From 7f8eb4de83b21183e631be360f8d4bd77f335bd2 Mon Sep 17 00:00:00 2001
From: Eric Fischer <enf@pobox.com>
Date: Tue, 15 Dec 2015 15:52:08 -0800
Subject: [PATCH] Add an option to guess an appropriately dense base zoom level

---
 CHANGELOG.md     |  5 +++
 README.md        |  3 +-
 geojson.c        | 83 +++++++++++++++++++++++++++++++++++++++++++++---
 man/tippecanoe.1 |  3 +-
 version.h        |  2 +-
 5 files changed, 89 insertions(+), 7 deletions(-)

diff --git a/CHANGELOG.md b/CHANGELOG.md
index 10e4a5d..0b13258 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -1,3 +1,8 @@
+## 1.5.0
+
+* Base zoom for dot-dropping can be specified independently of
+  maxzoom for tiling
+
 ## 1.4.3
 
 * Encode numeric attributes as integers instead of floating point if possible
diff --git a/README.md b/README.md
index 04f86e3..71b7d77 100644
--- a/README.md
+++ b/README.md
@@ -68,7 +68,8 @@ Options
 
  * -z _zoom_: Maxzoom: the highest zoom level for which tiles are generated (default 14)
  * -Z _zoom_: Minzoom: the lowest zoom level for which tiles are generated (default 0)
- * -B _zoom_: Base zoom, the level at and above which all points are included in the tiles (default maxzoom)
+ * -B _zoom_: Base zoom, the level at and above which all points are included in the tiles (default maxzoom).
+   If you use -Bg, it will guess a zoom level that will keep less than 50,000 features in the densest tile.
  * -d _detail_: Detail at max zoom level (default 12, for tile resolution of 4096)
  * -D _detail_: Detail at lower zoom levels (default 12, for tile resolution of 4096)
  * -m _detail_: Minimum detail that it will try if tiles are too big at regular detail (default 7)
diff --git a/geojson.c b/geojson.c
index a9e43bf..17bb30d 100644
--- a/geojson.c
+++ b/geojson.c
@@ -1035,13 +1035,84 @@ int read_json(int argc, char **argv, char *fname, const char *layername, int max
 		close(indexfd);
 	}
 
-	/* Copy geometries to a new file in index order */
-
 	indexfd = open(indexname, O_RDONLY);
 	if (indexfd < 0) {
 		perror("reopen sorted index");
 		exit(EXIT_FAILURE);
 	}
+
+	if (basezoom < 0) {
+		struct index *map = mmap(NULL, indexpos, PROT_READ, MAP_PRIVATE, indexfd, 0);
+		if (map == MAP_FAILED) {
+			perror("mmap");
+			exit(EXIT_FAILURE);
+		}
+
+#define MAX_ZOOM 30
+		struct tile {
+			unsigned x;
+			unsigned y;
+			long long count;
+		} tile[MAX_ZOOM + 1], max[MAX_ZOOM + 1];
+
+		{
+			int i;
+			for (i = 0; i <= MAX_ZOOM; i++) {
+				tile[i].x = tile[i].y = tile[i].count = 0;
+				max[i].x = max[i].y = max[i].count = 0;
+			}
+		}
+
+		long long indices = indexpos / sizeof(struct index);
+		long long i;
+		for (i = 0; i < indices; i++) {
+			unsigned xx, yy;
+			decode(map[i].index, &xx, &yy);
+
+			int z;
+			for (z = 0; z <= MAX_ZOOM; z++) {
+				unsigned xxx = 0, yyy = 0;
+				if (z != 0) {
+					xxx = xx >> (32 - z);
+					yyy = yy >> (32 - z);
+				}
+
+				if (tile[z].x != xxx || tile[z].y != yyy) {
+					if (tile[z].count > max[z].count) {
+						max[z] = tile[z];
+					}
+
+					tile[z].x = xxx;
+					tile[z].y = yyy;
+					tile[z].count = 0;
+				}
+
+				tile[z].count++;
+			}
+		}
+
+		basezoom = MAX_ZOOM;
+
+		int z;
+		for (z = MAX_ZOOM; z >= 0; z--) {
+			if (tile[z].count > max[z].count) {
+				max[z] = tile[z];
+			}
+
+			if (max[z].count < 50000) {
+				basezoom = z;
+			}
+
+			// printf("%d/%u/%u %lld\n", z, max[z].x, max[z].y, max[z].count);
+		}
+
+		fprintf(stderr, "Choosing a base zoom of -B%d to keep %lld features in tile %d/%u/%u.\n", basezoom, max[basezoom].count, basezoom, max[basezoom].x, max[basezoom].y);
+
+		munmap(map, indexpos);
+	}
+
+	/* Copy geometries to a new file in index order */
+
 	struct index *index_map = mmap(NULL, indexpos, PROT_READ, MAP_PRIVATE, indexfd, 0);
 	if (index_map == MAP_FAILED) {
 		perror("mmap index");
@@ -1262,7 +1333,11 @@ int main(int argc, char **argv) {
 			break;
 
 		case 'B':
-			basezoom = atoi(optarg);
+			if (strcmp(optarg, "g") == 0) {
+				basezoom = -2;
+			} else {
+				basezoom = atoi(optarg);
+			}
 			break;
 
 		case 'd':
@@ -1347,7 +1422,7 @@ int main(int argc, char **argv) {
 		exit(EXIT_FAILURE);
 	}
 
-	if (basezoom < 0) {
+	if (basezoom == -1) {
 		basezoom = maxzoom;
 	}
 
diff --git a/man/tippecanoe.1 b/man/tippecanoe.1
index 8754e4b..6f677bd 100644
--- a/man/tippecanoe.1
+++ b/man/tippecanoe.1
@@ -75,7 +75,8 @@ specified, the files are all merged into the single named layer.
 .IP \(bu 2
 \-Z \fIzoom\fP: Minzoom: the lowest zoom level for which tiles are generated (default 0)
 .IP \(bu 2
-\-B \fIzoom\fP: Base zoom, the level at and above which all points are included in the tiles (default maxzoom)
+\-B \fIzoom\fP: Base zoom, the level at and above which all points are included in the tiles (default maxzoom).
+If you use \-Bg, it will guess a zoom level that will keep less than 50,000 features in the densest tile.
 .IP \(bu 2
 \-d \fIdetail\fP: Detail at max zoom level (default 12, for tile resolution of 4096)
 .IP \(bu 2
diff --git a/version.h b/version.h
index 5adfc75..955f192 100644
--- a/version.h
+++ b/version.h
@@ -1 +1 @@
-#define VERSION "tippecanoe v1.4.3\n"
+#define VERSION "tippecanoe v1.5.0\n"