labwc/include/edges.h
John Lindgren ef766d16f0 common: flesh out enum lab_edge and prefer over wlr_edges/wlr_direction
I like the new common/edge.h. I don't like how inconsistently we use it.

Current situation:

 - enum wlr_edges and wlr_direction are designed to be used as bitset,
   and are defined compatibly

 - enum lab_edge is *also* designed to be used as bitset, but
   incompatible with the others (LEFT/RIGHT come before UP/DOWN)

 - we use an inconsistent mix of all three *AND* uint32_t (usually with
   the WLR_EDGE constants rather than the LAB_EDGE constants), and
   convert between them on an ad-hoc basis, sometimes implicitly

Let's clean this up:

 - reorder enum lab_edge to be compatible with the two wlr enums
   (check this by static_assert)

 - use TOP/BOTTOM naming rather than UP/DOWN (matches wlr_edges)

 - add constants for the remaining possible combinations of the 4 edges

 - use lab_edge for all internal edge/direction fields, consistently

 - add lab_edge_is_cardinal() as a sanity check before casting to
   enum wlr_direction, and then eliminate all of direction.c/h

Instead of "enum wlr_edges direction", we now have
"enum lab_edge direction" which is not that much better. At least we
are now clear that we're overloading one enum with two meanings.
2025-08-26 20:36:43 -04:00

122 lines
3.9 KiB
C

/* SPDX-License-Identifier: GPL-2.0-only */
#ifndef LABWC_EDGES_H
#define LABWC_EDGES_H
#include <limits.h>
#include <stdbool.h>
#include "common/edge.h"
#include "common/macros.h"
struct border;
struct output;
struct server;
struct view;
struct wlr_box;
static inline int
clipped_add(int a, int b)
{
if (b > 0) {
return a >= (INT_MAX - b) ? INT_MAX : (a + b);
} else if (b < 0) {
return a <= (INT_MIN - b) ? INT_MIN : (a + b);
}
return a;
}
static inline int
clipped_sub(int a, int b)
{
if (b > 0) {
return a <= (INT_MIN + b) ? INT_MIN : (a - b);
} else if (b < 0) {
return a >= (INT_MAX + b) ? INT_MAX : (a - b);
}
return a;
}
static inline int
edge_get_best(int next, int edge, bool decreasing)
{
if (!BOUNDED_INT(next)) {
/* Any bounded edge beats an unbounded next */
return BOUNDED_INT(edge) ? edge : next;
}
/* No unbounded edge ever beats next */
if (!BOUNDED_INT(edge)) {
return next;
}
/* Max edge wins for decreasing moves, min edge for increasing */
return decreasing ? MAX(next, edge) : MIN(next, edge);
}
struct edge {
/* Position of an edge along the axis perpendicular to it */
int offset;
/* Limits of edge along axis parallel to it */
int min;
int max;
};
/*
* edge_validator_t - edge validator signature
* @best: pointer to the current "best" edge
* @current: current position of a moving edge
* @target: position to which the moving edge will be moved
* @oppose: opposing edge of encountered region
* @align: aligned edge of encountered region
* @lesser: true if the moving edge is top or left, false otherwise
*
* This function will be used by edge_find_neighbors and edge_find_outputs to
* validate and select the "best" output or neighbor edge against which a
* moving edge should be snapped. The moving edge has current position
* "current" and desired position "target". The validator should determine
* whether motion of the crosses the given opposed and aligned edges of a trial
* region and should be considered a snap point.
*
* Opposing edges are on the opposite side of the target region from the moving
* edge (i.e., left <-> right, top <-> bottom). When the moving edge snaps to
* an opposing edge, the view should maintain the configured gap. Aligned edges
* are on the same side of the target region from the moving edge (i.e.,
* left <-> left, right <-> right, top <-> top, bottom <-> bottom). When the
* moving edge snaps to an aligned edge, the view should *not* include a gap.
*
* If window gaps are configured, all edges will be offset as appropriate to
* reflect the desired padding. Thus, the validator should generally compare
* the given current or target values directly to the opposing and aligned edge
* without regard for rc.gap.
*
* Any edge may take the values INT_MIN or INT_MAX to indicate that the edge
* should be effectively ignored. Should the validator decide that a given
* region edge (oppose or align) should be a preferred snap point, it should
* update the value of *best accordingly.
*/
typedef void (*edge_validator_t)(int *best, struct edge current,
struct edge target, struct edge oppose, struct edge align, bool lesser);
void edges_initialize(struct border *edges);
void edges_find_neighbors(struct border *nearest_edges, struct view *view,
struct wlr_box origin, struct wlr_box target,
struct output *output, edge_validator_t validator, bool ignore_hidden);
void edges_find_outputs(struct border *nearest_edges, struct view *view,
struct wlr_box origin, struct wlr_box target,
struct output *output, edge_validator_t validator);
void edges_adjust_move_coords(struct view *view, struct border edges,
int *x, int *y, bool use_pending);
void edges_adjust_resize_geom(struct view *view, struct border edges,
enum lab_edge resize_edges, struct wlr_box *geom, bool use_pending);
bool edges_traverse_edge(struct edge current, struct edge target, struct edge edge);
void edges_calculate_visibility(struct server *server, struct view *ignored_view);
#endif /* LABWC_EDGES_H */