src/cl_macros/ext/defmacro_parser.py

"""Extract defmacro definitions from Common Lisp source files.

Handles:
- String literals with escaped quotes
- Character literals (#\", #\\, #\(, etc.)
- Line comments (;) and block comments (#| ... |#)
- Reader macros (#' , #(, #., #+, #-, etc.)
- Nested parens
- defmacro, define-compiler-macro, defmacro!, defmacro/g!
"""

from __future__ import annotations

import re
from dataclasses import dataclass, field
from pathlib import Path


@dataclass
class DefmacroDef:
    macro_name: str
    args: str
    body: str
    full_form: str
    docstring: str | None = None
    line_number: int = 0
    source_file: Path | None = None
    form_type: str = "defmacro"


MACRO_STARTERS = [
    "defmacro",
    "define-compiler-macro",
    "defmacro!",
    "defmacro/g!",
]


class DefmacroParser:
    def __init__(self):
        self.errors: list[str] = []

    def extract_all(self, source_text: str) -> list[DefmacroDef]:
        """Find all macro definitions in source text."""
        self.errors = []
        results = []
        i = 0
        while i < len(source_text):
            form_start = self._find_next_macro_form(source_text, i)
            if form_start is None:
                break
            form = self._extract_balanced(source_text, form_start)
            if form:
                parsed = self._parse_defmacro(form, form_start)
                if parsed:
                    results.append(parsed)
                i = form_start + len(form)
            else:
                i = form_start + 1
        return results

    def _find_next_macro_form(self, text: str, start: int) -> int | None:
        """Find the opening paren of the next defmacro form."""
        i = start
        while i < len(text):
            ch = text[i]

            # Skip string literals
            if ch == '"':
                i = self._skip_string(text, i)
                continue

            # Skip character literals
            if ch == '#' and i + 1 < len(text) and text[i + 1] == '\\':
                i = self._skip_char_literal(text, i)
                continue

            # Skip line comments
            if ch == ';':
                i = self._skip_line_comment(text, i)
                continue

            # Skip block comments
            if ch == '#' and i + 1 < len(text) and text[i + 1] == '|':
                i = self._skip_block_comment(text, i)
                continue

            # Found opening paren — check if it's a macro form
            if ch == '(':
                name_start = self._find_form_name(text, i + 1)
                if name_start is not None:
                    name = text[name_start : name_start + 30]
                    for starter in MACRO_STARTERS:
                        if name.startswith(starter):
                            # Verify it's a full word boundary
                            after = name_start + len(starter)
                            if after >= len(text) or text[after] in ' \t\n\r(':
                                return i
                i += 1
                continue

            i += 1
        return None

    def _find_form_name(self, text: str, pos: int) -> int | None:
        """Skip whitespace after opening paren, return position of first non-whitespace char."""
        while pos < len(text) and text[pos] in ' \t\n\r':
            pos += 1
        return pos if pos < len(text) else None

    def _extract_balanced(self, text: str, start: int) -> str | None:
        """Extract a balanced s-expression starting at 'start'.

        Returns the full form including the outer parens, or None if unbalanced.
        """
        if start >= len(text) or text[start] != '(':
            return None

        depth = 0
        i = start
        while i < len(text):
            ch = text[i]

            if ch == '"':
                i = self._skip_string(text, i)
                continue

            if ch == '#' and i + 1 < len(text) and text[i + 1] == '\\':
                i = self._skip_char_literal(text, i)
                continue

            if ch == ';':
                i = self._skip_line_comment(text, i)
                continue

            if ch == '#' and i + 1 < len(text) and text[i + 1] == '|':
                i = self._skip_block_comment(text, i)
                continue

            # Handle #( vectors and #' function refs
            if ch == '#' and i + 1 < len(text) and text[i + 1] in "'(.":
                i += 2  # skip # and the next char
                continue

            # Handle #+ and #- reader conditionals
            if ch == '#' and i + 1 < len(text) and text[i + 1] in '+-':
                i += 2
                continue

            if ch == '(':
                depth += 1
            elif ch == ')':
                depth -= 1
                if depth == 0:
                    return text[start : i + 1]
            i += 1

        return None

    def _skip_string(self, text: str, start: int) -> int:
        """Skip past a string literal starting at the opening quote."""
        i = start + 1  # skip opening quote
        while i < len(text):
            if text[i] == '\\' and i + 1 < len(text):
                i += 2  # skip escaped char
                continue
            if text[i] == '"':
                return i + 1  # skip closing quote
            i += 1
        return i

    def _skip_char_literal(self, text: str, start: int) -> int:
        """Skip past a character literal #\\X. Returns position after the literal."""
        i = start + 2  # skip #\
        # The char literal is the next character (or a name like #\Space, #\Newline)
        if i >= len(text):
            return i
        if text[i] in ' \t\n\r()':
            return i  # bare #\ — edge case
        i += 1
        # If followed by more alpha chars, it's a named char like #\Space
        while i < len(text) and text[i].isalpha():
            i += 1
        return i

    def _skip_line_comment(self, text: str, start: int) -> int:
        """Skip past a semicolon comment to end of line."""
        i = start + 1
        while i < len(text) and text[i] != '\n':
            i += 1
        return i

    def _skip_block_comment(self, text: str, start: int) -> int:
        """Skip past a #| ... |# block comment. Handles nesting."""
        depth = 1
        i = start + 2  # skip #|
        while i < len(text) - 1:
            if text[i] == '|' and text[i + 1] == '#':
                depth -= 1
                i += 2
                if depth == 0:
                    return i
                continue
            if text[i] == '#' and text[i + 1] == '|':
                depth += 1
                i += 2
                continue
            i += 1
        return i

    def _parse_defmacro(self, form: str, pos: int) -> DefmacroDef | None:
        """Parse a defmacro form into its components."""
        line_number = form[:pos].count('\n') + 1 if pos > 0 else 1

        # Extract form type and macro name
        inner = self._strip_outer_parens(form)
        if inner is None:
            return None

        parts = self._split_top_level(inner, 3)
        if len(parts) < 2:
            return None

        form_type = parts[0].strip()
        if form_type not in MACRO_STARTERS:
            return None

        macro_name = parts[1].strip()
        if len(parts) >= 3:
            remaining = parts[2].strip()
            # Split args from body
            args_end = self._find_args_end(remaining)
            args = remaining[:args_end].strip() if args_end >= 0 else remaining
            body = remaining[args_end:].strip() if args_end >= 0 else ""
        else:
            args = "()"
            body = ""

        docstring = self._extract_docstring(body)

        return DefmacroDef(
            macro_name=macro_name,
            args=args,
            body=body,
            full_form=form,
            docstring=docstring,
            line_number=line_number,
            form_type=form_type,
        )

    def _strip_outer_parens(self, form: str) -> str | None:
        """Remove outer parens from a balanced form."""
        form = form.strip()
        if not form or form[0] != '(' or form[-1] != ')':
            return None
        return form[1:-1].strip()

    def _split_top_level(self, text: str, n: int) -> list[str]:
        """Split text into at most n parts at top-level whitespace."""
        parts = []
        depth = 0
        start = 0
        i = 0
        while i < len(text) and len(parts) < n - 1:
            ch = text[i]
            if ch == '"':
                i = self._skip_string(text, i)
                continue
            if ch == '#' and i + 1 < len(text) and text[i + 1] == '\\':
                i = self._skip_char_literal(text, i)
                continue
            if ch == ';':
                i = self._skip_line_comment(text, i)
                continue
            if ch == '#' and i + 1 < len(text) and text[i + 1] == '|':
                i = self._skip_block_comment(text, i)
                continue
            if ch == '(':
                depth += 1
            elif ch == ')':
                depth -= 1
            elif depth == 0 and ch in ' \t\n\r':
                part = text[start:i].strip()
                if part:
                    parts.append(part)
                    start = i + 1
            i += 1

        if start < len(text) and len(parts) < n:
            parts.append(text[start:].strip())

        return parts

    def _find_args_end(self, text: str) -> int:
        """Find the end of the lambda list (args). Returns index after closing paren."""
        text = text.lstrip()
        if not text or text[0] != '(':
            return 0

        depth = 0
        i = 0
        while i < len(text):
            ch = text[i]
            if ch == '"':
                i = self._skip_string(text, i)
                continue
            if ch == '#' and i + 1 < len(text) and text[i + 1] == '\\':
                i = self._skip_char_literal(text, i)
                continue
            if ch == ';':
                i = self._skip_line_comment(text, i)
                continue
            if ch == '#' and i + 1 < len(text) and text[i + 1] == '|':
                i = self._skip_block_comment(text, i)
                continue
            if ch == '(':
                depth += 1
            elif ch == ')':
                depth -= 1
                if depth == 0:
                    return i + 1
            i += 1
        return -1

    def _extract_docstring(self, body: str) -> str | None:
        """Extract docstring if first form in body is a string literal."""
        body = body.strip()
        if body and body[0] == '"':
            end = body.index('"', 1)
            while end > 0 and body[end - 1] == '\\':
                end = body.index('"', end + 1)
            if end > 0:
                return body[1:end]
        return None

    def extract_file(self, filepath: Path) -> list[DefmacroDef]:
        """Extract all macro definitions from a .lisp file."""
        try:
            text = filepath.read_text(errors="replace")
        except (OSError, UnicodeDecodeError) as e:
            self.errors.append(f"Failed to read {filepath}: {e}")
            return []

        results = self.extract_all(text)
        for r in results:
            r.source_file = filepath
        return results