The issue was that col.decks.save() and col.models.save() now require an argument — using them without an argument has no effect.
Slight issue with children
There's still a slight, partial issue related to the export of decks which have subdecks, but it seems that it was also present when exporting with Anki 2.1.26 and it only occurs for the very first export. It's a matter of a uuid being generated for the same deck config (assuming that the parent and child decks share it) when dealing with each of the parent and child decks (it's not saved in between), so phantom copies of the deck config are created in deck.json. The deck_config_uuid that persists is the one that was last applied. After the first export, all the phantom copies are cleared out and all the (sub)decks that should share a deck config do indeed have the same deck_config_uuid.
I'll try to fix this, as well, eventually, but since it was already previously present, it's far less urgent and I'm not sure how much time I'll have, I think that it's definitely worth merging the current PR (obviously after addressing any feedback) without a fix for this.
Fix #109.
The issue was that col.decks.save() and col.models.save() now require an argument — using them without an argument has no effect.
Slight issue with children
There's still a slight, partial issue related to the export of decks which have subdecks, but it seems that it was also present when exporting with Anki 2.1.26 and it only occurs for the very first export. It's a matter of a uuid being generated for the same deck config (assuming that the parent and child decks share it) when dealing with each of the parent and child decks (it's not saved in between), so phantom copies of the deck config are created in
deck.json
. Thedeck_config_uuid
that persists is the one that was last applied. After the first export, all the phantom copies are cleared out and all the (sub)decks that should share a deck config do indeed have the samedeck_config_uuid
.I'll try to fix this, as well, eventually, but since it was already previously present, it's far less urgent and I'm not sure how much time I'll have, I think that it's definitely worth merging the current PR (obviously after addressing any feedback) without a fix for this.