Repository navigation
layers panel
Home > The Layers panel
< Previous: Reference images | Next: File >
The running list of everything in your construction - browse, select, group, style, rename, reorder, lock, hide, and redefine.
The Layers panel is the running list of everything you've drawn - points, segments, circles, arcs, regions, and the groups that organize them. It mirrors your canvas: click a row to select that object on the canvas, and click an object on the canvas to light up its row. Selection flows both ways, so the panel is how you reach things that are small, stacked, or tucked inside a group.
Any row that contains other things shows a disclosure chevron - expand a region to see its boundary edges, a circle to see its defining points, or a group to see its contents.
Each row packs a lot onto one line. From left to right: a disclosure chevron (only when the object is built from others), a type icon, the object's name, and - toward the right - a type label in words: point, line, circle, perpendicular, and so on. Between the label and the right edge sit the object's lock and eye controls. A glance down the list tells you what each object is and how it relates to the rest. Free points you placed yourself sit at the top level under a plain name like A or B.
Expand an object and its rows reveal the objects it depends on - its sources - each tagged with a gold role chip naming the part it plays. A circle lists its CENTER and RADIUS points; a line its two points; a perpendicular its BASE line and the point it passes THROUGH. This is also where you rewire a definition (see Redefining an object's source, below).
When one of those child rows is a link to an object that lives elsewhere in the list - a pointer, not a separate copy - it's drawn with a small elbow arrow and its name set in italics. In the picture, the circle's RADIUS is the very point B that also sits on its own near the bottom of the list; the italic B under RADIUS points at that same object. Change the shared object once and every italic reference to it follows. By contrast, an object's own home row - where it's actually defined - shows its name upright.
A line-type object carries a LINE chip near its type label. Click it to switch the line's form - for example from a full Line to a Ray - and back. The change is live, so anything built on that line updates to match.
A search field sits at the top of the panel. Type into it and the list filters live to objects whose name contains what you typed - case doesn't matter. Matches are highlighted and their parent groups stay visible for context, so you can see where each result sits in the tree. Buried matches surface even if their group was collapsed, and the tree restores itself when you clear the search. A count shows how many objects matched, and an x button clears the field.
Tip
Press Enter to select every current match on the canvas at once. Hold ⌘ (or Ctrl) and press Enter to add the matches to whatever you already have selected.
It works the other way round too: select an object on the canvas and the Layers panel scrolls to its row and reveals it - expanding any collapsed group it lives in - so you can jump from a shape on screen straight to its place in the tree.
Grouping tidies the panel without changing any geometry - a group is just a folder. Select the objects you want to collect and press ⌘G. To take a group apart again, select it and press ⇧⌘G; its members drop back into place and nothing is lost. You can also rearrange membership by dragging rows: drop a row onto a group to file it inside, or drag it out to the top level to remove it.
Groups also decide where new objects land. When you select something inside a group and then start a drawing tool, whatever you create joins that same group instead of appearing at the top level. What matters is the selection at the moment you pick the tool - select a shape inside a group, choose the Circle tool, then place a circle, and the circle joins the group. If the objects you had selected live in different groups, the new object lands in the nearest group that contains all of them; if they share no group, it stays at the top level.
Tip
If you want a new object at the top level, click empty space to clear your selection before arming the tool.
Drag a row to move it - and if you've selected several rows, the whole selection travels together. As you drag, drop lines show exactly where the items will land: before or after a row, or inside a group. You can drop items into a specific slot inside an open group, or drag them all the way out to the top level.
Order in this list is also draw order: an item higher in the stack paints over the ones below it, both its fill and its outline. Reordering is how you control what covers what.
Select one or more objects and use the style strip at the bottom of the panel to set their look:
- Fill: the interior color of a closed shape. Click the swatch to open the color picker (hue and saturation wheel, lightness, RGB, HSL, alpha, and hex - all linked).
- Stroke: the outline color. Lines, rays, and even points can take a stroke color, not just closed shapes.
- Stroke width: the outline thickness. Set it to zero for no outline at all.
- Alpha: transparency, applied literally: fully opaque at the top of the range, see-through as you lower it.
A group folder can carry a style of its own. Style a group and it stamps every current member - and anything that joins the group later inherits that look automatically.
Strokes can be dashed. The Dashes control offers presets - Solid, Dashed, Dotted, Dash-dot - plus a Custom option where you set your own dash and gap lengths in millimeters. Two things to know: dash lengths are absolute millimeters (they don't scale with stroke width), and dashes are purely visual. They export cleanly to SVG, but they never affect geometry - intersections, fills, and snapping always read the solid underlying curve.
Note
Objects generated by the node graph are read-only here - the style strip shows them as styled by their node. Change their look at the node that made them.
Double-click a row to rename it. If that object is referenced elsewhere - for instance by a node that points back at it - the name updates there too.
A new object is named from a three-letter prefix and a running number: Seg1
for the first segment, Cir1 for the first circle, Arc1 for the first arc. The
three letters are the opening of the word for that kind, so a name reads as what
it is rather than as a code.
Placed points are the exception and keep the lettering a geometry text uses, A,
B, C and on to A1, B1, C1. Only a crossing takes a prefix of its own, Int1,
so a constructed intersection is tellable from a point you placed by name alone.
The prefix length is what keeps those apart: a point name is one letter and
digits, so a three-letter prefix can never collide with one.
A boolean you make on the canvas is named from the same table, as B plus two
letters of the word: BUn1 for a union, BIn1 for an intersect, BDi1 for a
difference, BSp1 for a split. The whole boolean family sorts together in the
panel and, unlike the old English-only "Union 1", it follows your language.
The prefixes follow the interface language. A circle is Cir1 in English and
Kre1 in German, because the letters come from that language's own word for the
kind. Switching language does not rename what is already in a file: existing
names stay put so that anything referring to them keeps working.
A prefix also means one kind of object across every language. Cer is the
circle in French and Italian and is nothing else anywhere, so a file drawn in one
language and opened in another can never be handed a name it is already using.
Files made before this scheme carry the older two-letter names, Sg1 and Ci1,
and their booleans carry English words. Open one and Locus says so, with a
reminder that stays put until you deal with it and a button that does the work.
The command itself is Update Object Names, in the command palette. It restates the auto-generated names and leaves alone any name you typed yourself, and it tells you how many it changed and how many of your own it kept. Choosing Later dismisses the reminder without changing anything.
It reads a generated name in any language and in either scheme, so a file
drawn in German, or one saved under the two-letter names, is brought onto your
own current scheme rather than left alone. It tells a drawn rectangle or square
from a plain group by the object itself rather than by its name. A design brought
in with Append carries a (2) suffix on its names; that still counts as
generated, so those are repaired too rather than treated as names you chose.
Anything a Reference node points at is renamed along with it, in this window
and in the split one, so a graph that names an object keeps working.
Each row has an eye control to hide or show the object. You can also hide the current selection from the Edit menu or with ⌘3; bring hidden objects back from the panel's eye.
Hiding a group hides everything inside it on the canvas - but without changing each member's own show/hide setting. So when you unhide the group, exactly what was showing before comes back, including any single item you'd hidden on its own beforehand. If a member is set to visible but its group is hidden, its eye appears muted - a reminder that a hidden parent is currently suppressing it.
Each object has a padlock, just to the left of its eye. Click it to lock the object: a locked object can't be moved, nudged with the arrow keys, edited, or deleted. It also drops out of reach on the canvas - clicking it, dragging a marquee over it, and Select All all skip a locked object, so it never gets caught up in an edit by accident. An open shackle means unlocked; a closed shackle means locked. Locking is handy for a background frame or a reference shape you want to build against without nudging by accident. If you've selected several objects, clicking one padlock locks or unlocks the whole selection.
Because a locked object can't be clicked on the canvas, its padlock here in the Layers panel is the way back - click the closed shackle on its row to unlock it. If you've locked something and lost the ability to grab it on screen, find its row and toggle the lock off.
Note
Objects owned by the node graph don't have a padlock - the node graph controls them.
Any row that contains other things - a group, a region, a boolean result - shows a disclosure chevron. It points right when the row is collapsed and rotates to point down when it's open. Click the chevron to open or close that one level. To open a group and everything nested inside it in one move, hold Option while you click.
Some objects can change how they're defined. Expand an object and you'll see its defining references - a segment's start and end points, a circle's center and radius point, an intersection's two source objects. Double-click one of those reference rows, then click a different point or object, and the definition swaps to the new source. The geometry and everything built on it update immediately, in a single undo step.
A few objects are "dual-form" - they can be defined more than one way and switch between them. This is why a shape stays live: change what it points at, and it follows.
Caution
Confirm against the build Which reference rows are editable depends on the object and how it was made. Objects owned by the node graph don't expose editable references here.
< Previous: Reference images | Next: File >