Initial basic functionality commit

What this commit adds:
* Ability to apply a theme, given a folder
* Extract all .zip files in /Themes/ into a relevant folder
* Get all theme metadata from info.smdh
* Proper unicode handling

What's still needed for initial release:
* Theme shuffling
* GUI

source/unicode.c

@@ -0,0 +1,130 @@
+#include <3ds.h>
+#include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
+
+#include "unicode.h"
+
+// Abandon all hope, ye who enter here
+
+// Convert a UTF16 encoded array into a UTF8 encoded string, stopping at a null terminator.
+// This properly null terminates, however you have to be sure your strings are 0d beforehand
+void wtoa(char *out, u16 *in)
+{
+    u8 lastByte = 1; // Initialize to 1 because
+    while(*in != 0 || lastByte != 0)
+    {
+        lastByte = *in;
+        if (*in == 0) continue;
+        *out = *in;
+        out++;
+        in++;
+    }
+    *out = 0;
+}
+
+// Convert standard ascii to a UTF16 encoded array.
+// Return the number of characters. Should be equal to strlen(in)
+ssize_t atow(u16 *out, char *in)
+{
+    u8 i = 0;
+    for (i = 0; i < strlen(in); i++) 
+    {
+        out[i] = in[i];
+    }
+    return i;
+}
+
+
+// Remove a file extension on a UTF16 encoded array. 
+// Returns the number of characters in the array. Should be equal to len - 4 in most cases.
+// Warning: This assumes everything after the first '.' is an extension and trashes it.
+ssize_t trim_extension(u16 *out, u16 *in)
+{
+    ssize_t i = 0;
+    while (true)
+    {
+        if (in[i] == '.' || in[i] == 0)
+        {
+            out[i] = '\0';
+            return i;
+        }
+        out[i] = in[i];
+        i++;
+    }
+    return i;
+}
+
+// Removes all Unicode characters (and extended ASCII) and replaces them with nothing
+// Returns a boolean indicated whether or not there's any change in the string
+bool strip_unicode(u16 *out, u16 *in, ssize_t len)
+{
+    bool changed = false;
+    int n = 0;
+    for (int i = 0; i < len; i++)
+    {
+        if (in[i] == 0)
+        {
+            out[i] = 0;
+            return changed;
+        } else if (in[i] > 127) // This isn't Chinese checkers - this isn't even regular checkers!
+        {
+            changed = true;
+        } else {
+            out[n] = in[i];
+            n++;
+        }
+    }
+    return changed;
+}
+
+
+// Copy one u16* into another u16*, without regard for the length.
+// source *must* be properly null terminated for this to work!
+u16 *strucpy(u16 *destination, const u16 *source)
+{
+    u16 pos = 0;
+    while (true)
+    {
+        if (source[pos] == 0)
+        {
+            break;  
+        }
+        destination[pos] = source[pos];
+        pos++;
+    }
+
+    return destination;
+}
+
+// Get the length of a UTF16 string
+// As always, the string must be properly null terminated
+ssize_t strulen(const u16 *str)
+{
+    ssize_t len = 0;
+    while (true) if (str[++len] == 0) return len; 
+}
+
+// Concatenate one u16* onto another.
+// Source and destination both must be properly null terminated to start with or this will *not* work!
+u16 *strucat(u16 *destination, const u16 *source)
+{
+    ssize_t dest_len = strulen(destination);
+
+    ssize_t source_len = strulen(source);
+
+    memcpy(&destination[dest_len], source, source_len * sizeof(u16));
+    return destination;
+}
+
+// Prints a u16* for debug purposes
+// Must be properly null-terminated
+void printu(u16 *input)
+{
+    ssize_t in_len = strulen(input);
+    ssize_t buf_len = in_len + 1; // Plus 1 for proper null termination
+    wchar_t *buf = calloc(buf_len, sizeof(wchar_t));
+    for (int i = 0; i < buf_len; i++) buf[i] = input[i];
+    printf("%ls\n", buf);
+    free(buf);
+}