noe/arthur
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

A position is an argument, not another command

Browse source 
Everything could only be added to the end, because `append` computed its own
position — the max end of the lane — and so had no opinion to state. Insert is
not a new command; it is the argument that function was missing. `:at` takes a
lane frame or `:end`, `:end` is the position where nothing has to move, and
appending stops being a separate operation from inserting. New, reused and
duplicated drawings all take it, because there was only ever one placement rule.

Placing ripples: occurrences at or after the position move later by the new
exposure's duration, and `:keep` against `:grow-symbol` still decides what
happens at the shot's end. OVERWRITE is deliberately not a policy argument yet.
Taking frames away from the occurrence already there is TRIMMING, and an
argument whose second value is unimplemented is worse than an argument that is
not there. A position strictly inside an existing exposure refuses and names
`split`, rather than splitting on the quiet: one command performing two is how
a command stops being predictable.

Then split, which turned out to cost almost nothing, and that is the
interesting part. The two pieces keep ONE `:time` and differ only in `:span`.
The right piece's own frames therefore carry on exactly where the left's
stopped, so its source clock, its keys and its corrections go on meaning what
they meant: a held drawing holds the same frame either side of the cut, and a
playing insert plays through it without a seam. There is no arithmetic on
in-points to get wrong, and no shot-length question, since the pieces occupy
the frames the one exposure occupied. The test samples every frame before and
after and asserts the picture is identical — for a hold, for an exposure with a
correction of its own, and for a playing insert.

That is not a clever split. It is `:span` being in the node's OWN coordinates,
which was decided long before there were lanes, paying for something it was not
designed for. The same property is why extending a hold leaves lane keys alone.

Both new commands act at the playhead, which needed `lane-frame` — the symbol's
frame as a frame of the lane's own time, nil through a stepped or looping lane
where one is not the other. Nil refuses; it does not snap to a nearby frame.

Two smaller things found while doing it. `placeable` promised "a whole lane
frame" in its refusal and then accepted 2.5, so both it and `split` now require
an integer, as `extend-hold` already did for its delta. And `lane-end` is
private: `:end` is the only way to ask for it.

401 tests, 5,612 assertions. The browser flow now splits an exposure at the
playhead and puts a drawing in the gap, and checks that six exposures are still
one row.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
This commit is contained in:
Your Name 2026-09-30 15:56:26 -04:00
parent 9446829774
commit 26517af2fd
6 changed files with 296 additions and 42 deletions
Download patch file Download diff file Expand all files Collapse all files

93
frontend/test/arthur/domain/lane_test.cljs
View file

@ -334,3 +334,96 @@
(is (= [0 10 20 30 40 50]
(mapv (fn [f] (let [[[id _] cx] (showing f)] (- cx (xs id)))) (range 6)))
"and the lane moves on every frame, odd ones included")))
(defn- drawn
"What every frame draws, as sorted values, so a picture can be compared
without naming the occurrences that produced it."
[doc fs]
(let [at (sample doc fs)]
(mapv #(sort (vals (get at %))) fs)))
(deftest a-drawing-goes-anywhere-in-the-lane-and-ripples-what-follows
(let [doc (document)
keys-of #(get-in % [:symbols :main :nodes :girl :channels [:xform :pos] :keys])
spans #(mapv (fn [id] (node/placed-span (get-in % [:symbols :main :nodes id])))
[:a :n :b :insert])
r (sequence/append-drawing doc :main :girl :n :drawing-n
{:at 4 :extent :grow-symbol})]
(is (= [[0 4] [4 5] [5 9] [9 13]] (spans (:clip r))))
(is (= 13 (get-in r [:clip :symbols :main :frames])))
(is (= (keys-of doc) (keys-of (:clip r))) "lane keys stay where they were authored")
(is (= :n (:selection r)))
(is (= 4 (:frame r)))
(is (empty? (clip/problems (:clip r))))
;; The same command with no room refuses, and says how much it needs.
(is (= 13 (:required-frames (sequence/append-drawing doc :main :girl :n :drawing-n {:at 4}))))
;; At the very front everything moves.
(is (= [[1 5] [0 1] [5 9] [9 13]]
(spans (:clip (sequence/append-drawing doc :main :girl :n :drawing-n
{:at 0 :extent :grow-symbol})))))
;; Inside an exposure is not a position for another one.
(is (re-find #"split it first"
(:refused (sequence/append-drawing doc :main :girl :n :drawing-n
{:at 2 :extent :grow-symbol}))))
(is (:refused (sequence/append-drawing doc :main :girl :n :drawing-n
{:at -1 :extent :grow-symbol})))
(is (:refused (sequence/append-drawing doc :main :girl :n :drawing-n
{:at ##Inf :extent :grow-symbol})))
;; Reuse and duplicate take a position too; it is one placement rule.
(is (= [4 5] (node/placed-span
(get-in (sequence/reuse-drawing doc :main :girl :n :drawing-b
{:at 4 :extent :grow-symbol})
[:clip :symbols :main :nodes :n]))))
(is (= [4 5] (node/placed-span
(get-in (sequence/duplicate-drawing doc :main :b :n
{:at 4 :extent :grow-symbol})
[:clip :symbols :main :nodes :n]))))))
(deftest splitting-an-exposure-changes-nothing-that-is-drawn
(let [doc (document)
fs (range 12)
before (drawn doc fs)]
(doseq [[label id cut] [["a held drawing" :a 2]
["an exposure with a correction of its own" :b 6]
["a playing insert" :insert 10]]]
(testing label
(let [r (sequence/split doc :main id cut :right)
after (:clip r)]
(is (= :right (:selection r)))
(is (= before (drawn after fs)) "the same picture, frame for frame")
(is (= (node/placed-span (get-in doc [:symbols :main :nodes id]))
[(first (node/placed-span (get-in after [:symbols :main :nodes id])))
(second (node/placed-span (get-in after [:symbols :main :nodes :right])))])
"the pieces occupy the frames the exposure did")
(is (= cut (second (node/placed-span (get-in after [:symbols :main :nodes id])))
(first (node/placed-span (get-in after [:symbols :main :nodes :right])))))
(is (= (:time (get-in doc [:symbols :main :nodes id]))
(:time (get-in after [:symbols :main :nodes :right])))
"one time map, so the right piece's own frames carry on")
(is (= (select-keys (get-in doc [:symbols :main :nodes id]) [:source :playback :channels])
(select-keys (get-in after [:symbols :main :nodes :right]) [:source :playback :channels])))
(is (= 12 (get-in after [:symbols :main :frames])) "and no shot-length question")
(is (empty? (clip/problems after))))))))
(deftest split-refuses-anything-that-is-not-one-cut-inside-one-exposure
(let [doc (document)]
(doseq [cut [0 4 8 12 -1 2.5 ##NaN nil]]
(is (:refused (sequence/split doc :main :b cut :right)) (str "cut at " (pr-str cut))))
(is (:refused (sequence/split doc :main :girl 2 :right)) "a lane is not an exposure")
(is (:refused (sequence/split doc :main :plate 2 :right)) "nor is a shape outside one")
(is (:refused (sequence/split doc :main :a 2 :b)) "the new ID has to be free")))
(deftest split-then-place-puts-a-drawing-inside-a-hold
;; The two commands the doc asks for, composed: neither one guesses.
(let [doc (document)
cut (:clip (sequence/split doc :main :a 2 :right))
r (sequence/append-drawing cut :main :girl :n :drawing-n
{:at 2 :extent :grow-symbol})
after (:clip r)]
(is (= [[0 2] [2 3] [3 5] [5 9] [9 13]]
(mapv #(node/placed-span (get-in after [:symbols :main :nodes %]))
[:a :n :right :b :insert])))
(is (= (get-in doc [:symbols :main :nodes :girl :channels])
(get-in after [:symbols :main :nodes :girl :channels]))
"the performance is still timed the way it was authored")
(is (empty? (clip/problems after)))))

27
frontend/test/browser/sequence.mjs
View file

@ -149,8 +149,33 @@ try {
assert.equal(new Set(cels.map(n => n.source.symbol)).size, 4,
'four exposures of four drawings: nothing is shared once every copy is made');
assert.equal(s.history.done.length, before.history.done.length + 3, 'three more commands, three more steps');
// A drawing into the middle of a hold: split, then insert. Both act at the
// playhead, and neither guesses what the other one is for.
const placed = s => instances(s)
.map(n => [n.time.at + n.span[0] / (n.time.rate ?? 1), n.time.at + n.span[1] / (n.time.rate ?? 1)])
.sort((a, b) => a[0] - b[0]);
assert.deepEqual(placed(s), [[0, 4], [4, 5], [5, 6], [6, 7]]);
await evaluate(`document.querySelector('.tl-cel').click()`);
await sleep(200);
assert.equal(await enabled('split'), false, 'the start of an exposure is not inside it');
await click('+1');
await click('+1');
assert.equal(await enabled('split'), true);
await click('split');
s = await shot();
assert.deepEqual(placed(s), [[0, 2], [2, 4], [4, 5], [5, 6], [6, 7]],
'one exposure became two, over the frames it had');
await click('insert');
s = await shot();
assert.deepEqual(placed(s), [[0, 2], [2, 3], [3, 5], [5, 6], [6, 7], [7, 8]],
'the new drawing took frame 2 and everything from there rippled later');
assert.equal(await evaluate('document.querySelectorAll(".tl-cel").length'), 6);
assert.equal(await evaluate('document.querySelectorAll(".tl-label:not(.tl-corner)").length'), 1,
'six exposures, still one row');
assert.equal(s.history.done.length, before.history.done.length + 5);
assert.equal(errors.length, 0, JSON.stringify(errors));
console.log('PASS: create lane/drawings, one-row cels, hold ripple, seek, explicit overflow, atomic undo, reuse/make unique/duplicate; no server writes');
console.log('PASS: create lane/drawings, one-row cels, hold ripple, seek, explicit overflow, atomic undo, reuse/make unique/duplicate, split and insert; no server writes');
} finally {
if (ws?.readyState === WebSocket.OPEN) {
ws.send(JSON.stringify({ id: 999999, method: 'Browser.close' }));