diff --git a/parse-pages.py b/parse-pages.py new file mode 100644 index 0000000..4545aef --- /dev/null +++ b/parse-pages.py @@ -0,0 +1,107 @@ +#!/usr/bin/env python3 +# -*- coding: utf-8 -*- +# +# parse-pages.py +# +# Copyright 2016 GOLDERWEB – Jonathan Golder +# +# This program is free software; you can redistribute it and/or modify +# it under the terms of the GNU General Public License as published by +# the Free Software Foundation; either version 2 of the License, or +# (at your option) any later version. +# +# This program is distributed in the hope that it will be useful, +# but WITHOUT ANY WARRANTY; without even the implied warranty of +# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the +# GNU General Public License for more details. +# +# You should have received a copy of the GNU General Public License +# along with this program; if not, write to the Free Software +# Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, +# MA 02110-1301, USA. +# +# +""" +Script to parse all redpages in configured categories +""" + +import pywikibot +from pywikibot import pagegenerators + +import jogobot + +import redpage +import redfam + + +def get_cat_pages( cat ): + """ + Generates a iteratable generator-object with all pages listet in given + category + + @param cat Category to request + @type cat str + + @returns generator Iteratable object with pages of given category + """ + + # Get site to work on from pywikibot config + site = pywikibot.Site() + + # Retrieve the content of given category + category = pywikibot.Category( site, cat ) + + # Build an iteratable generator object with page objects for given category + generator = pagegenerators.CategorizedPageGenerator( category ) + + return generator + + +def main(*args): + """ + Handles process + """ + + try: + jogobot.output( "BEGINN – parser-pages.py" ) + + # Iterate over configured categories + for cat in ( jogobot.config["redundances"]["redpage_cats"] ): + + # Iterate over pages in current cat + for page in get_cat_pages( cat ): + + # For pages configured to exclude, go on with next page + if page.title() in ( + jogobot.config["redundances"]["redpage_exclude"] ): + + continue + + # Initiate RedPage object + red_page = redpage.RedPage( page ) + + # Check whether parsing is needed + if red_page.is_parsing_needed(): + + # Iterate over returned generator with redfam sections + for fam in red_page.parse(): + + # Run RedFamParser on section text + redfam.RedFamParser.parser( fam, red_page.page._pageid, + red_page.is_archive() ) + else: + # If successfully parsed whole page, flush + # db write cache + redfam.RedFamParser.flush_db_cache() + jogobot.output( "Page '%s' parsed" % + red_page.page.title() ) + else: + # If successfully parsed all pages in cat, flush db write cache + redpage.RedPage.flush_db_cache() + + finally: + jogobot.output( "END – parser-pages.py" ) + pywikibot.stopme() + +if( __name__ == "__main__" ): + main() diff --git a/redfam.py b/redfam.py index 519fe81..0e316a7 100644 --- a/redfam.py +++ b/redfam.py @@ -30,7 +30,9 @@ import locale import re from datetime import datetime -import pywikibot +import mwparserfromhell as mwparser # noqa +import pywikibot # noqa +from pywikibot.tools import deprecated # noqa import jogobot from mysqlred import MysqlRedFam @@ -40,12 +42,12 @@ class RedFam: """ Basic class for RedFams, containing the basic data structure """ - + def __init__( self, articlesList, beginning, ending=None, red_page_id=None, status=0, fam_hash=None, heading=None ): """ Generates a new RedFam object - + @param articlesList list List of articles @param beginning datetime Beginning date @param ending datetime Ending date @@ -54,7 +56,7 @@ class RedFam: @param fam_hash str SHA1 hash of articlesList @param heading str Original heading of RedFam (Link) """ - + # Initial attribute values self._articlesList = articlesList self._beginning = beginning @@ -63,18 +65,18 @@ class RedFam: self._status = status self._fam_hash = fam_hash self._heading = heading - + # Calculates the sha1 hash over self._articlesList to # rediscover known redundance families self.calc_fam_hash() - + def __repr__( self ): """ Returns repression str of RedFam object - + @returns str repr() string """ - + __repr = "RedFam( " + \ "articlesList=" + repr( self._articlesList ) + \ ", heading=" + repr( self._heading ) + \ @@ -84,28 +86,28 @@ class RedFam: ", status=" + repr( self._status ) + \ ", fam_hash=" + repr( self._fam_hash ) + \ " )" - + return __repr def calc_fam_hash( self ): """ Calculates the SHA-1 hash for the articlesList of redundance family. Since we don't need security SHA-1 is just fine. - + @returns str String with the hexadecimal hash digest """ - + h = hashlib.sha1() h.update( str( self._articlesList[:8] ).encode('utf-8') ) - + if self._fam_hash and h.hexdigest() != self._fam_hash: raise RedFamHashError( self._fam_hash, h.hexdigest() ) - + elif self._fam_hash: return else: self._fam_hash = h.hexdigest() - + @classmethod def flush_db_cache( cls ): """ @@ -119,28 +121,28 @@ class RedFamParser( RedFam ): Provides an interface to RedFam for adding/updating redundance families while parsig redundance pages """ - + # Define the timestamp format __timestamp_format = jogobot.config['redundances']['timestamp_format'] - + # Define section heading re.pattern - __sectionhead_pat = re.compile( r"^(=+)(.*\[\[.+\]\].*\[\[.+\]\].*)\1" ) - + __sectionhead_pat = re.compile( r"^(.*\[\[.+\]\].*\[\[.+\]\].*)" ) + # Define timestamp re.pattern __timestamp_pat = re.compile( jogobot.config['redundances'] ['timestamp_regex'] ) - + # Textpattern for recognisation of done-notices __done_notice = ":Archivierung dieses Abschnittes \ wurde gewünscht von:" __done_notice2 = "{{Erledigt|" - + def __init__( self, heading, red_page_id, red_page_archive, beginning, ending=None ): """ Creates a RedFam object based on data collected while parsing red_pages combined with possibly former known data from db - + @param red_fam_heading str Wikitext heading of section @param red_page_id int MediaWiki page_id @param red_page_archive bool Is red_page an archive @@ -149,132 +151,132 @@ class RedFamParser( RedFam ): @param ending datetime Timestamp of ending str strptime parseable string """ - + # Set object attributes: self._red_page_id = red_page_id self._red_page_archive = red_page_archive self._fam_hash = None - + # Method self.add_beginning sets self._beginning directly self.add_beginning( beginning ) - + # Method self.add_ending sets self._ending directly if( ending ): self.add_ending( ending ) else: # If no ending was provided set to None self._ending = None - + self._status = None - + # Parse the provided heading of redundance section # to set self._articlesList self.heading_parser( heading ) - + # Calculates the sha1 hash over self._articlesList to # rediscover known redundance families - + self.calc_fam_hash() - + # Open database connection, ask for data if existing, # otherwise create entry self.__handle_db() - + # Check status changes self.status() - + # Triggers db update if anything changed self.changed() - + def __handle_db( self ): """ Handles opening of db connection """ - + # We need a connection to our mysqldb self.__mysql = MysqlRedFam( ) self.__mysql.get_fam( self._fam_hash ) - + if not self.__mysql.data: self.__mysql.add_fam( self._articlesList, self._heading, self._red_page_id, self._beginning, self._ending ) - + def heading_parser( self, heading ): """ Parses given red_fam_heading string and saves articles list + + @param heading Heading of RedFam-Section + @type heading wikicode or mwparser-parseable """ - - # Predefine a pattern for wikilinks' destination - wikilink_pat = re.compile( r"\[\[([^\[\]\|]+)(?:\]\]|\|)" ) - - # Parse content of heading for generating section links later - match = type( self ).__sectionhead_pat.search( heading ) - if match: - self._heading = match.group(2).strip() - else: - raise RedFamHeadingError( heading ) - - # We get the pages in first [0] element iterating over - # wikilink_pat.findall( line ) - # Strip leading and trailing whitespace in Links to prevent wrong - # fam_hashes (when receiving redfam from db) since MySQL drops it - self._articlesList = [ link.strip() for link - in wikilink_pat.findall( self._heading ) ] - + + # Parse heading with mwparse if needed + if not isinstance( heading, mwparser.wikicode.Wikicode ): + heading = mwparser.parse( heading ) + + # Save heading as string + self._heading = str( heading ) + + # Save destinations of wikilinks in headings + self._articlesList = [ str( link.title ) for link + in heading.ifilter_wikilinks() ] + # Catch sections with more then 8 articles, print error if len( self._articlesList ) > 8: # For repression in output we need to know the fam hash self.calc_fam_hash() - pywikibot.output( "\ -{datetime} – \03{{lightred}}[WARNING] – \ -Maximum number of articles in red_fam exceeded, maximum number is 8, \ -{number:d} were given \n {repress}".format( - datetime=datetime.now().strftime( "%Y-%m-%d %H:%M:%S" ), - number=len( self._articlesList ), repress=repr( self ) ) ) - + jogobot.output( + ( "\03{{lightred}}" + + "Maximum number of articles in red_fam exceeded, " + + "maximum number is 8, {number:d} were given \n {repress}" + ).format( datetime=datetime.now().strftime( + "%Y-%m-%d %H:%M:%S" ), number=len( self._articlesList ), + repress=repr( self ) ), + "WARNING" ) + + # Only save the first 8 articles self._articlesList = self._articlesList[:8] def add_beginning( self, beginning ): """ Adds the beginning date of a redundance diskussion to the object - + @param datetime datetime Beginning date """ - + self._beginning = self.__datetime( beginning ) - + def add_ending( self, ending ): """ Adds the ending date of a redundance diskussion to the object. - + @param datetime datetime Ending date """ - + self._ending = self.__datetime( ending ) - + def __datetime( self, timestamp ): """ Decides wether given timestamp is a parseable string or a datetime object and returns a datetime object in both cases - + @param datetime timestamp Datetime object str timestamp Parseable string with timestamp - + @returns datetime Datetime object """ - + # Make sure locale is set to 'de_DE.UTF-8' to prevent problems # with wrong month abreviations in strptime locale.setlocale(locale.LC_ALL, 'de_DE.UTF-8') - + if( isinstance( timestamp, datetime ) ): return timestamp else: result = datetime.strptime( timestamp, type( self ).__timestamp_format ) return result - + def status( self ): """ Handles detection of correct status @@ -284,10 +286,10 @@ Maximum number of articles in red_fam exceeded, maximum number is 8, \ - 2 Discussion archived --> ending (normaly) present, page is archive - 3 and greater status was set by worker script, do not change it """ - + # Do not change stati set by worker script etc. if not self.__mysql.data['status'] > 2: - + # No ending, discussion is running: # Sometimes archived discussions also have no detectable ending if not self._ending and not self._red_page_archive: @@ -299,101 +301,150 @@ Maximum number of articles in red_fam exceeded, maximum number is 8, \ self._status = 2 else: self._status = self.__mysql.data[ 'status' ] - + def changed( self ): """ Checks wether anything has changed and maybe triggers db update """ - + # On archived red_fams do not delete possibly existing ending if( not self._ending and self._status > 1 and self.__mysql.data[ 'ending' ] ): - + self._ending = self.__mysql.data[ 'ending' ] - + # Since status change means something has changed, update database if( self._status != self.__mysql.data[ 'status' ] or self._beginning != self.__mysql.data[ 'beginning' ] or self._ending != self.__mysql.data[ 'ending' ] or self._red_page_id != self.__mysql.data[ 'red_page_id' ] or self._heading != self.__mysql.data[ 'heading' ]): - + self.__mysql.update_fam( self._red_page_id, self._heading, self._beginning, self._ending, self._status ) - + @classmethod + @deprecated def is_sectionheading( cls, line ): """ Checks wether given line is a red_fam section heading - + @param str line String to check - + @returns bool Returns True if it is a section heading """ - - if cls.__sectionhead_pat.search( line ): + + if cls.__sectionhead_pat.search( str(line) ): return True else: return False - + + @classmethod + def parser( cls, text, pageid, isarchive=False ): + """ + Handles parsing of redfam section + + @param text Text of RedFam-Section + @type text wikicode or mwparser-parseable + """ + + # Parse heading with mwparse if needed + if not isinstance( text, mwparser.wikicode.Wikicode ): + text = mwparser.parse( text ) + + # Extract heading text + heading = next( text.ifilter_headings() ).title + + # Extract beginnig and maybe ending + (beginning, ending) = RedFamParser.extract_dates( text, isarchive ) + + # Create the RedFam object + RedFamParser( heading, pageid, isarchive, beginning, ending ) + + @classmethod + def extract_dates( cls, text, isarchive=False ): + """ + Returns tuple of the first and maybe last timestamp of a section. + Last timestamp is only returned if there is a done notice or param + *isarchiv* is set to 'True' + + @param text Text to search in + @type line Any Type castable to str + @param isarchive If true skip searching done notice (on archivepages) + @type isarchive bool + + @returns Timestamps, otherwise None + @returntype tuple of strs + """ + + # Match all timestamps + matches = cls.__timestamp_pat.findall( str( text ) ) + if matches: + + # First one is beginning + # Since some timestamps are broken we need to reconstruct them + # by regex match groups + beginning = ( matches[0][0] + ", " + matches[0][1] + ". " + + matches[0][2] + ". " + matches[0][3] ) + + # Last one maybe is ending + # Done notice format 1 + # Done notice format 2 + # Or on archivepages + if ( cls.__done_notice in text or + cls.__done_notice2 in text or + isarchive ): + + ending = ( matches[-1][0] + ", " + matches[-1][1] + ". " + + matches[-1][2] + ". " + matches[-1][3] ) + + else: + ending = None + + return (beginning, ending) + @classmethod + @deprecated( extract_dates ) def is_beginning( cls, line ): """ Returns the first timestamp found in line, otherwise None - + @param str line String to search in - + @returns str Timestamp, otherwise None """ - - match = cls.__timestamp_pat.search( line ) - if match: - # Since some timestamps are broken we need to reconstruct them - # by regex match groups - result = match.group(1) + ", " + match.group(2) + ". " +\ - match.group(3) + ". " + match.group(4) - return result - else: - return None - + + return cls.extract_dates( line )[0] + @classmethod - def is_ending( cls, line ): + @deprecated( extract_dates ) + def is_ending( cls, line, isarchive=False ): """ Returns the timestamp of done notice ( if one ), otherwise None - @param str line String to search in - - @returns str Timestamp, otherwise None + + @param line String to search in + @type line str + @param isarchive If true skip searching done notice (on archivepages) + @type isarchive bool + + @returns Timestamp, otherwise None + @returntype str """ - - if ( cls.__done_notice in line ) or ( cls.__done_notice2 in line ): - match = cls.__timestamp_pat.search( line ) - if match: - # Since some timestamps are broken we need to reconstruct them - # by regex match groups - result = match.group(1) + ", " + match.group(2) + ". " +\ - match.group(3) + ". " + match.group(4) - return result - return None - + + return cls.extract_dates( line )[1] + @classmethod + @deprecated( extract_dates ) def is_ending2( cls, line ): """ Returns the last timestamp found in line, otherwise None @param str line String to search in - + @returns str Timestamp, otherwise None """ - - matches = cls.__timestamp_pat.findall( line ) - if matches: - # Since some timestamps are broken we need to reconstruct them - # by regex match groups - result = matches[-1][0] + ", " + matches[-1][1] + ". " +\ - matches[-1][2] + ". " + matches[-1][3] - return result - else: - return None + + return cls.extract_dates( line, True )[1] class RedFamWorker( RedFam ): @@ -402,17 +453,17 @@ class RedFamWorker( RedFam ): where discussion is finished """ def __init__( self, mysql_data ): - + articlesList = [] for key in sorted( mysql_data.keys() ): if 'article' in key and mysql_data[ key ]: articlesList.append( mysql_data[ key ] ) - + super().__init__( articlesList, mysql_data[ 'beginning' ], mysql_data[ 'ending' ], mysql_data[ 'red_page_id' ], mysql_data[ 'status' ], mysql_data[ 'fam_hash' ], mysql_data[ 'heading' ] ) - + @classmethod def list_by_status( cls, status ): """ @@ -426,12 +477,12 @@ class RedFamWorker( RedFam ): print(fam) raise - + class RedFamError( Exception ): """ Base class for all Errors of RedFam-Module """ - + def __init__( self, message=None ): """ Handles Instantiation of RedFamError's @@ -440,12 +491,12 @@ class RedFamError( Exception ): self.message = "An Error occured while executing a RedFam action" else: self.message = message - + def __str__( self ): """ Output of error message """ - + return self.message @@ -453,12 +504,12 @@ class RedFamHashError( RedFamError ): """ Raised when given RedFamHash does not match with calculated """ - + def __init__( self, givenHash, calculatedHash ): - + message = "Given fam_hash ('{given}') does not match with \ calculated ('{calc}'".format( given=givenHash, calc=calculatedHash ) - + super().__init__( message ) @@ -467,8 +518,8 @@ class RedFamHeadingError ( RedFamError ): Raised when given RedFamHeading does not match __sectionhead_pat Regex """ def __init__( self, heading ): - + message = "Error while trying to parse section heading. Given heading \ '{heading}' does not match RegEx".format( heading=heading ) - + super().__init__( message ) diff --git a/redpage.py b/redpage.py index 96a853d..2b93ae8 100644 --- a/redpage.py +++ b/redpage.py @@ -26,53 +26,51 @@ Provides a class for handling redundance discussion pages and archives """ import pywikibot # noqa +import mwparserfromhell as mwparser + +import jogobot from mysqlred import MysqlRedPage -from redfam import RedFamParser class RedPage: """ Class for handling redundance discussion pages and archives """ - + def __init__( self, page, archive=False ): """ Generate a new RedPage object based on the given pywikibot page object - + @param page page Pywikibot/MediaWiki page object for page """ - + # Safe the pywikibot page object self.page = page self._archive = archive - + self.__handle_db( ) self.is_page_changed() - + self._parsed = None - if( self._changed or self.__mysql.data[ 'status' ] == 0 ): - self.parse() - - self.__update_db() - + def __handle_db( self ): """ Handles opening of db connection """ - + # We need a connection to our mysqldb self.__mysql = MysqlRedPage( self.page._pageid ) - + if not self.__mysql.data: self.__mysql.add_page( self.page.title(), self.page._revid ) - + def is_page_changed( self ): """ Check wether the page was changed since last run """ - + if( self.__mysql.data != { 'page_id': self.page._pageid, 'rev_id': self.page._revid, 'page_title': self.page.title(), @@ -85,102 +83,65 @@ class RedPage: """ Detects wether current page is an archive of discussions """ - + if( self._archive or ( u"/Archiv" in self.page.title() ) or ( "{{Archiv}}" in self.page.text ) or ( "{{Archiv|" in self.page.text ) ): - + return True else: return False - - def parse( self ): # noqa + + def is_parsing_needed( self ): + """ + Decides wether current RedPage needs to be parsed or not + """ + + if( self._changed or self.__mysql.data[ 'status' ] == 0 ): + return True + else: + return False + + def parse( self ): """ Handles the parsing process """ - # Since @param text is a string we need to split it in lines - text_lines = self.page.text.split( "\n" ) - length = len( text_lines ) - - # Initialise line counter - i = 0 - fam_heading = None - beginning = None - ending = None - - # Set line for last detected Redundance-Family to 0 - last_fam = 0 - - # Iterate over the lines of the page - for line in text_lines: - - # Check wether we have an "Redundance-Family"-Section heading - if RedFamParser.is_sectionheading( line ): - - # Save line number for last detected Redundance-Family - last_fam = i - # Save heading - fam_heading = line - - # Defined (re)initialisation of dates - beginning = None - ending = None - - # Check wether we are currently in an "Redundance-Family"-Section - if i > last_fam and last_fam > 0: - - # Check if we have alredy recognized the beginning date of the - # discussion (in former iteration) or if we have a done-notice - if not beginning: - beginning = RedFamParser.is_beginning( line ) - elif not ending: - ending = RedFamParser.is_ending( line ) - - # Detect end of red_fam section (next line is new sectionheading) - # or end of file - # Prevent from running out of index - if i < (length - 1): - test = RedFamParser.is_sectionheading( text_lines[ i + 1 ] ) - else: - test = False - if ( test or ( length == ( i + 1 ) ) ): - - # Create the red_fam object - if( fam_heading and beginning ): - - # Maybe we can find a ending by feed if we have None yet - # (No done notice on archive pages) - if not ending and self.is_archive(): - j = i - while (j > last_fam) and not ending: - j -= 1 - ending = RedFamParser.is_ending2( text_lines[ j ] ) - - # Create the RedFam object - RedFamParser( fam_heading, self.page._pageid, - self.is_archive(), beginning, ending ) - - # Increment line counter - i += 1 + # Generate Wikicode object + self.wikicode = mwparser.parse( self.page.text ) + + # Select RedFam-sections + # matches=Regexp or + # function( gets heading content as wikicode as param 1) + # include_lead = if true include first section (intro) + # include_heading = if true include heading + fams = self.wikicode.get_sections( + matches=jogobot.config["redundances"]["section_heading_regex"], + include_lead=False, include_headings=True ) + + # Iterate over RedFam + for fam in fams: + + yield fam + else: - RedFamParser.flush_db_cache() self._parsed = True - + self.__update_db() + def __update_db( self ): """ Updates the page meta data in mysql db """ if( self._parsed or not self._changed ): status = 1 - + if( self.is_archive() ): status = 2 else: status = 0 - + self.__mysql.update_page( self.page._revid, self.page.title(), status ) - + @classmethod def flush_db_cache( cls ): """ diff --git a/tox.ini b/tox.ini deleted file mode 100644 index 9236f4f..0000000 --- a/tox.ini +++ /dev/null @@ -1,2 +0,0 @@ -[flake8] -ignore = E129,E201,E202,W293