File size: 24,190 Bytes
1c6109f
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
import numpy as np
import networkx as nx
import matplotlib.pyplot as plt
import random
import time


class NetworkGenerator:
    def __init__(self, size="S", variant="F", topology="highly_connected",
                 node_drop_fraction=0.2,
                 bottleneck_cluster_count=None,
                 bottleneck_edges_per_link=1):
        """
        node_drop_fraction:
            Fraction of all (grid+1)^2 possible positions that are deactivated (not allowed as nodes).
            Example: 0.2 -> remove 1/5 of all grid positions.

        bottleneck_cluster_count:
            If None, chosen automatically based on size.
            Larger => more small dense clusters.

        bottleneck_edges_per_link:
            Number of edges connecting consecutive clusters (these are the bottlenecks).
            Keep this small (1 or 2) to preserve bottleneck behavior.
        """
        self.size = size.upper()
        self.variant = variant.upper()
        self.topology = topology.lower()

        if self.topology not in ["highly_connected", "bottlenecks", "linear"]:
            raise ValueError("topology must be: 'highly_connected', 'bottlenecks', or 'linear'")

        self.size_config = {
            "S": {"grid": 4, "node_factor": 0.4, "diag_weights": [1, 4]},
            "M": {"grid": 8, "node_factor": 0.4, "diag_weights": [1, 4]},
            "L": {"grid": 16, "node_factor": 0.4, "diag_weights": [1, 8]},
        }

        if self.size not in self.size_config:
            raise ValueError("Invalid size. Choose 'S', 'M', or 'L'.")
        if self.variant not in ["F", "R"]:
            raise ValueError("Invalid variant. Choose 'F' (fixed) or 'R' (random).")

        self.grid_size = self.size_config[self.size]["grid"]
        self.node_factor = self.size_config[self.size]["node_factor"]
        self.weight_dist = self.size_config[self.size]["diag_weights"]

        self.node_drop_fraction = float(node_drop_fraction)
        if not (0.0 <= self.node_drop_fraction < 1.0):
            raise ValueError("node_drop_fraction must be in [0.0, 1.0).")

        if bottleneck_cluster_count is None:
            self.bottleneck_cluster_count = {"S": 3, "M": 5, "L": 8}[self.size]
        else:
            self.bottleneck_cluster_count = int(bottleneck_cluster_count)
            if self.bottleneck_cluster_count < 2:
                raise ValueError("bottleneck_cluster_count must be >= 2.")

        self.bottleneck_edges_per_link = int(bottleneck_edges_per_link)
        if self.bottleneck_edges_per_link < 1:
            raise ValueError("bottleneck_edges_per_link must be >= 1.")

        self.graph = None
        self.nodes_list = None
        self.active_positions = None  # allowed grid positions


    # --------------------
    # Public API
    # --------------------
    def generate(self):
        """Generate a connected network representing rooms in a building."""
        max_attempts = 8

        for _ in range(max_attempts):
            self._build_node_mask()
            self._initialize_graph()
            self._add_nodes()

            nodes = list(self.graph.nodes())
            if len(nodes) < 2:
                continue

            # Topology-specific edge construction
            if self.topology == "bottlenecks":
                # Replace the usual step-1 connectivity with a cluster+bottleneck design.
                self._build_bottleneck_clusters(nodes)
            else:
                # --- STEP 1: CONNECTIVITY (NEARBY ROOMS ONLY) ---
                self._connect_all_nodes_by_nearby_growth(nodes)

                # --- STEP 2: ADD TOPOLOGY-SPECIFIC EXTRA EDGES ---
                self._add_edges()

            # --- STEP 3: REMOVE INTERSECTIONS & RECONNECT ---
            self._remove_intersections()

            # --- STEP 4: FINAL CONNECTIVITY CHECK ---
            if nx.is_connected(self.graph):
                return self.graph

        raise RuntimeError("Failed to generate a connected network after several attempts")


    def plot(self):
        plt.figure(figsize=(8, 8))
        pos = {node: (node[1], -node[0]) for node in self.graph.nodes()}
        nx.draw(self.graph, pos, with_labels=True, node_size=300, font_size=8)
        plt.title(f"Generated Network ({self.size}, {self.variant}, {self.topology})")
        plt.grid(True)
        plt.show()


    # --------------------
    # Modification 1: deactivate 1/5 of all possible nodes
    # --------------------
    def _build_node_mask(self):
        """Deactivate node_drop_fraction of all (grid+1)^2 positions."""
        all_positions = [
            (x, y)
            for x in range(self.grid_size + 1)
            for y in range(self.grid_size + 1)
        ]
        drop = int(self.node_drop_fraction * len(all_positions))
        deactivated = set(random.sample(all_positions, drop)) if drop > 0 else set()
        self.active_positions = set(all_positions) - deactivated


    # --------------------
    # Node initialization and placement
    # --------------------
    def _initialize_graph(self):
        self.graph = nx.Graph()

        # Prefer to seed from the middle region, but only from active positions.
        margin = max(1, self.grid_size // 4)
        low, high = margin, self.grid_size - margin

        middle_active = [(x, y) for (x, y) in self.active_positions if low <= x <= high and low <= y <= high]
        if middle_active:
            seed = random.choice(middle_active)
        else:
            seed = random.choice(list(self.active_positions))

        coords = np.array([seed[0], seed[1]])
        flags = np.zeros(4, dtype=int)
        self.nodes_list = [[coords, flags]]
        self.graph.add_node(tuple(coords))


    def _compute_nodes(self):
        total_possible = (self.grid_size + 1) ** 2

        # Important: total_possible is still the full grid size;
        # the mask reduces available positions and _add_nodes enforces that.
        if self.variant == "F":
            return int(self.node_factor * total_possible)
        return int(random.uniform(0.4, 0.7) * total_possible)


    def _add_nodes(self):
        """Place nodes mostly in the middle region (cluster logic), respecting active_positions."""
        total_nodes = self._compute_nodes()

        margin = max(1, self.grid_size // 4)
        low, high = margin, self.grid_size - margin

        attempts = 0
        while len(self.graph.nodes()) < total_nodes and attempts < 8000:
            attempts += 1
            x = random.randint(low, high)
            y = random.randint(low, high)

            if (x, y) not in self.active_positions:
                continue
            if (x, y) not in self.graph:
                self.graph.add_node((x, y))


    # --------------------
    # Connectivity for non-bottleneck modes
    # --------------------
    def _connect_all_nodes_by_nearby_growth(self, nodes):
        """Original connectivity step (nearby growth), unchanged except refactoring."""
        connected = set()
        remaining = set(nodes)

        current = random.choice(nodes)
        connected.add(current)
        remaining.remove(current)

        while remaining:
            candidates = [
                n for n in remaining
                if any(abs(n[0] - c[0]) <= 2 and abs(n[1] - c[1]) <= 2 for c in connected)
            ]

            candidate = random.choice(candidates) if candidates else random.choice(list(remaining))

            neighbors = [
                c for c in connected
                if abs(c[0] - candidate[0]) <= 2 and abs(c[1] - candidate[1]) <= 2
            ]
            n = random.choice(neighbors) if neighbors else random.choice(list(connected))

            # Keep your existing intersection checks (but connectivity is forced anyway)
            if n[0] == candidate[0] or n[1] == candidate[1]:
                _ = self._straight_edge_intersects(n, candidate)
            elif abs(n[0] - candidate[0]) == abs(n[1] - candidate[1]):
                _ = self._diagonal_intersects(n, candidate)

            self.graph.add_edge(n, candidate)

            connected.add(candidate)
            remaining.remove(candidate)


    # --------------------
    # Modification 2: Bottleneck = multiple small dense clusters connected by bottleneck edges
    # --------------------
    def _build_bottleneck_clusters(self, nodes):
        """
        Build a number of small, internally dense "grids" (clusters),
        then connect clusters with a small number of inter-cluster edges
        which become the bottlenecks.
        """
        # Clear any edges that might exist (seed node has no edges, but be explicit)
        self.graph.remove_edges_from(list(self.graph.edges()))

        clusters, centers = self._spatial_cluster_nodes(nodes, k=self.bottleneck_cluster_count)

        # Make each cluster internally dense.
        # We do "dense-without-intersections-when-possible" using your dense edge routine on subsets.
        for cluster in clusters:
            if len(cluster) < 2:
                continue

            # Ensure cluster is connected first using nearby growth inside the cluster
            self._connect_cluster_by_nearby_growth(cluster)

            # Then densify within the cluster
            max_edges = max(1, int(3.0 * len(cluster)))  # dense-ish without becoming fully complete
            self._add_cluster_dense(list(cluster), max_edges=max_edges)

        # Connect clusters in a chain (or near-chain) by centers.
        order = sorted(range(len(clusters)), key=lambda i: (centers[i][0], centers[i][1]))
        for a_idx, b_idx in zip(order[:-1], order[1:]):
            self._add_bottleneck_links(clusters[a_idx], clusters[b_idx], self.bottleneck_edges_per_link)

        # If something ended up disconnected (e.g., tiny clusters), reconnect lightly
        if not nx.is_connected(self.graph):
            self._attempt_reconnect_components(prefer_max_distance=2)


    def _spatial_cluster_nodes(self, nodes, k):
        """
        Simple spatial clustering:
        - pick k random centers
        - assign each node to closest center by Chebyshev distance
        - return clusters + centers
        """
        def cheb(a, b):
            return max(abs(a[0] - b[0]), abs(a[1] - b[1]))

        nodes = list(nodes)
        if k >= len(nodes):
            # each node its own cluster (degenerate)
            return [[n] for n in nodes], nodes[:]

        centers = random.sample(nodes, k)
        clusters = [[] for _ in range(k)]

        for n in nodes:
            best_i = min(range(k), key=lambda i: cheb(n, centers[i]))
            clusters[best_i].append(n)

        # Recompute centers as medoid-ish: pick node closest to mean
        new_centers = []
        for c in clusters:
            if not c:
                new_centers.append(random.choice(nodes))
                continue
            mx = sum(p[0] for p in c) / len(c)
            my = sum(p[1] for p in c) / len(c)
            new_centers.append(min(c, key=lambda p: (p[0] - mx) ** 2 + (p[1] - my) ** 2))

        # Remove empty clusters by merging them into nearest non-empty cluster
        non_empty = [(c, ctr) for c, ctr in zip(clusters, new_centers) if len(c) > 0]
        clusters = [c for c, _ in non_empty]
        centers = [ctr for _, ctr in non_empty]

        return clusters, centers


    def _connect_cluster_by_nearby_growth(self, cluster_nodes):
        """Connectivity step restricted to a cluster."""
        cluster_nodes = list(cluster_nodes)
        connected = set()
        remaining = set(cluster_nodes)

        current = random.choice(cluster_nodes)
        connected.add(current)
        remaining.remove(current)

        while remaining:
            candidates = [
                n for n in remaining
                if any(abs(n[0] - c[0]) <= 2 and abs(n[1] - c[1]) <= 2 for c in connected)
            ]
            candidate = random.choice(candidates) if candidates else random.choice(list(remaining))

            neighbors = [
                c for c in connected
                if abs(c[0] - candidate[0]) <= 2 and abs(c[1] - candidate[1]) <= 2
            ]
            n = random.choice(neighbors) if neighbors else random.choice(list(connected))

            self.graph.add_edge(n, candidate)
            connected.add(candidate)
            remaining.remove(candidate)


    def _add_bottleneck_links(self, cluster_a, cluster_b, m):
        """
        Add m inter-cluster edges as bottlenecks. Keep m small.
        Prefer edges that do not create intersections, but will force-connect if needed.
        """
        cluster_a = list(cluster_a)
        cluster_b = list(cluster_b)

        def cheb(a, b):
            return max(abs(a[0] - b[0]), abs(a[1] - b[1]))

        # Candidate pairs sorted by distance
        pairs = []
        for u in cluster_a:
            for v in cluster_b:
                pairs.append((cheb(u, v), u, v))
        pairs.sort(key=lambda t: t[0])

        added = 0
        used = set()
        for _, u, v in pairs:
            if added >= m:
                break
            if (u, v) in used or (v, u) in used:
                continue
            if self.graph.has_edge(u, v):
                continue

            # Prefer non-intersecting links
            if not self._would_create_intersection(u, v):
                self.graph.add_edge(u, v)
                used.add((u, v))
                added += 1

        # If we couldn't add enough without intersections, force the closest remaining
        if added < m:
            for _, u, v in pairs:
                if added >= m:
                    break
                if self.graph.has_edge(u, v):
                    continue
                self.graph.add_edge(u, v)
                added += 1


    # --------------------
    # Topology-specific extra edges (non-bottleneck modes)
    # --------------------
    def _compute_edge_count(self):
        total_nodes = len(self.graph.nodes())
        if self.variant == "F":
            return int(1.5 * total_nodes)
        return int(random.uniform(1.5, 2.5) * total_nodes)


    def _add_edges(self):
        nodes = list(self.graph.nodes())
        total_edges = self._compute_edge_count()

        if self.topology == "highly_connected":
            self._add_cluster_dense(nodes, total_edges)

        elif self.topology == "linear":
            self._make_linear(nodes)

        # Note: bottlenecks are built in _build_bottleneck_clusters(), not here.


    # --------------------
    # Dense / sparse edge routines (existing)
    # --------------------
    def _add_cluster_dense(self, nodes, max_edges):
        edges_added = 0
        nodes = list(nodes)
        random.shuffle(nodes)

        for i in range(len(nodes)):
            for j in range(i + 1, len(nodes)):
                if edges_added >= max_edges:
                    return
                n1, n2 = nodes[i], nodes[j]

                # Straight edge
                if (n1[0] == n2[0] or n1[1] == n2[1]):
                    if not self._straight_edge_intersects(n1, n2):
                        self.graph.add_edge(n1, n2)
                        edges_added += 1
                        continue

                # Diagonal edge
                if abs(n1[0] - n2[0]) == abs(n1[1] - n2[1]):
                    if not self._diagonal_intersects(n1, n2):
                        self.graph.add_edge(n1, n2)
                        edges_added += 1


    def _make_linear(self, nodes):
        nodes_sorted = sorted(nodes, key=lambda x: (x[0], x[1]))
        if not nodes_sorted:
            return

        prev = nodes_sorted[0]
        for nxt in nodes_sorted[1:]:
            x1, y1 = prev
            x2, y2 = nxt

            if x1 == x2 or y1 == y2:
                self.graph.add_edge(prev, nxt)
                prev = nxt
            else:
                if x1 != x2:
                    step = (x1 + (1 if x2 > x1 else -1), y1)
                    if step in nodes:
                        self.graph.add_edge(prev, step)
                        self.graph.add_edge(step, nxt)
                        prev = nxt
                        continue

                if y1 != y2:
                    step = (x1, y1 + (1 if y2 > y1 else -1))
                    if step in nodes:
                        self.graph.add_edge(prev, step)
                        self.graph.add_edge(step, nxt)
                        prev = nxt
                        continue

        for node in nodes_sorted:
            if random.random() < 0.15:
                x, y = node
                candidates = [(x + 1, y), (x - 1, y), (x, y + 1), (x, y - 1)]
                random.shuffle(candidates)

                for c in candidates:
                    if c in nodes and not self.graph.has_edge(node, c):
                        if self.graph.degree(node) < 3 and self.graph.degree(c) < 3:
                            self.graph.add_edge(node, c)
                        break


    # --------------------
    # Intersection checks (existing + used by reconnect)
    # --------------------
    def _straight_edge_intersects(self, n1, n2):
        x1, y1 = n1
        x2, y2 = n2

        if not (x1 == x2 or y1 == y2):
            return True

        if (x1, y1) > (x2, y2):
            n1, n2 = n2, n1
            x1, y1 = n1
            x2, y2 = n2

        for a, b in self.graph.edges():
            if {a, b} == {n1, n2}:
                continue

            ax, ay = a
            bx, by = b

            if y1 == y2:  # horizontal
                if ay == by == y1:
                    if max(ax, bx) >= min(x1, x2) and min(ax, bx) <= max(x1, x2):
                        return True

            if x1 == x2:  # vertical
                if ax == bx == x1:
                    if max(ay, by) >= min(y1, y2) and min(ay, by) <= max(y1, y2):
                        return True

        return False


    def _diagonal_intersects(self, n1, n2):
        x1, y1 = n1
        x2, y2 = n2

        for a, b in self.graph.edges():
            ax, ay = a
            bx, by = b

            if abs(ax - bx) == abs(ay - by):
                if not (max(x1, x2) < min(ax, bx) or min(x1, x2) > max(ax, bx)):
                    if not (max(y1, y2) < min(ay, by) or min(y1, y2) > max(ay, by)):
                        return True

        return False


    def _orientation(self, p, q, r):
        (px, py), (qx, qy), (rx, ry) = p, q, r
        val = (qy - py) * (rx - qx) - (qx - px) * (ry - qy)
        if val == 0:
            return 0
        return 1 if val > 0 else 2


    def _on_segment(self, p, q, r):
        (px, py), (qx, qy), (rx, ry) = p, q, r
        return (min(px, rx) <= qx <= max(px, rx) and
                min(py, ry) <= qy <= max(py, ry))


    def _segments_intersect(self, a, b, c, d):
        if a in (c, d) or b in (c, d):
            return False

        o1 = self._orientation(a, b, c)
        o2 = self._orientation(a, b, d)
        o3 = self._orientation(c, d, a)
        o4 = self._orientation(c, d, b)

        if o1 != o2 and o3 != o4:
            return True

        if o1 == 0 and self._on_segment(a, c, b):
            return True
        if o2 == 0 and self._on_segment(a, d, b):
            return True
        if o3 == 0 and self._on_segment(c, a, d):
            return True
        if o4 == 0 and self._on_segment(c, b, d):
            return True

        return False


    def _would_create_intersection(self, u, v):
        for x, y in self.graph.edges():
            if u in (x, y) or v in (x, y):
                continue
            if self._segments_intersect(u, v, x, y):
                return True
        return False


    def _remove_intersections(self):
        max_passes = 10
        pass_no = 0
        total_removed = 0

        while pass_no < max_passes:
            pass_no += 1
            edges = list(self.graph.edges())
            intersections = []

            for i in range(len(edges)):
                a, b = edges[i]
                for j in range(i + 1, len(edges)):
                    c, d = edges[j]
                    if self._segments_intersect(a, b, c, d):
                        intersections.append((a, b, c, d))

            if not intersections:
                break

            removed_this_pass = 0
            for a, b, c, d in intersections:
                if not self.graph.has_edge(a, b) or not self.graph.has_edge(c, d):
                    continue

                len1 = (a[0] - b[0]) ** 2 + (a[1] - b[1]) ** 2
                len2 = (c[0] - d[0]) ** 2 + (c[1] - d[1]) ** 2

                if len1 >= len2:
                    try:
                        self.graph.remove_edge(a, b)
                        removed_this_pass += 1
                    except Exception:
                        pass
                else:
                    try:
                        self.graph.remove_edge(c, d)
                        removed_this_pass += 1
                    except Exception:
                        pass

            total_removed += removed_this_pass
            self._attempt_reconnect_components(prefer_max_distance=2)

        if not nx.is_connected(self.graph):
            self._attempt_reconnect_components(prefer_max_distance=self.grid_size)

        final_edges = list(self.graph.edges())
        for i in range(len(final_edges)):
            a, b = final_edges[i]
            for j in range(i + 1, len(final_edges)):
                c, d = final_edges[j]
                if self._segments_intersect(a, b, c, d):
                    len1 = (a[0] - b[0]) ** 2 + (a[1] - b[1]) ** 2
                    len2 = (c[0] - d[0]) ** 2 + (c[1] - d[1]) ** 2
                    if len1 >= len2 and self.graph.has_edge(a, b):
                        self.graph.remove_edge(a, b)
                        total_removed += 1
                    elif self.graph.has_edge(c, d):
                        self.graph.remove_edge(c, d)
                        total_removed += 1

        print(f"[cleanup] Removed {total_removed} intersecting edges after {pass_no} passes.")


    def _attempt_reconnect_components(self, prefer_max_distance=2):
        comps = list(nx.connected_components(self.graph))
        if len(comps) <= 1:
            return

        def cheb(a, b):
            return max(abs(a[0] - b[0]), abs(a[1] - b[1]))

        comp_nodes = [list(c) for c in comps]
        max_relax = self.grid_size
        relax = prefer_max_distance

        while relax <= max_relax and len(comp_nodes) > 1:
            made_connection = False

            i = 0
            while i < len(comp_nodes) - 1:
                j = i + 1
                connected_this_round = False
                while j < len(comp_nodes):
                    best_pair = None
                    best_dist = None

                    for u in comp_nodes[i]:
                        for v in comp_nodes[j]:
                            if u == v:
                                continue
                            d = cheb(u, v)
                            if d <= relax and (best_dist is None or d < best_dist):
                                best_pair = (u, v)
                                best_dist = d

                    if best_pair is not None:
                        u, v = best_pair
                        if not self.graph.has_edge(u, v):
                            if not self._would_create_intersection(u, v):
                                self.graph.add_edge(u, v)
                            else:
                                # force if no clean option
                                self.graph.add_edge(u, v)

                            made_connection = True
                            connected_this_round = True
                            comp_nodes[i].extend(comp_nodes[j])
                            comp_nodes.pop(j)
                            break
                    else:
                        j += 1

                if not connected_this_round:
                    i += 1

            if not made_connection:
                relax += 1
            else:
                comps = list(nx.connected_components(self.graph))
                comp_nodes = [list(c) for c in comps]