baraba/bux-lang
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Phase 7.10: multi-file project build, module braces, else-if fix, UTF-8 lexer

Browse Source 
Runtime (stdlib/runtime.c):
- bux_list_dir() — directory listing with extension filter
- bux_run_cc() — invoke C compiler on generated C code
- bux_mkdir_if_needed() — create build directory
- bux_dir_exists() — check if directory exists
- Removed duplicate bux_path_join

Parser (src_bux/parser.bux):
- Module parsing with braces: module X { ... }
- else-if parsing: wraps inner if in synthetic Block
- Infinite-loop safeguards in module body and struct body parsing

Lexer (src_bux/lexer.bux):
- lexPeek: mask with & 255 to handle UTF-8 signed char bytes

CLI (src_bux/cli.bux):
- Cli_BuildProject(): multi-file build from src/ directory
  - Lists .bux files, parses each, merges declarations inline
  - Runs sema + HIR + C codegen on merged module
  - Invokes C compiler to produce binary
- New extern declarations: bux_file_exists, bux_dir_exists, bux_path_join,
  bux_mkdir_if_needed, bux_run_cc, bux_list_dir
- FileExists(), DirExists() wrapper functions
- "project" command dispatch

Known issue: else-if chain not fully emitted by buxc2 C backend
(lexSkipWhitespace missing comment handling in generated C).

All 18 examples pass, all unit tests pass.
This commit is contained in:
dimgigov
2026-05-31 18:14:35 +03:00
parent 35b856bab4
commit 74117595d2
7 changed files with 627 additions and 8 deletions
Download Patch File Download Diff File Expand all files Collapse all files
_selfhost/src/cli.bux
+185
View File
@@ -6,6 +6,12 @@ extern func PrintLine(s: String);
extern func Print(s: String);
extern func bux_read_file(path: String) -> String;
extern func bux_write_file(path: String, content: String) -> bool;
extern func bux_file_exists(path: String) -> int;
extern func bux_dir_exists(path: String) -> int;
extern func bux_path_join(a: String, b: String) -> String;
extern func bux_mkdir_if_needed(path: String) -> int;
extern func bux_run_cc(c_file: String, out_bin: String, runtime_c: String, io_c: String, math_lib: String) -> int;
extern func bux_list_dir(dir: String, ext: String, out_count: *int) -> *String;
func ReadFile(path: String) -> String {
return bux_read_file(path);
@@ -15,6 +21,14 @@ func WriteFile(path: String, content: String) -> bool {
return bux_write_file(path, content);
}
func FileExists(path: String) -> bool {
return bux_file_exists(path) != 0;
}
func DirExists(path: String) -> bool {
return bux_dir_exists(path) != 0;
}
// Import the compiler pipeline
// In self-hosting mode, these are compiled together from src_bux/
@@ -127,6 +141,171 @@ func Cli_Check(srcPath: String) -> int {
return 0;
}
// ---------------------------------------------------------------------------
// Project build — compile all .bux files in src/ directory
// ---------------------------------------------------------------------------
func Cli_CompileSource(source: String, sourceName: String) -> *HirModule {
// Phase 1: Lex
let lex: *Lexer = Lexer_Tokenize(source);
if Lexer_DiagCount(lex) > 0 {
Print("Lex errors in ");
PrintLine(sourceName);
return null as *HirModule;
}
// Phase 2: Parse
let mod: *Module = Parser_Parse(lex.tokens, lex.tokenCount);
if mod == null as *Module {
Print("Parse failed for ");
PrintLine(sourceName);
return null as *HirModule;
}
// Phase 3: Semantic analysis
let sema: *Sema = Sema_Analyze(mod);
if Sema_HasError(sema) {
Print("Sema errors in ");
PrintLine(sourceName);
return null as *HirModule;
}
// Phase 4: HIR lowering
let hirMod: *HirModule = HirLower_LowerModule(mod, sema);
return hirMod;
}
func Cli_BuildProject(projectDir: String) -> int {
let srcDir: String = bux_path_join(projectDir, "src");
if !DirExists(srcDir) {
Print("Error: no src/ directory in ");
PrintLine(projectDir);
return 1;
}
Print("Scanning ");
PrintLine(srcDir);
// List all .bux files in src/
var fileCount: int = 0;
let files: *String = bux_list_dir(srcDir, ".bux", &fileCount);
if fileCount == 0 {
PrintLine("Error: no .bux files found in src/");
return 1;
}
Print("Found ");
PrintInt(fileCount);
PrintLine(" source files");
// Create merged module
let merged: *Module = bux_alloc(sizeof(Module)) as *Module;
merged.name = "main";
merged.path = "";
merged.itemCount = 0;
merged.firstItem = null as *Decl;
// Parse each file and merge declarations into merged module
var i: int = 0;
while i < fileCount {
let path: String = files[i];
Print(" Parsing ");
PrintLine(path);
let source: String = ReadFile(path);
if String_Eq(source, "") {
Print(" Error: cannot read ");
PrintLine(path);
return 1;
}
let lex: *Lexer = Lexer_Tokenize(source);
if Lexer_DiagCount(lex) > 0 {
Print(" Lex errors in ");
PrintLine(path);
return 1;
}
let mod: *Module = Parser_Parse(lex.tokens, lex.tokenCount);
if mod == null as *Module {
Print(" Parse failed for ");
PrintLine(path);
return 1;
}
// Merge declarations from this module into merged
var decl: *Decl = mod.firstItem;
while decl != null as *Decl {
let next: *Decl = decl.childDecl2;
decl.childDecl2 = null as *Decl;
// Flatten module declarations
if decl.kind == dkModule {
var inner: *Decl = decl.childDecl1;
while inner != null as *Decl {
let innerNext: *Decl = inner.childDecl2;
inner.childDecl2 = merged.firstItem;
merged.firstItem = inner;
merged.itemCount = merged.itemCount + 1;
inner = innerNext;
}
} else {
decl.childDecl2 = merged.firstItem;
merged.firstItem = decl;
merged.itemCount = merged.itemCount + 1;
}
decl = next;
}
i = i + 1;
}
Print("Merged ");
PrintInt(merged.itemCount);
PrintLine(" declarations");
// Semantic analysis
PrintLine("Running sema...");
let sema: *Sema = Sema_Analyze(merged);
if Sema_HasError(sema) {
PrintLine("Sema errors found");
return 1;
}
// HIR lowering
PrintLine("Lowering to HIR...");
let hirMod: *HirModule = HirLower_LowerModule(merged, sema);
// C code generation
PrintLine("Generating C code...");
let cCode: String = CBackend_Generate(hirMod);
// Create build directory
let buildDir: String = bux_path_join(projectDir, "build");
discard bux_mkdir_if_needed(buildDir);
// Write C output
let cFile: String = bux_path_join(buildDir, "main.c");
if !WriteFile(cFile, cCode) {
PrintLine("Error: cannot write main.c");
return 1;
}
Print("C code written to ");
PrintLine(cFile);
// Find stdlib files
let runtimeC: String = bux_path_join(bux_path_join(projectDir, ".."), "stdlib/runtime.c");
let ioC: String = bux_path_join(bux_path_join(projectDir, ".."), "stdlib/io.c");
// Invoke C compiler
PrintLine("Invoking C compiler...");
let outBin: String = bux_path_join(buildDir, "buxc3");
let rc: int = bux_run_cc(cFile, outBin, runtimeC, ioC, "-lm");
if rc != 0 {
PrintLine("Error: C compilation failed");
return 1;
}
Print("Build successful: ");
PrintLine(outBin);
return 0;
}
// ---------------------------------------------------------------------------
// Main entry — dispatch based on args
// ---------------------------------------------------------------------------
@@ -169,6 +348,12 @@ func Cli_Run(args: *String, argCount: int) -> int {
return Cli_Build(src, out);
}
if String_Eq(cmd, "project") {
let dir: String = ".";
if argCount >= 3 { dir = args[2]; }
return Cli_BuildProject(dir);
}
Print("Unknown command: ");
PrintLine(cmd);
return 1;
_selfhost/src/lexer.bux
+1 -1
View File
@@ -101,7 +101,7 @@ func lexIsAtEnd(lex: *Lexer) -> bool {
func lexPeek(lex: *Lexer, ahead: int) -> uint32 {
let i: int = lex.pos + ahead;
if i < lex.sourceLen {
return lex.source[i] as uint32;
return (lex.source[i] as uint32) & 255;
}
return 0;
}
_selfhost/src/parser.bux
+88 -3
View File
@@ -447,8 +447,15 @@ func parserParseStmt(p: *Parser) -> *Stmt {
var elseBlock: *Block = null as *Block;
if parserMatch(p, tkElse) {
if parserCheck(p, tkIf) {
// else if → recursively parse an if stmt inside else
elseBlock = parserParseBlock(p);
// else if → parse the if statement, wrap in a synthetic block
let innerIf: *Stmt = parserParseStmt(p);
elseBlock = bux_alloc(sizeof(Block)) as *Block;
elseBlock.line = innerIf.line;
elseBlock.column = innerIf.column;
elseBlock.stmtCount = 1;
elseBlock.firstStmt = innerIf;
elseBlock.lastStmt = innerIf;
innerIf.nextStmt = null as *Stmt;
} else {
elseBlock = parserParseBlock(p);
}
@@ -739,10 +746,17 @@ func parserParseStructDecl(p: *Parser, isPublic: bool) -> *Decl {
while !parserCheck(p, tkRBrace) && parserPeek(p, 0) != tkEndOfFile {
if d.fieldCount >= 8 { break; }
if parserCheck(p, tkNewLine) { discard parserAdvance(p); continue; }
if parserCheck(p, tkSemicolon) { discard parserAdvance(p); continue; }
let beforePos: int = p.pos;
let fName: LexToken = parserExpect(p, tkIdent, "expected field name");
discard parserExpect(p, tkColon, "expected ':' in struct field");
let fType: *TypeExpr = parserParseType(p);
parserMatch(p, tkSemicolon);
// Infinite-loop safeguard
if p.pos == beforePos {
discard parserAdvance(p);
continue;
}
if d.fieldCount == 0 {
d.field0.name = fName.text;
@@ -939,10 +953,46 @@ func parserParseDecl(p: *Parser) -> *Decl {
if kind == tkModule {
discard parserAdvance(p);
let name: LexToken = parserExpect(p, tkIdent, "expected module name");
parserMatch(p, tkSemicolon);
// Parse optional path segments (::Name)
while parserCheck(p, tkColonColon) {
discard parserAdvance(p);
discard parserExpect(p, tkIdent, "expected module path segment");
}
let d: *Decl = bux_alloc(sizeof(Decl)) as *Decl;
d.kind = dkModule;
d.strValue = name.text;
// Parse module body with braces
if parserCheck(p, tkLBrace) {
discard parserAdvance(p); // consume {
var items: *Decl = null as *Decl;
var lastItem: *Decl = null as *Decl;
while !parserCheck(p, tkRBrace) && parserPeek(p, 0) != tkEndOfFile {
if parserCheck(p, tkNewLine) || parserCheck(p, tkSemicolon) {
discard parserAdvance(p);
continue;
}
let beforePos: int = p.pos;
let item: *Decl = parserParseDecl(p);
if item != null as *Decl {
item.childDecl2 = null as *Decl;
if items == null as *Decl {
items = item;
lastItem = item;
} else {
lastItem.childDecl2 = item;
lastItem = item;
}
}
// Infinite-loop safeguard: if no progress, skip token
if p.pos == beforePos {
discard parserAdvance(p);
}
}
parserMatch(p, tkRBrace); // consume }
d.childDecl1 = items; // first item of module
} else {
parserMatch(p, tkSemicolon);
}
return d;
}
@@ -1013,4 +1063,39 @@ func Parser_Free(p: *Parser) {
bux_free(p.diags as *void);
bux_free(p as *void);
}
// ---------------------------------------------------------------------------
// Module merging — merge multiple parsed modules into one
// ---------------------------------------------------------------------------
// Add a single declaration to the merged module (push front to linked list)
func parserMergeAddDecl(mod: *Module, decl: *Decl) {
if decl == null as *Decl { return; }
// Flatten module declarations: extract inner items
if decl.kind == dkModule {
var inner: *Decl = decl.childDecl1;
while inner != null as *Decl {
let next: *Decl = inner.childDecl2;
inner.childDecl2 = mod.firstItem;
mod.firstItem = inner;
mod.itemCount = mod.itemCount + 1;
inner = next;
}
} else {
decl.childDecl2 = mod.firstItem;
mod.firstItem = decl;
mod.itemCount = mod.itemCount + 1;
}
}
func Parser_MergeModules(modules: *void, count: int) -> *Module {
// This function is kept for future use but not called by Cli_BuildProject
// which does inline merging for simplicity
let merged: *Module = bux_alloc(sizeof(Module)) as *Module;
merged.name = "main";
merged.path = "";
merged.itemCount = 0;
merged.firstItem = null as *Decl;
return merged;
}
}