Anki/anki/importing/__init__.py
Damien Elmes 9c247f45bd remove q/a cache, tags in fields, rewrite remaining ids, more
Anki used random 64bit IDs for cards, facts and fields. This had some nice
properties:
- merging data in syncs and imports was simply a matter of copying each way,
  as conflicts were astronomically unlikely
- it made it easy to identify identical cards and prevent them from being
  reimported
But there were some negatives too:
- they're more expensive to store
- javascript can't handle numbers > 2**53, which means AnkiMobile, iAnki and
  so on have to treat the ids as strings, which is slow
- simply copying data in a sync or import can lead to corruption, as while a
  duplicate id indicates the data was originally the same, it may have
  diverged. A more intelligent approach is necessary.
- sqlite was sorting the fields table based on the id, which meant the fields
  were spread across the table, and costly to fetch

So instead, we'll move to incremental ids. In the case of model changes we'll
declare that a schema change and force a full sync to avoid having to deal
with conflicts, and in the case of cards and facts, we'll need to update the
ids on one end to merge. Identical cards can be detected by checking to see if
their id is the same and their creation time is the same.

Creation time has been added back to cards and facts because it's necessary
for sync conflict merging. That means facts.pos is not required.

The graves table has been removed. It's not necessary for schema related
changes, and dead cards/facts can be represented as a card with queue=-4 and
created=0. Because we will record schema modification time and can ensure a
full sync propagates to all endpoints, it means we can remove the dead
cards/facts on schema change.

Tags have been removed from the facts table and are represented as a field
with ord=-1 and fmid=0. Combined with the locality improvement for fields, it
means that fetching fields is not much more expensive than using the q/a
cache.

Because of the above, removing the q/a cache is a possibility now. The q and a
columns on cards has been dropped. It will still be necessary to render the
q/a on fact add/edit, since we need to record media references. It would be
nice to avoid this in the future. Perhaps one way would be the ability to
assign a type to fields, like "image", "audio", or "latex". LaTeX needs
special consider anyway, as it was being rendered into the q/a cache.
2011-04-28 09:23:53 +09:00

375 lines
13 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
"""\
Importing support
==============================
To import, a mapping is created of the form: [FieldModel, ...]. The mapping
may be extended by calling code if a file has more fields. To ignore a
particular FieldModel, replace it with None. A special number 0 donates a tags
field. The same field model should not occur more than once."""
import time
#from anki.cards import cardsTable
#from anki.facts import factsTable, fieldsTable
from anki.lang import _
from anki.utils import canonifyTags, fieldChecksum
from anki.utils import canonifyTags, ids2str
from anki.errors import *
#from anki.deck import NEW_CARDS_RANDOM
# Base importer
##########################################################################
class ForeignCard(object):
"An temporary object storing fields and attributes."
def __init__(self):
self.fields = []
self.tags = u""
class Importer(object):
needMapper = True
tagDuplicates = False
# if set, update instead of regular importing
# (foreignCardFieldIndex, fieldModelId)
updateKey = None
multipleCardsAllowed = True
needDelimiter = False
def __init__(self, deck, file):
self.file = file
self._model = deck.currentModel
self._mapping = None
self.log = []
self.deck = deck
self.total = 0
self.tagsToAdd = u""
def doImport(self):
"Import."
if self.updateKey is not None:
return self.doUpdate()
random = self.deck.newCardOrder == NEW_CARDS_RANDOM
num = 6
if random:
num += 1
self.deck.startProgress(num)
self.deck.updateProgress(_("Importing..."))
c = self.foreignCards()
if self.importCards(c):
self.deck.updateProgress()
self.deck.updateCardTags(self.cardIds)
if random:
self.deck.updateProgress()
self.deck.randomizeNewCards(self.cardIds)
self.deck.finishProgress()
if c:
self.deck.setModified()
def doUpdate(self):
self.deck.startProgress(7)
# grab the data from the external file
self.deck.updateProgress(_("Updating..."))
cards = self.foreignCards()
# grab data from db
self.deck.updateProgress()
fields = self.deck.db.all("""
select factId, value from fields where fieldModelId = :id
and value != ''""",
id=self.updateKey[1])
# hash it
self.deck.updateProgress()
vhash = {}
fids = []
for (fid, val) in fields:
fids.append(fid)
vhash[val] = fid
# prepare tags
tagsIdx = None
try:
tagsIdx = self.mapping.index(0)
for c in cards:
c.tags = canonifyTags(self.tagsToAdd + " " + c.fields[tagsIdx])
except ValueError:
pass
# look for matches
self.deck.updateProgress()
upcards = []
newcards = []
for c in cards:
v = c.fields[self.updateKey[0]]
if v in vhash:
# ignore empty keys
if v:
# fid, card
upcards.append((vhash[v], c))
else:
newcards.append(c)
# update fields
for fm in self.model.fieldModels:
if fm.id == self.updateKey[1]:
# don't update key
continue
try:
index = self.mapping.index(fm)
except ValueError:
# not mapped
continue
data = [{'fid': fid,
'fmid': fm.id,
'v': c.fields[index],
'chk': self.maybeChecksum(c.fields[index], fm.unique)}
for (fid, c) in upcards]
self.deck.db.execute("""
update fields set value = :v, chksum = :chk where factId = :fid
and fieldModelId = :fmid""", data)
# update tags
self.deck.updateProgress()
if tagsIdx is not None:
data = [{'fid': fid,
't': c.fields[tagsIdx]}
for (fid, c) in upcards]
self.deck.db.execute(
"update facts set tags = :t where id = :fid",
data)
# rebuild caches
self.deck.updateProgress()
cids = self.deck.db.column0(
"select id from cards where factId in %s" %
ids2str(fids))
self.deck.updateCardTags(cids)
self.deck.updateProgress()
self.deck.updateCardsFromFactIds(fids)
self.total = len(cards)
self.deck.setModified()
self.deck.finishProgress()
def fields(self):
"The number of fields."
return 0
def maybeChecksum(self, data, unique):
if not unique:
return ""
return fieldChecksum(data)
def foreignCards(self):
"Return a list of foreign cards for importing."
assert 0
def resetMapping(self):
"Reset mapping to default."
numFields = self.fields()
m = [f for f in self.model.fieldModels]
m.append(0)
rem = max(0, self.fields() - len(m))
m += [None] * rem
del m[numFields:]
self._mapping = m
def getMapping(self):
if not self._mapping:
self.resetMapping()
return self._mapping
def setMapping(self, mapping):
self._mapping = mapping
mapping = property(getMapping, setMapping)
def getModel(self):
return self._model
def setModel(self, model):
self._model = model
# update the mapping for the new model
self._mapping = None
self.getMapping()
model = property(getModel, setModel)
def importCards(self, cards):
"Convert each card into a fact, apply attributes and add to deck."
# ensure all unique and required fields are mapped
for fm in self.model.fieldModels:
if fm.required or fm.unique:
if fm not in self.mapping:
raise ImportFormatError(
type="missingRequiredUnique",
info=_("Missing required/unique field '%(field)s'") %
{'field': fm.name})
active = 0
for cm in self.model.cardModels:
if cm.active: active += 1
if active > 1 and not self.multipleCardsAllowed:
raise ImportFormatError(type="tooManyCards",
info=_("""\
The current importer only supports a single active card template. Please disable\
all but one card template."""))
# strip invalid cards
cards = self.stripInvalid(cards)
cards = self.stripOrTagDupes(cards)
self.cardIds = []
if cards:
self.addCards(cards)
return cards
def addCards(self, cards):
"Add facts in bulk from foreign cards."
# map tags field to attr
try:
idx = self.mapping.index(0)
for c in cards:
c.tags += " " + c.fields[idx]
except ValueError:
pass
# add facts
self.deck.updateProgress()
factIds = [genID() for n in range(len(cards))]
factCreated = {}
def fudgeCreated(d, tmp=[]):
if not tmp:
tmp.append(time.time())
else:
tmp[0] += 0.0001
d['created'] = tmp[0]
factCreated[d['id']] = d['created']
return d
self.deck.db.execute(factsTable.insert(),
[fudgeCreated({'modelId': self.model.id,
'tags': canonifyTags(self.tagsToAdd + " " + cards[n].tags),
'id': factIds[n]}) for n in range(len(cards))])
self.deck.db.execute("""
delete from factsDeleted
where factId in (%s)""" % ",".join([str(s) for s in factIds]))
# add all the fields
self.deck.updateProgress()
for fm in self.model.fieldModels:
try:
index = self.mapping.index(fm)
except ValueError:
index = None
data = [{'factId': factIds[m],
'fieldModelId': fm.id,
'ordinal': fm.ordinal,
'id': genID(),
'value': (index is not None and
cards[m].fields[index] or u""),
'chksum': self.maybeChecksum(
index is not None and
cards[m].fields[index] or u"", fm.unique)
}
for m in range(len(cards))]
self.deck.db.execute(fieldsTable.insert(),
data)
# and cards
self.deck.updateProgress()
active = 0
for cm in self.model.cardModels:
if cm.active:
active += 1
data = [self.addMeta({
'id': genID(),
'factId': factIds[m],
'factCreated': factCreated[factIds[m]],
'cardModelId': cm.id,
'ordinal': cm.ordinal,
'question': u"",
'answer': u""
},cards[m]) for m in range(len(cards))]
self.deck.db.execute(cardsTable.insert(),
data)
self.deck.updateProgress()
self.deck.updateCardsFromFactIds(factIds)
self.total = len(factIds)
def addMeta(self, data, card):
"Add any scheduling metadata to cards"
if 'fields' in card.__dict__:
del card.fields
t = data['factCreated'] + data['ordinal'] * 0.00001
data['created'] = t
data['modified'] = t
data['due'] = t
data.update(card.__dict__)
data['tags'] = u""
self.cardIds.append(data['id'])
data['combinedDue'] = data['due']
if data.get('successive', 0):
t = 1
elif data.get('reps', 0):
t = 0
else:
t = 2
data['type'] = t
data['queue'] = t
return data
def stripInvalid(self, cards):
return [c for c in cards if self.cardIsValid(c)]
def cardIsValid(self, card):
fieldNum = len(card.fields)
for n in range(len(self.mapping)):
if self.mapping[n] and self.mapping[n].required:
if fieldNum <= n or not card.fields[n].strip():
self.log.append("Fact is missing field '%s': %s" %
(self.mapping[n].name,
", ".join(card.fields)))
return False
return True
def stripOrTagDupes(self, cards):
# build a cache of items
self.uniqueCache = {}
for field in self.mapping:
if field and field.unique:
self.uniqueCache[field.id] = self.getUniqueCache(field)
return [c for c in cards if self.cardIsUnique(c)]
def getUniqueCache(self, field):
"Return a dict with all fields, to test for uniqueness."
return dict(self.deck.db.all(
"select value, 1 from fields where fieldModelId = :fmid",
fmid=field.id))
def cardIsUnique(self, card):
fieldsAsTags = []
for n in range(len(self.mapping)):
if self.mapping[n] and self.mapping[n].unique:
if card.fields[n] in self.uniqueCache[self.mapping[n].id]:
if not self.tagDuplicates:
self.log.append("Fact has duplicate '%s': %s" %
(self.mapping[n].name,
", ".join(card.fields)))
return False
fieldsAsTags.append(self.mapping[n].name.replace(" ", "-"))
else:
self.uniqueCache[self.mapping[n].id][card.fields[n]] = 1
if fieldsAsTags:
card.tags += u" Duplicate:" + (
"+".join(fieldsAsTags))
card.tags = canonifyTags(card.tags)
return True
# Export modules
##########################################################################
from anki.importing.csvfile import TextImporter
from anki.importing.anki10 import Anki10Importer
from anki.importing.mnemosyne10 import Mnemosyne10Importer
from anki.importing.wcu import WCUImporter
from anki.importing.supermemo_xml import SupermemoXmlImporter
from anki.importing.dingsbums import DingsBumsImporter
Importers = (
(_("Text separated by tabs or semicolons (*)"), TextImporter),
(_("Anki Deck (*.anki)"), Anki10Importer),
(_("Mnemosyne Deck (*.mem)"), Mnemosyne10Importer),
(_("CueCard Deck (*.wcu)"), WCUImporter),
(_("Supermemo XML export (*.xml)"), SupermemoXmlImporter),
(_("DingsBums?! Deck (*.dbxml)"), DingsBumsImporter),
)