From 4cfb4f74a64ab59023f55675c389d758db8d81c9 Mon Sep 17 00:00:00 2001
From: IonutParau <ionut.alex.parau@gmail.com>
Date: Thu, 29 May 2025 12:12:32 +0200
Subject: [PATCH] start of unicode lib

---
 build.zig        |  1 +
 src/emulator.c   |  6 +++++
 src/neonucleus.h | 14 ++++++++++++
 src/unicode.c    | 57 ++++++++++++++++++++++++++++++++++++++++++++++++
 4 files changed, 78 insertions(+)
 create mode 100644 src/unicode.c

diff --git a/build.zig b/build.zig
index cd8a907..a08ee32 100644
--- a/build.zig
+++ b/build.zig
@@ -12,6 +12,7 @@ fn addEngineSources(c: *std.Build.Step.Compile) void {
             "src/component.c",
             "src/computer.c",
             "src/universe.c",
+            "src/unicode.c",
             // components
             "src/components/eeprom.c",
             "src/components/filesystem.c",
diff --git a/src/emulator.c b/src/emulator.c
index 440f44a..795addf 100644
--- a/src/emulator.c
+++ b/src/emulator.c
@@ -191,6 +191,12 @@ bool ne_fs_isDirectory(nn_component *component, ne_fs *fs, const char *path) {
     return DirectoryExists(p);
 }
 
+bool ne_fs_makeDirectory(nn_component *component, ne_fs *fs, const char *path) {
+    const char *p = ne_fs_diskPath(component, path);
+
+    return MakeDirectory(p) == 0;
+}
+
 int main() {
     printf("Setting up universe\n");
     nn_universe *universe = nn_newUniverse();
diff --git a/src/neonucleus.h b/src/neonucleus.h
index 51a1a6d..6a9327a 100644
--- a/src/neonucleus.h
+++ b/src/neonucleus.h
@@ -153,6 +153,20 @@ bool nn_removeRef(nn_refc *refc, size_t count);
 /* Returns true if the object should be freed */
 bool nn_decRef(nn_refc *refc);
 
+bool nn_unicode_validate(const char *s);
+// returned string must be nn_free()'d
+char *nn_unicode_char(int *codepoints, size_t codepointCount);
+// returned array must be nn_free()'d
+int *nn_unicode_codepoints(const char *s);
+size_t nn_unicode_len(const char *s);
+int nn_unicode_codepointAt(const char *s, size_t byteOffset);
+size_t nn_unicode_codepointSize(int codepoint);
+const char *nn_unicode_codepointToChar(int codepoint, size_t *len);
+size_t nn_unicode_charWidth(int codepoint);
+size_t nn_unicode_wlen(const char *s);
+void nn_unicode_upper(char *s);
+void nn_unicode_lower(char *s);
+
 double nn_realTime();
 double nn_realTimeClock(void *_);
 /* Will busy-loop until the time passes. This is meant for computed latencies in components. */
diff --git a/src/unicode.c b/src/unicode.c
new file mode 100644
index 0000000..a4cf285
--- /dev/null
+++ b/src/unicode.c
@@ -0,0 +1,57 @@
+#include "neonucleus.h"
+#include <string.h>
+
+bool nn_unicode_validate(const char *s) {
+    // TODO: validate UTF-8-ness
+    return true;
+}
+
+// A general unicode library, which assumes unicode encoding.
+// It is used to power the Lua architecture's Unicode API re-implementation.
+// It can also just be used to deal with unicode.
+
+char *nn_unicode_char(int *codepoints, size_t codepointCount) {
+    size_t len = 0;
+    for(size_t i = 0; i < codepointCount; i++) {
+        int codepoint = codepoints[i];
+        len += nn_unicode_codepointSize(codepoint);
+    }
+
+    char *buf = nn_malloc(len+1);
+    if(buf == NULL) return buf;
+    buf[len] = '\0';
+
+    size_t j = 0;
+    for(size_t i = 0; i < codepointCount; i++) {
+        int codepoint = codepoints[i];
+        size_t codepointLen = 0;
+        const char *c = nn_unicode_codepointToChar(codepoint, &codepointLen);
+        memcpy(buf + j, c, codepointLen);
+        j += codepointLen;
+    }
+
+    return buf;
+}
+
+int *nn_unicode_codepoints(const char *s);
+
+size_t nn_unicode_len(const char *s) {
+    size_t count = 0;
+    while (*s) {
+        count += (*s++ & 0xC0) != 0x80;
+    }
+    return count;
+}
+
+int nn_unicode_codepointAt(const char *s, size_t byteOffset);
+
+size_t nn_unicode_codepointSize(int codepoint);
+
+const char *nn_unicode_codepointToChar(int codepoint, size_t *len);
+
+size_t nn_unicode_charWidth(int codepoint);
+
+size_t nn_unicode_wlen(const char *s);
+
+void nn_unicode_upper(char *s);
+void nn_unicode_lower(char *s);