mirror of
https://github.com/ankitects/anki.git
synced 2025-09-19 14:32:22 -04:00
2f27133705
SQLAlchemy is a great tool, but it wasn't a great fit for Anki: - We often had to drop down to raw SQL for performance reasons. - The DB cursors and results were wrapped, which incurred a sizable performance hit due to introspection. Operations like fetching 50k records from a hot cache were taking more than twice as long to complete. - We take advantage of sqlite-specific features, so SQL language abstraction is useless to us. - The anki schema is quite small, so manually saving and loading objects is not a big burden. In the process of porting to DBAPI, I've refactored the database schema: - App configuration data that we don't need in joins or bulk updates has been moved into JSON objects. This simplifies serializing, and means we won't need DB schema changes to store extra options in the future. This change obsoletes the deckVars table. - Renamed tables: -- fieldModels -> fields -- cardModels -> templates -- fields -> fdata - a number of attribute names have been shortened Classes like Card, Fact & Model remain. They maintain a reference to the deck. To write their state to the DB, call .flush(). Objects no longer have their modification time manually updated. Instead, the modification time is updated when they are flushed. This also applies to the deck. Decks will now save on close, because various operations that were done at deck load will be moved into deck close instead. Operations like undoing buried card are cheap on a hot cache, but expensive on startup. Programmatically you can call .close(save=False) to avoid a save and a modification bump. This will be useful for generating due counts. Because of the new saving behaviour, the save and save as options will be removed from the GUI in the future. The q/a cache and field cache generating has been centralized. Facts will automatically rebuild the cache on flush; models can do so with model.updateCache(). Media handling has also been reworked. It has moved into a MediaRegistry object, which the deck holds. Refcounting has been dropped - it meant we had to compare old and new value every time facts or models were changed, and existed for the sole purpose of not showing errors on a missing media download. Instead we just media.registerText(q+a) when it's updated. The download function will be expanded to ask the user if they want to continue after a certain number of files have failed to download, which should be an adequate alternative. And we now add the file into the media DB when it's copied to th emedia directory, not when the card is commited. This fixes duplicates a user would get if they added the same media to a card twice without adding the card. The old DeckStorage object had its upgrade code split in a previous commit; the opening and upgrading code has been merged back together, and put in a separate storage.py file. The correct way to open a deck now is import anki; d = anki.Deck(path). deck.getCard() -> deck.sched.getCard() same with answerCard deck.getCard(id) returns a Card object now. And the DB wrapper has had a few changes: - sql statements are a more standard DBAPI: - statement() -> execute() - statements() -> executemany() - called like execute(sql, 1, 2, 3) or execute(sql, a=1, b=2, c=3) - column0 -> list
556 lines
19 KiB
Python
556 lines
19 KiB
Python
# -*- coding: utf-8 -*-
|
|
# Copyright: Damien Elmes <anki@ichi2.net>
|
|
# License: GNU GPL, version 3 or later; http://www.gnu.org/copyleft/gpl.html
|
|
|
|
import time, datetime, simplejson, random
|
|
from operator import itemgetter
|
|
from heapq import *
|
|
#from anki.cards import Card
|
|
from anki.utils import parseTags, ids2str
|
|
from anki.lang import _
|
|
from anki.consts import *
|
|
|
|
# the standard Anki scheduler
|
|
class Scheduler(object):
|
|
def __init__(self, deck):
|
|
self.deck = deck
|
|
self.db = deck.db
|
|
self.name = "main"
|
|
self.queueLimit = 200
|
|
self.learnLimit = 1000
|
|
self.updateCutoff()
|
|
|
|
def getCard(self):
|
|
"Pop the next card from the queue. None if finished."
|
|
self.checkDay()
|
|
id = self.getCardId()
|
|
if id:
|
|
return self.deck.getCard(id)
|
|
|
|
def reset(self):
|
|
self.resetConf()
|
|
t = time.time()
|
|
self.resetLearn()
|
|
print "lrn %0.2fms" % ((time.time() - t)*1000); t = time.time()
|
|
self.resetReview()
|
|
print "rev %0.2fms" % ((time.time() - t)*1000); t = time.time()
|
|
self.resetNew()
|
|
print "new %0.2fms" % ((time.time() - t)*1000); t = time.time()
|
|
|
|
def answerCard(self, card, ease):
|
|
if card.queue == 0:
|
|
self.answerLearnCard(card, ease)
|
|
elif card.queue == 1:
|
|
self.revCount -= 1
|
|
self.answerRevCard(card, ease)
|
|
elif card.queue == 2:
|
|
# put it in the learn queue
|
|
card.queue = 0
|
|
self.newCount -= 1
|
|
self.answerLearnCard(card, ease)
|
|
else:
|
|
raise Exception("Invalid queue")
|
|
card.flushSched()
|
|
|
|
def counts(self):
|
|
# FIXME: should learn count include new cards due today, or be separate?
|
|
return (self.learnCount, self.revCount)
|
|
|
|
def cardQueue(self, card):
|
|
return card.queue
|
|
|
|
# Getting the next card
|
|
##########################################################################
|
|
|
|
def getCardId(self):
|
|
"Return the next due card id, or None."
|
|
# learning card due?
|
|
id = self.getLearnCard()
|
|
if id:
|
|
return id
|
|
# new first, or time for one?
|
|
if self.timeForNewCard():
|
|
return self.getNewCard()
|
|
# card due for review?
|
|
id = self.getReviewCard()
|
|
if id:
|
|
return id
|
|
# new cards left?
|
|
id = self.getNewCard()
|
|
if id:
|
|
return id
|
|
# collapse or finish
|
|
return self.getLearnCard(collapse=True)
|
|
|
|
# New cards
|
|
##########################################################################
|
|
|
|
# need to keep track of reps for timebox and new card introduction
|
|
|
|
def resetNew(self):
|
|
l = self.deck.qconf
|
|
if l['newToday'][0] != self.today:
|
|
# it's a new day; reset counts
|
|
l['newToday'] = [self.today, 0]
|
|
lim = min(self.queueLimit, l['newPerDay'] - l['newToday'][1])
|
|
if lim <= 0:
|
|
self.newQueue = []
|
|
self.newCount = 0
|
|
else:
|
|
self.newQueue = self.db.all("""
|
|
select id %s from cards where
|
|
queue = 2 %s order by due limit %d""" % (self.newOrder(), self.groupLimit('new'),
|
|
lim))
|
|
self.newQueue.sort(key=itemgetter(1), reverse=True)
|
|
self.newCount = len(self.newQueue)
|
|
self.updateNewCardRatio()
|
|
|
|
def getNewCard(self):
|
|
if self.newQueue:
|
|
return self.newQueue.pop()[0]
|
|
|
|
def newOrder(self):
|
|
return (",ord", "")[self.deck.qconf['newTodayOrder']]
|
|
|
|
def updateNewCardRatio(self):
|
|
if self.deck.qconf['newCardSpacing'] == NEW_CARDS_DISTRIBUTE:
|
|
if self.newCount:
|
|
self.newCardModulus = (
|
|
(self.newCount + self.revCount) / self.newCount)
|
|
# if there are cards to review, ensure modulo >= 2
|
|
if self.revCount:
|
|
self.newCardModulus = max(2, self.newCardModulus)
|
|
return
|
|
self.newCardModulus = 0
|
|
|
|
def timeForNewCard(self):
|
|
"True if it's time to display a new card when distributing."
|
|
if not self.newCount:
|
|
return False
|
|
if self.deck.qconf['newCardSpacing'] == NEW_CARDS_LAST:
|
|
return False
|
|
elif self.deck.qconf['newCardSpacing'] == NEW_CARDS_FIRST:
|
|
return True
|
|
elif self.newCardModulus:
|
|
return self.deck.reps and self.deck.reps % self.newCardModulus == 0
|
|
|
|
# Learning queue
|
|
##########################################################################
|
|
|
|
def resetLearn(self):
|
|
self.learnQueue = self.db.all("""
|
|
select due, id from cards where
|
|
queue = 0 and due < :lim order by due
|
|
limit %d""" % self.learnLimit, lim=self.dayCutoff)
|
|
self.learnCount = len(self.learnQueue)
|
|
|
|
def getLearnCard(self, collapse=False):
|
|
if self.learnQueue:
|
|
cutoff = time.time()
|
|
if collapse:
|
|
cutoff -= self.deck.collapseTime
|
|
if self.learnQueue[0][0] < cutoff:
|
|
return heappop(self.learnQueue)[1]
|
|
|
|
def answerLearnCard(self, card, ease):
|
|
# ease 1=no, 2=yes, 3=remove
|
|
conf = self.learnConf(card)
|
|
if ease == 3:
|
|
self.removeLearnCard(card, conf)
|
|
return
|
|
card.cycles += 1
|
|
if ease == 2:
|
|
card.grade += 1
|
|
else:
|
|
card.grade = 0
|
|
if card.grade >= len(conf['delays']):
|
|
self.graduateLearnCard(card, conf)
|
|
else:
|
|
card.due = time.time() + conf['delays'][card.grade]*60
|
|
|
|
def learnConf(self, card):
|
|
conf = self.confForCard(card)
|
|
if card.type == 2:
|
|
return conf['new']
|
|
else:
|
|
return conf['lapse']
|
|
|
|
def removeLearnCard(self, card, conf):
|
|
if card.type == 1:
|
|
int_ = None
|
|
elif not card.cycles:
|
|
# first time bonus
|
|
int_ = conf['ints'][1]
|
|
else:
|
|
# normal remove
|
|
int_ = conf['ints'][2]
|
|
self.rescheduleAsReview(card, conf, int_)
|
|
|
|
def graduateLearnCard(self, card, conf):
|
|
if card.type == 1:
|
|
int_ = None
|
|
else:
|
|
int_ = conf['ints'][0]
|
|
self.rescheduleAsReview(card, conf, int_)
|
|
|
|
def rescheduleAsReview(self, card, conf, int_):
|
|
card.queue = 1
|
|
card.factor = conf['initialFactor']
|
|
if int_:
|
|
# new card
|
|
card.type = 1
|
|
card.interval = int_
|
|
print "handle log, etc"
|
|
|
|
# Reviews
|
|
##########################################################################
|
|
|
|
def resetReview(self):
|
|
self.revQueue = self.db.all("""
|
|
select id from cards where
|
|
queue = 1 %s and due < :lim order by %s limit %d""" % (
|
|
self.groupLimit("rev"), self.revOrder(), self.queueLimit),
|
|
lim=self.dayCutoff)
|
|
if self.deck.qconf['revCardOrder'] == REV_CARDS_RANDOM:
|
|
random.shuffle(self.revQueue)
|
|
else:
|
|
self.revQueue.reverse()
|
|
self.revCount = len(self.revQueue)
|
|
|
|
def getReviewCard(self):
|
|
if self.haveRevCards():
|
|
return self.revQueue.pop()
|
|
|
|
def haveRevCards(self):
|
|
if self.revCount:
|
|
if not self.revQueue:
|
|
self.fillRevQueue()
|
|
return self.revQueue
|
|
|
|
def revOrder(self):
|
|
return ("interval desc",
|
|
"interval",
|
|
"due")[self.deck.qconf['revCardOrder']]
|
|
|
|
# FIXME: rewrite
|
|
def showFailedLast(self):
|
|
return self.collapseTime or not self.delay0
|
|
|
|
# Answering a card
|
|
##########################################################################
|
|
|
|
def _answerCard(self, card, ease):
|
|
undoName = _("Answer Card")
|
|
self.setUndoStart(undoName)
|
|
now = time.time()
|
|
# old state
|
|
oldState = self.cardState(card)
|
|
oldQueue = self.cardQueue(card)
|
|
lastDelaySecs = time.time() - card.due
|
|
lastDelay = lastDelaySecs / 86400.0
|
|
oldSuc = card.successive
|
|
# update card details
|
|
last = card.interval
|
|
card.interval = self.nextInterval(card, ease)
|
|
card.lastInterval = last
|
|
if card.reps:
|
|
# only update if card was not new
|
|
card.lastDue = card.due
|
|
card.due = self.nextDue(card, ease, oldState)
|
|
if not self.finishScheduler:
|
|
# don't update factor in custom schedulers
|
|
self.updateFactor(card, ease)
|
|
# spacing
|
|
self.spaceCards(card)
|
|
# adjust counts for current card
|
|
if ease == 1:
|
|
if card.due < self.dayCutoff:
|
|
self.learnCount += 1
|
|
if oldQueue == 0:
|
|
self.learnCount -= 1
|
|
elif oldQueue == 1:
|
|
self.revCount -= 1
|
|
else:
|
|
self.newAvail -= 1
|
|
# card stats
|
|
self.updateCardStats(card, ease, oldState)
|
|
# update type & ensure past cutoff
|
|
card.type = self.cardType(card)
|
|
card.queue = card.type
|
|
if ease != 1:
|
|
card.due = max(card.due, self.dayCutoff+1)
|
|
# allow custom schedulers to munge the card
|
|
if self.answerPreSave:
|
|
self.answerPreSave(card, ease)
|
|
# save
|
|
card.due = card.due
|
|
card.saveSched()
|
|
# review history
|
|
print "make sure flags is set correctly when reviewing early"
|
|
logReview(self.db, card, ease, 0)
|
|
self.modified = now
|
|
# leech handling - we need to do this after the queue, as it may cause
|
|
# a reset()
|
|
isLeech = self.isLeech(card)
|
|
if isLeech:
|
|
self.handleLeech(card)
|
|
runHook("cardAnswered", card.id, isLeech)
|
|
self.setUndoEnd(undoName)
|
|
|
|
def updateCardStats(self, card, ease, state):
|
|
card.reps += 1
|
|
if ease == 1:
|
|
card.successive = 0
|
|
card.lapses += 1
|
|
else:
|
|
card.successive += 1
|
|
# if not card.firstAnswered:
|
|
# card.firstAnswered = time.time()
|
|
|
|
def spaceCards(self, card):
|
|
new = time.time() + self.newSpacing
|
|
self.db.execute("""
|
|
update cards set
|
|
due = (case
|
|
when queue = 1 then due + 86400 * (case
|
|
when interval*:rev < 1 then 0
|
|
else interval*:rev
|
|
end)
|
|
when queue = 2 then :new
|
|
end),
|
|
modified = :now
|
|
where id != :id and fid = :fid
|
|
and due < :cut
|
|
and queue between 1 and 2""",
|
|
id=card.id, now=time.time(), fid=card.fid,
|
|
cut=self.dayCutoff, new=new, rev=self.revSpacing)
|
|
# update local cache of seen facts
|
|
self.spacedFacts[card.fid] = new
|
|
|
|
# Interval management
|
|
##########################################################################
|
|
|
|
def nextInterval(self, card, ease):
|
|
"Return the next interval for CARD given EASE."
|
|
delay = self.adjustedDelay(card, ease)
|
|
return self._nextInterval(card, delay, ease)
|
|
|
|
def _nextInterval(self, card, delay, ease):
|
|
interval = card.interval
|
|
factor = card.factor
|
|
# if cramming / reviewing early
|
|
if delay < 0:
|
|
interval = max(card.lastInterval, card.interval + delay)
|
|
if interval < self.midIntervalMin:
|
|
interval = 0
|
|
delay = 0
|
|
# if interval is less than mid interval, use presets
|
|
if ease == 1:
|
|
interval *= self.delay2
|
|
if interval < self.hardIntervalMin:
|
|
interval = 0
|
|
elif interval == 0:
|
|
if ease == 2:
|
|
interval = random.uniform(self.hardIntervalMin,
|
|
self.hardIntervalMax)
|
|
elif ease == 3:
|
|
interval = random.uniform(self.midIntervalMin,
|
|
self.midIntervalMax)
|
|
elif ease == 4:
|
|
interval = random.uniform(self.easyIntervalMin,
|
|
self.easyIntervalMax)
|
|
else:
|
|
# if not cramming, boost initial 2
|
|
if (interval < self.hardIntervalMax and
|
|
interval > 0.166):
|
|
mid = (self.midIntervalMin + self.midIntervalMax) / 2.0
|
|
interval = mid / factor
|
|
# multiply last interval by factor
|
|
if ease == 2:
|
|
interval = (interval + delay/4) * 1.2
|
|
elif ease == 3:
|
|
interval = (interval + delay/2) * factor
|
|
elif ease == 4:
|
|
interval = (interval + delay) * factor * self.factorFour
|
|
fuzz = random.uniform(0.95, 1.05)
|
|
interval *= fuzz
|
|
return interval
|
|
|
|
def nextIntervalStr(self, card, ease, short=False):
|
|
"Return the next interval for CARD given EASE as a string."
|
|
int = self.nextInterval(card, ease)
|
|
return anki.utils.fmtTimeSpan(int*86400, short=short)
|
|
|
|
def nextDue(self, card, ease, oldState):
|
|
"Return time when CARD will expire given EASE."
|
|
if ease == 1:
|
|
# 600 is a magic value which means no bonus, and is used to ease
|
|
# upgrades
|
|
cram = self.scheduler == "cram"
|
|
if (not cram and oldState == "mature"
|
|
and self.delay1 and self.delay1 != 600):
|
|
# user wants a bonus of 1+ days. put the failed cards at the
|
|
# start of the future day, so that failures that day will come
|
|
# after the waiting cards
|
|
return self.dayCutoff + (self.delay1 - 1)*86400
|
|
else:
|
|
due = 0
|
|
else:
|
|
due = card.interval * 86400.0
|
|
return due + time.time()
|
|
|
|
def updateFactor(self, card, ease):
|
|
"Update CARD's factor based on EASE."
|
|
print "update cardIsBeingLearnt()"
|
|
if not card.reps:
|
|
# card is new, inherit beginning factor
|
|
card.factor = self.averageFactor
|
|
if card.successive and not self.cardIsBeingLearnt(card):
|
|
if ease == 1:
|
|
card.factor -= 0.20
|
|
elif ease == 2:
|
|
card.factor -= 0.15
|
|
if ease == 4:
|
|
card.factor += 0.10
|
|
card.factor = max(1.3, card.factor)
|
|
|
|
def adjustedDelay(self, card, ease):
|
|
"Return an adjusted delay value for CARD based on EASE."
|
|
if self.cardIsNew(card):
|
|
return 0
|
|
if card.due <= self.dayCutoff:
|
|
return (self.dayCutoff - card.due) / 86400.0
|
|
else:
|
|
return (self.dayCutoff - card.due) / 86400.0
|
|
|
|
# Leeches
|
|
##########################################################################
|
|
|
|
def isLeech(self, card):
|
|
no = card.lapses
|
|
fmax = self.getInt('leechFails')
|
|
if not fmax:
|
|
return
|
|
return (
|
|
# failed
|
|
not card.successive and
|
|
# greater than fail threshold
|
|
no >= fmax and
|
|
# at least threshold/2 reps since last time
|
|
(fmax - no) % (max(fmax/2, 1)) == 0)
|
|
|
|
def handleLeech(self, card):
|
|
scard = self.cardFromId(card.id, True)
|
|
tags = scard.fact.tags
|
|
tags = addTags("Leech", tags)
|
|
scard.fact.tags = canonifyTags(tags)
|
|
scard.fact.setModified(textChanged=True, deck=self)
|
|
self.updateFactTags([scard.fact.id])
|
|
self.db.expunge(scard)
|
|
if self.getBool('suspendLeeches'):
|
|
self.suspendCards([card.id])
|
|
self.reset()
|
|
|
|
# Tools
|
|
##########################################################################
|
|
|
|
def resetConf(self):
|
|
"Update group conf cache."
|
|
self.groupConfs = dict(self.db.all("select id, gcid from groups"))
|
|
self.confCache = {}
|
|
|
|
def confForCard(self, card):
|
|
id = self.groupConfs[card.gid]
|
|
if id not in self.confCache:
|
|
self.confCache[id] = simplejson.loads(
|
|
self.db.scalar("select conf from gconf where id = :id",
|
|
id=id))
|
|
return self.confCache[id]
|
|
|
|
def resetSchedBuried(self):
|
|
"Put temporarily suspended cards back into play."
|
|
self.db.execute(
|
|
"update cards set queue = type where queue = -3")
|
|
|
|
def groupLimit(self, type):
|
|
l = self.deck.qconf[type+"Groups"]
|
|
if not l:
|
|
# everything
|
|
return ""
|
|
return " and gid in %s" % ids2str(l)
|
|
|
|
# Daily cutoff
|
|
##########################################################################
|
|
|
|
def updateCutoff(self):
|
|
d = datetime.datetime.utcfromtimestamp(
|
|
time.time() - self.deck.utcOffset) + datetime.timedelta(days=1)
|
|
d = datetime.datetime(d.year, d.month, d.day)
|
|
newday = self.deck.utcOffset - time.timezone
|
|
d += datetime.timedelta(seconds=newday)
|
|
cutoff = time.mktime(d.timetuple())
|
|
# cutoff must not be in the past
|
|
while cutoff < time.time():
|
|
cutoff += 86400
|
|
# cutoff must not be more than 24 hours in the future
|
|
cutoff = min(time.time() + 86400, cutoff)
|
|
self.dayCutoff = cutoff
|
|
self.today = int(cutoff/86400 - self.deck.created/86400)
|
|
|
|
def checkDay(self):
|
|
# check if the day has rolled over
|
|
if time.time() > self.dayCutoff:
|
|
self.updateCutoff()
|
|
self.reset()
|
|
|
|
# Review early
|
|
##########################################################################
|
|
|
|
def setupReviewEarlyScheduler(self):
|
|
self.fillRevQueue = self._fillRevEarlyQueue
|
|
self.rebuildRevCount = self._rebuildRevEarlyCount
|
|
self.finishScheduler = self.setupStandardScheduler
|
|
self.answerPreSave = self._reviewEarlyPreSave
|
|
self.scheduler = "reviewEarly"
|
|
|
|
def _reviewEarlyPreSave(self, card, ease):
|
|
if ease > 1:
|
|
# prevent it from appearing in next queue fill
|
|
card.queue = -3
|
|
|
|
def _rebuildRevEarlyCount(self):
|
|
# in the future it would be nice to skip the first x days of due cards
|
|
self.revCount = self.db.scalar(
|
|
self.cardLimit(
|
|
"revActive", "revInactive", """
|
|
select count() from cards c where queue = 1 and due > :now
|
|
"""), now=self.dayCutoff)
|
|
|
|
def _fillRevEarlyQueue(self):
|
|
if self.revCount and not self.revQueue:
|
|
self.revQueue = self.db.all(
|
|
self.cardLimit(
|
|
"revActive", "revInactive", """
|
|
select id, fid from cards c where queue = 1 and due > :lim
|
|
order by due limit %d""" % self.queueLimit), lim=self.dayCutoff)
|
|
self.revQueue.reverse()
|
|
|
|
# Learn more
|
|
##########################################################################
|
|
|
|
def setupLearnMoreScheduler(self):
|
|
self.rebuildNewCount = self._rebuildLearnMoreCount
|
|
self.updateNewCountToday = self._updateLearnMoreCountToday
|
|
self.finishScheduler = self.setupStandardScheduler
|
|
self.scheduler = "learnMore"
|
|
|
|
def _rebuildLearnMoreCount(self):
|
|
self.newAvail = self.db.scalar(
|
|
self.cardLimit(
|
|
"newActive", "newInactive",
|
|
"select count(*) from cards c where queue = 2 "
|
|
"and due < :lim"), lim=self.dayCutoff)
|
|
self.spacedCards = []
|
|
|
|
def _updateLearnMoreCountToday(self):
|
|
self.newCount = self.newAvail
|