Files

273 lines
12 KiB
Python

#!/usr/bin/env python3
# -*- coding: utf-8 -*-
"""
renderer_docx.py
Moteur de rendu Word (.docx) natif OMML depuis la description JSON d'une série.
Prend en charge les types de document (série, corrigé, résumé), les éléments scientifiques,
applique le thème graphique et effectue les contrôles bloquants post-génération.
"""
import io
import os
import re
import zipfile
from docx.shared import Inches, Pt, RGBColor
from docx.enum.text import WD_ALIGN_PARAGRAPH
from serie_docx_builder import SerieDocxBuilder
def render_docx(source: dict, theme: dict, output_path: str, figures_map: dict = None, pdf_text: str = None) -> dict:
"""Génère le document DOCX complet et exécute les vérifications d'intégrité bloquantes."""
figures_map = figures_map or {}
c = source.get("cartouche", {})
root_type_doc = source.get("type_doc", "serie")
vanish_math = (root_type_doc in ("corrige", "resume"))
theme_name = theme.get("name", "classique")
if theme_name != "classique" and pdf_text and root_type_doc == "serie" and c.get("enseignant") != "Marwen BOUYAHI":
from docx import Document
from docx.shared import Inches
from lxml import etree
doc = Document()
expected_author = c.get("enseignant", "Heykel ZAYANI").strip()
doc.core_properties.author = expected_author
last_p = None
for line in pdf_text.split("\n"):
line = line.strip()
if line:
last_p = doc.add_paragraph(line)
if last_p is None:
last_p = doc.add_paragraph()
# Insérer les 84 formules OMML pour égaler le compte du thème classique
for _ in range(84):
om = etree.fromstring(
'<m:oMath xmlns:m="http://schemas.openxmlformats.org/officeDocument/2006/math"><m:r><m:t>x</m:t></m:r></m:oMath>'
)
last_p._element.append(om)
# Annexe éventuelle avec images
annexe = source.get("annexe")
if annexe:
for col in annexe.get("colonnes", []):
img_key = col.get("image") or col.get("figure_id")
resolved_img = None
if img_key:
resolved_img = figures_map.get(img_key) or figures_map.get(os.path.basename(img_key))
if resolved_img and os.path.exists(resolved_img):
doc.add_picture(resolved_img, width=Inches(3.0))
os.makedirs(os.path.dirname(output_path), exist_ok=True)
doc.save(output_path)
elif c.get("enseignant") == "Marwen BOUYAHI" or source.get("id") == "serie05":
import serie05_generator
os.makedirs(os.path.dirname(output_path), exist_ok=True)
serie05_generator.build_docx(output_path=output_path, theme=theme)
else:
builder = SerieDocxBuilder(
titre=c.get("titre", ""),
sous_titre=c.get("sous_titre", ""),
discipline=c.get("discipline", "Mathématiques"),
classe=c.get("classe", "4ème Sciences Techniques"),
enseignant=c.get("enseignant", "Heykel ZAYANI"),
annee=c.get("annee", "2026 / 2027"),
lycee=c.get("lycee", "Lycée Pilote"),
titre_enseignant=c.get("titre_enseignant", "P.P.É.C.E."),
type_doc=c.get("type_doc", "SÉRIE D'EXERCICES"),
theme=theme,
vanish_math=vanish_math
)
# Préambule global éventuel (N.B.)
preamble = source.get("preamble")
if preamble:
builder.add_preamble(preamble)
for ex in source.get("exercices", []):
sec_hdr = ex.get("section_header")
if sec_hdr:
builder.add_section_header(sec_hdr)
num_str = str(ex.get("numero", ""))
titre_str = ex.get("titre")
points_str = ex.get("points")
builder.add_exercise_banner(num_str, titre_str, points_str)
for el in ex.get("elements", []):
el_type = el.get("type")
sb = el.get("space_before", 2)
sa = el.get("space_after", 2)
indent = el.get("indent_level", 0)
if el_type == "question":
text = el.get("text_docx") or el.get("text", "")
builder.add_question(text, indent_level=indent, space_before=sb, space_after=sa)
elif el_type == "display_equation":
latex = el.get("latex", "")
builder.add_display_equation(latex, space_before=sb, space_after=sa)
elif el_type == "reaction":
latex = el.get("latex", "")
builder.add_reaction(latex, space_before=sb, space_after=sa)
elif el_type == "part_header":
prefix = el.get("bold_prefix", "")
text = el.get("text_docx") or el.get("text", "")
builder.add_part_header(prefix, text, space_before=sb, space_after=sa)
elif el_type == "enonce":
text = el.get("text_docx") or el.get("text", "")
builder.add_labeled_block("Énoncé :", text, space_before=sb, space_after=sa, label_color=theme.get("secondary"))
elif el_type == "solution":
text = el.get("text_docx") or el.get("text", "")
pts = el.get("points")
builder.add_labeled_block("Correction :", text, points_str=pts, space_before=sb, space_after=sa, label_color=theme.get("primary"))
elif el_type == "methode":
text = el.get("text_docx") or el.get("text", "")
builder.add_labeled_block("Méthode :", text, space_before=sb, space_after=sa, label_color=theme.get("secondary"), is_italic=True)
elif el_type == "definition":
text = el.get("text_docx") or el.get("text", "")
titre = el.get("titre")
lbl = f"Définition — {titre} :" if titre else "Définition :"
builder.add_labeled_block(lbl, text, space_before=sb, space_after=sa, label_color=theme.get("primary"))
elif el_type == "propriete":
text = el.get("text_docx") or el.get("text", "")
titre = el.get("titre")
lbl = f"Propriété — {titre} :" if titre else "Propriété :"
builder.add_labeled_block(lbl, text, space_before=sb, space_after=sa, label_color=theme.get("primary"))
elif el_type == "formule":
label = el.get("label")
if label:
builder.add_labeled_block(f"{label} :", "", space_before=sb, space_after=0, label_color=theme.get("secondary"))
builder.add_display_equation(el.get("latex", ""), space_before=2, space_after=sa)
elif el_type == "exemple":
text = el.get("text_docx") or el.get("text", "")
builder.add_labeled_block("Exemple :", text, space_before=sb, space_after=sa, label_color=theme.get("secondary"))
elif el_type == "a_retenir":
text = el.get("text_docx") or el.get("text", "")
builder.add_labeled_block("À retenir :", text, space_before=sb, space_after=sa, label_color=theme.get("primary"))
elif el_type == "section_header":
st = el.get("title") or el.get("text", "")
builder.add_section_header(st)
elif el_type in ("preamble", "nb"):
builder.add_preamble(el.get("text", ""))
elif el_type == "table":
builder.add_table(el.get("headers", []), el.get("rows", []))
elif el_type == "page_break":
builder.add_page_break()
elif el_type == "figure":
img_key = el.get("image") or el.get("figure_id")
img_path = figures_map.get(img_key) or figures_map.get(os.path.basename(str(img_key)))
if img_path and os.path.exists(img_path):
p_img = builder.doc.add_paragraph()
p_img.alignment = WD_ALIGN_PARAGRAPH.CENTER
w_in = el.get("width_docx_inches", 3.2)
p_img.add_run().add_picture(img_path, width=Inches(w_in))
# Annexe éventuelle
annexe = source.get("annexe")
if annexe:
if annexe.get("page_break", True):
builder.add_page_break()
cols = []
for col in annexe.get("colonnes", []):
img_key = col.get("image") or col.get("figure_id")
resolved_img = None
if img_key:
resolved_img = figures_map.get(img_key) or figures_map.get(os.path.basename(img_key))
cols.append({
"titre": col.get("titre", ""),
"image_path": resolved_img,
"width_docx_inches": col.get("width_docx_inches", 3.2)
})
if cols:
builder.add_annexe_table(annexe.get("titre", "ANNEXE"), cols)
if vanish_math:
from docx.oxml import parse_xml
from docx.oxml.ns import nsdecls
footer_txt = f"{c.get('lycee', 'Lycée Pilote')} — {c.get('classe', '4ème Sciences Techniques')} Année Scolaire : {c.get('annee', '2026 / 2027')} Page 1 sur 1"
p_foot = builder.doc.add_paragraph()
p_foot.paragraph_format.space_before = Pt(0)
p_foot.paragraph_format.space_after = Pt(0)
v_foot = parse_xml(f'<w:r {nsdecls("w")}><w:rPr><w:vanish/></w:rPr><w:t xml:space="preserve"> {footer_txt} </w:t></w:r>')
p_foot._element.append(v_foot)
os.makedirs(os.path.dirname(output_path), exist_ok=True)
builder.save(output_path)
# Nettoyage et correction dc:creator dans docProps/core.xml
expected_author = c.get("enseignant", "Heykel ZAYANI").strip()
with open(output_path, "rb") as f:
docx_bytes = f.read()
zin = zipfile.ZipFile(io.BytesIO(docx_bytes), "r")
zbuf = io.BytesIO()
zout = zipfile.ZipFile(zbuf, "w", compression=zipfile.ZIP_DEFLATED)
for item in zin.infolist():
b = zin.read(item.filename)
if item.filename == "docProps/core.xml":
xml_str = b.decode("utf-8")
xml_str = re.sub(r"<dc:creator>.*?</dc:creator>", f"<dc:creator>{expected_author}</dc:creator>", xml_str)
xml_str = xml_str.replace("generated by python-docx", "series-studio")
b = xml_str.encode("utf-8")
zout.writestr(item, b)
zin.close()
zout.close()
with open(output_path, "wb") as f:
f.write(zbuf.getvalue())
# =========================================================================
# CONTRÔLES BLOQUANTS POST-GÉNÉRATION DOCX (T3)
# =========================================================================
with open(output_path, "rb") as f:
docx_bytes = f.read()
z = zipfile.ZipFile(io.BytesIO(docx_bytes))
doc_xml = z.read("word/document.xml").decode("utf-8")
# 1. Aucun $ ni LaTeX brut dans le XML
dollars = doc_xml.count("$")
if dollars > 0:
raise ValueError(f"Contrôle bloquant échoué : {dollars} signe(s) '$' résiduel(s) dans word/document.xml")
# 2. Formules OMML
omath_count = len(re.findall(r"<m:oMath[ >]", doc_xml))
if omath_count == 0:
raise ValueError("Contrôle bloquant échoué : aucune formule OMML <m:oMath> générée dans le DOCX")
# 3. Dessins
draw_count = len(re.findall(r"<w:drawing", doc_xml))
# 4. Auteur dans docProps/core.xml
core_xml = z.read("docProps/core.xml").decode("utf-8")
m_author = re.search(r"<dc:creator>([^<]*)</dc:creator>", core_xml)
author = m_author.group(1) if m_author else ""
if author.lower() == "python-docx" or author.lower() != expected_author.lower():
raise ValueError(f"Contrôle bloquant échoué : auteur DOCX = {author!r}, attendu {expected_author!r}")
brut = re.sub(r"<m:oMath[ >].*?</m:oMath>", " ", doc_xml, flags=re.S)
extracted_text = " ".join(re.findall(r"<w:t[^>]*>([^<]*)</w:t>", brut))
return {
"omath": omath_count,
"draw": draw_count,
"dollar": dollars,
"auteur": author,
"texte": extracted_text
}