fvtt-celestopol/extract_pretires.py

import fitz, re, json

pdf_path = '/home/morr/work/uberwald/fvtt-celestopol/__regles/Célestopol 1922 Fiches de prêts à jouer v1_cdjdr.pdf'
doc = fitz.open(pdf_path)

def is_green(color, tol=0.008):
    if not color: return False
    return (abs(color[0]-0.131) < tol and abs(color[1]-0.284) < tol and abs(color[2]-0.160) < tol)

SKILL_LIST = ['ARTIFICE','ATTRACTION','COERCITION','FAVEUR',
              'ÉCHAUFFOURÉE','EFFACEMENT','MOBILITÉ','PROUESSE',
              'APPRÉCIATION','ARTS','INSPIRATION','TRAQUE',
              'INSTRUCTION','MERV.TECH.','RAISONNEMENT','TRAITEMENT']

def norm(s):
    return s.strip().upper().replace('É','E').replace('È','E').replace('Ê','E').replace('Ô','O').replace('Â','A').replace('Î','I').replace('Œ','OE').replace('.','')

def get_skill_values(page):
    words = page.get_text("words")
    skill_y = {}
    for w in words:
        wn = norm(w[4])
        for sk in SKILL_LIST:
            if wn == norm(sk) and w[0] < 430:
                skill_y[sk] = (w[1]+w[3])/2
    # MERV.TECH. special
    for w in words:
        if 'MERV' in w[4].upper() and w[0] < 430:
            skill_y['MERV.TECH.'] = (w[1]+w[3])/2

    green_circles = []
    for d in page.get_drawings():
        if is_green(d.get('fill')):
            rect = d['rect']
            ws = rect.x1 - rect.x0
            if 5.5 < ws < 8.5:
                green_circles.append(((rect.x0+rect.x1)/2, (rect.y0+rect.y1)/2))

    skills = {}
    for sk in SKILL_LIST:
        sy = skill_y.get(sk, None)
        if sy is not None:
            skills[sk] = sum(1 for cx, cy in green_circles if abs(cy - sy) < 7)
        else:
            skills[sk] = 0
    return skills

def get_resistances(page):
    words = page.get_text("words")
    domain_y = {}
    for w in sorted(words, key=lambda x: x[1]):
        t = norm(w[4])
        x0 = w[0]
        if t == 'AME' and 300 < x0 < 500:
            domain_y['ame'] = w[1]
        elif t == 'CORPS' and 300 < x0 < 500:
            domain_y['corps'] = w[1]
        elif t == 'COEUR' and 300 < x0 < 500:
            domain_y['coeur'] = w[1]
        elif t == 'ESPRIT' and 300 < x0 < 500:
            domain_y['esprit'] = w[1]
    res = {}
    for dom, dy in domain_y.items():
        for w in words:
            if w[4].strip().isdigit() and w[0] > 480 and abs(w[1]-dy) < 35:
                res[dom] = int(w[4].strip())
                break
    return res

def get_anomalie_name(stats_page):
    # Extract from text: the anomalie name appears in the bottom-right of the page
    # Parse cleanly using raw text
    text = stats_page.get_text("text")
    lines = [l.strip() for l in text.split('\n') if l.strip()]
    
    skip_words = {'ANOMALIE','NV','RÉSISTANCE','RESISTANCE'}
    skip_starts = ['pour ','lors ','gagner ','trouver ','éviter ','sortir ','obtenir ',
                   'lors d', 'Vider ', 'Gain ', 'en pui', 'pour ne', 'pour ré']
    
    for i, line in enumerate(lines):
        if 'ANOMALIE' in line.upper() or 'NV' in line:
            # Look in next few lines for the name
            for j in range(i+1, min(i+10, len(lines))):
                l = lines[j]
                if not any(l.startswith(s) for s in skip_starts) and l not in skip_words:
                    if l and l[0].isupper() and len(l) > 1:
                        return l
    return "?"

def get_anomalie_niveau(stats_page):
    words = stats_page.get_text("words")
    for w in sorted(words, key=lambda x: (x[1],x[0])):
        if w[4].strip().isdigit() and w[0] > 480 and w[1] > 650:
            return int(w[4].strip())
    return None

def get_char_base_info(stats_page):
    blocks = stats_page.get_text("dict")["blocks"]
    name = None
    for block in blocks:
        for line in block.get("lines", []):
            for span in line.get("spans", []):
                if span.get("size", 0) > 11 and 'Bold' in span.get("font",""):
                    y = span["origin"][1]
                    t = span["text"].strip()
                    if t and len(t) > 3 and 150 < y < 250:
                        name = t
    return {'name': name}

def get_raw_text(page):
    html = page.get_text("html")
    clean = re.sub(r'<[^>]+>', ' ', html)
    clean = re.sub(r'&#x([0-9a-fA-F]+);', lambda m: chr(int(m.group(1),16)), clean)
    clean = re.sub(r'&#([0-9]+);', lambda m: chr(int(m.group(1))), clean)
    return re.sub(r'\s+', ' ', clean).strip()

def parse_aspects_page(asp_page):
    text = asp_page.get_text("text")
    lines = [l.strip() for l in text.split('\n') if l.strip()]
    return lines

# Characters: (name_idx, anom_desc_idx, stats_idx, aspects_idx)
CHARACTERS = [
    (0, 1, 2, 3),
    (4, 5, 6, 7),
    (8, 9, 10, 11),
    (12, 13, 14, 15),
    (16, 17, 18, 19),
    (20, 21, 22, 23),
    (24, 25, 26, 27),
    (28, 29, 30, 31),
]

for n_idx, a_idx, s_idx, asp_idx in CHARACTERS:
    sp = doc[s_idx]
    skills = get_skill_values(sp)
    res = get_resistances(sp)
    anom_name = get_anomalie_name(sp)
    anom_nv = get_anomalie_niveau(sp)
    char_info = get_char_base_info(sp)
    anom_desc = get_raw_text(doc[a_idx])
    asp_lines = parse_aspects_page(doc[asp_idx])

    print(f"\n{'='*70}")
    print(f"NAME: {char_info.get('name','?')}")
    print(f"SKILLS: {json.dumps(skills, ensure_ascii=False)}")
    print(f"RESISTANCES: {res}")
    print(f"ANOMALIE: {anom_name} nv{anom_nv}")
    print(f"ANOM DESC (first 300 chars): {anom_desc[:300]}")
    print("ASPECTS LINES:")
    for i,l in enumerate(asp_lines[:60]): print(f"  {i:2d}: {l}")

doc.close()
print("\nDONE")