def __init__(self,year,activation=5,overall=False):
self.year = year
'''The year the cohort started.
'''
self.activation = activation
'''Minimum number of edits per month to be included in the cohort
'''
self.overall = overall
ts,ts_i = utils.create_time_stamps_month(fromym='%s01'%self.year,toym='201012')
self.time_stamps = ts
self.time_stamps_index = ts_i
'''Only take time_stamps starting with self.year
'''
self.cohorts = [self.year,'others']
'''Cohort definition
'''
self.cohort_labels = ['%s cohort'% i for i in self.cohorts]
'''Cohort labels
'''
if self.overall:
self.sqlQuery = "SELECT * FROM fabian WHERE user_id IN (SELECT user_id FROM fabian WHERE first_edit_year=%s GROUP BY user_id HAVING SUM(add_edits)>%s);"%(self.year,self.activation)
else:
self.sqlQuery = 'SELECT * FROM fabian WHERE add_edits > %s AND first_edit_year=%s'%(self.activation,self.year)
Cohort.__init__(self)
def __init__(self, period=3):
# We don't take into consideration people who have 0 edits as the data is coded sparse and we don't have
# this information
# Note that len(self.cohorts) has to be equal the number of bins in the numpy.array
# self.cohorts = [1,3,5,7,10,20,40,60,80,100,200,500,1000,5000,10000,'>10000 edits']
self.cohorts = [1, 5, 10, 50, 100, 500, 1000, '>1000 edits']
'''Cohort definition
'''
self.cohort_labels = self.cohorts[:]
self.cohort_labels[0] = '%s edit' % (self.cohort_labels[0])
for i in range(1, len(self.cohorts) - 1):
self.cohort_labels[i] = '%s-%s edits' % (self.cohorts[i - 1] + 1,
self.cohorts[i])
# self.cohort_labels = ['<%s edits'%(e) for e in self.cohorts]
'''Cohort labels
'''
self.period = period
'''The number of month an editor is considered new
'''
self.old_user_id = None
'''The user_id of the previously encountered editor as we iterate through the table
'''
self.lastym = None
'''The ym at the end of the `period` months after the first edit of an editor
'''
self.firstedit = None
self.fe_index = None
self.edits = 0
self.added = 0
self.removed = 0
self.net = 0
Cohort.__init__(self)
def __init__(self, minedits=1, maxedits=None):
self.cohorts = [int(i) for i in range(0, len(settings.time_stamps))]
'''Cohort definition
'''
self.cohort_labels = ['%s month old' % i for i in self.cohorts]
'''Cohort labels
'''
self.sqlQuery = 'SELECT * FROM %s;' % tables.EDITOR_YEAR_MONTH
'''The SQL query returns edit information for each editor for each ym she has edited.'''
self.minedits = minedits
'''Minimum number of edits by editor in a given month to be included'''
self.maxedits = maxedits
'''Maximum number of edits by editor in a given month to be included'''
self.ncolors = utils.numberOfMonths(settings.time_stamps[0],
settings.time_stamps[-1]) / 6
'''
Number of visible colors in the wikipride plots. E.g. one color for every six month for wikipride plots
'''
Cohort.__init__(self)
def __init__(self):
if self.cohorts is None or self.cohort_labels is None:
logger.error(
"self.cohorts or self.cohort_labels not properly defined")
# raise Exception("self.cohorts or self.cohort_labels not properly defined")
Cohort.__init__(self)
def __init__(self,reverttype,activation=1):
'''
arg reverttype: str, type of revert: 'reverting','reverted','revertedto'
'''
self.reverttype = reverttype
self.activation = activation
'''Minimum number of reverts by editor to be included'''
Cohort.__init__(self)
def __init__(self):
# We don't take into consideration people who have 0 edits as the data is coded sparse and we don't have
# this information
# Note that len(self.cohorts) has to be equal the number of bins in the numpy.array
# self.cohorts = [1,3,5,7,10,20,40,60,80,100,200,500,1000,5000,10000,'>10000 edits']
self.cohorts = [1,5,10,50,100,500,1000,'>1000 edits']
'''Cohort definition
'''
self.cohort_labels = self.cohorts[:]
self.cohort_labels[0] = '%s edit'%(self.cohort_labels[0])
for i in range(1,len(self.cohorts)-1):
self.cohort_labels[i] = '%s-%s edits'%(self.cohorts[i-1]+1,self.cohorts[i])
# self.cohort_labels = ['<%s edits'%(e) for e in self.cohorts]
'''Cohort labels
'''
Cohort.__init__(self)
def __init__(self,activation=1):
# editor_types = ['Administrator','Bureaucrat','Eliminator','Huggler','Bot','Multiple','Other']
editor_types = ['Administrator','Huggler','Bot','Admin & Huggle','Other']
self.editor_types_dict = {a : i for i,a in enumerate(editor_types)}
'''Dictionary mapping editory type to numpy.array matrix
'''
self.cohorts = [int(i) for i in range(0,len(editor_types))]
'''Cohort definition
'''
self.cohort_labels = editor_types
'''Cohort labels
'''
#initialize helper structures
try:
from db import sql
cur = sql.getSSCursor()
# administrators
cur.execute("select user_id from ptwiki_p.user u join ptwiki_p.user_groups ug on (u.user_id=ug.ug_user) where ug.ug_group='sysop';")
self.administrators = set(i[0] for i in cur)
# bureaucrat
cur.execute("select user_id from ptwiki_p.user u join ptwiki_p.user_groups ug on (u.user_id=ug.ug_user) where ug.ug_group='bureaucrat';")
self.bureaucrats = set(i[0] for i in cur)
# eliminators
cur.execute("select user_id from ptwiki_p.user u join ptwiki_p.user_groups ug on (u.user_id=ug.ug_user) where ug.ug_group='eliminator';")
self.eliminators = set(i[0] for i in cur)
# hugglers
cur.execute("select user_id from u_declerambaul.ptwiki_huggle;")
self.hugglers = set(i[0] for i in cur)
# bots
cur.execute("select user_id from u_declerambaul.ptwiki_bots;")
self.bots = set(i[0] for i in cur)
self.sqlQuery = 'SELECT * FROM u_declerambaul.ptwiki_%s_editor_year_month;'%reverttype
except:
logging.error("Could not establish SQL connection to initialize RevertsByEditorType.")
Cohort.__init__(self)
def __init__(self,activitylevels=[5,10,100]):
self.cohorts = activitylevels
'''Cohort definition
'''
self.cohort_labels = ['%s+ edits'%l for l in self.cohorts]
'''Cohort labels
'''
self.sqlQuery = 'SELECT * FROM %s;'%tables.EDITOR_YEAR_MONTH_NS0_NOREDIRECT
'''The SQL query returns edit information for each editor for each ym she has edited.'''
# self.ncolors = utils.numberOfMonths(settings.time_stamps[0],settings.time_stamps[-1])/6
'''
Number of visible colors in the wikipride plots.
'''
Cohort.__init__(self)
def __init__(self):
self.cohorts = ('0', '1', '2', '3', '4', '5','other')
'''Cohort definition
'''
self.cohort_labels = ['%s namespace'% i for i in self.cohorts]
'''Cohort labels
'''
self.cohort_index = {'0':0, '1':1, '2':2, '3':3, '4':4, '5':5}
self.sqlQuery = 'SELECT * FROM %s;'%tables.EDITOR_YEAR_MONTH_NAMESPACE
'''The SQL query returns edit information for each editor for each ym she has edited.'''
Cohort.__init__(self)
def __init__(self, activitylevels=[5, 10, 100]):
self.cohorts = activitylevels
'''Cohort definition
'''
self.cohort_labels = ['%s+ edits' % l for l in self.cohorts]
'''Cohort labels
'''
self.sqlQuery = 'SELECT * FROM %s;' % tables.EDITOR_YEAR_MONTH_NS0_NOREDIRECT
'''The SQL query returns edit information for each editor for each ym she has edited.'''
# self.ncolors = utils.numberOfMonths(settings.time_stamps[0],settings.time_stamps[-1])/6
'''
Number of visible colors in the wikipride plots.
'''
Cohort.__init__(self)
def __init__(self):
# We don't take into consideration people who have 0 edits as the data is coded sparse and we don't have
# this information
# Note that len(self.cohorts) has to be equal the number of bins in the numpy.array
# self.cohorts = [1,3,5,7,10,20,40,60,80,100,200,500,1000,5000,10000,'>10000 edits']
self.cohorts = [1, 5, 10, 50, 100, 500, 1000, '>1000 edits']
'''Cohort definition
'''
self.cohort_labels = self.cohorts[:]
self.cohort_labels[0] = '%s edit' % (self.cohort_labels[0])
for i in range(1, len(self.cohorts) - 1):
self.cohort_labels[i] = '%s-%s edits' % (self.cohorts[i - 1] + 1,
self.cohorts[i])
# self.cohort_labels = ['<%s edits'%(e) for e in self.cohorts]
'''Cohort labels
'''
Cohort.__init__(self)
def __init__(self):
# We don't take into consideration people who have 0 edits as the data is coded sparse and we don't have
# this information
self.cohorts = [1, 5, 50, 100, 500, 1000, '>1000 edits']
'''Cohort definition
'''
self.cohort_labels = self.cohorts[:]
self.cohort_labels[0] = '%s edit' % (self.cohort_labels[0])
for i in range(1, len(self.cohorts) - 1):
self.cohort_labels[i] = '%s-%s edits' % (self.cohorts[i - 1] + 1,
self.cohorts[i])
# self.cohort_labels = ['<%s edits'%(e) for e in self.cohorts]
'''Cohort labels
'''
self.sqlQuery = 'SELECT * FROM %s;' % tables.EDITOR_YEAR_MONTH
'''The SQL query returns edit information for each editor for each ym she has edited.'''
Cohort.__init__(self)
def __init__(self):
# We don't take into consideration people who have 0 edits as the data is coded sparse and we don't have
# this information
self.cohorts = [1,5,50,100,500,1000,'>1000 edits']
'''Cohort definition
'''
self.cohort_labels = self.cohorts[:]
self.cohort_labels[0] = '%s edit'%(self.cohort_labels[0])
for i in range(1,len(self.cohorts)-1):
self.cohort_labels[i] = '%s-%s edits'%(self.cohorts[i-1]+1,self.cohorts[i])
# self.cohort_labels = ['<%s edits'%(e) for e in self.cohorts]
'''Cohort labels
'''
self.sqlQuery = 'SELECT * FROM %s;'%tables.EDITOR_YEAR_MONTH
'''The SQL query returns edit information for each editor for each ym she has edited.'''
Cohort.__init__(self)
def __init__(self):
self.cohorts = ['New Editors']
'''Cohort definition
'''
self.cohort_labels = self.cohorts
'''Cohort labels
'''
self.sqlQuery = """SELECT
first_edit_year,
first_edit_month,
count(*) AS recruits
FROM
%s
GROUP BY
first_edit_year,
first_edit_month;"""%tables.USER_COHORT
'''The SQL query returns the new editor count for each ym.'''
Cohort.__init__(self)
def __init__(self,minedits=1,maxedits=None):
self.cohorts = [int(i) for i in range(0,len(settings.time_stamps))]
'''Cohort definition
'''
self.cohort_labels = ['%s month old'% i for i in self.cohorts]
'''Cohort labels
'''
self.sqlQuery = 'SELECT * FROM %s;'%tables.EDITOR_YEAR_MONTH
'''The SQL query returns edit information for each editor for each ym she has edited.'''
self.minedits = minedits
'''Minimum number of edits by editor in a given month to be included'''
self.maxedits = maxedits
'''Maximum number of edits by editor in a given month to be included'''
self.ncolors = utils.numberOfMonths(settings.time_stamps[0],settings.time_stamps[-1])/6
'''
Number of visible colors in the wikipride plots. E.g. one color for every six month for wikipride plots
'''
Cohort.__init__(self)
def __init__(self,activation=5):
self.activation = activation
self.NS = ( '4', '5' )
ts,ts_i = utils.create_time_stamps_month(fromym='200401',toym='201012')
self.time_stamps = ts
self.time_stamps_index = ts_i
'''Only take time_stamps starting with self.year
'''
self.cohorts = range(2004,2011)
'''Cohort definition
'''
self.cohort_labels = ['%s cohort'% i for i in self.cohorts]
'''Cohort labels
'''
# self.sqlQuery = 'SELECT * FROM fabian WHERE namespace IN (4,5) AND add_edits > %s AND first_edit_year in (%s)'%(self.activation,','.join([str(c) for c in self.cohorts]))
Cohort.__init__(self)
def __init__(self,period=3 ):
# We don't take into consideration people who have 0 edits as the data is coded sparse and we don't have
# this information
# Note that len(self.cohorts) has to be equal the number of bins in the numpy.array
# self.cohorts = [1,3,5,7,10,20,40,60,80,100,200,500,1000,5000,10000,'>10000 edits']
self.cohorts = [1,5,10,50,100,500,1000,'>1000 edits']
'''Cohort definition
'''
self.cohort_labels = self.cohorts[:]
self.cohort_labels[0] = '%s edit'%(self.cohort_labels[0])
for i in range(1,len(self.cohorts)-1):
self.cohort_labels[i] = '%s-%s edits'%(self.cohorts[i-1]+1,self.cohorts[i])
# self.cohort_labels = ['<%s edits'%(e) for e in self.cohorts]
'''Cohort labels
'''
self.period = period
'''The number of month an editor is considered new
'''
self.old_user_id = None
'''The user_id of the previously encountered editor as we iterate through the table
'''
self.lastym = None
'''The ym at the end of the `period` months after the first edit of an editor
'''
self.firstedit = None
self.fe_index = None
self.edits = 0
self.added = 0
self.removed = 0
self.net = 0
Cohort.__init__(self)
def __init__(self):
if self.cohorts is None or self.cohort_labels is None:
logger.error("self.cohorts or self.cohort_labels not properly defined")
# raise Exception("self.cohorts or self.cohort_labels not properly defined")
Cohort.__init__(self)