2009-05-13 05:35:59 +01:00
#.-*- coding: utf-8 -*-
2009-07-02 22:31:28 +01:00
from django . conf import settings
import core . models as models
import csv , re , datetime , os , shutil
2009-07-03 05:31:49 +01:00
from utils import save_carefully
2009-05-13 05:53:37 +01:00
def saveMugShot ( mugShotPath , mugShotFilename , person ) :
if mugShotFilename . startswith ( r ' i/ ' ) : #if filename in cell has the directory attached (I think they all do), remove it
mugShotFilename = mugShotFilename [ 2 : ]
else :
mugShotFilename = mugShotFilename # just in case one doesn't
2009-05-19 06:32:42 +01:00
dummyObj = models . Photo ( file = mugShotFilename )
#Put a copy of the file in the right place. mugShotObj.file.path is determined by the django filesystemstorage specified in models.py
if not os . path . exists ( dummyObj . file . path ) :
shutil . copy ( mugShotPath , dummyObj . file . path )
mugShotObj , created = save_carefully (
models . Photo ,
lookupAttribs = { ' is_mugshot ' : True , ' file ' : mugShotFilename } ,
nonLookupAttribs = { ' caption ' : " Mugshot for " + person . first_name + " " + person . last_name }
)
2009-05-13 05:53:37 +01:00
2009-05-19 06:32:42 +01:00
if created :
mugShotObj . contains_person . add ( person )
mugShotObj . save ( )
2009-05-13 05:53:37 +01:00
def parseMugShotAndBlurb ( personline , header , person ) :
2009-07-03 05:31:49 +01:00
""" create mugshot Photo instance """
2009-05-13 05:53:37 +01:00
mugShotFilename = personline [ header [ " Mugshot " ] ]
mugShotPath = os . path . join ( settings . EXPOWEB , " folk " , mugShotFilename )
2009-05-13 05:35:59 +01:00
if mugShotPath [ - 3 : ] == ' jpg ' : #if person just has an image, add it
2009-05-13 05:53:37 +01:00
saveMugShot ( mugShotPath = mugShotPath , mugShotFilename = mugShotFilename , person = person )
2009-05-13 05:35:59 +01:00
elif mugShotPath [ - 3 : ] == ' htm ' : #if person has an html page, find the image(s) and add it. Also, add the text from the html page to the "blurb" field in his model instance.
personPageOld = open ( mugShotPath , ' r ' ) . read ( )
2009-05-19 06:32:42 +01:00
if not person . blurb :
person . blurb = re . search ( ' <body>.*<hr ' , personPageOld , re . DOTALL ) . group ( ) #this needs to be refined, take care of the HTML and make sure it doesn't match beyond the blurb
for mugShotFilename in re . findall ( ' i/.*?jpg ' , personPageOld , re . DOTALL ) :
mugShotPath = os . path . join ( settings . EXPOWEB , " folk " , mugShotFilename )
saveMugShot ( mugShotPath = mugShotPath , mugShotFilename = mugShotFilename , person = person )
2009-05-13 05:53:37 +01:00
person . save ( )
2009-05-13 05:35:59 +01:00
def LoadPersonsExpos ( ) :
persontab = open ( os . path . join ( settings . EXPOWEB , " noinfo " , " folk.csv " ) )
personreader = csv . reader ( persontab )
headers = personreader . next ( )
header = dict ( zip ( headers , range ( len ( headers ) ) ) )
2009-05-13 05:48:47 +01:00
# make expeditions
print " Loading expeditions "
2009-05-13 05:35:59 +01:00
years = headers [ 5 : ]
2009-05-13 05:53:37 +01:00
2009-05-13 05:35:59 +01:00
for year in years :
2009-05-19 06:32:42 +01:00
lookupAttribs = { ' year ' : year }
nonLookupAttribs = { ' name ' : " CUCC expo %s " % year }
save_carefully ( models . Expedition , lookupAttribs , nonLookupAttribs )
2009-05-13 05:35:59 +01:00
2009-05-13 05:48:47 +01:00
# make persons
print " Loading personexpeditions "
2009-05-13 06:15:48 +01:00
#expoers2008 = """Edvin Deadman,Kathryn Hopkins,Djuke Veldhuis,Becka Lawson,Julian Todd,Natalie Uomini,Aaron Curtis,Tony Rooke,Ollie Stevens,Frank Tully,Martin Jahnke,Mark Shinwell,Jess Stirrups,Nial Peters,Serena Povia,Olly Madge,Steve Jones,Pete Harley,Eeva Makiranta,Keith Curtis""".split(",")
#expomissing = set(expoers2008)
2009-05-13 05:35:59 +01:00
2009-05-13 05:48:47 +01:00
for personline in personreader :
name = personline [ header [ " Name " ] ]
2009-05-13 05:35:59 +01:00
name = re . sub ( " <.*?> " , " " , name )
mname = re . match ( " ( \ w+)(?: \ s((?:van |ten )? \ w+))?(?: \ s \ (([^)]*) \ ))? " , name )
2009-05-13 05:48:47 +01:00
nickname = mname . group ( 3 ) or " "
2009-05-13 06:15:48 +01:00
lookupAttribs = { ' first_name ' : mname . group ( 1 ) , ' last_name ' : ( mname . group ( 2 ) or " " ) }
nonLookupAttribs = { ' is_vfho ' : personline [ header [ " VfHO member " ] ] , }
2009-05-19 06:32:42 +01:00
person , created = save_carefully ( models . Person , lookupAttribs , nonLookupAttribs )
2009-05-13 06:15:48 +01:00
2009-05-13 06:02:42 +01:00
parseMugShotAndBlurb ( personline = personline , header = header , person = person )
2009-05-13 05:35:59 +01:00
2009-05-13 05:48:47 +01:00
# make person expedition from table
for year , attended in zip ( headers , personline ) [ 5 : ] :
expedition = models . Expedition . objects . get ( year = year )
2009-05-13 05:35:59 +01:00
if attended == " 1 " or attended == " -1 " :
2009-05-19 06:32:42 +01:00
lookupAttribs = { ' person ' : person , ' expedition ' : expedition }
nonLookupAttribs = { ' nickname ' : nickname , ' is_guest ' : ( personline [ header [ " Guest " ] ] == " 1 " ) }
save_carefully ( models . PersonExpedition , lookupAttribs , nonLookupAttribs )
2009-05-13 05:35:59 +01:00
2009-05-13 05:53:37 +01:00
2009-05-13 05:35:59 +01:00
# this fills in those people for whom 2008 was their first expo
2009-05-13 06:15:48 +01:00
#print "Loading personexpeditions 2008"
#for name in expomissing:
# firstname, lastname = name.split()
# is_guest = name in ["Eeva Makiranta", "Keith Curtis"]
# print "2008:", name
# persons = list(models.Person.objects.filter(first_name=firstname, last_name=lastname))
# if not persons:
# person = models.Person(first_name=firstname, last_name = lastname, is_vfho = False, mug_shot = "")
# #person.Sethref()
# person.save()
# else:
# person = persons[0]
# expedition = models.Expedition.objects.get(year="2008")
# personexpedition = models.PersonExpedition(person=person, expedition=expedition, nickname="", is_guest=is_guest)
# personexpedition.save()
2009-05-13 05:48:47 +01:00
2009-05-13 05:53:37 +01:00
#Notability is now a method of person. Makes no sense to store it in the database; it would need to be recalculated every time something changes. - AC 16 Feb 09
2009-05-13 05:48:47 +01:00
# could rank according to surveying as well
2009-05-13 05:53:37 +01:00
#print "Setting person notability"
#for person in models.Person.objects.all():
#person.notability = 0.0
#for personexpedition in person.personexpedition_set.all():
#if not personexpedition.is_guest:
#person.notability += 1.0 / (2012 - int(personexpedition.expedition.year))
#person.bisnotable = person.notability > 0.3 # I don't know how to filter by this
#person.save()
2009-05-13 05:48:47 +01:00
# used in other referencing parser functions
2009-05-13 05:39:52 +01:00
# expedition name lookup cached for speed (it's a very big list)
Gpersonexpeditionnamelookup = { }
def GetPersonExpeditionNameLookup ( expedition ) :
global Gpersonexpeditionnamelookup
res = Gpersonexpeditionnamelookup . get ( expedition . name )
if res :
return res
2009-05-13 05:48:47 +01:00
res = { }
2009-05-13 05:39:52 +01:00
duplicates = set ( )
2009-05-13 05:48:47 +01:00
print " Calculating GetPersonExpeditionNameLookup for " , expedition . year
2009-05-13 05:39:52 +01:00
personexpeditions = models . PersonExpedition . objects . filter ( expedition = expedition )
for personexpedition in personexpeditions :
possnames = [ ]
f = personexpedition . person . first_name . lower ( )
l = personexpedition . person . last_name . lower ( )
if l :
possnames . append ( f + " " + l )
possnames . append ( f + " " + l [ 0 ] )
possnames . append ( f + l [ 0 ] )
possnames . append ( f [ 0 ] + " " + l )
possnames . append ( f )
if personexpedition . nickname :
possnames . append ( personexpedition . nickname . lower ( ) )
for possname in possnames :
if possname in res :
duplicates . add ( possname )
else :
res [ possname ] = personexpedition
for possname in duplicates :
del res [ possname ]
Gpersonexpeditionnamelookup [ expedition . name ] = res
return res
