One hundred fifty grams of carbohydrate is most likely too high for ketosis, would be my guess. It might just be possible for your friend to be in ketosis when she eats only 90-100 g, however.
I believe the 150-gram figure was calculated by Dr. George Cahill as the brain’s minimum glucose need in his landmark study, Starvation in Man, published in the late 1960’s or early 1970’s. But that is clearly wrong, since many people do fine eating far less carbohydrate than 150 g. Prof. Benjamin Bikman questions whether the brain actually needs any glucose at all.
What is not in question is that the brain does very nicely getting a very high percentage, if not all, of its energy from ketones. It is also a known fact that the liver is capable of supplying all the glucose needed by the rest of the body (which is not all that much, actually), even when we eat no carbohydrate at all.
Eighty grams of protein seems okay, but 800 calories is starvation level. If your friend agrees to eat 150 g of carbohydrate, that will at least bring her intake up to 960 calories, which is still starvation.
All in all, I think it unlikely that your friend will lose any fat. Eating at a starvation level of calories prevents the body from shedding excess fat until a point late in the process of starvation. The body actually consumes muscle for a while first, before starting in on the fat. She may lose weight, but not fat.
Your friend is probably constipated because she isn’t getting enough salt. If she consumed between 10 and 15 grams a day (including the salt already present in her food) and drinks enough liquid to satisfy her thirst, she will most likely stop being constipated. Too much salt, by the way, brings a problem very much the opposite of constipation, so she wants enough salt, but not too much.