Files
open-school/backend/app/program_content.py
2026-05-01 21:12:05 +02:00

221 lines
7.0 KiB
Python

import base64
import html
import os
import re
from functools import lru_cache
from pathlib import Path
from urllib.parse import quote
def default_content_root() -> Path:
current = Path(__file__).resolve()
candidates = [
current.parents[1] / "contenus_pedagogiques",
current.parents[1] / "Programme" / "contenus_pedagogiques",
]
candidates.extend(parent / "Programme" / "contenus_pedagogiques" for parent in current.parents)
candidates.extend(
[
Path("/programme/contenus_pedagogiques"),
Path("/app/contenus_pedagogiques"),
Path("/app/Programme/contenus_pedagogiques"),
]
)
return next((candidate for candidate in candidates if candidate.exists()), candidates[0])
CONTENT_ROOT = Path(os.getenv("PROGRAM_CONTENT_ROOT", default_content_root())).resolve()
GRADE_ALIASES = {
"CM1": "cm1",
"CM2": "cm2",
"6e": "6e",
"6eme": "6e",
"6ieme": "6e",
"5e": "5e",
"5eme": "5e",
"5ieme": "5e",
"4e": "4e",
"4eme": "4e",
"4ieme": "4e",
"3e": "3e",
"3eme": "3e",
"3ieme": "3e",
}
def normalize_grade(grade: str) -> str:
compact = (grade or "").strip().replace("è", "e").replace("é", "e")
return GRADE_ALIASES.get(compact, compact.lower())
def title_from_slug(slug: str) -> str:
parts = slug.split("_")
if parts and parts[0][:2].isdigit():
parts = parts[1:]
return " ".join(part.capitalize() for part in parts if part)
def encode_asset_path(relative_path: str) -> str:
payload = relative_path.encode("utf-8")
return base64.urlsafe_b64encode(payload).decode("ascii").rstrip("=")
def decode_asset_token(token: str) -> Path:
padding = "=" * (-len(token) % 4)
relative = base64.urlsafe_b64decode(f"{token}{padding}").decode("utf-8")
path = (CONTENT_ROOT / relative).resolve()
if CONTENT_ROOT not in path.parents:
raise ValueError("Chemin de ressource invalide")
return path
def build_asset_url(relative_path: str) -> str:
return f"/api/program/assets/{quote(encode_asset_path(relative_path))}"
def build_section_url(relative_path: str, section_index: int) -> str:
token = quote(encode_asset_path(relative_path))
return f"/api/program/assets/{token}/sections/{section_index}"
def extract_svg_sections(svg_path: Path) -> list[dict]:
text = svg_path.read_text(encoding="utf-8", errors="ignore")
sections: list[dict] = []
pattern = re.compile(
r'<g\s+transform="translate\((?P<x>[-\d.]+)\s+(?P<y>[-\d.]+)\)">\s*'
r'<rect\s+x="0"\s+y="0"\s+width="(?P<w>[-\d.]+)"\s+height="(?P<h>[-\d.]+)"[^>]*>'
r'(?P<body>.*?)</g>',
re.DOTALL,
)
for match in pattern.finditer(text):
width = float(match.group("w"))
height = float(match.group("h"))
if width < 250 or height < 120:
continue
title_match = re.search(r'<text[^>]*>(?P<title>.*?)</text>', match.group("body"), re.DOTALL)
raw_title = re.sub(r"<[^>]+>", "", title_match.group("title") if title_match else "")
title = html.unescape(raw_title).strip() or f"Étape {len(sections) + 1}"
margin = 28
sections.append(
{
"index": len(sections),
"title": title,
"view_box": [
max(float(match.group("x")) - margin, 0),
max(float(match.group("y")) - margin, 0),
width + margin * 2,
height + margin * 2,
],
}
)
return sections
def render_svg_section(path: Path, section_index: int) -> str:
sections = extract_svg_sections(path)
if section_index < 0 or section_index >= len(sections):
raise IndexError("Section introuvable")
x, y, width, height = sections[section_index]["view_box"]
text = path.read_text(encoding="utf-8", errors="ignore")
text = re.sub(r'\swidth="[^"]+"', f' width="{int(width)}"', text, count=1)
text = re.sub(r'\sheight="[^"]+"', f' height="{int(height)}"', text, count=1)
text = re.sub(
r'\sviewBox="[^"]+"',
f' viewBox="{x:g} {y:g} {width:g} {height:g}"',
text,
count=1,
)
return text
@lru_cache(maxsize=1)
def list_lessons() -> list[dict]:
lessons: list[dict] = []
if not CONTENT_ROOT.exists():
return lessons
for svg_dir in sorted(CONTENT_ROOT.glob("cycle_*/*/*/**/svg")):
if not svg_dir.is_dir():
continue
svg_files = sorted(svg_dir.glob("*.svg"))
if not svg_files:
continue
lesson_dir = svg_dir.parent
relative_lesson = lesson_dir.relative_to(CONTENT_ROOT).as_posix()
parts = relative_lesson.split("/")
if len(parts) < 4:
continue
cycle, subject, grade = parts[0], parts[1], parts[2]
title = title_from_slug(lesson_dir.name)
readme = lesson_dir / "README.md"
if readme.exists():
first_heading = next(
(
line.lstrip("#").strip()
for line in readme.read_text(encoding="utf-8", errors="ignore").splitlines()
if line.startswith("#")
),
"",
)
title = first_heading or title
assets = []
for svg_path in svg_files:
relative_path = svg_path.relative_to(CONTENT_ROOT).as_posix()
sections = extract_svg_sections(svg_path)
assets.append(
{
"name": svg_path.stem,
"title": title_from_slug(svg_path.stem),
"path": relative_path,
"url": build_asset_url(relative_path),
"sections": [
{
**section,
"url": build_section_url(relative_path, section["index"]),
}
for section in sections
],
}
)
lessons.append(
{
"id": relative_lesson,
"title": title,
"subject": title_from_slug(subject),
"grade": grade,
"cycle": cycle,
"asset_count": len(assets),
"assets": assets,
}
)
return lessons
def list_lessons_for_grade(grade: str) -> list[dict]:
normalized = normalize_grade(grade)
return [lesson for lesson in list_lessons() if lesson["grade"] == normalized]
def get_content_status() -> dict:
lessons = list_lessons()
by_grade: dict[str, int] = {}
for lesson in lessons:
by_grade[lesson["grade"]] = by_grade.get(lesson["grade"], 0) + 1
return {
"content_root": str(CONTENT_ROOT),
"content_root_exists": CONTENT_ROOT.exists(),
"lesson_count": len(lessons),
"lessons_by_grade": by_grade,
}
def get_lesson(lesson_id: str) -> dict | None:
return next((lesson for lesson in list_lessons() if lesson["id"] == lesson_id), None)