gh-18: Add lib\diff.asmln

Author: python-processing-unit

.gitignore

@@ -1,3 +1,5 @@
 # Ignore Python bytecode cache directories
 __pycache__/
 ext/__pycache__/
+# Ignore MyPy cache directory
+.mypy_cache/

docs/SPECIFICATION.html

@@ -775,9 +775,9 @@
 
 ### 12.13 File operations:
 
-- `READFILE(STR: path, STR: coding=\"UTF-8\"):STR` — Reads the file at `path` and returns its contents as a `STR`. Supported `coding` values (case-insensitive) are `UTF-8`, `UTF-8 BOM`, `UTF-16 LE`, `UTF-16 BE`, `ANSI` (Windows-1252 on Windows, Latin-1 elsewhere), `binary` (alias `bin`), and `hexadecimal` (alias `hex`). Text codings decode with replacement on invalid bytes; `UTF-8` tolerates and strips a BOM if present. `binary` returns an 8-bit-per-byte bitstring; `hexadecimal` returns a lowercase hexadecimal string. If the file cannot be opened or read, the interpreter raises a runtime error (rewrite: `READFILE`).
+- `READFILE(STR: path, STR: coding = "UTF-8"):STR` — Reads the file at `path` and returns its contents as a `STR`. Supported `coding` values (case-insensitive) are `UTF-8`, `UTF-8 BOM`, `UTF-16 LE`, `UTF-16 BE`, `ANSI` (Windows-1252 on Windows, Latin-1 elsewhere), `binary` (alias `bin`), and `hexadecimal` (alias `hex`). Text codings decode with replacement on invalid bytes; `UTF-8` tolerates and strips a BOM if present. `binary` returns an 8-bit-per-byte bitstring; `hexadecimal` returns a lowercase hexadecimal string. If the file cannot be opened or read, the interpreter raises a runtime error (rewrite: `READFILE`).
 
-- `WRITEFILE(STR: blob, STR: path, STR: coding=\"UTF-8\"):INT` — Writes `blob` to `path` using the same `coding` options as `READFILE`. Text codings write in the specified encoding (`UTF-8 BOM` emits a BOM; `ANSI` maps to Windows-1252 on Windows, Latin-1 elsewhere). `binary` expects a bitstring of 0/1 characters whose length is a multiple of 8; `hexadecimal` expects valid hexadecimal. On I/O failure the call returns `INT` 0; invalid coding or malformed binary/hex data raises a runtime error (rewrite: `WRITEFILE`).
+- `WRITEFILE(STR: blob, STR: path, STR: coding = "UTF-8"):INT` — Writes `blob` to `path` using the same `coding` options as `READFILE`. Text codings write in the specified encoding (`UTF-8 BOM` emits a BOM; `ANSI` maps to Windows-1252 on Windows, Latin-1 elsewhere). `binary` expects a bitstring of 0/1 characters whose length is a multiple of 8; `hexadecimal` expects valid hexadecimal. On I/O failure the call returns `INT` 0; invalid coding or malformed binary/hex data raises a runtime error (rewrite: `WRITEFILE`).
 
 - `EXISTFILE(STR: path):INT` — Returns `INT` 1 when a filesystem object exists at `path`, otherwise returns `INT` 0. The argument must be a `STR`.
 

lib/diff.asmln

@@ -0,0 +1,72 @@
+# diff utilities for ASM-Lang
+#
+# API:
+# UNIFIED(modified: STR, source: STR):STR
+# CONTEXT(modified: STR, source: STR):STR
+# SIDE_BY_SIDE(modified: STR, source: STR):STR
+
+FUNC _append_line(STR: acc, STR: line):STR{
+    IF( EQ(acc, "") ){ RETURN(line) } ELSE { RETURN( JOIN(acc, "\n", line) ) }
+}
+
+FUNC UNIFIED(STR: mod, STR: src):STR{
+    TNS: a = SPLIT(mod, "\n")
+    TNS: b = SPLIT(src, "\n")
+    INT: la = TLEN(a, 1)
+    INT: lb = TLEN(b, 1)
+    INT: n = MAX(la, lb)
+    STR: out = ""
+    INT: i = 1
+    WHILE( LTE(i, n) ){
+        STR: left = ""
+        IF( LTE(i, la) ){ left = a[i] }
+        STR: right = ""
+        IF( LTE(i, lb) ){ right = b[i] }
+
+        IF( EQ(left, right) ){
+            IF( NOT(EQ(left, "")) ){
+                STR: line = JOIN(" ", left)
+                out = _append_line(out, line)
+            }
+        } ELSE {
+            IF( NOT(EQ(left, "")) ){
+                out = _append_line(out, JOIN("-", left))
+            }
+            IF( NOT(EQ(right, "")) ){
+                out = _append_line(out, JOIN("+", right))
+            }
+        }
+        i = ADD(i, 1)
+    }
+    RETURN(out)
+}
+
+FUNC CONTEXT(STR: mod, STR: src):STR{
+    # For simplicity, use the same output as UNIFIED but with a ' ' prefix
+    STR: u = UNIFIED(mod, src)
+    IF( EQ(u, "") ){ RETURN("") }
+    # Prepend a simple header to resemble context diff
+    RETURN( JOIN("*** a", "\n", u) )
+}
+
+FUNC SIDE_BY_SIDE(STR: mod, STR: src):STR{
+    TNS: a = SPLIT(mod, "\n")
+    TNS: b = SPLIT(src, "\n")
+    INT: la = TLEN(a, 1)
+    INT: lb = TLEN(b, 1)
+    INT: n = MAX(la, lb)
+    STR: out = ""
+    INT: i = 1
+    WHILE( LTE(i, n) ){
+        STR: left = ""
+        IF( LTE(i, la) ){ left = a[i] }
+        STR: right = ""
+        IF( LTE(i, lb) ){ right = b[i] }
+        STR: marker = " "
+        IF( NOT(EQ(left, right)) ){ marker = "|" }
+        STR: line = JOIN(left, " ", marker, " ", right)
+        out = _append_line(out, line)
+        i = ADD(i, 1)
+    }
+    RETURN(out)
+}

test.asmln

@@ -7,6 +7,7 @@ IMPORT(image)
 IMPORT(gui)
 IMPORT(waveforms)
 IMPORT(stats)
+IMPORT(diff)
 
 FUNC BUILTIN_TESTS():INT{
     PRINT("Running built-in tests...")
@@ -1119,6 +1120,17 @@ FUNC STDLIB_TESTS():INT{
     } ELSE {
         PRINT("Skipping GUI tests on non-Windows host.")
     }
+
+    # diff library
+    STR: _d_a = "line1\nline2\nline3"
+    STR: _d_b = "line1\nlineX\nline3"
+    ASSERT( EQ( BOOL(diff.UNIFIED(_d_a, _d_b)), 1 ) )
+    ASSERT( EQ( BOOL(diff.CONTEXT(_d_a, _d_b)), 1 ) )
+    ASSERT( EQ( BOOL(diff.SIDE_BY_SIDE(_d_a, _d_b)), 1 ) )
+    DEL(_d_a)
+    DEL(_d_b)
+    PRINT("Diff extension tests passed.")
+
     PRINT("All standard library tests passed.\n")
     RETURN(0)
 }