentxpath == '' and site.get('contentcss', '') != '': - # CSS - contentxpath = GenericTranslator().css_to_xpath(site.get('contentcss')) - titlexpath = site.get('titlexpath', '') - if titlexpath == '' and site.get('titlecss', '') != '': - titlexpath = GenericTranslator().css_to_xpath(site.get('titlecss')) - - try: - - if uri.startswith(cmdscheme): - # run command and retrieve output - process = subprocess.Popen(uri[len(cmdscheme):], stdout=subprocess.PIPE, shell=True, close_fds=True) - file = process.stdout - else: - # open website - req = urllib.request.Request(uri) - if 'user-agent' in site: - req.add_header('User-Agent', site['user-agent']) - if 'accept' in site: - req.add_header('Accept', site['accept']) - file = urllib.request.urlopen(req) - - - if contenttype == 'text' or (contentxpath == '' and titlexpath == ''): - thefullcontent = file.read().decode(enc, errors='ignore') - contents = [thefullcontent] - if splitregex != '': - contents = thefullcontent.split(splitregex) - titles = [] - else: - baseuri = uri - if contenttype == 'html': - parser = etree.HTMLParser(encoding=enc) - else: - parser = etree.XMLParser(recover=True, encoding=enc) - - tree = etree.parse(file, parser) - - # xpath - contentresult = tree.xpath(contentxpath) if contentxpath else [] - titleresult = tree.xpath(titlexpath) if titlexpath else [] - - # translate relative URIs to absolute URIs - if contenttype == 'html': - basetaglist = tree.xpath('/html/head/base') - if len(basetaglist) != 0: - baseuri = basetaglist[0].attrib['href'] - if len(contentresult) != 0: - toAbsoluteURIs(contentresult, baseuri) - if len(titleresult) != 0: - toAbsoluteURIs(titleresult, baseuri) - - if contentxpath != '' and titlexpath != '' and len(contentresult) != len(titleresult): - warning = 'WARNING: number of title blocks (' + str(len(titleresult)) + ') does not match number of content blocks (' + str(len(contentresult)) + ')' - elif contentxpath and len(contentresult) == 0: - warning = 'WARNING: content selector became invalid!' - elif titlexpath and len(titleresult) == 0: - warning = 'WARNING: title selector became invalid!' - else: - if len(contentresult) == 0: - contentresult = titleresult - if len(titleresult) == 0: - titleresult = contentresult - - if isinstance(contentresult, str): - contents = [contentresult] - else: - contents = [etree.tostring(s, encoding=enc, pretty_print=True).decode(enc, errors='ignore') for s in contentresult] - if isinstance(titleresult, str): - titles = [getSubject(titleresult)] - else: - titles = [getSubject(etree.tostring(s, method='text', encoding=enc).decode(enc, errors='ignore')) for s in titleresult] - - except IOError as e: - warning = 'WARNING: could not open URL; maybe content was moved?\n\n' + str(e) - - if file is not None: - file.close() - - if uri.startswith(cmdscheme) and process.wait() != 0: - warning = 'WARNING: process terminated with an error' - - if warning: - return {'content': content, 'titles': titles, 'warning': warning} - - # parse regex - if contentregex: - contents = [x for y in [re.findall(r'' + contentregex, c, re.S) for c in contents] for x in y] - if titleregex: - titles = [x for y in [re.findall(r'' + titleregex, c, re.S) for c in titles] for x in y] - - if contentregex and titleregex and len(contents) != len(titles): - warning = 'WARNING: number of title blocks (' + str(len(titles)) + ') does not match number of content blocks (' + str(len(contents)) + ') after regex' - elif contentregex and len(contents) == 0: - warning = 'WARNING: content regex became invalid!' - elif titleregex and len(titles) == 0: - warning = 'WARNING: title regex became invalid!' - else: - if len(contents) == 0: - contents = titles - if len(titles) == 0: - titles = [getSubject(c) for c in contents] - - return {'contents': contents, 'titles': titles, 'warning': warning} - - -# returns a short subject line -def getSubject(textContent): - if textContent == None or textContent == '': - return config.subjectPostfix - textContent = re.sub(' +', ' ', re.sub('\s', ' ', textContent)).strip() - return (textContent[:maxTitleLength] + ' [..]') if len(textContent) > maxTitleLength else textContent - - # generates a new RSS feed item def genFeedItem(subject, content, link, change): feeditem = etree.Element('item') @@ -213,23 +63,26 @@ def genFeedItem(subject, content, link, change): # sends mail notification -def sendmail(receiver, subject, content, sendAsHtml, link): +def sendmail(receiver, subject, content, sendAsHtml, link, encoding=None): global mailsession, defaultEncoding + if encoding is None: + encoding = defaultEncoding + if sendAsHtml: baseurl = None - if link != None: + if link is not None: content = '

' + subject + '

\n' + content baseurl = urljoin(link, '/') - mail = MIMEText('' + subject + '' + ('' if baseurl else '') + '' + content + '', 'html', defaultEncoding) + mail = MIMEText('' + subject + '' + ('' if baseurl else '') + '' + content + '', 'html', encoding) else: - if link != None: + if link is not None: content = link + '\n\n' + content - mail = MIMEText(content, 'text', defaultEncoding) + mail = MIMEText(content, 'text', encoding) mail['From'] = config.sender mail['To'] = receiver - mail['Subject'] = Header(subject, defaultEncoding) + mail['Subject'] = Header(subject, encoding) # initialize session once, not each time this method gets called if mailsession is None: @@ -244,103 +97,114 @@ def sendmail(receiver, subject, content, sendAsHtml, link): # returns a list of all content that is stored locally for a specific site -def getStoredHashes(shortname): - +def getStoredHashes(name): result = [] - filename = shortname + ".txt" + filename = os.path.join(config.workingDirectory, name + ".txt") if os.path.exists(filename): - with open(filename, 'r') as file: - for line in file: + with open(filename, 'r') as thefile: + for line in thefile: result.append(line.rstrip()) return result # updates list of content that is stored locally for a specific site -def storeHashes(shortname, contentHashes): - - with open(shortname + '.txt', 'w') as file: +def storeHashes(name, contentHashes): + with open(os.path.join(config.workingDirectory, name + '.txt'), 'w') as thefile: for h in contentHashes: - file.write(h + "\n") + thefile.write(h + "\n") + + +def runParsers(parsers, contentList=None): + if contentList is None: + contentList = [] + + for parser in parsers: + contentList = parser.performAction(contentList) + + return contentList def pollWebsites(): global defaultEncoding # parse existing feed or create a new one + rssfile = config.rssfile + if not os.path.isabs(rssfile): + rssfile = os.path.join(config.workingDirectory, rssfile) + if config.enableRSSFeed: - if os.path.isfile(config.rssfile): - feedXML = etree.parse(config.rssfile) + if os.path.isfile(rssfile): + feedXML = etree.parse(rssfile) else: feedXML = etree.parse(io.StringIO(emptyfeed)) # start polling sites mailsSent = 0 for site in config.sites: - print('polling site [' + site['shortname'] + '] ...') - sessionHashes = [] - parseResult = parseSite(site) + print('polling site [' + site['name'] + '] ...') receiver = site.get('receiver', config.receiver) - # if something went wrong, notify the user - if parseResult['warning']: - subject = '[' + site['shortname'] + '] WARNING' - print('WARNING: ' + parseResult['warning']) + try: + contentList = runParsers(site['parsers']) + except Exception as e: + # if something went wrong, notify the user + subject = '[' + site['name'] + '] WARNING' + print('WARNING: ' + str(e)) if config.enableMailNotifications: if config.maxMailsPerSession == -1 or mailsSent < config.maxMailsPerSession: - sendmail(receiver, subject, parseResult['warning'], False, None) + sendmail(receiver=receiver, subject=subject, content=str(e), sendAsHtml=False, link=None) mailsSent = mailsSent + 1 if config.enableRSSFeed: - feedXML.xpath('//channel')[0].append(genFeedItem(subject, parseResult['warning'], site['uri'], 0)) - else: - # otherwise, check which parts of the site were updated - changes = 0 - fileHashes = getStoredHashes(site['shortname']) - i = 0 - for content in parseResult['contents']: - - contenthash = hashlib.md5(content.encode(defaultEncoding)).hexdigest() - if contenthash not in fileHashes: - if config.maxMailsPerSession == -1 or mailsSent < config.maxMailsPerSession: - changes += 1 - sessionHashes.append(contenthash) - - subject = '[' + site['shortname'] + '] ' + parseResult['titles'][i] - print(' ' + subject) - if config.enableMailNotifications and len(fileHashes) > 0: - sendmail(receiver, subject, content, (site.get('type', 'html') == 'html'), site['uri']) - mailsSent = mailsSent + 1 - - if config.enableRSSFeed: - feedXML.xpath('//channel')[0].append(genFeedItem(subject, content, site['uri'], changes)) - else: + feedXML.xpath('//channel')[0].append(genFeedItem(subject, str(e), "", 0)) + continue + + sessionHashes = [] + changedContents = [] + fileHashes = getStoredHashes(site['name']) + for content in contentList: + + contenthash = hashlib.md5(content.content.encode(content.encoding)).hexdigest() + if contenthash not in fileHashes: + if config.maxMailsPerSession == -1 or mailsSent < config.maxMailsPerSession: sessionHashes.append(contenthash) + changedContents.append(content) - i += 1 + subject = '[' + site['name'] + '] ' + content.title + print(' ' + subject) + if config.enableMailNotifications and len(fileHashes) > 0: + sendAsHtml = (content.contenttype == 'html') + sendmail(receiver=receiver, subject=subject, content=content.content, sendAsHtml=sendAsHtml, link=content.uri, encoding=content.encoding) + mailsSent = mailsSent + 1 + if config.enableRSSFeed: + feedXML.xpath('//channel')[0].append(genFeedItem(subject, content.content, content.uri, len(changedContents))) + else: + sessionHashes.append(contenthash) + + if 'postRun' in site: + runParsers(site['postRun'], changedContents) - if changes > 0: - storeHashes(site['shortname'], sessionHashes) - print(' ' + str(changes) + ' updates') + if len(changedContents) > 0: + storeHashes(site['name'], sessionHashes) + print(' ' + str(len(changedContents)) + ' updates') # store feed if config.enableRSSFeed: for o in feedXML.xpath('//channel/item[position() +# Copyright: (2013-2017) Michael Till Beck # License: GPL-2.0+ + import http.server import socketserver import importlib import sys import getopt + bind = 'localhost' port = 8000 configMod = 'config' try: - opts, args = getopt.getopt(sys.argv[1:], 'hc:b:p:', ['help', 'config=', 'bind=', 'port=']) + opts, args = getopt.getopt(sys.argv[1:], 'hc:b:p:', ['help', 'config=', 'bind=', 'port=']) except getopt.GetoptError: - print('Usage: FeedServer.py --config=config --port=8000') - sys.exit(1) + print('Usage: FeedServer.py --config=config --port=8000 --bind=localhost') + sys.exit(1) for opt, arg in opts: - if opt == '-h': - print('Usage: FeedServer.py --config=config --bind=localhost --port=8000') - exit() - elif opt in ('-c', '--config'): - configMod = arg - elif opt in ('-b', '--bind'): - bind = arg - elif opt in ('-p', '--port'): - port = int(arg) + if opt == '-h': + print('Usage: FeedServer.py --config=config --bind=localhost --port=8000') + exit() + elif opt in ('-c', '--config'): + configMod = arg + elif opt in ('-b', '--bind'): + bind = arg + elif opt in ('-p', '--port'): + port = int(arg) config = importlib.import_module(configMod) diff --git a/mwctools.py b/mwctools.py new file mode 100755 index 0000000..cefbbf0 --- /dev/null +++ b/mwctools.py @@ -0,0 +1,239 @@ +#!/usr/bin/env python3 +# -*- coding: utf-8 -*- + +# Copyright: (2013-2017) Michael Till Beck +# License: GPL-2.0+ + + +import urllib.request +import urllib.error +import urllib.parse +import subprocess + +from lxml import etree +from cssselect import GenericTranslator +import re + + +# Attributes in HTML files storing URI values. These values are automatically translated to absolute URIs. +uriAttributes = [['//img[@src]', 'src'], ['//a[@href]', 'href']] + +maxTitleLength = 150 + + +class Parser: + # input: [Content], output: [Content] + def performAction(self, contentList): + pass + + +class Receiver(Parser): + def __init__(self, uri): + self.uri = uri + + +class Content: + def __init__(self, uri, encoding, title, content, contenttype): + self.uri = uri + self.encoding = encoding + self.title = title + self.content = content + self.contenttype = contenttype + + +# returns a short subject line +def getSubject(textContent): + global maxTitleLength + + if textContent is None or len(textContent.strip()) == 0: + return 'Website has been updated' + textContent = re.sub(' +', ' ', re.sub('\s', ' ', textContent)).strip() + return (textContent[:maxTitleLength] + ' [..]') if len(textContent) > maxTitleLength else textContent + + +# translates all relative URIs found in trees to absolute URIs +def toAbsoluteURIs(trees, baseuri): + global uriAttributes + + for tree in trees: + if isinstance(tree, str): + continue + for uriAttribute in uriAttributes: + tags = tree.xpath(uriAttribute[0]) + for tag in tags: + if tag.attrib.get(uriAttribute[1]) is not None: + if urllib.parse.urlparse(tag.attrib[uriAttribute[1]]).scheme == '': + tag.attrib[uriAttribute[1]] = urllib.parse.urljoin(baseuri, tag.attrib[uriAttribute[1]]) + + +class URLReceiver(Receiver): + def __init__(self, uri, contenttype='html', encoding='utf-8', userAgent=None, accept=None): + super().__init__(uri) + self.contenttype = contenttype + self.encoding = encoding + self.userAgent = userAgent + self.accept = accept + + # input: [Content], output: [Content] + def performAction(self, contentList=None): + if contentList is None: + contentList = [] + + # open website + req = urllib.request.Request(self.uri) + if self.userAgent is not None: + req.add_header('User-Agent', self.userAgent) + if self.accept is not None: + req.add_header('Accept', self.accept) + + with urllib.request.urlopen(req) as thefile: + filecontent = thefile.read().decode(self.encoding, errors='ignore') + contentList.append(Content(uri=self.uri, encoding=self.encoding, title=None, content=filecontent, contenttype=self.contenttype)) + + return contentList + + +class CommandReceiver(Receiver): + def __init__(self, command, contenttype='text', encoding='utf-8'): + super().__init__(command) + self.encoding = encoding + self.command = command + self.contenttype = contenttype + + # input: [Content], output: [Content] + def performAction(self, contentList=None): + if contentList is None: + contentList = [] + + # run command and retrieve output + process = subprocess.Popen(self.command, stdout=subprocess.PIPE, shell=True, close_fds=True) + thefile = process.stdout + result = thefile.read().decode(self.encoding, errors='ignore') + thefile.close() + + if process.wait() != 0: + raise Exception("process terminated with an error") + + contentList.append(Content(uri=None, encoding=self.encoding, title=None, content=result, contenttype=self.contenttype)) + return contentList + + +class XPathParser(Parser): + def __init__(self, contentxpath, titlexpath=None): + self.contentxpath = contentxpath + self.titlexpath = titlexpath + + # input: [Content], output: [Content] + def performAction(self, contentList): + result = [] + for content in contentList: + result.extend(self.parseOneObject(content)) + return result + + # input: Content, output: [Content] + def parseOneObject(self, content): + baseuri = content.uri + if content.contenttype == 'html': + parser = etree.HTMLParser(encoding=content.encoding) + else: + parser = etree.XMLParser(recover=True, encoding=content.encoding) + + tree = etree.fromstring(content.content, parser=parser) + + # xpath + contentresult = [] if self.contentxpath is None else tree.xpath(self.contentxpath) + titleresult = [] if self.titlexpath is None else tree.xpath(self.titlexpath) + + # translate relative URIs to absolute URIs + if content.contenttype == 'html': + basetaglist = tree.xpath('/html/head/base') + if len(basetaglist) != 0: + baseuri = basetaglist[0].attrib['href'] + if len(contentresult) != 0: + toAbsoluteURIs(contentresult, baseuri) + if len(titleresult) != 0: + toAbsoluteURIs(titleresult, baseuri) + + if self.contentxpath and len(contentresult) == 0: + raise Exception('WARNING: content selector became invalid!') + if self.titlexpath and len(titleresult) == 0: + raise Exception('WARNING: title selector became invalid!') + + contents = [] + titles = [] + if isinstance(contentresult, str): + contents = [contentresult] + else: + if len(contentresult) == 0: + contentresult = titleresult + contents = [etree.tostring(s, encoding=content.encoding, pretty_print=True).decode(content.encoding, errors='ignore') for s in contentresult] + + if isinstance(titleresult, str): + titles = [getSubject(titleresult)]*len(contents) + else: + if len(titleresult) == 0 or len(titleresult) != len(contentresult): + titleresult = contentresult + titles = [getSubject(etree.tostring(s, method='text', encoding=content.encoding).decode(content.encoding, errors='ignore')) for s in titleresult] + + result = [] + for i in range(0, len(contents)): + result.append(Content(uri=content.uri, encoding=content.encoding, title=titles[i], content=contents[i], contenttype=content.contenttype)) + + return result + + +class CSSParser(Parser): + def __init__(self, contentcss, titlecss=None): + contentxpath = GenericTranslator().css_to_xpath(contentcss) + titlexpath = None + if titlecss is not None: + titlexpath = GenericTranslator().css_to_xpath(titlecss) + + self.xpathparser = XPathParser(contentxpath=contentxpath, titlexpath=titlexpath) + + # input: [Content], output: [Content] + def performAction(self, contentList): + return self.xpathparser.performAction(contentList) + + +class RegExParser(Parser): + def __init__(self, contentregex, titleregex=None): + self.contentregex = contentregex + self.titleregex = titleregex + + # input: [Content], output: [Content] + def performAction(self, contentList): + result = [] + for content in contentList: + result.extend(self.parseOneObject(content)) + return result + + # input: Content, output: [Content] + def parseOneObject(self, content): + contents = [] + titles = [] + if self.contentregex is not None: + for c in re.findall(r'' + self.contentregex, content.content, re.M): + if len(c.strip()) != 0: + contents.append(c) + if self.titleregex is not None: + for c in re.findall(r'' + self.titleregex, content.title, re.M): + if len(c.strip()) != 0: + titles.append(c) + + if self.contentregex is not None and len(contents) == 0: + raise Exception('WARNING: content regex became invalid!') + elif self.titleregex is not None and len(titles) == 0: + raise Exception('WARNING: title regex became invalid!') + else: + if len(contents) == 0: + contents = titles + if len(titles) == 0 or len(titles) != len(contents): + titles = [getSubject(c) for c in contents] + + result = [] + for i in range(0, len(contents)): + result.append(Content(uri=content.uri, encoding=content.encoding, title=titles[i], content=contents[i], contenttype=content.contenttype)) + + return result + -- cgit v1.2.3