Clean up numeric.h and split FacePositionCache from it

I also optiized FacePositionCache a bit: I removed a map
lookup and vector copy from both branches of getFacePosition.

17 files changed, 205 insertions, 157 deletions

diff --git a/build/android/jni/Android.mk b/build/android/jni/Android.mk
index d4a8d988f..44d98fada 100644
--- a/build/android/jni/Android.mk
+++ b/build/android/jni/Android.mk
@@ -143,6 +143,7 @@ LOCAL_SRC_FILES := \
 		jni/src/dungeongen.cpp                    \
 		jni/src/emerge.cpp                        \
 		jni/src/environment.cpp                   \
+		jni/src/face_position_cache.cpp           \
 		jni/src/filecache.cpp                     \
 		jni/src/filesys.cpp                       \
 		jni/src/fontengine.cpp                    \
diff --git a/src/CMakeLists.txt b/src/CMakeLists.txt
index 4c705cec8..7c1a4eee0 100644
--- a/src/CMakeLists.txt
+++ b/src/CMakeLists.txt
@@ -388,6 +388,7 @@ set(common_SRCS
 	dungeongen.cpp
 	emerge.cpp
 	environment.cpp
+	face_position_cache.cpp
 	filesys.cpp
 	genericobject.cpp
 	gettext.cpp
diff --git a/src/clientenvironment.cpp b/src/clientenvironment.cpp
index 36e4437b6..47df49b84 100644
--- a/src/clientenvironment.cpp
+++ b/src/clientenvironment.cpp
@@ -30,6 +30,7 @@ with this program; if not, write to the Free Software Foundation, Inc.,
 #include "raycast.h"
 #include "voxelalgorithms.h"
 #include "settings.h"
+#include <algorithm>
 
 /*
 	ClientEnvironment
diff --git a/src/clientiface.cpp b/src/clientiface.cpp
index 7f476f0ea..78339055f 100644
--- a/src/clientiface.cpp
+++ b/src/clientiface.cpp
@@ -20,7 +20,6 @@ with this program; if not, write to the Free Software Foundation, Inc.,
 #include <sstream>
 
 #include "clientiface.h"
-#include "util/numeric.h"
 #include "remoteplayer.h"
 #include "settings.h"
 #include "mapblock.h"
@@ -32,6 +31,7 @@ with this program; if not, write to the Free Software Foundation, Inc.,
 #include "log.h"
 #include "network/serveropcodes.h"
 #include "util/srp.h"
+#include "face_position_cache.h"
 
 const char *ClientInterface::statenames[] = {
 	"Invalid",
diff --git a/src/content_cao.cpp b/src/content_cao.cpp
index d18a0233f..d0d3eb84f 100644
--- a/src/content_cao.cpp
+++ b/src/content_cao.cpp
@@ -44,6 +44,7 @@ with this program; if not, write to the Free Software Foundation, Inc.,
 #include "camera.h" // CameraModes
 #include "wieldmesh.h"
 #include "log.h"
+#include <algorithm>
 
 class Settings;
 struct ToolCapabilities;
diff --git a/src/face_position_cache.cpp b/src/face_position_cache.cpp
new file mode 100644
index 000000000..f57e75da9
--- /dev/null
+++ b/src/face_position_cache.cpp
@@ -0,0 +1,110 @@
+/*
+Minetest
+Copyright (C) 2015 Nerzhul, Loic Blot <loic.blot@unix-experience.fr>
+
+This program is free software; you can redistribute it and/or modify
+it under the terms of the GNU Lesser General Public License as published by
+the Free Software Foundation; either version 2.1 of the License, or
+(at your option) any later version.
+
+This program is distributed in the hope that it will be useful,
+but WITHOUT ANY WARRANTY; without even the implied warranty of
+MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+GNU Lesser General Public License for more details.
+
+You should have received a copy of the GNU Lesser General Public License along
+with this program; if not, write to the Free Software Foundation, Inc.,
+51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA.
+*/
+
+#include "face_position_cache.h"
+#include "threading/mutex_auto_lock.h"
+
+
+UNORDERED_MAP<u16, std::vector<v3s16> > FacePositionCache::cache;
+Mutex FacePositionCache::cache_mutex;
+
+// Calculate the borders of a "d-radius" cube
+const std::vector<v3s16> &FacePositionCache::getFacePositions(u16 d)
+{
+	MutexAutoLock lock(cache_mutex);
+	UNORDERED_MAP<u16, std::vector<v3s16> >::iterator it = cache.find(d);
+	if (it != cache.end())
+		return it->second;
+
+	return generateFacePosition(d);
+}
+
+const std::vector<v3s16> &FacePositionCache::generateFacePosition(u16 d)
+{
+	cache[d] = std::vector<v3s16>();
+	std::vector<v3s16> &c = cache[d];
+	if (d == 0) {
+		c.push_back(v3s16(0,0,0));
+		return c;
+	}
+	if (d == 1) {
+		// This is an optimized sequence of coordinates.
+		c.push_back(v3s16( 0, 1, 0)); // Top
+		c.push_back(v3s16( 0, 0, 1)); // Back
+		c.push_back(v3s16(-1, 0, 0)); // Left
+		c.push_back(v3s16( 1, 0, 0)); // Right
+		c.push_back(v3s16( 0, 0,-1)); // Front
+		c.push_back(v3s16( 0,-1, 0)); // Bottom
+		// 6
+		c.push_back(v3s16(-1, 0, 1)); // Back left
+		c.push_back(v3s16( 1, 0, 1)); // Back right
+		c.push_back(v3s16(-1, 0,-1)); // Front left
+		c.push_back(v3s16( 1, 0,-1)); // Front right
+		c.push_back(v3s16(-1,-1, 0)); // Bottom left
+		c.push_back(v3s16( 1,-1, 0)); // Bottom right
+		c.push_back(v3s16( 0,-1, 1)); // Bottom back
+		c.push_back(v3s16( 0,-1,-1)); // Bottom front
+		c.push_back(v3s16(-1, 1, 0)); // Top left
+		c.push_back(v3s16( 1, 1, 0)); // Top right
+		c.push_back(v3s16( 0, 1, 1)); // Top back
+		c.push_back(v3s16( 0, 1,-1)); // Top front
+		// 18
+		c.push_back(v3s16(-1, 1, 1)); // Top back-left
+		c.push_back(v3s16( 1, 1, 1)); // Top back-right
+		c.push_back(v3s16(-1, 1,-1)); // Top front-left
+		c.push_back(v3s16( 1, 1,-1)); // Top front-right
+		c.push_back(v3s16(-1,-1, 1)); // Bottom back-left
+		c.push_back(v3s16( 1,-1, 1)); // Bottom back-right
+		c.push_back(v3s16(-1,-1,-1)); // Bottom front-left
+		c.push_back(v3s16( 1,-1,-1)); // Bottom front-right
+		// 26
+		return c;
+	}
+
+	// Take blocks in all sides, starting from y=0 and going +-y
+	for (s16 y = 0; y <= d - 1; y++) {
+		// Left and right side, including borders
+		for (s16 z =- d; z <= d; z++) {
+			c.push_back(v3s16( d, y, z));
+			c.push_back(v3s16(-d, y, z));
+			if (y != 0) {
+				c.push_back(v3s16( d, -y, z));
+				c.push_back(v3s16(-d, -y, z));
+			}
+		}
+		// Back and front side, excluding borders
+		for (s16 x = -d + 1; x <= d - 1; x++) {
+			c.push_back(v3s16(x, y, d));
+			c.push_back(v3s16(x, y, -d));
+			if (y != 0) {
+				c.push_back(v3s16(x, -y, d));
+				c.push_back(v3s16(x, -y, -d));
+			}
+		}
+	}
+
+	// Take the bottom and top face with borders
+	// -d < x < d, y = +-d, -d < z < d
+	for (s16 x = -d; x <= d; x++)
+	for (s16 z = -d; z <= d; z++) {
+		c.push_back(v3s16(x, -d, z));
+		c.push_back(v3s16(x, d, z));
+	}
+	return c;
+}
diff --git a/src/face_position_cache.h b/src/face_position_cache.h
new file mode 100644
index 000000000..c1d2841c4
--- /dev/null
+++ b/src/face_position_cache.h
@@ -0,0 +1,44 @@
+/*
+Minetest
+Copyright (C) 2015 Nerzhul, Loic Blot <loic.blot@unix-experience.fr>
+
+This program is free software; you can redistribute it and/or modify
+it under the terms of the GNU Lesser General Public License as published by
+the Free Software Foundation; either version 2.1 of the License, or
+(at your option) any later version.
+
+This program is distributed in the hope that it will be useful,
+but WITHOUT ANY WARRANTY; without even the implied warranty of
+MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+GNU Lesser General Public License for more details.
+
+You should have received a copy of the GNU Lesser General Public License along
+with this program; if not, write to the Free Software Foundation, Inc.,
+51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA.
+*/
+
+#ifndef FACE_POSITION_CACHE_HEADER
+#define FACE_POSITION_CACHE_HEADER
+
+#include "irr_v3d.h"
+#include "threading/mutex.h"
+#include "util/cpp11_container.h"
+
+#include <map>
+#include <vector>
+
+/*
+ * This class permits caching getFacePosition call results.
+ * This reduces CPU usage and vector calls.
+ */
+class FacePositionCache {
+public:
+	static const std::vector<v3s16> &getFacePositions(u16 d);
+
+private:
+	static const std::vector<v3s16> &generateFacePosition(u16 d);
+	static UNORDERED_MAP<u16, std::vector<v3s16> > cache;
+	static Mutex cache_mutex;
+};
+
+#endif
diff --git a/src/mg_decoration.cpp b/src/mg_decoration.cpp
index 51e4fbbcc..ec31a9c01 100644
--- a/src/mg_decoration.cpp
+++ b/src/mg_decoration.cpp
@@ -24,6 +24,8 @@ with this program; if not, write to the Free Software Foundation, Inc.,
 #include "map.h"
 #include "log.h"
 #include "util/numeric.h"
+#include <algorithm>
+
 
 FlagDesc flagdesc_deco[] = {
 	{"place_center_x",  DECO_PLACE_CENTER_X},
diff --git a/src/mg_ore.cpp b/src/mg_ore.cpp
index d840d745a..c6c1c45a7 100644
--- a/src/mg_ore.cpp
+++ b/src/mg_ore.cpp
@@ -20,9 +20,10 @@ with this program; if not, write to the Free Software Foundation, Inc.,
 #include "mg_ore.h"
 #include "mapgen.h"
 #include "noise.h"
-#include "util/numeric.h"
 #include "map.h"
 #include "log.h"
+#include <algorithm>
+
 
 FlagDesc flagdesc_ore[] = {
 	{"absheight",                 OREFLAG_ABSHEIGHT},
diff --git a/src/script/lua_api/l_env.cpp b/src/script/lua_api/l_env.cpp
index 0ded12658..b8b6bc5ba 100644
--- a/src/script/lua_api/l_env.cpp
+++ b/src/script/lua_api/l_env.cpp
@@ -35,6 +35,7 @@ with this program; if not, write to the Free Software Foundation, Inc.,
 #include "treegen.h"
 #include "emerge.h"
 #include "pathfinder.h"
+#include "face_position_cache.h"
 
 struct EnumString ModApiEnvMod::es_ClearObjectsMode[] =
 {
diff --git a/src/script/lua_api/l_server.cpp b/src/script/lua_api/l_server.cpp
index ea993d7b7..a0e475dec 100644
--- a/src/script/lua_api/l_server.cpp
+++ b/src/script/lua_api/l_server.cpp
@@ -26,6 +26,7 @@ with this program; if not, write to the Free Software Foundation, Inc.,
 #include "environment.h"
 #include "player.h"
 #include "log.h"
+#include <algorithm>
 
 // request_shutdown()
 int ModApiServer::l_request_shutdown(lua_State *L)
diff --git a/src/unittest/test_noderesolver.cpp b/src/unittest/test_noderesolver.cpp
index 55acece6a..b009f5d2e 100644
--- a/src/unittest/test_noderesolver.cpp
+++ b/src/unittest/test_noderesolver.cpp
@@ -24,6 +24,8 @@ with this program; if not, write to the Free Software Foundation, Inc.,
 #include "gamedef.h"
 #include "nodedef.h"
 
+#include <algorithm>
+
 
 class TestNodeResolver : public TestBase {
 public:
diff --git a/src/util/basic_macros.h b/src/util/basic_macros.h
index bd4b890eb..687d7cf85 100644
--- a/src/util/basic_macros.h
+++ b/src/util/basic_macros.h
@@ -20,14 +20,13 @@ with this program; if not, write to the Free Software Foundation, Inc.,
 #ifndef BASICMACROS_HEADER
 #define BASICMACROS_HEADER
 
-#include <algorithm>
-
 #define ARRLEN(x) (sizeof(x) / sizeof((x)[0]))
 
 #define MYMIN(a, b) ((a) < (b) ? (a) : (b))
 
 #define MYMAX(a, b) ((a) > (b) ? (a) : (b))
 
+// Requires <algorithm>
 #define CONTAINS(c, v) (std::find((c).begin(), (c).end(), (v)) != (c).end())
 
 // To disable copy constructors and assignment operations for some class
diff --git a/src/util/numeric.cpp b/src/util/numeric.cpp
index 87f1040ea..e6a9cb75e 100644
--- a/src/util/numeric.cpp
+++ b/src/util/numeric.cpp
@@ -24,100 +24,9 @@ with this program; if not, write to the Free Software Foundation, Inc.,
 #include "../noise.h" // PseudoRandom, PcgRandom
 #include "../threading/mutex_auto_lock.h"
 #include <string.h>
-#include <iostream>
 
-UNORDERED_MAP<u16, std::vector<v3s16> > FacePositionCache::m_cache;
-Mutex FacePositionCache::m_cache_mutex;
-// Calculate the borders of a "d-radius" cube
-// TODO: Make it work without mutex and data races, probably thread-local
-std::vector<v3s16> FacePositionCache::getFacePositions(u16 d)
-{
-	MutexAutoLock cachelock(m_cache_mutex);
-	if (m_cache.find(d) != m_cache.end())
-		return m_cache[d];
-
-	generateFacePosition(d);
-	return m_cache[d];
-
-}
-
-void FacePositionCache::generateFacePosition(u16 d)
-{
-	m_cache[d] = std::vector<v3s16>();
-	if(d == 0) {
-		m_cache[d].push_back(v3s16(0,0,0));
-		return;
-	}
-	if(d == 1) {
-		/*
-			This is an optimized sequence of coordinates.
-		*/
-		m_cache[d].push_back(v3s16( 0, 1, 0)); // top
-		m_cache[d].push_back(v3s16( 0, 0, 1)); // back
-		m_cache[d].push_back(v3s16(-1, 0, 0)); // left
-		m_cache[d].push_back(v3s16( 1, 0, 0)); // right
-		m_cache[d].push_back(v3s16( 0, 0,-1)); // front
-		m_cache[d].push_back(v3s16( 0,-1, 0)); // bottom
-		// 6
-		m_cache[d].push_back(v3s16(-1, 0, 1)); // back left
-		m_cache[d].push_back(v3s16( 1, 0, 1)); // back right
-		m_cache[d].push_back(v3s16(-1, 0,-1)); // front left
-		m_cache[d].push_back(v3s16( 1, 0,-1)); // front right
-		m_cache[d].push_back(v3s16(-1,-1, 0)); // bottom left
-		m_cache[d].push_back(v3s16( 1,-1, 0)); // bottom right
-		m_cache[d].push_back(v3s16( 0,-1, 1)); // bottom back
-		m_cache[d].push_back(v3s16( 0,-1,-1)); // bottom front
-		m_cache[d].push_back(v3s16(-1, 1, 0)); // top left
-		m_cache[d].push_back(v3s16( 1, 1, 0)); // top right
-		m_cache[d].push_back(v3s16( 0, 1, 1)); // top back
-		m_cache[d].push_back(v3s16( 0, 1,-1)); // top front
-		// 18
-		m_cache[d].push_back(v3s16(-1, 1, 1)); // top back-left
-		m_cache[d].push_back(v3s16( 1, 1, 1)); // top back-right
-		m_cache[d].push_back(v3s16(-1, 1,-1)); // top front-left
-		m_cache[d].push_back(v3s16( 1, 1,-1)); // top front-right
-		m_cache[d].push_back(v3s16(-1,-1, 1)); // bottom back-left
-		m_cache[d].push_back(v3s16( 1,-1, 1)); // bottom back-right
-		m_cache[d].push_back(v3s16(-1,-1,-1)); // bottom front-left
-		m_cache[d].push_back(v3s16( 1,-1,-1)); // bottom front-right
-		// 26
-		return;
-	}
 
-	// Take blocks in all sides, starting from y=0 and going +-y
-	for(s16 y=0; y<=d-1; y++) {
-		// Left and right side, including borders
-		for(s16 z=-d; z<=d; z++) {
-			m_cache[d].push_back(v3s16(d,y,z));
-			m_cache[d].push_back(v3s16(-d,y,z));
-			if(y != 0) {
-				m_cache[d].push_back(v3s16(d,-y,z));
-				m_cache[d].push_back(v3s16(-d,-y,z));
-			}
-		}
-		// Back and front side, excluding borders
-		for(s16 x=-d+1; x<=d-1; x++) {
-			m_cache[d].push_back(v3s16(x,y,d));
-			m_cache[d].push_back(v3s16(x,y,-d));
-			if(y != 0) {
-				m_cache[d].push_back(v3s16(x,-y,d));
-				m_cache[d].push_back(v3s16(x,-y,-d));
-			}
-		}
-	}
-
-	// Take the bottom and top face with borders
-	// -d<x<d, y=+-d, -d<z<d
-	for(s16 x=-d; x<=d; x++)
-	for(s16 z=-d; z<=d; z++) {
-		m_cache[d].push_back(v3s16(x,-d,z));
-		m_cache[d].push_back(v3s16(x,d,z));
-	}
-}
-
-/*
-    myrand
-*/
+// myrand
 
 PcgRandom g_pcgrand;
 
diff --git a/src/util/numeric.h b/src/util/numeric.h
index 760657171..0954b7151 100644
--- a/src/util/numeric.h
+++ b/src/util/numeric.h
@@ -26,28 +26,20 @@ with this program; if not, write to the Free Software Foundation, Inc.,
 #include "../irr_v3d.h"
 #include "../irr_aabb3d.h"
 #include "../threading/mutex.h"
-#include "cpp11_container.h"
-#include <list>
-#include <vector>
 
+#define rangelim(d, min, max) ((d) < (min) ? (min) : ((d) > (max) ? (max) : (d)))
+#define myfloor(x) ((x) > 0.0 ? (int)(x) : (int)(x) - 1)
+// The naive swap performs better than the xor version
+#define SWAP(t, x, y) do { \
+	t temp = x; \
+	x = y; \
+	y = temp; \
+} while (0)
 
-/*
- * This class permits to cache getFacePosition call results
- * This reduces CPU usage and vector calls
- */
-class FacePositionCache
-{
-public:
-	static std::vector<v3s16> getFacePositions(u16 d);
-private:
-	static void generateFacePosition(u16 d);
-	static UNORDERED_MAP<u16, std::vector<v3s16> > m_cache;
-	static Mutex m_cache_mutex;
-};
 
 inline s16 getContainerPos(s16 p, s16 d)
 {
-	return (p>=0 ? p : p-d+1) / d;
+	return (p >= 0 ? p : p - d + 1) / d;
 }
 
 inline v2s16 getContainerPos(v2s16 p, s16 d)
@@ -130,16 +122,6 @@ inline bool isInArea(v3s16 p, v3s16 d)
 	);
 }
 
-#define rangelim(d, min, max) ((d) < (min) ? (min) : ((d)>(max)?(max):(d)))
-#define myfloor(x) ((x) > 0.0 ? (int)(x) : (int)(x) - 1)
-
-// The naive swap performs better than the xor version
-#define SWAP(t, x, y) do { \
-	t temp = x;            \
-	x = y;                 \
-	y = temp;              \
-} while (0)
-
 inline void sortBoxVerticies(v3s16 &p1, v3s16 &p2) {
 	if (p1.X > p2.X)
 		SWAP(s16, p1.X, p2.X);
@@ -266,11 +248,10 @@ inline s32 myround(f32 f)
 */
 inline v3s16 floatToInt(v3f p, f32 d)
 {
-	v3s16 p2(
-		(p.X + (p.X>0 ? d/2 : -d/2))/d,
-		(p.Y + (p.Y>0 ? d/2 : -d/2))/d,
-		(p.Z + (p.Z>0 ? d/2 : -d/2))/d);
-	return p2;
+	return v3s16(
+		(p.X + (p.X > 0 ? d / 2 : -d / 2)) / d,
+		(p.Y + (p.Y > 0 ? d / 2 : -d / 2)) / d,
+		(p.Z + (p.Z > 0 ? d / 2 : -d / 2)) / d);
 }
 
 /*
@@ -278,34 +259,31 @@ inline v3s16 floatToInt(v3f p, f32 d)
 */
 inline v3f intToFloat(v3s16 p, f32 d)
 {
-	v3f p2(
+	return v3f(
 		(f32)p.X * d,
 		(f32)p.Y * d,
 		(f32)p.Z * d
 	);
-	return p2;
 }
 
 // Random helper. Usually d=BS
 inline aabb3f getNodeBox(v3s16 p, float d)
 {
 	return aabb3f(
-		(float)p.X * d - 0.5*d,
-		(float)p.Y * d - 0.5*d,
-		(float)p.Z * d - 0.5*d,
-		(float)p.X * d + 0.5*d,
-		(float)p.Y * d + 0.5*d,
-		(float)p.Z * d + 0.5*d
+		(float)p.X * d - 0.5 * d,
+		(float)p.Y * d - 0.5 * d,
+		(float)p.Z * d - 0.5 * d,
+		(float)p.X * d + 0.5 * d,
+		(float)p.Y * d + 0.5 * d,
+		(float)p.Z * d + 0.5 * d
 	);
 }
 
+
 class IntervalLimiter
 {
 public:
-	IntervalLimiter():
-		m_accumulator(0)
-	{
-	}
+	IntervalLimiter() : m_accumulator(0) {}
 	/*
 		dtime: time from last call to this method
 		wanted_interval: interval wanted
@@ -316,15 +294,17 @@ public:
 	bool step(float dtime, float wanted_interval)
 	{
 		m_accumulator += dtime;
-		if(m_accumulator < wanted_interval)
+		if (m_accumulator < wanted_interval)
 			return false;
 		m_accumulator -= wanted_interval;
 		return true;
 	}
-protected:
+
+private:
 	float m_accumulator;
 };
 
+
 /*
 	Splits a list into "pages". For example, the list [1,2,3,4,5] split
 	into two pages would be [1,2,3],[4,5]. This function computes the
@@ -340,29 +320,21 @@ protected:
 */
 inline void paging(u32 length, u32 page, u32 pagecount, u32 &minindex, u32 &maxindex)
 {
-	if(length < 1 || pagecount < 1 || page < 1 || page > pagecount)
-	{
+	if (length < 1 || pagecount < 1 || page < 1 || page > pagecount) {
 		// Special cases or invalid parameters
 		minindex = maxindex = 0;
-	}
-	else if(pagecount <= length)
-	{
+	} else if(pagecount <= length) {
 		// Less pages than entries in the list:
 		// Each page contains at least one entry
 		minindex = (length * (page-1) + (pagecount-1)) / pagecount;
 		maxindex = (length * page + (pagecount-1)) / pagecount;
-	}
-	else
-	{
+	} else {
 		// More pages than entries in the list:
 		// Make sure the empty pages are at the end
-		if(page < length)
-		{
+		if (page < length) {
 			minindex = page-1;
 			maxindex = page;
-		}
-		else
-		{
+		} else {
 			minindex = 0;
 			maxindex = 0;
 		}
@@ -371,14 +343,14 @@ inline void paging(u32 length, u32 page, u32 pagecount, u32 &minindex, u32 &maxi
 
 inline float cycle_shift(float value, float by = 0, float max = 1)
 {
-    if (value + by < 0) return max + by + value;
+    if (value + by < 0)   return value + by + max;
     if (value + by > max) return value + by - max;
     return value + by;
 }
 
 inline bool is_power_of_two(u32 n)
 {
-	return n != 0 && (n & (n-1)) == 0;
+	return n != 0 && (n & (n - 1)) == 0;
 }
 
 // Compute next-higher power of 2 efficiently, e.g. for power-of-2 texture sizes.
diff --git a/src/util/string.cpp b/src/util/string.cpp
index 141068512..d41b91f24 100644
--- a/src/util/string.cpp
+++ b/src/util/string.cpp
@@ -25,6 +25,7 @@ with this program; if not, write to the Free Software Foundation, Inc.,
 #include "hex.h"
 #include "../porting.h"
 
+#include <algorithm>
 #include <sstream>
 #include <iomanip>
 #include <map>
diff --git a/util/travis/clang-format-whitelist.txt b/util/travis/clang-format-whitelist.txt
index cac2faf2d..0b290ae87 100644
--- a/util/travis/clang-format-whitelist.txt
+++ b/util/travis/clang-format-whitelist.txt
@@ -72,6 +72,8 @@ src/environment.cpp
 src/event.h
 src/event_manager.h
 src/exceptions.h
+src/face_position_cache.cpp
+src/face_position_cache.h
 src/filecache.cpp
 src/filesys.cpp
 src/filesys.h