7 files changed, 117 insertions, 53 deletions
diff --git a/src/luma/arghandl.cc b/src/luma/arghandl.cc
deleted file mode 100644
index f9f08c0..0000000
--- a/src/luma/arghandl.cc
+++ /dev/null
@@ -1,12 +0,0 @@
-# include <cstdio>
-# include <luma.hh>
-# include <string>
-using namespace std::literals::string_literals;
-void luma::arghandl(int const argc,char const * * argv) {
-	for(int pos = 0x1;pos < argc; ++pos) {
-		std::string arg = argv[pos];
-		if(pos == 0x1) {
-			this->lumafile = arg;
-		}
-	}
-}
diff --git a/src/luma/luma.cc b/src/luma/luma.cc
deleted file mode 100644
index 5b72788..0000000
--- a/src/luma/luma.cc
+++ /dev/null
@@ -1,37 +0,0 @@
-# include <cstdio>
-# include <cstdlib>
-# include <luma.hh>
-# include <fcntl.h>
-# include <sys/stat.h>
-# include <unistd.h>
-[[noreturn]] luma::luma(int const argc,char const * * argv) {
-	this->arghandl(argc,argv);
-	std::printf("Will open \"%s\".\u000A",this->lumafile.c_str());
-	int file      = ::open(this->lumafile.c_str(),O_RDONLY);
-	char tok      = '\0';
-	char word[0x1000];
-	int wordpos = 0x0;
-	struct ::stat fst;
-	::fstat(file,&fst);
-	printf("Size of file is %ld characters.\u000A",fst.st_size);
-	::sleep(0x1);
-	for(int filepos = 0x0;filepos < fst.st_size;++filepos) {
-		if(::read(file,&tok,0x1) < 0x0) {
-			std::printf("Error reading file.\u000A");
-			::_exit(EXIT_FAILURE);
-		}
-		if(tok == '\u000A' || tok == '\u0020') {
-			if(wordpos == 0x0) continue;
-			word[wordpos] = '\0';
-			std::printf("The word is \"%s\".\u000A",word);
-			word[0x0] = '\0';
-			wordpos   = 0x0;
-		}
-		else {
-			word[wordpos] = tok;
-			std::printf("Got character '%c' and set word[%d] to it.\u000A",tok,wordpos);
-			++wordpos;
-		}
-	}
-	::_exit(EXIT_SUCCESS);
-}
diff --git a/src/luma/print.c b/src/luma/print.c
new file mode 100644
index 0000000..df53d8a
--- /dev/null
+++ b/src/luma/print.c
@@ -0,0 +1,16 @@
+# include <luma/utf8enc.h>
+# include <stdio.h>
+# include <string.h>
+void luma_print([[maybe_unused]] char * str,...) {
+	/*for(size_t n = 0x0;;++n) {
+		if(str[n] == 0x0) {
+			fwrite(&(char){0xA},0x1,0x1,stdout);
+			break;
+		}
+		if(!strcmp(&str[n],"\uFFFD")) {
+			fwrite(&(char){0x20},0x1,0x1,stdout);
+			continue;
+		}
+		fwrite(&str[n],0x1,0x1,stdout);
+	}*/
+}
diff --git a/src/luma/utf8dec.c b/src/luma/utf8dec.c
new file mode 100644
index 0000000..35ef07e
--- /dev/null
+++ b/src/luma/utf8dec.c
@@ -0,0 +1,8 @@
+# include <luma/utf8dec.h>
+# include <stdint.h>
+# include <stdlib.h>
+uint32_t * luma_utf8dec([[maybe_unused]] char const * str) {
+	uint32_t * utf = malloc(0x4);
+	utf[0x0] = (uint32_t){0x0};
+	return utf;
+}
diff --git a/src/luma/utf8enc.c b/src/luma/utf8enc.c
new file mode 100644
index 0000000..cd1edff
--- /dev/null
+++ b/src/luma/utf8enc.c
@@ -0,0 +1,61 @@
+# include <luma/utf8enc.h>
+# include <stdint.h>
+# include <stdio.h>
+# include <stdlib.h>
+uint8_t const * luma_utf8enc(uint32_t * codeps) {
+	size_t sz    = (size_t){0x0}; // Size of input array (bytes).
+	size_t outsz = (size_t){0x0}; // Size of output array /bytes).
+	for(size_t n = (size_t){0x0};;n += (size_t){0x1}) { // First pass: get size of input array, and determine size of output array.
+		uint32_t codep = codeps[n]; // Current Unicode codepoint.
+		if(codep == (uint32_t){0x0}) { // U+0000 is Null.
+			sz = n;
+			break;
+		}
+		if(codep > 0x10FFFF) { // Codepoint out of range.
+			return NULL;
+		}
+		if(codep > 0xFFFF) { // 4 bytes.
+			outsz += (size_t){0x2};
+			continue;
+		}
+		if(codep > 0x7FF) { // 3 bytes.
+			outsz += (size_t){0x3};
+			continue;
+		}
+		if(codep > 0x7F) { // 2 bytes.
+			outsz += (size_t){0x2};
+			continue;
+		}
+		// 1 byte.
+		outsz += (size_t){0x1};
+	}
+	outsz += (size_t){0x1}; // Add space for null-terminator.
+	printf("There are %zu element(s).\n",sz);
+	printf("The output will have %zu element(s).\n",outsz);
+	uint8_t * outstr              = malloc(outsz); // Allocate space for output array.
+	outstr[outsz - (size_t){0x1}] = (uint8_t){0x0}; // Create null-terminator on output array.
+	size_t outn = (size_t){0x0}; // Keep track of position in output array.
+	for(size_t n = (size_t){0x0};n < sz;n += (size_t){0x1}) {
+		uint32_t codep = codeps[n]; // Current Unicode codepoint.
+		if(codep > 0xFFFF) {
+			outstr[outn] = (uint8_t){0x3F};
+			outn += (size_t){0x1};
+			continue;
+		}
+		if(codep > 0x7FF) {
+			outstr[outn] = (uint8_t){0x3F};
+			outn += (size_t){0x1};
+			continue;
+		}
+		if(codep > 0x7F) {
+			outstr[outn] = (uint8_t){0xC0 + (codep >> 0x6)};
+			outn += (size_t){0x1};
+			outstr[outn] = (uint8_t){0x80 + ((uint8_t){codep << 0x2} >> 0x2)};
+			outn += (size_t){0x1};
+			continue;
+		}
+		outstr[outn] = codep;
+		outn += (size_t){0x1};
+	}
+	return (uint8_t const *){outstr};
+}
diff --git a/src/main.c b/src/main.c
new file mode 100644
index 0000000..8f55eec
--- /dev/null
+++ b/src/main.c
@@ -0,0 +1,32 @@
+# include <locale.h>
+# include <luma/arch.h>
+# include <luma/print.h>
+# include <luma/utf8enc.h>
+# include <stdint.h>
+# include <stdio.h>
+# include <stdlib.h>
+int main(void) {
+	setlocale(LC_ALL,"en_GB.UTF-8");
+	enum luma_arch code[] = {
+		luma_arch_lab,
+		luma_arch_hello,
+	};
+	for(size_t i = (size_t){0x0};i < sizeof code / sizeof code[0x0];++i) {
+		printf("Got code %d.\n",code[i]);
+	}
+	uint8_t const * msg = luma_utf8enc((uint32_t[]){0x00A2,0x2C,0x20Ac,0x2C,0x218A,0x2C,0x1F44B,0x0});
+	printf("%u\n",msg[0x0]);
+	printf("%u\n",msg[0x1]);
+	printf("%s\n",msg);
+	//uint32_t * utf = luma_utf8dec(msg);
+	free((void *)msg);
+	/*for(size_t n = (size_t){0x0};;n += (size_t){0x1}) {
+		if(utf[n] == (uint32_t){0x0}) {
+			break;
+		}
+		printf("%d\n",utf[n]);
+	}
+	free((void *)utf);*/
+	//luma_print("Hello world. �👋");
+	exit(EXIT_SUCCESS);
+}
diff --git a/src/main.cc b/src/main.cc
deleted file mode 100644
index 5de90c4..0000000
--- a/src/main.cc
+++ /dev/null
@@ -1,4 +0,0 @@
-# include <luma.hh>
-int main(int const argc,char const * * argv) {
-	::luma luma(argc,argv);
-}