#!/usr/bin/env python3
"""PPPA MI stratēģijas lapa (pppa.lv/mi-strategija) -> mašīnlasāms XML.

Lietojums:
    python3 tools/mi_strategija_no_html.py source/mi-strategija-v2.2.html data/

Teksts tiek pārņemts burtiski; HTML formatējums (treknraksts, saites) tiek
noņemts, saišu adreses saglabātas atribūtā @href.
"""
import html
import re
import sys
from pathlib import Path
from xml.sax.saxutils import escape, quoteattr

NS = "urn:pppa:cac:mi-strategija:0.1"
ROMAN = {"I": 1, "II": 2, "III": 3, "IV": 4, "V": 5}


def text(fragment):
    """HTML fragments -> tīrs teksts vienā rindā."""
    s = re.sub(r"<svg.*?</svg>", "", fragment, flags=re.S)
    s = re.sub(r"<br\s*/?>", " ", s)
    s = re.sub(r"<[^>]+>", "", s)
    s = html.unescape(s)
    return re.sub(r"\s+", " ", s).strip()


def one(pattern, s, default=None):
    m = re.search(pattern, s, re.S)
    if not m:
        if default is not None:
            return default
        raise ValueError("Nav atrasts: " + pattern)
    return m.group(1)


def absurl(u):
    return "https://pppa.lv" + u if u.startswith("/") else u


def date_iso(lv):
    m = re.fullmatch(r"(\d{2})\.(\d{2})\.(\d{4})\.?", lv.strip())
    return f"{m.group(3)}-{m.group(2)}-{m.group(1)}" if m else None


def el(name, txt, indent, **attrs):
    a = "".join(f" {k}={quoteattr(str(v))}" for k, v in attrs.items() if v not in (None, ""))
    return f"{' ' * indent}<{name}{a}>{escape(txt)}</{name}>"


def main(src, out_dir):
    h = Path(src).read_text(encoding="utf-8")
    h = re.sub(r"<script.*?</script>|<style.*?</style>", "", h, flags=re.S)

    # --- Galvene ---
    hero = one(r'<section class="strat-hero">(.*?)</section>', h)
    title = text(one(r"<h1>(.*?)</h1>", hero))
    status = text(one(r'<p class="eyebrow">(.*?)</p>', hero))
    slogan = [text(x) for x in re.split(r'<span class="hs-sep">.*?</span>',
                                        one(r'<p class="hero-slogan">(.*?)</p>', hero))]
    vision = text(one(r'<p class="lead">(.*?)</p>', hero))
    summary = one(r'<div class="summary-box" id="kopsavilkums">(.*?)</div>', h)
    upd = one(r"Pēdējoreiz atjaunots:\s*([\d.]+)", summary).rstrip(".")
    version = one(r"Darba versija\s*(v[\d.]+)", summary)
    intake = one(r"Priekšlikumu saņemšana uzsākta <strong[^>]*>([\d.]+)</strong>", summary).rstrip(".")
    doc_id = f"pppa-mis-{date_iso(upd)}"

    # --- Pīlāru kartītes (kopsavilkums) ---
    cards = {}
    for c in re.finditer(r'<button type="button" class="pillar-card[^"]*" data-tab="(\d)">(.*?)</button>', h, re.S):
        cards[int(c.group(1))] = {
            "name": text(one(r'<span class="pc-jumbo">(.*?)</span>', c.group(2))),
            "short": text(one(r'<span class="pc-sub">(.*?)</span>', c.group(2))),
            "summary": text(one(r'<span class="pc-text">(.*?)</span>', c.group(2))),
        }

    # --- Pīlāri, mērķi, rīcības ---
    panels = one(r'<div class="pillar-panels">(.*?)</div><!-- /pillar-panels -->', h)
    pillars = re.split(r'<section class="pillar pillar-panel[^"]*"', panels)[1:]
    xml_p, refs_all, action_ids = [], [], set()
    for p in pillars:
        rn = one(r'<span class="pillar-rn">(\w+)</span>', p)
        pn = ROMAN[rn]
        pid = f"{doc_id}.p{pn}"
        head = one(r'<div class="pillar-head">(.*?)</div>', p)
        out = [f'  <Pillar id="{pid}" number="{rn}">',
               el("Name", cards[pn]["name"], 4),
               el("ShortTitle", cards[pn]["short"], 4),
               el("Heading", text(one(r"<h2>(.*?)</h2>", head)), 4),
               el("Summary", cards[pn]["summary"], 4),
               el("Description", text(one(r"</h2>\s*<p>(.*?)</p>", head)), 4)]
        for g in re.split(r'<div class="goal-block">', p)[1:]:
            gn = int(one(r'<span class="goal-num">(\d+)\.</span>', g))
            gid = f"{doc_id}.m{gn}"
            out += [f'    <Goal id="{gid}" number="{gn}.">',
                    el("Title", text(one(r"<h3>(.*?)</h3>", g)), 6),
                    el("Description", text(one(r"</h3>\s*<p>(.*?)</p>", g)), 6)]
            for a in re.finditer(r'<div class="action-card" id="pasakums-(\d+)">(.*?)\n\s*</div>\s*(?=<div class="action-card"|</div>\s*</div>)', g, re.S):
                an, body = int(a.group(1)), a.group(2)
                aid = f"{doc_id}.r{an:02d}"
                action_ids.add(an)
                printed = text(one(r'<span class="ac-num">(.*?)</span>', body))
                assert printed == f"Rīcība {an}", (printed, an)
                dl = text(one(r'<span class="deadline-badge">(.*?)</span>', re.sub(r"<svg.*?</svg>", "", body, flags=re.S)))
                out.append(f'      <Action id="{aid}" number="{an}">')
                for t in re.finditer(r'<span class="ac-tag( rec)?">(.*?)</span>', body):
                    out.append(el("Tag", text(t.group(2)), 8,
                                  kind="parvaldibas-rekomendacija" if t.group(1) else None))
                out += [el("Title", text(one(r"<h4>(.*?)</h4>", body)), 8),
                        el("Text", text(one(r'<p class="ac-body">(.*?)</p>', body)), 8),
                        el("Measurement", text(re.sub(r"<strong>.*?</strong>", "",
                           one(r'<div class="ac-measure">(.*?)</div>', body), count=1)), 8),
                        el("Deadline", dl, 8, value=date_iso(dl))]
                for r in re.findall(r'<p class="ac-ref">(.*?)</p>', body, re.S):
                    t = text(r)
                    kind = {"Saistīts ar:": "saistits", "Avots:": "avots", "Piemērs:": "piemers"}
                    k = next((v for p_, v in kind.items() if t.startswith(p_)), None)
                    if k:
                        t = t.split(":", 1)[1].strip()
                    hrefs = " ".join(absurl(u) for u in re.findall(r'href="([^"]+)"', r))
                    nums = re.findall(r"Rīcība (\d+)", t)
                    refs_all.append((an, nums))
                    acts = " ".join(f"{doc_id}.r{int(n):02d}" for n in nums)
                    out.append(el("Reference", t, 8, kind=k, actions=acts, href=hrefs))
                out.append("      </Action>")
            out.append("    </Goal>")
        out.append("  </Pillar>")
        xml_p += out

    # --- Principi ---
    pr_sec = one(r'id="principi">(.*?)<hr class="section-divider">', h)
    pr_intro = text(one(r'<p class="section-intro">(.*?)</p>', pr_sec))
    xml_pr = ['  <Principles>', el("Heading", text(one(r"^(.*?)</h2>", pr_sec)), 4),
              el("Intro", pr_intro, 4)]
    for i, b in enumerate(re.findall(r'<div class="mg-box">(.*?)</div>', pr_sec, re.S), 1):
        xml_pr += [f'    <Principle id="{doc_id}.pr{i}">',
                   el("Label", text(one(r'<p class="mg-label">(.*?)</p>', b)), 6),
                   el("Text", text(one(r'</p>\s*<p>(.*?)</p>', b)), 6),
                   "    </Principle>"]
    xml_pr.append("  </Principles>")

    # --- Saņemtie priekšlikumi ---
    pk = one(r'id="priekslikumi">.*?<p class="section-intro">(.*?)</p>', h)
    pk_nums = sorted({int(n) for n in re.findall(r"\((?:Rīcība )?(\d+)\)", text(pk))})
    xml_pk = [el("ProposalsNote", text(pk), 2,
                 actions=" ".join(f"{doc_id}.r{n:02d}" for n in pk_nums))]

    # --- Starptautiskā pieredze ---
    xml_rs = ["  <Resources>"]
    for i, c in enumerate(re.findall(r'<div class="resource-card">(.*?)</div>', h, re.S), 1):
        xml_rs += [f'    <Resource id="{doc_id}.a{i}" href={quoteattr(one(r'href="([^"]+)"', c))}>',
                   el("Country", text(one(r'<p class="rc-country">(.*?)</p>', c)), 6),
                   el("Title", text(one(r"<h3>(.*?)</h3>", c)), 6),
                   el("Text", text(one(r"</h3>\s*<p>(.*?)</p>", c)), 6),
                   "    </Resource>"]
    xml_rs.append("  </Resources>")

    # --- Pārbaudes ---
    expect = set(range(1, max(action_ids) + 1))
    missing = expect - action_ids
    bad = [(a, n) for a, ns in refs_all for n in ns if int(n) not in action_ids]
    print(f"{doc_id}: {len(pillars)} pīlāri, {len(action_ids)} rīcības; trūkst {sorted(missing) or '—'}; "
          f"nederīgas atsauces {bad or '—'}")

    head = [
        '<?xml version="1.0" encoding="UTF-8"?>',
        f'<AIStrategy xmlns="{NS}" id="{doc_id}" version="{version}">',
        el("Title", title, 2),
        el("Author", "PPP Asociācija (PPPA)", 2),
        el("Status", status, 2),
        el("Updated", upd + ".", 2, value=date_iso(upd)),
        el("ProposalsOpened", intake + ".", 2, value=date_iso(intake)),
        el("SourceUrl", "https://pppa.lv/mi-strategija", 2),
        "  <Slogan>" + "".join(el("Word", w, 0) for w in slogan) + "</Slogan>",
        el("Vision", vision, 2),
    ]
    xml = "\n".join(head + xml_p + xml_pr + xml_pk + xml_rs + ["</AIStrategy>", ""])
    out = Path(out_dir) / f"{doc_id}.xml"
    out.parent.mkdir(parents=True, exist_ok=True)
    out.write_text(xml, encoding="utf-8")
    print("->", out)


if __name__ == "__main__":
    main(sys.argv[1], sys.argv[2] if len(sys.argv) > 2 else "data")