chemenu/tools/chemenu/frontmatter_io.py

"""Read/write markdown files with YAML frontmatter, matching the formatting
conventions already used across wiki/ (inline flow-style lists, unquoted
dates, two-decimal confidence values).

We deliberately avoid a generic yaml.dump() for the frontmatter block because
PyYAML's default block-style output does not match the existing convention
(e.g. `tags: [a, b, c]` on one line). Instead we serialize each top-level key
explicitly, preserving dict insertion order.
"""
from __future__ import annotations

import datetime
import re
from pathlib import Path
from typing import Any

import yaml

from chemenu.errors import ValidationError

try:  # pragma: no cover - which branch runs depends on the host's libyaml
    from yaml import CSafeLoader as _Loader
except ImportError:  # pragma: no cover
    from yaml import SafeLoader as _Loader

FRONTMATTER_RE = re.compile(r"\A---\n(.*?)\n---\n?(.*)\Z", re.DOTALL)

# Frontmatter is a flat block of scalars and one-line lists. Real pages sit
# well under a kilobyte, so this is not a budget anyone writes against - it is
# the bound that keeps parse cost proportional to the corpus rather than to
# whatever a single file claims to be. It matters because the parser is on the
# read path, and the read path is what gets exposed.
MAX_FRONTMATTER_BYTES = 64 * 1024

# YAML anchors and aliases are refused outright rather than budgeted. A page's
# frontmatter has no use for them, and alias expansion is where a small
# document becomes an enormous object graph: 267 bytes of nested aliases
# compose in 0.2 ms into 672,603 nodes, growing 9**n with nesting depth at
# constant parse time. A size limit alone does not touch that, because the
# input stays small - see `ALIAS_BOMB` in the tests.
#
# The check runs on the *event* stream (`yaml.parse`), which is streaming and
# resolves nothing - so asking the question costs O(text) and never triggers
# the expansion it is asking about. `"*"` is a necessary character in any alias
# node, so its absence proves absence without parsing at all, which is the case
# every real page takes.
_ALIAS_HINT = "*"


class FrontmatterError(ValidationError):
    """Frontmatter that cannot be used: missing, malformed, oversized, or
    refused by a limit. Raised only by the strict entry points - the permissive
    ones report it as a string instead."""


def _load_frontmatter(fm_text: str) -> tuple[dict[str, Any] | None, str | None]:
    """Parse one frontmatter block into (mapping, error). Exactly one is None.

    The single parser behind both `read_page()` and `frontmatter_error()`. They
    used to have one each, which is how the permissive path could degrade to
    `{}` for a reason the strict path described differently - and how every
    caller wanting both answers read and parsed the file twice.
    """
    encoded = len(fm_text.encode("utf-8"))
    if encoded > MAX_FRONTMATTER_BYTES:
        return None, (
            f"frontmatter is {encoded} bytes, over the {MAX_FRONTMATTER_BYTES}-byte limit"
        )
    if _ALIAS_HINT in fm_text:
        try:
            for event in yaml.parse(fm_text, Loader=_Loader):
                if isinstance(event, yaml.AliasEvent):
                    return None, (
                        "frontmatter uses a YAML alias (`*"
                        f"{event.anchor}`); anchors and aliases are not allowed here"
                    )
        except yaml.YAMLError as exc:
            return None, f"invalid YAML frontmatter: {_first_line(exc)}"
        except RecursionError:
            return None, "frontmatter is nested too deeply to parse"
    try:
        parsed = yaml.load(fm_text, Loader=_Loader)
    except yaml.YAMLError as exc:
        return None, f"invalid YAML frontmatter: {_first_line(exc)}"
    except RecursionError:
        # PyYAML composes recursively, so deep nesting exhausts the interpreter
        # stack rather than raising a YAMLError. Unbounded nesting is bounded by
        # MAX_FRONTMATTER_BYTES; this catches what fits under it.
        return None, "frontmatter is nested too deeply to parse"
    if parsed is None:
        return None, "empty frontmatter block"
    if not isinstance(parsed, dict):
        return None, f"frontmatter is {type(parsed).__name__}, expected a mapping"
    return parsed, None


def _first_line(exc: Exception) -> str:
    text = str(exc)
    return text.splitlines()[0] if text else exc.__class__.__name__


def read_page_with_error(path: Path) -> tuple[dict[str, Any], str, str | None]:
    """(frontmatter, body, error) - the permissive read, with the reason it was
    permissive handed back instead of dropped.

    A page whose YAML is broken reads as `{}`, and a `{}` page then has no
    `confidence` and no `kind`: it drops out of `--field confidence<0.6` -
    precisely the query meant to find pages in bad shape - while looking to the
    caller like a page that simply did not match. Returning the reason is what
    lets a caller say so instead of losing the page quietly.
    """
    text = path.read_text(encoding="utf-8")
    match = FRONTMATTER_RE.match(text)
    if not match:
        return {}, text, "no `---` frontmatter block"
    fm_text, body = match.group(1), match.group(2)
    parsed, error = _load_frontmatter(fm_text)
    if error == "empty frontmatter block":
        # An empty block is a legitimate shape for the permissive read - it
        # carries no fields, and there is nothing to lose. Only `lint` treats
        # it as a finding.
        return {}, body, error
    return (parsed or {}), body, error


def read_page(path: Path) -> tuple[dict[str, Any], str]:
    """Return (frontmatter_dict, body) for a markdown file. If the file has no
    frontmatter block, returns ({}, full_text).

    Deliberately permissive: malformed YAML degrades to an empty dict so bulk
    operations never crash on one bad page. Use `read_page_with_error()` (search
    does) or `frontmatter_error()` (`wikitool lint` does) to surface those pages
    instead of losing them silently.
    """
    frontmatter, body, _ = read_page_with_error(path)
    return frontmatter, body


def read_page_strict(path: Path) -> tuple[dict[str, Any], str]:
    """`read_page()` that raises `FrontmatterError` instead of degrading.

    For any path that ingests frontmatter this instance did not write itself.
    The permissive read is right for bulk operations over a corpus the operator
    committed; it is wrong the moment the document arrives from outside, where
    "unparseable" must stop the document rather than empty it.
    """
    frontmatter, body, error = read_page_with_error(path)
    if error is not None:
        raise FrontmatterError(f"{path}: {error}")
    return frontmatter, body


def frontmatter_error(path: Path) -> str | None:
    """Return a human-readable reason why `path`'s frontmatter can't be used,
    or None if it parses into a dict.

    This is the strict counterpart to `read_page()`: without it, a page whose
    YAML is malformed (or whose frontmatter block is missing entirely) reads
    back as `{}` and then quietly slips past every frontmatter-driven check.
    """
    return read_page_with_error(path)[2]


def _format_scalar(value: Any, flow: bool = False) -> str:
    """Render one frontmatter value. `flow` says it is going inside a `[...]`
    sequence, where more characters are indicators than at document level."""
    if value is None:
        return '""'
    if isinstance(value, bool):
        return "true" if value else "false"
    if isinstance(value, float):
        return f"{value:.2f}"
    if isinstance(value, int):
        return str(value)
    if isinstance(value, (datetime.date, datetime.datetime)):
        return value.isoformat()
    text = str(value)
    needs_quoting = (
        text == ""
        or text[0] in "[{'\"#&*!|>%@`"
        or text.strip() != text
        or ": " in text
        or not _round_trips_as_string(text, flow=flow)
    )
    if needs_quoting:
        return _quote(text)
    return text


def _quote(text: str) -> str:
    """Render `text` as a quoted YAML scalar valid in any context.

    The dumper is asked for a one-element flow sequence and the brackets are
    taken off again, rather than a bare scalar: a bare plain scalar comes back
    with a `...` document-end marker attached, which is correct YAML for a
    document and nonsense inside a `[...]` list. Going through the library
    either way keeps the escaping rules where they belong - the same reason
    `_round_trips_as_string` asks the loader instead of listing cases.
    """
    dumped = yaml.safe_dump([text], default_flow_style=True, width=10**9).strip()
    return dumped[1:-1].strip()


def _round_trips_as_string(text: str, flow: bool = False) -> bool:
    """Whether writing `text` unquoted would read back as the same string.

    A string that merely *looks* like another YAML type comes back as that type:
    `"1945"` written bare is an int on the next read, and a schema expecting a
    string then rejects a page nothing visibly changed. The same holds for
    floats, YAML 1.1's `yes`/`no`/`on`/`off` booleans, `null`, and forms like
    `0x1F` or `1_000`.

    Asking the loader instead of listing the cases is deliberate: the reader and
    the writer then agree by construction, and a resolver rule this code never
    heard of cannot drift out from under it.

    Dates need no exception here. A date field holds a `datetime.date`, which
    `_format_scalar` renders bare before ever reaching this function - see
    `normalize_dates` for the other half of that contract.

    `flow` asks the question in the context list values are actually written
    in. At document level a comma is an ordinary character, so the probe says
    "safe to write bare"; inside the `[...]` this file emits for every list it
    is an indicator, and the value silently comes back as two elements. That is
    how a `raw_files:` entry naming a file with a comma in its name lost half
    of itself on write - after `--set` had parsed it correctly.
    """
    probe, expected = (f"[{text}]", [text]) if flow else (text, text)
    try:
        return yaml.load(probe, Loader=_Loader) == expected
    except yaml.YAMLError:
        # Unparseable bare - quoting is exactly the fix.
        return False


def normalize_dates(frontmatter: dict[str, Any]) -> dict[str, Any]:
    """A copy of `frontmatter` with every `datetime.date` rendered as an ISO
    string, at the top level and one list deep.

    Frontmatter carries dates as `datetime.date` - that is what `yaml.safe_load`
    produces for a bare `2026-08-29`, what `_format_scalar` writes back bare, and
    therefore what the whole corpus stores. The schemas nonetheless declare those
    fields `type: string`, so anything validating raw frontmatter has to convert
    first.

    This lives here rather than beside one validator because there are two, and
    only one of them used to do it: `TypeResolver.validate_frontmatter` (behind
    `lint`) normalized, while `touch`'s `validate_fields` did not. The gap was
    invisible only because `touch` happened to write date *strings*.
    """
    normalized: dict[str, Any] = {}
    for key, value in frontmatter.items():
        if isinstance(value, datetime.date):
            normalized[key] = value.isoformat()
        elif isinstance(value, list):
            normalized[key] = [
                item.isoformat() if isinstance(item, datetime.date) else item for item in value
            ]
        else:
            normalized[key] = value
    return normalized


def _format_list(items: list[Any]) -> str:
    if not items:
        return "[]"
    return "[" + ", ".join(_format_scalar(v, flow=True) for v in items) + "]"


def dump_frontmatter(frontmatter: dict[str, Any]) -> str:
    lines = []
    for key, value in frontmatter.items():
        if isinstance(value, list):
            lines.append(f"{key}: {_format_list(value)}")
        else:
            lines.append(f"{key}: {_format_scalar(value)}")
    return "\n".join(lines)


def write_page(path: Path, frontmatter: dict[str, Any], body: str) -> None:
    path.parent.mkdir(parents=True, exist_ok=True)
    fm_text = dump_frontmatter(frontmatter)
    if not body.startswith("\n"):
        body = "\n" + body
    content = f"---\n{fm_text}\n---{body}"
    if not content.endswith("\n"):
        content += "\n"
    path.write_text(content, encoding="utf-8")