From f207ba01fe5f1cb4ca150ff748598da4541c3151 Mon Sep 17 00:00:00 2001 From: Jim Miller Date: Fri, 16 Mar 2012 12:26:45 -0500 Subject: [PATCH] Adding Sam's code for nfacommunity.com & midnightwhispers.ca. --- fanficdownloader/adapters/__init__.py | 2 + .../adapters/adapter_midnightwhispersca.py | 285 +++++++++++++++++ .../adapters/adapter_nfacommunitycom.py | 289 ++++++++++++++++++ 3 files changed, 576 insertions(+) create mode 100644 fanficdownloader/adapters/adapter_midnightwhispersca.py create mode 100644 fanficdownloader/adapters/adapter_nfacommunitycom.py diff --git a/fanficdownloader/adapters/__init__.py b/fanficdownloader/adapters/__init__.py index e2d0f896..cfe37666 100644 --- a/fanficdownloader/adapters/__init__.py +++ b/fanficdownloader/adapters/__init__.py @@ -48,6 +48,8 @@ import adapter_portkeyorg import adapter_mugglenetcom import adapter_hpfandomnet import adapter_thequidditchpitchorg +import adapter_nfacommunitycom +import adapter_midnightwhispersca ## This bit of complexity allows adapters to be added by just adding ## importing. It eliminates the long if/else clauses we used to need diff --git a/fanficdownloader/adapters/adapter_midnightwhispersca.py b/fanficdownloader/adapters/adapter_midnightwhispersca.py new file mode 100644 index 00000000..d638ecc0 --- /dev/null +++ b/fanficdownloader/adapters/adapter_midnightwhispersca.py @@ -0,0 +1,285 @@ +# -*- coding: utf-8 -*- + +# Copyright 2011 Fanficdownloader team +# +# Licensed under the Apache License, Version 2.0 (the "License"); +# you may not use this file except in compliance with the License. +# You may obtain a copy of the License at +# +# http://www.apache.org/licenses/LICENSE-2.0 +# +# Unless required by applicable law or agreed to in writing, software +# distributed under the License is distributed on an "AS IS" BASIS, +# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +# See the License for the specific language governing permissions and +# limitations under the License. +# + +import time +import logging +import re +import urllib2 + +from .. import BeautifulSoup as bs +from ..htmlcleanup import stripHTML +from .. import exceptions as exceptions + +from base_adapter import BaseSiteAdapter, makeDate + +# Search for XXX comments--that's where things are most likely to need changing. + +# This function is called by the downloader in all adapter_*.py files +# in this dir to register the adapter class. So it needs to be +# updated to reflect the class below it. That, plus getSiteDomain() +# take care of 'Registering'. +def getClass(): + return MidnightwhispersCaAdapter # XXX + +# Class name has to be unique. Our convention is camel case the +# sitename with Adapter at the end. www is skipped. +class MidnightwhispersCaAdapter(BaseSiteAdapter): # XXX + + def __init__(self, config, url): + BaseSiteAdapter.__init__(self, config, url) + + self.decode = ["Windows-1252", + "utf8"] # 1252 is a superset of iso-8859-1. + # Most sites that claim to be + # iso-8859-1 (and some that claim to be + # utf8) are really windows-1252. + self.is_adult=False + + # get storyId from url--url validation guarantees query is only sid=1234 + self.story.setMetadata('storyId',self.parsedUrl.query.split('=',)[1]) + logging.debug("storyId: (%s)"%self.story.getMetadata('storyId')) + + # normalized story URL. + # XXX Most sites don't have the /fanfic part. Replace all to remove it usually. + self._setURL('http://' + self.getSiteDomain() + '/viewstory.php?sid='+self.story.getMetadata('storyId')) + + # Each adapter needs to have a unique site abbreviation. + self.story.setMetadata('siteabbrev','mw') # XXX + + # If all stories from the site fall into the same category, + # the site itself isn't likely to label them as such, so we + # do. + self.story.addToList("category","Queer as Folk") # XXX + + # The date format will vary from site to site. + # http://docs.python.org/library/datetime.html#strftime-strptime-behavior + self.dateformat = "%B %d, %Y" # XXX + + @staticmethod # must be @staticmethod, don't remove it. + def getSiteDomain(): + # The site domain. Does have www here, if it uses it. + return 'www.midnightwhispers.ca' # XXX + + def getSiteExampleURLs(self): + return "http://"+self.getSiteDomain()+"/viewstory.php?sid=1234" + + def getSiteURLPattern(self): + return re.escape("http://"+self.getSiteDomain()+"/viewstory.php?sid=")+r"\d+$" + + ## Getting the chapter list and the meta data, plus 'is adult' checking. + def extractChapterUrlsAndMetadata(self): + + if self.is_adult or self.getConfig("is_adult"): + # Weirdly, different sites use different warning numbers. + # If the title search below fails, there's a good chance + # you need a different number. print data at that point + # and see what the 'click here to continue' url says. + + # Furthermore, there's a couple sites now with more than + # one warning level for different ratings. And they're + # fussy about it. midnightwhispers has three: 10, 3 & 5. + # we'll try 5 first. + addurl = "&ageconsent=ok&warning=5" # XXX + else: + addurl="" + + # index=1 makes sure we see the story chapter index. Some + # sites skip that for one-chapter stories. + url = self.url+'&index=1'+addurl + logging.debug("URL: "+url) + + try: + data = self._fetchUrl(url) + except urllib2.HTTPError, e: + if e.code == 404: + raise exceptions.StoryDoesNotExist(self.url) + else: + raise e + + # The actual text that is used to announce you need to be an + # adult varies from site to site. Again, print data before + # the title search to troubleshoot. + + # Since the warning text can change by warning level, let's + # look for the warning pass url. nfacommunity uses + # &warning= -- actually, so do other sites. Must be an + # eFiction book. + + # viewstory.php?sid=1882&warning=4 + # viewstory.php?sid=1654&ageconsent=ok&warning=5 + #print data + #m = re.search(r"'viewstory.php\?sid=1882(&warning=4)'",data) + m = re.search(r"'viewstory.php\?sid=\d+((?:&ageconsent=ok)?&warning=\d+)'",data) + if m != None: + if self.is_adult or self.getConfig("is_adult"): + # We tried the default and still got a warning, so + # let's pull the warning number from the 'continue' + # link and reload data. + addurl = m.group(1) + # correct stupid & error in url. + addurl = addurl.replace("&","&") + url = self.url+'&index=1'+addurl + logging.debug("URL 2nd try: "+url) + + try: + data = self._fetchUrl(url) + except urllib2.HTTPError, e: + if e.code == 404: + raise exceptions.StoryDoesNotExist(self.url) + else: + raise e + else: + raise exceptions.AdultCheckRequired(self.url) + + if "Access denied. This story has not been validated by the adminstrators of this site." in data: + raise exceptions.FailedToDownload(self.getSiteDomain() +" says: Access denied. This story has not been validated by the adminstrators of this site.") + + # use BeautifulSoup HTML parser to make everything easier to find. + soup = bs.BeautifulSoup(data) + # print data + + + # Now go hunting for all the meta data and the chapter list. + + ## Title + a = soup.find('a', href=re.compile(r'viewstory.php\?sid='+self.story.getMetadata('storyId')+"$")) + self.story.setMetadata('title',stripHTML(a)) # title's inside a tag. + + # Find authorid and URL from... author url. + a = soup.find('a', href=re.compile(r"viewuser.php\?uid=\d+")) + self.story.setMetadata('authorId',a['href'].split('=')[1]) + self.story.setMetadata('authorUrl','http://'+self.host+'/'+a['href']) + self.story.setMetadata('author',a.string) + + # Find the chapters: + for chapter in soup.findAll('a', href=re.compile(r'viewstory.php\?sid='+self.story.getMetadata('storyId')+"&chapter=\d+$")): + # just in case there's tags, like in chapter titles. + self.chapterUrls.append((stripHTML(chapter),'http://'+self.host+'/'+chapter['href']+addurl)) + + self.story.setMetadata('numChapters',len(self.chapterUrls)) + + # eFiction sites don't help us out a lot with their meta data + # formating, so it's a little ugly. + + # utility method + def defaultGetattr(d,k): + try: + return d[k] + except: + return "" + + # Rated: NC-17
etc + labels = soup.findAll('span',{'class':'label'}) + for labelspan in labels: + value = labelspan.nextSibling + label = labelspan.string + + if 'Summary' in label: + ## Everything until the next span class='label' + svalue = "" + while not defaultGetattr(value,'class') == 'label': + svalue += str(value) + value = value.nextSibling + self.setDescription(url,svalue) + #self.story.setMetadata('description',stripHTML(svalue)) + + if 'Rated' in label: + self.story.setMetadata('rating', value) + + if 'Word count' in label: + self.story.setMetadata('numWords', value) + + if 'Categories' in label: + cats = labelspan.parent.findAll('a',href=re.compile(r'browse.php\?type=categories')) + catstext = [cat.string for cat in cats] + for cat in catstext: + self.story.addToList('category',cat.string) + + if 'Characters' in label: + chars = labelspan.parent.findAll('a',href=re.compile(r'browse.php\?type=characters')) + charstext = [char.string for char in chars] + for char in charstext: + self.story.addToList('characters',char.string) + + ## Not all sites use Genre, but there's no harm to + ## leaving it in. Check to make sure the type_id number + ## is correct, though--it's site specific. + if 'Genre' in label: + genres = labelspan.parent.findAll('a',href=re.compile(r'browse.php\?type=class&type_id=2')) # XXX + genrestext = [genre.string for genre in genres] + self.genre = ', '.join(genrestext) + for genre in genrestext: + self.story.addToList('genre',genre.string) + + ## Not all sites use Warnings, but there's no harm to + ## leaving it in. Check to make sure the type_id number + ## is correct, though--it's site specific. + if 'Warnings' in label: + warnings = labelspan.parent.findAll('a',href=re.compile(r'browse.php\?type=class&type_id=2')) # XXX + warningstext = [warning.string for warning in warnings] + self.warning = ', '.join(warningstext) + for warning in warningstext: + self.story.addToList('warnings',warning.string) + + if 'Completed' in label: + if 'Yes' in value: + self.story.setMetadata('status', 'Completed') + else: + self.story.setMetadata('status', 'In-Progress') + + if 'Published' in label: + self.story.setMetadata('datePublished', makeDate(stripHTML(value), self.dateformat)) + + if 'Updated' in label: + # there's a stray [ at the end. + #value = value[0:-1] + self.story.setMetadata('dateUpdated', makeDate(stripHTML(value), self.dateformat)) + + try: + # Find Series name from series URL. + a = soup.find('a', href=re.compile(r"viewseries.php\?seriesid=\d+")) + series_name = a.string + series_url = 'http://'+self.host+'/'+a['href'] + + # use BeautifulSoup HTML parser to make everything easier to find. + seriessoup = bs.BeautifulSoup(self._fetchUrl(series_url)) + storyas = seriessoup.findAll('a', href=re.compile(r'^viewstory.php\?sid=\d+$')) + i=1 + for a in storyas: + if a['href'] == ('viewstory.php?sid='+self.story.getMetadata('storyId')): + self.setSeries(series_name, i) + break + i+=1 + + except: + # I find it hard to care if the series parsing fails + pass + + # grab the text for an individual chapter. + def getChapterText(self, url): + + logging.debug('Getting chapter text from: %s' % url) + + soup = bs.BeautifulStoneSoup(self._fetchUrl(url), + selfClosingTags=('br','hr')) # otherwise soup eats the br/hr tags. + + div = soup.find('div', {'id' : 'story'}) + + if None == div: + raise exceptions.FailedToDownload("Error downloading Chapter: %s! Missing required element!" % url) + + return self.utf8FromSoup(url,div) diff --git a/fanficdownloader/adapters/adapter_nfacommunitycom.py b/fanficdownloader/adapters/adapter_nfacommunitycom.py new file mode 100644 index 00000000..07951450 --- /dev/null +++ b/fanficdownloader/adapters/adapter_nfacommunitycom.py @@ -0,0 +1,289 @@ +# -*- coding: utf-8 -*- + +# Copyright 2011 Fanficdownloader team +# +# Licensed under the Apache License, Version 2.0 (the "License"); +# you may not use this file except in compliance with the License. +# You may obtain a copy of the License at +# +# http://www.apache.org/licenses/LICENSE-2.0 +# +# Unless required by applicable law or agreed to in writing, software +# distributed under the License is distributed on an "AS IS" BASIS, +# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +# See the License for the specific language governing permissions and +# limitations under the License. +# + +import time +import logging +import re +import urllib2 + +from .. import BeautifulSoup as bs +from ..htmlcleanup import stripHTML +from .. import exceptions as exceptions + +from base_adapter import BaseSiteAdapter, makeDate + +# Search for XXX comments--that's where things are most likely to need changing. + +# This function is called by the downloader in all adapter_*.py files +# in this dir to register the adapter class. So it needs to be +# updated to reflect the class below it. That, plus getSiteDomain() +# take care of 'Registering'. +def getClass(): + return NfaCommunityComAdapter # XXX + +# Class name has to be unique. Our convention is camel case the +# sitename with Adapter at the end. www is skipped. +class NfaCommunityComAdapter(BaseSiteAdapter): # XXX + + def __init__(self, config, url): + BaseSiteAdapter.__init__(self, config, url) + + self.decode = ["Windows-1252", + "utf8"] # 1252 is a superset of iso-8859-1. + # Most sites that claim to be + # iso-8859-1 (and some that claim to be + # utf8) are really windows-1252. + self.is_adult=False + + # get storyId from url--url validation guarantees query is only sid=1234 + self.story.setMetadata('storyId',self.parsedUrl.query.split('=',)[1]) + logging.debug("storyId: (%s)"%self.story.getMetadata('storyId')) + + # normalized story URL. + # XXX Most sites don't have the /fanfic part. Replace all to remove it usually. + self._setURL('http://' + self.getSiteDomain() + '/viewstory.php?sid='+self.story.getMetadata('storyId')) + + # Each adapter needs to have a unique site abbreviation. + self.story.setMetadata('siteabbrev','nfa') # XXX + + # If all stories from the site fall into the same category, + # the site itself isn't likely to label them as such, so we + # do. + self.story.addToList("category","NCIS") # XXX + + # The date format will vary from site to site. + # http://docs.python.org/library/datetime.html#strftime-strptime-behavior + self.dateformat = "%m/%d/%Y" # XXX + + @classmethod + def getAcceptDomains(cls): + return ['www.nfacommunity.com','nfacommunity.com'] + + @staticmethod # must be @staticmethod, don't remove it. + def getSiteDomain(): + # The site domain. Does have www here, if it uses it. + return 'nfacommunity.com' # XXX + + def getSiteExampleURLs(self): + return "http://"+self.getSiteDomain()+"/viewstory.php?sid=1234" + + def getSiteURLPattern(self): + return "http://(www.)?"+re.escape(self.getSiteDomain()+"/viewstory.php?sid=")+r"\d+$" + + ## Getting the chapter list and the meta data, plus 'is adult' checking. + def extractChapterUrlsAndMetadata(self): + + if self.is_adult or self.getConfig("is_adult"): + # Weirdly, different sites use different warning numbers. + # If the title search below fails, there's a good chance + # you need a different number. print data at that point + # and see what the 'click here to continue' url says. + + # Furthermore, there's a couple sites now with more than + # one warning level for different ratings. And they're + # fussy about it. nfacommunity has two: 4 & 5. + # we'll try 5 first. + addurl = "&ageconsent=ok&warning=5" # XXX + else: + addurl="" + + # index=1 makes sure we see the story chapter index. Some + # sites skip that for one-chapter stories. + url = self.url+'&index=1'+addurl + logging.debug("URL: "+url) + + try: + data = self._fetchUrl(url) + except urllib2.HTTPError, e: + if e.code == 404: + raise exceptions.StoryDoesNotExist(self.url) + else: + raise e + + # The actual text that is used to announce you need to be an + # adult varies from site to site. Again, print data before + # the title search to troubleshoot. + + # Since the warning text can change by warning level, let's + # look for the warning pass url. nfacommunity uses + # &warning= -- actually, so do other sites. Must be an + # eFiction book. + + # viewstory.php?sid=1882&warning=4 + # viewstory.php?sid=1654&ageconsent=ok&warning=5 + #print data + #m = re.search(r"'viewstory.php\?sid=1882(&warning=4)'",data) + m = re.search(r"'viewstory.php\?sid=\d+((?:&ageconsent=ok)?&warning=\d+)'",data) + if m != None: + if self.is_adult or self.getConfig("is_adult"): + # We tried the default and still got a warning, so + # let's pull the warning number from the 'continue' + # link and reload data. + addurl = m.group(1) + # correct stupid & error in url. + addurl = addurl.replace("&","&") + url = self.url+'&index=1'+addurl + logging.debug("URL 2nd try: "+url) + + try: + data = self._fetchUrl(url) + except urllib2.HTTPError, e: + if e.code == 404: + raise exceptions.StoryDoesNotExist(self.url) + else: + raise e + else: + raise exceptions.AdultCheckRequired(self.url) + + if "Access denied. This story has not been validated by the adminstrators of this site." in data: + raise exceptions.FailedToDownload(self.getSiteDomain() +" says: Access denied. This story has not been validated by the adminstrators of this site.") + + # use BeautifulSoup HTML parser to make everything easier to find. + soup = bs.BeautifulSoup(data) + + # Now go hunting for all the meta data and the chapter list. + + ## Title + a = soup.find('a', href=re.compile(r'viewstory.php\?sid='+self.story.getMetadata('storyId')+"$")) + self.story.setMetadata('title',a.string) + + # Find authorid and URL from... author url. + a = soup.find('a', href=re.compile(r"viewuser.php\?uid=\d+")) + self.story.setMetadata('authorId',a['href'].split('=')[1]) + self.story.setMetadata('authorUrl','http://'+self.host+'/'+a['href']) + self.story.setMetadata('author',a.string) + + # Find the chapters: + for chapter in soup.findAll('a', href=re.compile(r'viewstory.php\?sid='+self.story.getMetadata('storyId')+"&chapter=\d+$")): + # just in case there's tags, like in chapter titles. + self.chapterUrls.append((stripHTML(chapter),'http://'+self.host+'/'+chapter['href']+addurl)) + + self.story.setMetadata('numChapters',len(self.chapterUrls)) + + + + # eFiction sites don't help us out a lot with their meta data + # formating, so it's a little ugly. + + # utility method + def defaultGetattr(d,k): + try: + return d[k] + except: + return "" + + # Rated: NC-17
etc + labels = soup.findAll('span',{'class':'label'}) + for labelspan in labels: + value = labelspan.nextSibling + label = labelspan.string + + if 'Summary' in label: + ## Everything until the next span class='label' + svalue = "" + while not defaultGetattr(value,'class') == 'label': + svalue += str(value) + value = value.nextSibling + self.setDescription(url,svalue) + #self.story.setMetadata('description',stripHTML(svalue)) + + if 'Rated' in label: + self.story.setMetadata('rating', value) + + if 'Word count' in label: + self.story.setMetadata('numWords', value) + + if 'Categories' in label: + cats = labelspan.parent.findAll('a',href=re.compile(r'browse.php\?type=categories')) + catstext = [cat.string for cat in cats] + for cat in catstext: + self.story.addToList('category',cat.string) + + if 'Characters' in label: + chars = labelspan.parent.findAll('a',href=re.compile(r'browse.php\?type=characters')) + charstext = [char.string for char in chars] + for char in charstext: + self.story.addToList('characters',char.string) + + ## Not all sites use Genre, but there's no harm to + ## leaving it in. Check to make sure the type_id number + ## is correct, though--it's site specific. + if 'Genre' in label: + genres = labelspan.parent.findAll('a',href=re.compile(r'browse.php\?type=class&type_id=2')) # XXX + genrestext = [genre.string for genre in genres] + self.genre = ', '.join(genrestext) + for genre in genrestext: + self.story.addToList('genre',genre.string) + + ## Not all sites use Warnings, but there's no harm to + ## leaving it in. Check to make sure the type_id number + ## is correct, though--it's site specific. + if 'Warnings' in label: + warnings = labelspan.parent.findAll('a',href=re.compile(r'browse.php\?type=class&type_id=2')) # XXX + warningstext = [warning.string for warning in warnings] + self.warning = ', '.join(warningstext) + for warning in warningstext: + self.story.addToList('warnings',warning.string) + + if 'Completed' in label: + if 'Yes' in value: + self.story.setMetadata('status', 'Completed') + else: + self.story.setMetadata('status', 'In-Progress') + + if 'Published' in label: + self.story.setMetadata('datePublished', makeDate(stripHTML(value), self.dateformat)) + + if 'Updated' in label: + # there's a stray [ at the end. + #value = value[0:-1] + self.story.setMetadata('dateUpdated', makeDate(stripHTML(value), self.dateformat)) + + try: + # Find Series name from series URL. + a = soup.find('a', href=re.compile(r"viewseries.php\?seriesid=\d+")) + series_name = a.string + series_url = 'http://'+self.host+'/'+a['href'] + + # use BeautifulSoup HTML parser to make everything easier to find. + seriessoup = bs.BeautifulSoup(self._fetchUrl(series_url)) + storyas = seriessoup.findAll('a', href=re.compile(r'^viewstory.php\?sid=\d+$')) + i=1 + for a in storyas: + if a['href'] == ('viewstory.php?sid='+self.story.getMetadata('storyId')): + self.setSeries(series_name, i) + break + i+=1 + + except: + # I find it hard to care if the series parsing fails + pass + + # grab the text for an individual chapter. + def getChapterText(self, url): + + logging.debug('Getting chapter text from: %s' % url) + + soup = bs.BeautifulStoneSoup(self._fetchUrl(url), + selfClosingTags=('br','hr')) # otherwise soup eats the br/hr tags. + + div = soup.find('div', {'id' : 'story'}) + + if None == div: + raise exceptions.FailedToDownload("Error downloading Chapter: %s! Missing required element!" % url) + + return self.utf8FromSoup(url,div)