diff --git a/data/main.lua b/data/main.lua
index 91753ab..1cde7fe 100644
--- a/data/main.lua
+++ b/data/main.lua
@@ -10,6 +10,14 @@ function _init()
     --turbo(false)
     --local perico = "péricòñ"
     --print(utf8.len(perico))
+    s = surface.load("tiles01.gif")
+    --surface.source(s)
+    p = palette.load("tiles01.gif")
+    palette.set(p)
+    print(#p)
+    surface.save(s, "data/copy.gif", p)
+    s = surface.load("copy.gif")
+    --draw.source(s)
 end
 
 function _update()
@@ -18,7 +26,8 @@ function _update()
     end
 
     draw.cls(5)
-    draw.print("HOLA",0,0)
+    draw.blit(s, 0, 0, 64, 64, 10, 10)
+    --draw.print(#p,0,0,2)
 end
 
 function normal_update()
diff --git a/data/tiles01.gif b/data/tiles01.gif
new file mode 100644
index 0000000..958bbf7
Binary files /dev/null and b/data/tiles01.gif differ
diff --git a/gif.c b/gif.c
index 8c19b14..631ae40 100644
--- a/gif.c
+++ b/gif.c
@@ -2,104 +2,31 @@
 #include <stdlib.h>
 #include <string.h>
 #include <fcntl.h>
-
-#define EXTENSION_INTRODUCER   0x21
-#define IMAGE_DESCRIPTOR       0x2C
-#define TRAILER                0x3B
-
-#define GRAPHIC_CONTROL        0xF9
-#define APPLICATION_EXTENSION  0xFF
-#define COMMENT_EXTENSION      0xFE
-#define PLAINTEXT_EXTENSION    0x01
+#include <stdint.h>
 
 #define READ(dst, size) memcpy(dst, buffer, size); buffer += size
 
-typedef struct
-{
-    unsigned short width;
-    unsigned short height;
-    unsigned char fields;
-    unsigned char background_color_index;
-    unsigned char pixel_aspect_ratio;
-}
-screen_descriptor_t;
+struct rgb { uint8_t r, g, b; };
 
-typedef struct
+struct image_descriptor_t
 {
-    unsigned char r;
-    unsigned char g;
-    unsigned char b;
-}
-rgb;
+    uint16_t left, top, width, height;
+    uint8_t  fields;
+};
 
-typedef struct
-{
-    unsigned short image_left_position;
-    unsigned short image_top_position;
-    unsigned short image_width;
-    unsigned short image_height;
-    unsigned char fields;
-}
-image_descriptor_t;
-
-typedef struct
+struct dictionary_entry_t
 {
     unsigned char byte;
-    int prev;
-    int len;
-}
-dictionary_entry_t;
+    int prev, len;
+};
 
-typedef struct
+void uncompress( int code_length, const uint8_t *input, int input_length, uint8_t *out )
 {
-    unsigned char extension_code;
-    unsigned char block_size;
-}
-extension_t;
-
-typedef struct
-{
-    unsigned char fields;
-    unsigned short delay_time;
-    unsigned char transparent_color_index;
-}
-graphic_control_extension_t;
-
-typedef struct
-{
-    unsigned char application_id[ 8 ];
-    unsigned char version[ 3 ];
-}
-application_extension_t;
-
-typedef struct
-{
-    unsigned short left;
-    unsigned short top;
-    unsigned short width;
-    unsigned short height;
-    unsigned char cell_width;
-    unsigned char cell_height;
-    unsigned char foreground_color;
-    unsigned char background_color;
-}
-plaintext_extension_t;
-
-//static unsigned short width = 0;
-//static unsigned short height = 0;
-//static unsigned char* uncompressed_data = NULL;
-
-void uncompress( int code_length,
-               const unsigned char *input,
-               int input_length,
-               unsigned char *out )
-{
-    //int maxbits;
     int i, bit;
     int code, prev = -1;
     dictionary_entry_t *dictionary;
     int dictionary_ind;
-    unsigned int mask = 0x01;
+    uint32_t mask = 0x01;
     int reset_code_length;
     int clear_code; // This varies depending on code_length
     int stop_code;  // one more than clear code
@@ -107,44 +34,24 @@ void uncompress( int code_length,
 
     clear_code = 1 << ( code_length );
     stop_code = clear_code + 1;
-    // To handle clear codes
     reset_code_length = code_length;
 
-    // Create a dictionary large enough to hold "code_length" entries.
-    // Once the dictionary overflows, code_length increases
     dictionary = ( dictionary_entry_t * )
     malloc( sizeof( dictionary_entry_t ) * ( 1 << ( code_length + 1 ) ) );
 
-    // Initialize the first 2^code_len entries of the dictionary with their
-    // indices.  The rest of the entries will be built up dynamically.
-
-    // Technically, it shouldn't be necessary to initialize the
-    // dictionary.  The spec says that the encoder "should output a
-    // clear code as the first code in the image data stream".  It doesn't
-    // say must, though...
-    for ( dictionary_ind = 0;
-         dictionary_ind < ( 1 << code_length );
-         dictionary_ind++ )
+    for ( dictionary_ind = 0; dictionary_ind < ( 1 << code_length ); dictionary_ind++ )
     {
         dictionary[ dictionary_ind ].byte = dictionary_ind;
-        // XXX this only works because prev is a 32-bit int (> 12 bits)
         dictionary[ dictionary_ind ].prev = -1;
         dictionary[ dictionary_ind ].len = 1;
     }
+    dictionary_ind+=2;
 
-    // 2^code_len + 1 is the special "end" code; don't give it an entry here
-    dictionary_ind++;
-    dictionary_ind++;
-
-    // TODO verify that the very last byte is clear_code + 1
     while ( input_length )
     {
         code = 0x0;
-        // Always read one more bit than the code length
-        for ( i = 0; i < ( code_length + 1 ); i++ )
+        for (i=0; i<(code_length + 1); i++)
         {
-            // This is different than in the file read example; that
-            // was a call to "next_bit"
             bit = ( *input & mask ) ? 1 : 0;
             mask <<= 1;
 
@@ -161,85 +68,54 @@ void uncompress( int code_length,
         if ( code == clear_code )
         {
             code_length = reset_code_length;
-            dictionary = ( dictionary_entry_t * ) realloc( dictionary,
-                                                          sizeof( dictionary_entry_t ) * ( 1 << ( code_length + 1 ) ) );
+            dictionary = ( dictionary_entry_t * ) realloc( dictionary, sizeof( dictionary_entry_t ) * ( 1 << ( code_length + 1 ) ) );
 
-            for ( dictionary_ind = 0;
-                 dictionary_ind < ( 1 << code_length );
-                 dictionary_ind++ )
+            for ( dictionary_ind = 0; dictionary_ind < ( 1 << code_length ); dictionary_ind++ )
             {
                 dictionary[ dictionary_ind ].byte = dictionary_ind;
-                // XXX this only works because prev is a 32-bit int (> 12 bits)
                 dictionary[ dictionary_ind ].prev = -1;
                 dictionary[ dictionary_ind ].len = 1;
             }
-            dictionary_ind++;
-            dictionary_ind++;
+            dictionary_ind+=2;
             prev = -1;
             continue;
         }
-        else if ( code == stop_code )
-        {
-            /*if ( input_length > 1 )
-            {
-                fprintf( stderr, "Malformed GIF (early stop code)\n" );
-                exit( 0 );
-            }*/
-            break;
-        }
+        else if ( code == stop_code ) break;
 
-        // Update the dictionary with this character plus the _entry_
-        // (character or string) that came before it
-        if ( ( prev > -1 ) && ( code_length < 12 ) )
+        if ( (prev > -1) && (code_length < 12) )
         {
-            if ( code > dictionary_ind )
+            if (code > dictionary_ind)
             {
-                fprintf( stderr, "code = %.02x, but dictionary_ind = %.02x\n",
-                        code, dictionary_ind );
+                fprintf( stderr, "code = %.02x, but dictionary_ind = %.02x\n", code, dictionary_ind );
                 exit( 0 );
             }
 
-            // Special handling for KwKwK
             if ( code == dictionary_ind )
             {
                 int ptr = prev;
-
-                while ( dictionary[ ptr ].prev != -1 )
-                {
-                    ptr = dictionary[ ptr ].prev;
-                }
+                while ( dictionary[ ptr ].prev != -1 ) ptr = dictionary[ ptr ].prev;
                 dictionary[ dictionary_ind ].byte = dictionary[ ptr ].byte;
             }
             else
             {
                 int ptr = code;
-                while ( dictionary[ ptr ].prev != -1 )
-                {
-                    ptr = dictionary[ ptr ].prev;
-                }
+                while ( dictionary[ ptr ].prev != -1 ) ptr = dictionary[ ptr ].prev;
                 dictionary[ dictionary_ind ].byte = dictionary[ ptr ].byte;
             }
 
             dictionary[ dictionary_ind ].prev = prev;
-
             dictionary[ dictionary_ind ].len = dictionary[ prev ].len + 1;
-
             dictionary_ind++;
 
-            // GIF89a mandates that this stops at 12 bits
-            if ( ( dictionary_ind == ( 1 << ( code_length + 1 ) ) ) &&
-                ( code_length < 11 ) )
+            if ( (dictionary_ind == (1 << (code_length + 1))) && (code_length < 11) )
             {
                 code_length++;
-
-                dictionary = ( dictionary_entry_t * ) realloc( dictionary,
-                                                              sizeof( dictionary_entry_t ) * ( 1 << ( code_length + 1 ) ) );
+                dictionary = ( dictionary_entry_t * ) realloc( dictionary, sizeof( dictionary_entry_t ) * ( 1 << ( code_length + 1 ) ) );
             }
         }
 
         prev = code;
 
-        // Now copy the dictionary entry backwards into "out"
         match_len = dictionary[ code ].len;
         while ( code != -1 )
         {
@@ -251,198 +127,107 @@ void uncompress( int code_length,
             }
             code = dictionary[ code ].prev;
         }
-
         out += match_len;
     }
 }
 
-static int read_sub_blocks( unsigned char* buffer, unsigned char **data )
+static int read_sub_blocks( uint8_t* buffer, uint8_t **data )
 {
-    int data_length;
-    int index;
-    unsigned char block_size;
-
-    // Everything following are data sub-blocks, until a 0-sized block is
-    // encountered.
-    data_length = 0;
+    int data_length = 0;
     *data = NULL;
-    index = 0;
+    int index = 0;
 
-    while ( 1 )
-    {
-		READ(&block_size, 1);
-
-        if ( block_size == 0 )  // end of sub-blocks
-        {
-            break;
-        }
+    while (1) {
+        uint8_t block_size = *(buffer++);
+        if ( block_size == 0 ) break;
 
         data_length += block_size;
-        *data = (unsigned char*)realloc( *data, data_length );
-
-        // TODO this could be split across block size boundaries
+        *data = (uint8_t*)realloc( *data, data_length );
 		READ(*data + index, block_size);
-
         index += block_size;
     }
 
     return data_length;
 }
 
-unsigned char* process_image_descriptor( unsigned char* buffer,
-                                    rgb *gct,
-                                    int gct_size,
-                                    int resolution_bits )
+uint8_t* process_image_descriptor( uint8_t* buffer, rgb *gct, int gct_size, int resolution_bits )
 {
     image_descriptor_t image_descriptor;
-    int compressed_data_length;
-    unsigned char *compressed_data = NULL;
-    unsigned char lzw_code_size;
-    int uncompressed_data_length = 0;
-    unsigned char *uncompressed_data = NULL;
-
-    // TODO there could actually be lots of these
 	READ(&image_descriptor, 9);
 
-    // TODO if LCT = true, read the LCT
-
+    uint8_t lzw_code_size;
 	READ(&lzw_code_size, 1);
 
-    compressed_data_length = read_sub_blocks( buffer, &compressed_data );
+    uint8_t *compressed_data = NULL;
+    const int compressed_data_length = read_sub_blocks( buffer, &compressed_data );
 
-//    width = image_descriptor.image_width;
-//    height = image_descriptor.image_height;
-    uncompressed_data_length = image_descriptor.image_width *
-    image_descriptor.image_height;
-    uncompressed_data = (unsigned char*)malloc( uncompressed_data_length );
+    const int uncompressed_data_length = image_descriptor.width * image_descriptor.height;
+    uint8_t *uncompressed_data = (uint8_t*)malloc( uncompressed_data_length );
 
-    uncompress( lzw_code_size, compressed_data, compressed_data_length,
-               uncompressed_data );
+    uncompress( lzw_code_size, compressed_data, compressed_data_length, uncompressed_data );
 
     if ( compressed_data ) free( compressed_data );
 
-    //if ( uncompressed_data )
-    //    free( uncompressed_data );
-
     return uncompressed_data;
 }
 
-/**
- * @param gif_file the file descriptor of a file containing a
- *  GIF-encoded file.  This should point to the first byte in
- *  the file when invoked.
- */
-#define rb (*(buffer++))
+uint32_t* LoadPalette(uint8_t *buffer, uint16_t *size = NULL )
+{
+    buffer += 10;
+    const uint8_t fields = *buffer;
 
-uint32_t* LoadPalette(unsigned char *buffer) {
-	unsigned char header[7];
-	screen_descriptor_t screen_descriptor;
-	//int color_resolution_bits;
+	uint32_t *global_color_table = (uint32_t *)calloc(1, 1024);;
 
-	int global_color_table_size = 0;  // number of entries in global_color_table
-	uint32_t *global_color_table = NULL;
+    if (size) *size = 0;
+	if (fields & 0x80) {
+		int global_color_table_size = 1 << (((fields & 0x07) + 1));
+        if (size) *size = global_color_table_size;
 
-	READ(header, 6);
-	READ(&screen_descriptor, 7);
-
-	//color_resolution_bits = ((screen_descriptor.fields & 0x70) >> 4) + 1;
-	global_color_table = (uint32_t *)calloc(1, 1024);
-
-	if (screen_descriptor.fields & 0x80) {
-		global_color_table_size = 1 << (((screen_descriptor.fields & 0x07) + 1));
-
-		//global_color_table = (rgb *)malloc(3 * global_color_table_size);
-		//READ(global_color_table, 3 * global_color_table_size);
         for (int i=0; i<global_color_table_size;++i) {
-            global_color_table[i] = (buffer[0]<<16) + (buffer[1]<<8) + buffer[2];
             buffer+=3;
+            global_color_table[i] = (buffer[0]<<16) + (buffer[1]<<8) + buffer[2];
         }
 	}
 	return global_color_table;
 }
 
-static unsigned char* process_gif_stream(unsigned char *buffer, unsigned short* w, unsigned short* h)
+static uint8_t* LoadGif(uint8_t *buffer, uint16_t* w, uint16_t* h)
 {
-    unsigned char header[ 7 ];
-    screen_descriptor_t screen_descriptor;
-    int color_resolution_bits;
-
-    int global_color_table_size =0;  // number of entries in global_color_table
+    int global_color_table_size = 0;  // number of entries in global_color_table
     rgb *global_color_table = NULL;
 
-    unsigned char block_type = 0x0;
+    buffer += 6; // Ignore header
+    *w = (uint16_t)*buffer; buffer+=2;
+    *h = (uint16_t)*buffer; buffer+=2;
+    const uint8_t fields = *buffer; buffer+=3;
 
-    // A GIF file starts with a Header (section 17)
-	READ(header, 6);
-    header[ 6 ] = 0x0;
+    const int color_resolution_bits = ( ( fields & 0x70 ) >> 4 ) + 1;
 
-    // XXX there's another format, GIF87a, that you may still find
-    // floating around.
-    /*if ( strcmp( "GIF89a", (char*)header ) )
+    if ( fields & 0x80 )
     {
-        fprintf( stderr,
-                "Invalid GIF file (header is '%s', should be 'GIF89a')\n",
-                header );
-        return NULL;
-    }*/
-
-    // Followed by a logical screen descriptor
-    // Note that this works because GIFs specify little-endian order; on a
-    // big-endian machine, the height & width would need to be reversed.
-
-    // Can't use sizeof here since GCC does byte alignment;
-    // sizeof( screen_descriptor_t ) = 8!
-	READ(&screen_descriptor, 7);
-    *w = screen_descriptor.width;
-    *h = screen_descriptor.height;
-
-    color_resolution_bits = ( ( screen_descriptor.fields & 0x70 ) >> 4 ) + 1;
-
-    if ( screen_descriptor.fields & 0x80 )
-    {
-        //int i;
-        // If bit 7 is set, the next block is a global color table; read it
-        global_color_table_size = 1 << 
-        ( ( ( screen_descriptor.fields & 0x07 ) + 1 ) );
-        
-        global_color_table = ( rgb * ) malloc( 3 * global_color_table_size );
-        
-        // XXX this could conceivably return a short count...
+        global_color_table_size = 1 << ( ( ( fields & 0x07 ) + 1 ) );
+        global_color_table = (rgb*)malloc( 3 * global_color_table_size );
 		READ(global_color_table, 3 * global_color_table_size);
     }
     
-    while ( block_type != TRAILER )
+    uint8_t block_type = 0x0;
+    while ( block_type != 0x3B )
     {
 		READ(&block_type, 1);
         
-        unsigned char size;
+        uint8_t size;
         switch ( block_type )
         {
-            case IMAGE_DESCRIPTOR:
-				return process_image_descriptor(buffer,
-					global_color_table,
-					global_color_table_size,
-					color_resolution_bits);
+            case 0x2C:
+				return process_image_descriptor(buffer, global_color_table, global_color_table_size, color_resolution_bits);
+            case 0x21:
+                buffer++; //size = *(buffer++); buffer += size;
+                do { size = *(buffer++); buffer += size; } while (size != 0);
                 break;
-            case EXTENSION_INTRODUCER:
-                buffer++;
-                size = *(buffer++);
-                buffer += size;
-                do {
-                    size = *(buffer++);
-                    buffer += size;
-                } while (size != 0);
-
-                /*if ( !process_extension( buffer ) )
-                {
-                    return NULL;
-                }*/
-                break;
-            case TRAILER:
+            case 0x3B:
                 break;
             default:
-                fprintf( stderr, "Bailing on unrecognized block type %.02x\n",
-                        block_type );
+                fprintf( stderr, "Bailing on unrecognized block type %.02x\n", block_type );
                 return NULL;
         }
     }
@@ -450,29 +235,6 @@ static unsigned char* process_gif_stream(unsigned char *buffer, unsigned short*
 }
 
 
-unsigned char* LoadGif(unsigned char *buffer, unsigned short* w, unsigned short* h) {
+/*unsigned char* LoadGif(unsigned char *buffer, unsigned short* w, unsigned short* h) {
 	return process_gif_stream(buffer, w, h);
-}
-
-/*int main( int argc, char *argv[] )
-{
-    FILE* gif_file;
-    
-    if ( argc < 2 )
-    {
-        fprintf( stderr, "Usage: %s <path-to-gif-file>\n", argv[ 0 ] );
-        exit( 0 );
-    }
-    
-    gif_file = fopen( argv[ 1 ], "rb" );
-    
-    if ( gif_file == NULL )
-    {
-        fprintf( stderr, "Unable to open file '%s'", argv[ 1 ] );
-        perror( ": " );
-    }
-    
-    process_gif_stream( gif_file );
-    
-    fclose( gif_file );
 }*/
diff --git a/gifenc.h b/gifenc.h
new file mode 100644
index 0000000..40a9ae6
--- /dev/null
+++ b/gifenc.h
@@ -0,0 +1,228 @@
+#include <stdio.h>
+#include <stdlib.h>
+#include <stdint.h>
+#include <string.h>
+
+namespace gif
+{
+    struct gif_t {
+        uint16_t w, h;
+        int depth;
+        int bgindex;
+        FILE *fd;
+        int offset;
+        int nframes;
+        uint8_t *frame, *back;
+        uint32_t partial;
+        uint8_t buffer[0xFF];
+    };
+
+    struct node_t {
+        uint16_t key;
+        node_t *children[];
+    };
+
+    static node_t *new_node(uint16_t key, int degree)
+    {
+        node_t *node = (node_t*)calloc(1, sizeof(*node) + degree * sizeof(node_t *));
+        if (node) node->key = key;
+        return node;
+    }
+
+    static node_t *new_trie(int degree, int *nkeys)
+    {
+        node_t *root = new_node(0, degree);
+        /* Create nodes for single pixels. */
+        for (*nkeys = 0; *nkeys < degree; (*nkeys)++) root->children[*nkeys] = new_node(*nkeys, degree);
+        *nkeys += 2; /* skip clear code and stop code */
+        return root;
+    }
+
+    static void del_trie(node_t *root, int degree)
+    {
+        if (!root) return;
+        for (int i = 0; i < degree; i++) del_trie(root->children[i], degree);
+        free(root);
+    }
+
+    static void put_loop(gif_t *gif, uint16_t loop);
+
+    gif_t *create(const char *fname, uint16_t width, uint16_t height, uint8_t *palette, uint8_t depth, uint8_t bgindex, int loop)
+    {
+        gif_t *gif = (gif_t*)calloc(1, sizeof(*gif) + (bgindex < 0 ? 2 : 1)*width*height);
+        gif->w = width; gif->h = height;
+        gif->bgindex = bgindex;
+        gif->frame = (uint8_t *) &gif[1];
+        gif->back = &gif->frame[width*height];
+        gif->fd = fopen(fname, "wb");
+        if (!gif->fd) { free(gif); return NULL; }
+        fwrite("GIF89a", 6, 1, gif->fd);
+        fwrite(&width, 2, 1, gif->fd);
+        fwrite(&height, 2, 1, gif->fd);
+        gif->depth = depth;
+        fputc((!palette?0x70:0xF0|(depth-1)), gif->fd);
+        fputc(bgindex, gif->fd); fputc(0, gif->fd);
+        if (palette) fwrite(palette, 3 << depth, 1, gif->fd);
+        if (loop >= 0 && loop <= 0xFFFF) put_loop(gif, (uint16_t)loop);
+        return gif;
+    }
+
+    static void put_loop(gif_t *gif, uint16_t loop)
+    {
+        fputc('!', gif->fd); fputc(0xFF, gif->fd); fputc(0x0B, gif->fd);
+        fwrite("NETSCAPE2.0", 11, 1, gif->fd);
+        fputc(0x03, gif->fd); fputc(0x01, gif->fd);
+        fwrite(&loop, 2, 1, gif->fd);
+        fputc(0, gif->fd);
+    }
+
+    /* Add packed key to buffer, updating offset and partial.
+    *   gif->offset holds position to put next *bit*
+    *   gif->partial holds bits to include in next byte */
+    static void put_key(gif_t *gif, uint16_t key, int key_size)
+    {
+        int byte_offset, bit_offset, bits_to_write;
+        byte_offset = gif->offset / 8;
+        bit_offset = gif->offset % 8;
+        gif->partial |= ((uint32_t) key) << bit_offset;
+        bits_to_write = bit_offset + key_size;
+        while (bits_to_write >= 8) {
+            gif->buffer[byte_offset++] = gif->partial & 0xFF;
+            if (byte_offset == 0xFF) {
+                fputc(0xFF, gif->fd);
+                fwrite(gif->buffer, 0xFF, 1, gif->fd);
+                byte_offset = 0;
+            }
+            gif->partial >>= 8;
+            bits_to_write -= 8;
+        }
+        gif->offset = (gif->offset + key_size) % (0xFF * 8);
+    }
+
+    static void end_key(gif_t *gif)
+    {
+        uint8_t byte_offset;
+        byte_offset = gif->offset >> 3;
+        if (gif->offset & 0x07) gif->buffer[byte_offset++] = gif->partial & 0xFF;
+        if (byte_offset)
+        {
+            fputc(byte_offset, gif->fd);
+            fwrite(gif->buffer, byte_offset, 1, gif->fd);
+        }
+        fputc(0, gif->fd);
+        gif->offset = gif->partial = 0;
+    }
+
+    static void put_image(gif_t *gif, uint16_t w, uint16_t h, uint16_t x, uint16_t y)
+    {
+        int nkeys, key_size, i, j;
+        node_t *node, *child, *root;
+        int degree = 1 << gif->depth;
+
+        fputc(',', gif->fd);
+        fwrite(&x, 2, 1, gif->fd);
+        fwrite(&y, 2, 1, gif->fd);
+        fwrite(&w, 2, 1, gif->fd);
+        fwrite(&h, 2, 1, gif->fd);
+        fputc(0, gif->fd); fputc(gif->depth, gif->fd);
+        root = node = new_trie(degree, &nkeys);
+        key_size = gif->depth + 1;
+        put_key(gif, degree, key_size); /* clear code */
+        for (i = y; i < y+h; i++) {
+            for (j = x; j < x+w; j++) {
+                uint8_t pixel = gif->frame[i*gif->w+j] & (degree - 1);
+                child = node->children[pixel];
+                if (child) {
+                    node = child;
+                } else {
+                    put_key(gif, node->key, key_size);
+                    if (nkeys < 0x1000) {
+                        if (nkeys == (1 << key_size))
+                            key_size++;
+                        node->children[pixel] = new_node(nkeys++, degree);
+                    } else {
+                        put_key(gif, degree, key_size); /* clear code */
+                        del_trie(root, degree);
+                        root = node = new_trie(degree, &nkeys);
+                        key_size = gif->depth + 1;
+                    }
+                    node = root->children[pixel];
+                }
+            }
+        }
+        put_key(gif, node->key, key_size);
+        put_key(gif, degree + 1, key_size); /* stop code */
+        end_key(gif);
+        del_trie(root, degree);
+    }
+
+    static int get_bbox(gif_t *gif, uint16_t *w, uint16_t *h, uint16_t *x, uint16_t *y)
+    {
+        int i, j, k;
+        int left, right, top, bottom;
+        uint8_t back;
+        left = gif->w; right = 0;
+        top = gif->h; bottom = 0;
+        k = 0;
+        for (i = 0; i < gif->h; i++) {
+            for (j = 0; j < gif->w; j++, k++) {
+                back = gif->bgindex >= 0 ? gif->bgindex : gif->back[k];
+                if (gif->frame[k] != back) {
+                    if (j < left)   left    = j;
+                    if (j > right)  right   = j;
+                    if (i < top)    top     = i;
+                    if (i > bottom) bottom  = i;
+                }
+            }
+        }
+        if (left != gif->w && top != gif->h) {
+            *x = left; *y = top;
+            *w = right - left + 1;
+            *h = bottom - top + 1;
+            return 1;
+        } else {
+            return 0;
+        }
+    }
+
+    static void add_graphics_control_extension(gif_t *gif, uint16_t d)
+    {
+        uint8_t flags = ((gif->bgindex >= 0 ? 2 : 1) << 2) + 1;
+        fputc('!', gif->fd); fputc(0xF9, gif->fd); fputc(0x04, gif->fd); fputc(flags, gif->fd);
+        fwrite(&d, 2, 1, gif->fd);
+        fputc(gif->bgindex, gif->fd); fputc(0, gif->fd);
+    }
+
+    void addFrame(gif_t *gif, uint16_t delay)
+    {
+        uint16_t w, h, x, y;
+        uint8_t *tmp;
+
+        if (delay || (gif->bgindex >= 0))
+            add_graphics_control_extension(gif, delay);
+        if (gif->nframes == 0) {
+            w = gif->w;
+            h = gif->h;
+            x = y = 0;
+        } else if (!get_bbox(gif, &w, &h, &x, &y)) {
+            /* image's not changed; save one pixel just to add delay */
+            w = h = 1;
+            x = y = 0;
+        }
+        put_image(gif, w, h, x, y);
+        gif->nframes++;
+        if (gif->bgindex < 0) {
+            tmp = gif->back;
+            gif->back = gif->frame;
+            gif->frame = tmp;
+        }
+    }
+
+    void close(gif_t* gif)
+    {
+        fputc(';', gif->fd);
+        fclose(gif->fd);
+        free(gif);
+    }
+    
+}
diff --git a/lua.cpp b/lua.cpp
index 5a4ab74..7b11909 100644
--- a/lua.cpp
+++ b/lua.cpp
@@ -24,13 +24,42 @@ extern "C" {
         return 1;
     }
 
+    static int cpp_surf_save(lua_State *L) {
+        uint8_t surface = luaL_checkinteger(L, 1);
+        const char* str = luaL_checkstring(L, 2);
+
+        int r,g,b;
+        uint8_t pal[256*3];
+        uint8_t *p=pal;
+        if (lua_istable(L, -1)) {
+            const int len = SDL_min(256, lua_rawlen(L, -1));
+            for (int i=1;i<=len;++i) {
+                lua_rawgeti(L, -1, i);
+                lua_getfield(L, -1, "r");
+                *(p++) = luaL_checknumber(L, -1);
+                lua_pop(L, 1);
+                lua_getfield(L, -1, "g");
+                *(p++) = luaL_checknumber(L, -1);
+                lua_pop(L, 1);
+                lua_getfield(L, -1, "b");
+                *(p++) = luaL_checknumber(L, -1);
+                lua_pop(L, 1);
+                lua_pop(L, 1);
+                //pal[i-1] = (r<<16)+(g<<8)+b;
+            }
+        }
+
+        savesurf(surface, str, pal);
+        return 0;
+    }
+
     static int cpp_surf_free(lua_State *L) {
         uint8_t surface = luaL_checkinteger(L, 1);
         freesurf(surface);
         return 0;
     }
 
-    static int cpp_surf_size(lua_State *L) {
+    static int cpp_surf_getSize(lua_State *L) {
         uint8_t surface = luaL_checkinteger(L, 1);
         lua_pushinteger(L, surfw(surface));
         lua_pushinteger(L, surfh(surface));
@@ -43,9 +72,10 @@ extern "C" {
 
     static int cpp_pal_load(lua_State *L) {
         const char* str = luaL_checkstring(L, 1);
-        uint32_t *pal = loadpal(str);
+        uint16_t size;
+        uint32_t *pal = loadpal(str, &size);
         lua_createtable(L, 2, 0);
-        for (int i=0;i<=256;++i) {
+        for (int i=0;i<size;++i) {
             uint32_t color = pal[i];
             lua_createtable(L, 0, 3);
 
@@ -69,8 +99,8 @@ extern "C" {
         if (lua_istable(L, -1)) {
             uint32_t pal[256];
             const int len = SDL_min(256, lua_rawlen(L, -1));
-            for (int i=1;i<=len;++i) {
-                lua_rawgeti(L, -1, i);
+            for (int i=0;i<len;++i) {
+                lua_rawgeti(L, -1, i+1);
                 lua_getfield(L, -1, "r");
                 r = luaL_checknumber(L, -1);
                 lua_pop(L, 1);
@@ -81,31 +111,31 @@ extern "C" {
                 b = luaL_checknumber(L, -1);
                 lua_pop(L, 1);
                 lua_pop(L, 1);
-                pal[i-1] = (r<<16)+(g<<8)+b;
+                pal[i] = (r<<16)+(g<<8)+b;
             }
             setpal(pal);
         }
         return 0;
     }
 
-    static int cpp_pal_color(lua_State *L) {
-        if (lua_gettop(L) >= 1) {
-            uint8_t index = luaL_checkinteger(L, 1);
-            if (lua_gettop(L) > 1) {
-                uint8_t r = luaL_checkinteger(L, 2);
-                uint8_t g = luaL_optinteger(L, 3, 0);
-                uint8_t b = luaL_optinteger(L, 4, 0);
-                uint32_t color = (r<<16) + (g<<8) + b;
-                setcolor(index, color);
-                return 0;
-            } else {
-                uint32_t color = getcolor(index);
-                lua_pushinteger(L, (color>>16)&0xff);
-                lua_pushinteger(L, (color>>8)&0xff);
-                lua_pushinteger(L, color&0xff);
-                return 3;
-            }
-        }
+    static int cpp_pal_getColor(lua_State *L) {
+        if (lua_gettop(L) != 1) return luaL_error(L, "Function 'mini.palette.getColor' Unexpected number of parameters.");
+        uint8_t index = luaL_checkinteger(L, 1);
+        uint32_t color = getcolor(index);
+        lua_pushinteger(L, (color>>16)&0xff);
+        lua_pushinteger(L, (color>>8)&0xff);
+        lua_pushinteger(L, color&0xff);
+        return 3;
+    }
+
+    static int cpp_pal_setColor(lua_State *L) {
+        if (lua_gettop(L) != 4) return luaL_error(L, "Function 'mini.palette.setColor' Unexpected number of parameters.");
+        uint8_t index = luaL_checkinteger(L, 1);
+        uint8_t r = luaL_checkinteger(L, 2);
+        uint8_t g = luaL_optinteger(L, 3, 0);
+        uint8_t b = luaL_optinteger(L, 4, 0);
+        uint32_t color = (r<<16) + (g<<8) + b;
+        setcolor(index, color);
         return 0;
     }
 
@@ -774,17 +804,35 @@ void push_lua_funcs() {
     lua_newtable(L);
     lua_pushcfunction(L,cpp_surf_new);      lua_setfield(L, -2, "new");
     lua_pushcfunction(L,cpp_surf_load);     lua_setfield(L, -2, "load");
+    lua_pushcfunction(L,cpp_surf_save);     lua_setfield(L, -2, "save");
     lua_pushcfunction(L,cpp_surf_free);     lua_setfield(L, -2, "free");
-    lua_pushcfunction(L,cpp_surf_size);     lua_setfield(L, -2, "size");
-    lua_setglobal(L, "surf");
+    lua_pushcfunction(L,cpp_surf_getSize);  lua_setfield(L, -2, "getSize");
+    // setTarget
+    // cls
+    // setPixel
+    // getPixel
+    lua_setglobal(L, "surface");
 
+    map.load
+    map.save
+    map.draw
+    map.getTile
+    map.setTile
+    
     lua_newtable(L);
-    lua_pushcfunction(L,cpp_pal_load);      lua_setfield(L, -2, "load");
-    lua_pushcfunction(L,cpp_pal_set);       lua_setfield(L, -2, "set");
-    lua_pushcfunction(L,cpp_pal_color);     lua_setfield(L, -2, "color");
-    lua_pushcfunction(L,cpp_pal_trans);     lua_setfield(L, -2, "trans");
-    lua_pushcfunction(L,cpp_pal_sub);       lua_setfield(L, -2, "sub");
-    lua_setglobal(L, "pal");
+    lua_pushcfunction(L,cpp_pal_load);          lua_setfield(L, -2, "load");
+    lua_pushcfunction(L,cpp_pal_set);           lua_setfield(L, -2, "set");
+    lua_pushcfunction(L,cpp_pal_getColor);      lua_setfield(L, -2, "getColor");
+    lua_pushcfunction(L,cpp_pal_setColor);      lua_setfield(L, -2, "setColor");
+    lua_pushcfunction(L,cpp_pal_trans);         lua_setfield(L, -2, "trans");
+    lua_pushcfunction(L,cpp_pal_sub);           lua_setfield(L, -2, "sub");
+    lua_setglobal(L, "palette");
+
+    subpalette.resetAll
+    subpalette.set
+    subpalette.reset
+    subpalette.setRange
+    subpalette.resetRange
 
     lua_newtable(L);
     lua_pushcfunction(L,cpp_draw_dest);     lua_setfield(L, -2, "dest");
@@ -999,6 +1047,7 @@ void push_lua_funcs() {
     lua_pushinteger(L, 20);                 lua_setfield(L, -2, "TOUCHPAD");
 
     lua_setglobal(L, "pad");
+
 }
 
 int MiniLoader(lua_State *L) {
diff --git a/mini.cpp b/mini.cpp
index 729e73a..236b553 100644
--- a/mini.cpp
+++ b/mini.cpp
@@ -3,6 +3,7 @@
 #include <string.h>
 #include "lua.h"
 #include "gif.c"
+#include "gifenc.h"
 //#include "SDL2/SDL_mixer.h"
 #include "jail_audio.h"
 
@@ -216,6 +217,14 @@ uint8_t loadsurf(const char* filename) {
     return i;
 }
 
+void savesurf(uint8_t surface, const char* filename, uint8_t *pal)
+{
+    gif::gif_t *file = gif::create(filename, surfaces[surface].w, surfaces[surface].h, pal, (pal?8:0), 0, -1);
+    memcpy(file->frame, surfaces[surface].p, surfaces[surface].w*surfaces[surface].h);
+    gif::addFrame(file, 0);
+    gif::close(file);
+}
+
 void freesurf(uint8_t surface) {
     if (surfaces[surface].p != NULL) free(surfaces[surface].p);
     surfaces[surface].p = NULL;
@@ -493,10 +502,10 @@ void bcolor(uint8_t color) {
     ds::back_color=color;
 }
 
-uint32_t *loadpal(const char* filename) {
+uint32_t *loadpal(const char* filename, uint16_t *palsize) {
     int size;
     uint8_t *buffer = (uint8_t*)file_getfilebuffer(filename, size);
-    uint32_t *pal = LoadPalette(buffer);
+    uint32_t *pal = LoadPalette(buffer, palsize);
     free(buffer);
     return pal;
     /*
diff --git a/mini.h b/mini.h
index d948802..985e796 100644
--- a/mini.h
+++ b/mini.h
@@ -119,6 +119,7 @@ int scrh();
 
 uint8_t newsurf(int w, int h);
 uint8_t loadsurf(const char* filename);
+void savesurf(uint8_t surface, const char* filename, uint8_t *pal);
 void freesurf(uint8_t surface);
 int surfw(uint8_t surface);
 int surfh(uint8_t surface);
@@ -131,7 +132,7 @@ void cls(uint8_t color=0);
 void color(uint8_t color=6);
 void bcolor(uint8_t color=0);
 
-uint32_t *loadpal(const char* filename);
+uint32_t *loadpal(const char* filename, uint16_t *palsize=NULL);
 void setpal(uint32_t *pal);
 void setcolor(uint8_t index, uint32_t color);
 uint32_t getcolor(uint8_t index);