"""Run the repeatable static gates for one completed CM108G unit.""" from __future__ import annotations from html import unescape from html.parser import HTMLParser from pathlib import Path from xml.etree import ElementTree import json import re import sys ROOT = Path(__file__).resolve().parents[1] SVG_BY_WEEK = {1:"evidence-inference-decision.svg",2:"managerial-role-map.svg",3:"decision-memo-flow.svg",4:"week-04-concept-map.svg",5:"goals-planning-architecture.svg",6:"strategy-diversification-architecture.svg"} class P(HTMLParser): def __init__(self): super().__init__() self.ids = {} self.frags = [] self.headings = [] self.heading_texts = [] self.fonts = [] self.actions = [] self.main = False self.tools = False self.skip_link = False self.concept_scroll_region = False self.notes_modal = False self.table_wrap_count = 0 self.accessible_table_wrap_count = 0 self.external_links = [] self.anchor_depth = 0 self.nested_anchor_count = 0 self._anchor_href = None self._anchor_parts = [] self._heading_level = None self._heading_parts = [] def handle_starttag(self, tag, attrs): a = {k.lower(): v or "" for k, v in attrs} classes = set(a.get("class", "").split()) if tag == "a": if self.anchor_depth: self.nested_anchor_count += 1 self.anchor_depth += 1 if "id" in a: self.ids[a["id"]] = self.ids.get(a["id"], 0) + 1 if tag == "a" and a.get("href", "").startswith("#"): self.frags.append(a["href"][1:]) if tag == "a" and a.get("href", "") and not a.get("href", "").startswith("#"): self._anchor_href = a["href"] self._anchor_parts = [] if tag == "a" and "skip-link" in classes and a.get("href") == "#reading": self.skip_link = True if tag in {"h1", "h2", "h3", "h4", "h5", "h6"}: self._heading_level = int(tag[1]) self._heading_parts = [] self.headings.append(self._heading_level) if tag == "button" and a.get("data-font-size"): self.fonts.append(a["data-font-size"]) if a.get("data-study-action"): self.actions.append(a["data-study-action"]) if "concept-figure__graphic" in classes: self.concept_scroll_region = ( a.get("tabindex") == "0" and a.get("role") == "region" and bool(a.get("aria-label")) ) if "table-wrap" in classes: self.table_wrap_count += 1 if a.get("tabindex") == "0" and a.get("role") == "region" and bool(a.get("aria-label")): self.accessible_table_wrap_count += 1 if a.get("id") == "notes-panel": self.notes_modal = a.get("role") == "dialog" and a.get("aria-modal") == "true" self.main |= a.get("data-count") == "main-body" self.tools |= "data-study-tools" in a def handle_data(self, data): if self._heading_level is not None: self._heading_parts.append(data) if self._anchor_href is not None: self._anchor_parts.append(data) def handle_endtag(self, tag): if tag == "a": self.anchor_depth = max(0, self.anchor_depth - 1) if tag == "a" and self._anchor_href is not None: self.external_links.append((self._anchor_href, " ".join("".join(self._anchor_parts).split()))) self._anchor_href = None self._anchor_parts = [] if self._heading_level is not None and tag == f"h{self._heading_level}": self.heading_texts.append((self._heading_level, " ".join("".join(self._heading_parts).split()))) self._heading_level = None self._heading_parts = [] def raw_markdown_words(text): """Legacy audit count; Markdown destinations may inflate this number.""" return len(re.findall(r"\b[A-Za-z]+(?:['’][A-Za-z]+)?\b", text)) def visible_markdown(text): """Return learner-visible Markdown text without URLs, tags, or comments.""" text = re.sub(r"", " ", text, flags=re.S) text = re.sub(r"!?\[([^\]]*)\]\((?:[^()]|\([^()]*\))*\)", r"\1", text) text = re.sub(r"]+>", " ", text) text = re.sub(r"https?://[^\s)\]>]+", " ", text) text = re.sub(r"<[^>]+>", " ", text) return text def visible_words(text): """Conservative English count: a hyphenated compound is one visible word.""" return len(re.findall(r"\b[A-Za-z]+(?:[-'’][A-Za-z]+)*\b", visible_markdown(text))) def normalized(text): return " ".join(unescape(re.sub(r"<[^>]+>", " ", text)).split()).casefold() def run(week): unit=f"week-{week:02d}"; module=ROOT/"04_modules"/unit; page=ROOT/"06_web/pages"/f"{unit}.html"; svg_path=ROOT/"06_web/assets/svg"/SVG_BY_WEEK.get(week,f"{unit}-concept-map.svg") manuscript=(module/"manuscript.md").read_text(encoding="utf-8"); resources=(module/"resources.md").read_text(encoding="utf-8"); glossary=(module/"glossary.md").read_text(encoding="utf-8"); assessment=(module/"assessment.md").read_text(encoding="utf-8"); m=re.search(r"(.*?)",manuscript,re.S) if not m: raise ValueError(f"{unit}: main-body markers missing") body=m.group(1); ptext=page.read_text(encoding="utf-8"); p=P(); p.feed(ptext); css=(ROOT/"06_web/css/textbook.css").read_text(encoding="utf-8"); reader_js=(ROOT/"06_web/js/textbook-reader-v2.js").read_text(encoding="utf-8"); svg=svg_path.read_text(encoding="utf-8"); root=ElementTree.fromstring(svg); a={k.rsplit("}",1)[-1].lower():v for k,v in root.attrib.items()} forbidden = ("Chief PM", "Research Agent", "Writing Agent", "Reviewer Agent", "Context Packet", "Claim Ledger", "C:\\Users\\", "pm_state.json") expected = {"toggle-notes", "print", "open-export", "highlight", "underline", "show-note", "download-notes", "download-reading", "clear-record"} heading_values = [value.casefold() for _, value in p.heading_texts] svg_labels = [] for node in root.iter(): if node.tag.rsplit("}", 1)[-1] == "text": value = " ".join("".join(node.itertext()).split()) if value and value not in svg_labels: svg_labels.append(value) alternative_match = re.search(r'
]*>(.*?)
', ptext, re.S) alternative_text = normalized(alternative_match.group(1)) if alternative_match else "" svg_labels_gate = bool(svg_labels) and all(normalized(label) in alternative_text for label in svg_labels) if week <= 5: wrapper = (ROOT / "scripts" / f"build_week{week}_html.py").read_text(encoding="utf-8") canonical_builder_gate = "def build_legacy" in wrapper and f"return build_unit({week})" in wrapper else: canonical_builder_gate = True controlled_reference_links = ( ("https://d21.co.jp/book/detail/978-4-7993-2555-1", "Discover 21 publisher record"), ("https://www.waseda.jp/fcom/wbs/faculty-en/15632", "Waseda Business School author profile"), ) reference_book_descriptive_link_gate = all( f"[{label}]({href})" in resources and any(link_href == href and link_text == label and link_text != link_href for link_href, link_text in p.external_links) for href, label in controlled_reference_links ) result = { "unit": unit, "word_count_method": "learner-visible English; link labels retained; URL destinations/tags/comments excluded; hyphenated compounds count as one", "main_body_word_count": visible_words(body), "raw_markdown_main_body_word_count": raw_markdown_words(body), "supporting_content_word_count": sum(visible_words(text) for text in (glossary, assessment, resources)), "raw_markdown_supporting_content_word_count": sum(raw_markdown_words(text) for text in (glossary, assessment, resources)), "missing_fragments": sorted(set(p.frags) - set(p.ids)), "duplicate_ids": sorted(key for key, value in p.ids.items() if value > 1), "word_gate": visible_words(body) >= 5000, "authoring_buffer_gate": visible_words(body) >= 5100, "heading_gate": p.headings.count(1) == 1, "heading_hierarchy_gate": all((later - earlier) <= 1 for earlier, later in zip(p.headings, p.headings[1:])), "one_learning_map_heading_gate": sum("learning map" in value for value in heading_values) == 1, "one_no_ai_heading_gate": sum("no ai task" in value for value in heading_values) == 1, "one_glossary_heading_gate": sum("glossary" in value for value in heading_values) == 1, "glossary_term_count": len(re.findall(r"(?m)^(?:\d+\.\s+\*\*|##\s+\d+\.)", glossary)), "glossary_count_gate": 1 <= len(re.findall(r"(?m)^(?:\d+\.\s+\*\*|##\s+\d+\.)", glossary)) <= 15, "assessment_question_count": len(re.findall(r"(?m)^##\s+(?:Question\s+)?\d+(?=\b|[.:|])", assessment)), "assessment_count_gate": len(re.findall(r"(?m)^##\s+(?:Question\s+)?\d+(?=\b|[.:|])", assessment)) == 3, "font_gate": set(p.fonts) >= {"xs", "s", "m", "l", "xl"}, "study_tools": p.tools, "study_action_gate": expected <= set(p.actions), "main_body_marker_gate": p.main, "single_column_css_gate": "column-count: 2" not in css and "column-count: 1" in css, "reader_v2_gate": "../js/textbook-reader-v2.js" in ptext, "stable_annotation_gate": all(token in reader_js for token in ("stableHash", "reanchorSegment", "blockId", "start", "end", "quote", "itemMatchesPage")), "keyboard_selection_gate": all(token in reader_js for token in ("keyboardSelectionPending", "selectionNavigationKeys", "focusToolbar", "selectionTools.contains(document.activeElement)")), "focus_management_gate": all(token in reader_js for token in ("focusableElements", "lastNotesFocus", "lastDialogFocus", 'setAttribute("aria-modal", "true")')), "local_date_export_gate": "localDateStamp(new Date())" in reader_js and "toISOString().slice(0, 10)" not in reader_js, "responsive_contents_gate": "Show page contents" in reader_js and "@media (max-width: 900px)" in css, "mobile_concept_scroll_gate": ".concept-figure img, .concept-figure svg { width: 760px" in css, "svg_font_control_gate": "--diagram-text-size" in css and ".concept-figure__graphic svg text" in css, "skip_link_gate": p.skip_link and ".skip-link:focus" in css, "concept_scroll_region_gate": p.concept_scroll_region, "table_wrap_count": p.table_wrap_count, "accessible_table_wrap_count": p.accessible_table_wrap_count, "table_accessibility_gate": p.table_wrap_count > 0 and p.accessible_table_wrap_count == p.table_wrap_count, "notes_modal_gate": p.notes_modal, "print_notes_gate": "print-notes" in reader_js and ".print-notes { display: block" in css and "break-after: avoid-page" in css, "privacy_gate": not any(value.casefold() in ptext.casefold() for value in forbidden), "learner_internal_status_gate": not any(value in ptext.casefold() for value in ( "course edition under review", "this unit is an english course edition under review", "reviewer-cleared", "instructor-approved", "review candidate", "pending instructor approval", "approved public-source boundary", )), "answer_key_leakage_gate": not any(value in ptext.casefold() for value in ( "answer key", "correct choices", "correct answers", "instructor answer", "scoring guide", )), "instructor_only_leakage_gate": not any(value in ptext.casefold() for value in ( "instructor-only:start", "instructor-only:end", "instructor rationale and scoring guide", "exclude the following instructor-only block", )), "book_url_not_exposed_gate": "docs.google.com/document" not in ptext and "docs.google.com/document" not in resources, "reference_book_title_gate": "the new business of management" not in (resources + "\n" + ptext).casefold(), "reference_book_attribution_gate": all(value in resources.casefold() for value in ( "koji mitani", "すべての働く人のための新しい経営学", "the instructor", "target", "value", "capability", "revenue model", "teacher-provided", "management theory", "research evidence", "application", "practice", )) and "inspir" in resources.casefold() and "not a translation" in resources.casefold() and "official english edition" in resources.casefold() and "substitute" in resources.casefold(), "reference_book_descriptive_link_gate": reference_book_descriptive_link_gate, "valid_anchor_nesting_gate": p.nested_anchor_count == 0 and p.anchor_depth == 0, "inline_citation_gate": len(re.findall(r"https://doi\.org/", body)) >= 3, "anti_ai_punctuation_gate": "—" not in body and "–" not in body, "svg_gate": root.tag.endswith("svg") and bool(a.get("viewbox")) and a.get("role") == "img" and bool(a.get("aria-label")), "svg_text_alternative_gate": bool(alternative_match) and "how the elements connect:" in alternative_text and svg_labels_gate, "canonical_builder_gate": canonical_builder_gate, } required = [ "word_gate", "heading_gate", "heading_hierarchy_gate", "one_learning_map_heading_gate", "one_no_ai_heading_gate", "one_glossary_heading_gate", "glossary_count_gate", "assessment_count_gate", "font_gate", "study_tools", "study_action_gate", "main_body_marker_gate", "single_column_css_gate", "reader_v2_gate", "stable_annotation_gate", "keyboard_selection_gate", "focus_management_gate", "local_date_export_gate", "responsive_contents_gate", "mobile_concept_scroll_gate", "svg_font_control_gate", "skip_link_gate", "concept_scroll_region_gate", "table_accessibility_gate", "notes_modal_gate", "print_notes_gate", "privacy_gate", "learner_internal_status_gate", "answer_key_leakage_gate", "instructor_only_leakage_gate", "book_url_not_exposed_gate", "reference_book_title_gate", "reference_book_attribution_gate", "reference_book_descriptive_link_gate", "valid_anchor_nesting_gate", "inline_citation_gate", "anti_ai_punctuation_gate", "svg_gate", "svg_text_alternative_gate", "canonical_builder_gate", ] result["total_counted_word_count"]=result["main_body_word_count"]+result["supporting_content_word_count"]; result["recommendation"]="pass-static-gates" if all(result[k] for k in required) and not result["missing_fragments"] and not result["duplicate_ids"] else "blocked" out=ROOT/"07_tests"; out.mkdir(exist_ok=True); (out/f"{unit}-static-qa.json").write_text(json.dumps(result,indent=2),encoding="utf-8",newline="\n"); (out/f"{unit}-static-qa.md").write_text(f"# {unit} static QA\n\n"+"\n".join(f"- **{k}:** {v}" for k,v in result.items())+"\n\nStatic QA does not replace independent Strict Review, browser interaction review, print review, or instructor approval.\n",encoding="utf-8",newline="\n"); print(json.dumps(result,indent=2)) if __name__=="__main__": if len(sys.argv)!=2 or not sys.argv[1].isdigit(): raise SystemExit("Usage: python scripts/qa_unit.py WEEK") run(int(sys.argv[1]))