feat: WP-0003 complete — LEVEL3 advanced features + error framework

Implements full LEVEL3 feature set: cross-references (xref.py), numbered
figures (figures.py), auto-diagrams (diagrams.py), bibliography/citations
(bibliography.py), LEVEL3 capability detection (level3.py), and structured
error/warning records (errors.py). Builder, importer, and differ updated for
LEVEL3 round-trip support. REST and MCP interfaces updated with structured
warning records. 259 tests passing.

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
This commit is contained in:
2026-03-16 10:51:38 +00:00
parent 760047b82b
commit ac442ea41f
26 changed files with 3713 additions and 74 deletions

View File

@@ -5,6 +5,8 @@ from __future__ import annotations
import re
from dataclasses import dataclass, field
from markidocx.errors import OutputState
HEADING_RE = re.compile(r"^(#{1,6})\s+(.+)$", re.MULTILINE)
LIST_ITEM_RE = re.compile(r"^(\s*[-*+]|\s*\d+\.)\s+(.+)$", re.MULTILINE)
TABLE_ROW_RE = re.compile(r"^\|.+\|$", re.MULTILINE)
@@ -19,6 +21,7 @@ class DriftReport:
degraded: list[str] = field(default_factory=list)
broken: list[str] = field(default_factory=list)
unsupported: list[str] = field(default_factory=list)
output_state: OutputState = OutputState.FINAL
def compare(original: str, reimported: str) -> DriftReport:
@@ -76,13 +79,29 @@ def compare(original: str, reimported: str) -> DriftReport:
else:
degraded.append(f"link:lost {link[:40]}")
# --- Cross-references (FR-531, FR-540) ---
_compare_xrefs(original, reimported, preserved, degraded, broken)
# --- Figures (FR-532, FR-541) ---
_compare_figures(original, reimported, preserved, degraded, broken)
# --- Citations & Bibliography (FR-535, FR-542) ---
from markidocx.bibliography import compare_citations
compare_citations(original, reimported, preserved, degraded, broken)
has_drift = bool(degraded or broken)
output_state = (
OutputState.FINAL if not has_drift
else (OutputState.DEGRADED if not broken else OutputState.PARTIAL)
)
return DriftReport(
has_drift=has_drift,
preserved=preserved,
degraded=degraded,
broken=broken,
unsupported=unsupported,
output_state=output_state,
)
@@ -104,6 +123,64 @@ def _count_tables(text: str) -> int:
return count
def _compare_figures(
original: str,
reimported: str,
preserved: list[str],
degraded: list[str],
broken: list[str],
) -> None:
"""Compare figure labels and captions (FR-532, FR-541)."""
from markidocx.figures import extract_figure_captions, extract_figure_labels
orig_labels = extract_figure_labels(original)
reim_labels = extract_figure_labels(reimported)
for label in orig_labels:
if label in reim_labels:
preserved.append(f"figure-label:{label}")
else:
broken.append(f"figure-label:missing '{label}'")
orig_captions = extract_figure_captions(original)
reim_captions = extract_figure_captions(reimported)
orig_set = set(orig_captions)
reim_set = set(reim_captions)
for caption in orig_set:
if caption in reim_set:
preserved.append(f"figure-caption:{caption[:40]}")
else:
degraded.append(f"figure-caption:lost '{caption[:40]}'")
def _compare_xrefs(
original: str,
reimported: str,
preserved: list[str],
degraded: list[str],
broken: list[str],
) -> None:
"""Compare cross-reference anchors and links (FR-531, FR-540)."""
from markidocx.xref import extract_anchors, extract_xref_links
orig_anchors = extract_anchors(original)
reim_anchors = extract_anchors(reimported)
for anchor in orig_anchors:
if anchor in reim_anchors:
preserved.append(f"xref-anchor:{anchor}")
else:
broken.append(f"xref-anchor:missing '{anchor}'")
orig_xrefs = extract_xref_links(original)
reim_xrefs = extract_xref_links(reimported)
for link_text, anchor in orig_xrefs:
if (link_text, anchor) in reim_xrefs:
preserved.append(f"xref-link:[{link_text}][{anchor}]")
elif anchor not in reim_anchors:
broken.append(f"xref-link:broken-target [{link_text}][{anchor}]")
else:
degraded.append(f"xref-link:degraded [{link_text}][{anchor}]")
def _compare_sets(
kind: str,
orig: list[str],