Result: done
Files: see commit
Commit: BTreeMap::entry, HashSet/BTreeSet::extend, Vec::extend of a list, set or sequence, extend(xs.clone()) as iter().cloned() (ADR-collections-1)
Part: #950
Pushed: 30bdbc6 and 9538c0b (index fix: the first lowered.rs held another session’s comptime_tree names)
Tests: nikaia test –both-settings –project tests/compiler 287 ok (4 new: a_btree_map_entry_or_insert_and_modify, a_set_extends_from_a_list_a_set_and_a_drain, a_vec_extends_from_a_set, a_set_extend_keeps_its_items NK2105); gate set 23 ok; clippy clean
Gate: green
Deviation: new tests failed on the unfixed compiler (NK1171 entry/extend, NK1102 Vec::extend of a set), seen before the fix, not by stash. One parameter type: items is ? on the three extends (no list|set|seq union in contracts), so element type is not checked at the call, rustc checks it. Checker needed one change only (check_core_call_on owned_entry_keys for BTreeMap::entry); other entry handling is name-agnostic. lower-std worked in the shared checkout; comptime_tree/text_tiers lowered files left unstaged (other agents).
Found: D4 now also rewrites existing tools code: prover_scope, prover_state, prover_walk, emit_core_args, emit_stmt lowered files changed (x.extend(y.clone()) -> iter().cloned()); new fact collection_clones
Needs: none (brief B may start)
Changelog: BTreeMap::entry, HashSet::extend and BTreeSet::extend exist; Vec::extend takes a list, a set or a sequence; s.extend(xs.clone()) copies the elements, not the list (ADR-collections-1, #950).
Owner: none
Disk freed: none
Rs2nika: none (std/compiler change, no move)
