troggle/databaseReset.py

import os
import time
import timeit
import settings
os.environ['PYTHONPATH'] = settings.PYTHON_PATH
os.environ.setdefault('DJANGO_SETTINGS_MODULE', 'settings')
from django.core import management
from django.db import connection
from django.contrib.auth.models import User
from django.http import HttpResponse
from django.core.urlresolvers import reverse
from troggle.core.models import Cave, Entrance
import troggle.flatpages.models
import json

databasename=settings.DATABASES['default']['NAME']
expouser=settings.EXPOUSER
expouserpass=settings.EXPOUSERPASS
expouseremail=settings.EXPOUSER_EMAIL

def reinit_db():
    """Rebuild database from scratch. Deletes the file first if sqlite is used,
    otherwise it drops the database and creates it.
    """
    if settings.DATABASES['default']['ENGINE'] == 'django.db.backends.sqlite3':
        try:
            os.remove(databasename)
        except OSError:
            pass
    else:
        cursor = connection.cursor()
        cursor.execute("DROP DATABASE %s" % databasename)
        cursor.execute("CREATE DATABASE %s" % databasename)
        cursor.execute("ALTER DATABASE %s CHARACTER SET=utf8" % databasename)
        cursor.execute("USE %s" % databasename)
    syncuser()

def syncuser():
    """Sync user - needed after reload
    """
    print("Synchronizing user")
    management.call_command('syncdb', interactive=False)
    user = User.objects.create_user(expouser, expouseremail, expouserpass)
    user.is_staff = True
    user.is_superuser = True
    user.save()

def dirsredirect():
    """Make directories that troggle requires and sets up page redirects
    """
    #should also deal with permissions here.
    if not os.path.isdir(settings.PHOTOS_ROOT):
        os.mkdir(settings.PHOTOS_ROOT)
    for oldURL, newURL in [("indxal.htm", reverse("caveindex"))]:
        f = troggle.flatpages.models.Redirect(originalURL = oldURL, newURL = newURL)
        f.save()

def import_caves():
    import parsers.caves
    print("Importing Caves")
    parsers.caves.readcaves()

def import_people():
    import parsers.people
    print("Importing People (folk.csv)")
    parsers.people.LoadPersonsExpos()

def import_logbooks():
    import parsers.logbooks
    print("Importing Logbooks")
    parsers.logbooks.LoadLogbooks()

def import_QMs():
    print("Importing QMs (old caves)")
    import parsers.QMs
    # import process itself runs on qm.csv in only 3 caves, not 264!
    
def import_survexblks():
    import parsers.survex
    print("Importing Survex Blocks")
    parsers.survex.LoadAllSurvexBlocks()

def import_survexpos(): 
    import parsers.survex
    print("Importing Survex x/y/z Positions")
    parsers.survex.LoadPos()

def import_surveyimgs():
    """This appears to store data in unused objects. The code is kept
    for future re-working to manage progress against notes, plans and elevs.
    """
    import parsers.surveys
    print("Importing survey images")
    parsers.surveys.parseSurveys(logfile=settings.LOGFILE)

def import_surveyscans():
    import parsers.surveys
    print("Importing Survey Scans")
    parsers.surveys.LoadListScans()

def import_tunnelfiles():
    import parsers.surveys
    print("Importing Tunnel files")
    parsers.surveys.LoadTunnelFiles()

# - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 
def import_auto_logbooks():
    import parsers.logbooks
    import os
    for pt in troggle.core.models.PersonTrip.objects.all():
        pt.delete()
    for lbe in troggle.core.models.LogbookEntry.objects.all():
        lbe.delete()
    for expedition in troggle.core.models.Expedition.objects.all():
        directory = os.path.join(settings.EXPOWEB,
                                 "years",
                                 expedition.year,
                                 "autologbook")
        for root, dirs, filenames in os.walk(directory):
            for filename in filenames:
                print(os.path.join(root, filename))
                parsers.logbooks.parseAutoLogBookEntry(os.path.join(root, filename))

#Temporary function until definitive source of data transfered.
from django.template.defaultfilters import slugify
from django.template import Context, loader
def dumplogbooks():
    def get_name(pe):
            if pe.nickname:
                return pe.nickname
            else:
                return pe.person.first_name
    for lbe in troggle.core.models.LogbookEntry.objects.all():
            dateStr = lbe.date.strftime("%Y-%m-%d")
            directory = os.path.join(settings.EXPOWEB, 
                                "years",
                                lbe.expedition.year, 
                                "autologbook")
            if not os.path.isdir(directory):
                os.mkdir(directory)
            filename = os.path.join(directory, 
                                    dateStr + "." + slugify(lbe.title)[:50] + ".html")
            if lbe.cave:
                print(lbe.cave.reference())
                trip = {"title": lbe.title, "html":lbe.text, "cave": lbe.cave.reference(), "caveOrLocation": "cave"}
            else:
                trip = {"title": lbe.title, "html":lbe.text, "location":lbe.place, "caveOrLocation": "location"}
            pts = [pt for pt in lbe.persontrip_set.all() if pt.personexpedition]
            persons = [{"name": get_name(pt.personexpedition), "TU": pt.time_underground, "author": pt.is_logbook_entry_author} for pt in pts]
            f = open(filename, "wb")
            template = loader.get_template('dataformat/logbookentry.html')
            context = Context({'trip': trip, 
                               'persons': persons,
                               'date': dateStr,
                               'expeditionyear': lbe.expedition.year})
            output = template.render(context)
            f.write(unicode(output).encode( "utf-8" ))
            f.close()
# - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 

class JobQueue():
    """A list of import operations to run. Always reports times
    in the same order. 
    """
    def __init__(self,run):
        self.runlabel = run
        self.queue = [] # tuples of (jobname, jobfunction)
        self.results = {}
        self.results_order=[
            "date","runlabel","reinit", "caves", "people",
            "logbooks", "QMs", "survexblks", "survexpos",
            "tunnel", "scans", "surveyimgs", "test", "dirsredirect", "syncuser"  ]
        for k in self.results_order:
            self.results[k]=[]
        self.tfile = "import_profile.json"
        self.htmlfile = "profile.html"

    #Adding elements to queue - enqueue
    def enq(self,label,func):
        self.queue.append((label,func))
        return True

    #Removing the last element from the queue - dequeue
    # def deq(self):
    #     if len(self.queue)>0:
    #         return self.queue.pop()
    #     return ("Queue Empty!")

    def run(self):
        if os.path.isfile(self.tfile):
            try:
                f = open(self.tfile, "r")
                data = json.load(f)
                for j in data:
                    self.results[j] = data[j]
            except:
                print "FAILURE parsing JSON file %s" % (self.tfile)
                # Python bug: https://github.com/ShinNoNoir/twitterwebsearch/issues/12
            f.close()
        
        for j in self.results_order:
            self.results[j].append(None) # append a placeholder
        
        print "** Running job ", self.runlabel
        jobstart = time.time()
        self.results["date"].pop()
        self.results["date"].append(jobstart)
        self.results["runlabel"].pop()
        self.results["runlabel"].append(self.runlabel)

        for i in self.queue:
            start = time.time()
            i[1]()    #  looks ugly but invokes function passed in the second item in the tuple
            duration = time.time()-start
            print "\n*- Ended \"",  i[0], "\"  %.1f seconds" % duration
            self.results[i[0]].pop()  # the null item
            self.results[i[0]].append(duration)
               
        with open(self.tfile, 'w') as f:
            json.dump(self.results, f)     

        jobend = time.time()
        jobduration = jobend-jobstart
        print "** Ended all jobs. %.1f seconds" % jobduration

        # currently uses django db whatever it was. CHANGE this to explicitly use
        # a new sqlite3 db and then import the sql dump of that into the troggle db
        # instead of loading directly into the troggle sqlite db.
        # in-memory ":memory:" sqlite is ~ 7x faster and all of troggle can be
        # loaded in 6 minutes that way
        djconn = django.db.connection
        from dump import _iterdump
        with open('memdump.sql', 'w') as f:
            for line in _iterdump(djconn):
                f.write('%s\n' % line.encode("utf8"))

        # now import the memory image sql into   
        ####(to do)   
        return True

    def showprofile(self):
        """Prints out the time it took to run the jobqueue"""
        for k in self.results_order:
            if k =="dirsredirect":
                break
            elif k =="syncuser":
                break
            elif k =="test":
                break
            elif k =="date":
                print "     days ago ",
            else:
                print '%10s (s)' % k,
            percen=0
            r = self.results[k]  
            #print "min=",min
            
            for i in range(len(r)):
                if k == "runlabel": 
                    if r[i]:
                        rp =   r[i]
                    else:
                        rp = "      - "
                    print '%8s' % rp,
                elif k =="date":
                    # Calculate dates as days before present
                    if r[i]:
                        if i == len(r)-1:
                            print "    this",
                        else:
                            # prints one place to the left of where you expect
                            days = (r[i]-r[len(r)-1])/(24*60*60)
                            print '%8.2f' % days,
                elif r[i]: 
                    print '%8.1f' % r[i],
                    if i == len(r)-1 and r[i-1]:
                        percen = 100* (r[i] - r[i-1])/r[i-1]
                        if abs(percen) >0.1:
                            print '%8.1f%%' % percen,
                        else:
                            print "      -  ",
            print ""
        return True


def usage():
    print("""Usage is 'python databaseReset.py <command> [runlabel]'
             where command is:
             reset     - normal usage: clear database and reread everything from files - time-consuming
             caves     - read in the caves 
             logbooks  - read in the logbooks
             people    - read in the people from folk.csv
             QMs       - read in the QM csv files (older caves only)
             reinit    - clear database (delete everything) and make empty tables. Import nothing.
             scans     - the survey scans in all the wallets
             survex    - read in the survex files - all the survex blocks but not the x/y/z positions
             survexpos - just the x/y/z Pos out of the survex files 

             tunnel    - read in the Tunnel files - which scans the survey scans too

             resetend     - (archaic?)
             writecaves   - *disabled* (archaic?)
             autologbooks - read in autologbooks (what are these?)
             dumplogbooks - write out autologbooks (not working?)
             syncuser     - needed after reloading database from SQL backup
             surveyimgs   - read in scans by expo, must run after "people". Not used.
             test         - testing...

             and [runlabel] is an optional string identifying this run of the script
             in the stored profiling data 'import-profile.json'
             
             caves and logbooks must be run on an empty db before the others as they
             set up db tables used by the others.
             """)

if __name__ == "__main__":
    import troggle.core.models
    import sys
    import django
    django.setup()

    if len(sys.argv)>2:
        runlabel = sys.argv[len(sys.argv)-1]
    else: 
        runlabel=None
        
    jq = JobQueue(runlabel)
    
    if len(sys.argv)==1:
        usage()
        exit()
    elif "test" in sys.argv:
        jq.enq("reinit",reinit_db)
        jq.enq("dirsredirect",dirsredirect)
        jq.enq("caves",import_caves)
        jq.enq("survexblks",import_survexblks)
        jq.enq("survexpos",import_survexpos)
    elif "caves" in sys.argv:
        jq.enq("caves",import_caves)
    elif "logbooks" in sys.argv:
        jq.enq("logbooks",import_logbooks)
    elif "people" in sys.argv:
        jq.enq("people",import_people)
    elif "QMs" in sys.argv:
        jq.enq("QMs",import_QMs)
    elif "reload_db" in sys.argv:
        jq.enq("reload",reload_db)
    elif "reset" in sys.argv:
        jq.enq("reinit",reinit_db)
        jq.enq("dirsredirect",dirsredirect)
        jq.enq("caves",import_caves)
        jq.enq("people",import_people)
        jq.enq("scans",import_surveyscans)
        jq.enq("logbooks",import_logbooks)
        jq.enq("QMs",import_QMs)
        jq.enq("survexblks",import_survexblks)
        jq.enq("survexpos",import_survexpos)
        jq.enq("tunnel",import_tunnelfiles)
    elif "scans" in sys.argv:
        jq.enq("scans",import_surveyscans)
    elif "survex" in sys.argv:
        jq.enq("survexblks",import_survexblks)
        jq.enq("survexpos",import_survexpos)
    elif "survexpos" in sys.argv:
        jq.enq("survexpos",import_survexpos)
    elif "surveys" in sys.argv:
        jq.enq("surveyimgs",import_surveyimgs)
    elif "tunnel" in sys.argv:
        jq.enq("tunnel",import_tunnelfiles)
    elif "help" in sys.argv:
        usage()
    elif "resetend" in sys.argv:
        jq.enq("QMs",import_QMs)
        jq.enq("tunnel",import_tunnelfiles)
        jq.enq("surveyimgs",import_surveyimgs)
        #import_descriptions() # no longer present
        #parse_descriptions() # no longer present
    # elif "writeCaves" in sys.argv:
        # writeCaves() # no longer present
    elif "autologbooks" in sys.argv:
        import_auto_logbooks()
    elif "dumplogbooks" in sys.argv:
        dumplogbooks()
    else:
        usage()
        print("%s not recognised as a command." % sys.argv[1])
        exit()

    jq.run()
    jq.showprofile()