All ingested activities were getting a hardcoded default scoring strategy of "average", causing unexpected and undesired behavior on multi-input questions in migrated/ingested courses. For example, on a two-part question with default 1 point per part, the score for two correct answers would be 1 out of 1, when instructors expect 2 out of 2. This corrects the code to set the "total" scoring strategy, the natural default, and the one which is applied as default elsewhere in the code on updating custom scoring flag (see maybe_update_scoring_strategy).
This remains a hardcoded default but a reasonable one: the main use for ingestion is migrated courses and they will always want this; it seems unlikely there will ever be a desire for anything other than "total" for activity level scoring; and it is no worse than what was already in the code, only an improvement. The strategy remains changeable after ingestion through custom scoring interface.
All ingested activities were getting a hardcoded default scoring strategy of "average", causing unexpected and undesired behavior on multi-input questions in migrated/ingested courses. For example, on a two-part question with default 1 point per part, the score for two correct answers would be 1 out of 1, when instructors expect 2 out of 2. This corrects the code to set the "total" scoring strategy, the natural default, and the one which is applied as default elsewhere in the code on updating custom scoring flag (see
maybe_update_scoring_strategy).This remains a hardcoded default but a reasonable one: the main use for ingestion is migrated courses and they will always want this; it seems unlikely there will ever be a desire for anything other than "total" for activity level scoring; and it is no worse than what was already in the code, only an improvement. The strategy remains changeable after ingestion through custom scoring interface.