"""Extract checkable claims from the management repo's docs -> claims.tsv.

Phase 1 only *extracts*; nothing here judges whether a claim is true.
Verification happens in Phase 2.

A script cannot recognise a claim semantically, so extraction is trigger
based: a doc line is emitted when it matches at least one pattern that
marks it as asserting something checkable about the world (a component,
a count, a status, a version, a path, a date, the mirror topology, an
issue). Recall is favoured over precision -- a false positive costs a
Phase-2 glance, a false negative loses evidence.

Only management-repo docs are scanned: the mandate scopes claims to the
management repo's statements about its components.
"""

import re

from common import DOC_SUFFIXES, MGMT_REPO, cell, days_since, last_commit, \
    tracked_files, write_tsv

COMPONENT_WORDS = (
    r"ThreadNet[- ](?:Web|Call|Git|Operating|Server Suite)|threadnet-call|"
    r"thread-net-git|threadnet-operating|threadnet-web|axion1337\.chat-gitops|"
    r"game-operating|gameserver|management"
)

TRIGGERS = [
    ("component-ref", re.compile(COMPONENT_WORDS, re.I)),
    ("count", re.compile(
        r"\b(ein|zwei|drei|vier|fünf|sechs|sieben|acht|neun|zehn|\d+)\s+"
        r"(Repos?|Produkt-Repos?|Projekte?|Issues?|Commits?|Pipelines?|Themes?|"
        r"Hosts?|Mirrors?)\b", re.I)),
    ("status", re.compile(
        r"\b(erledigt|offen|live|aktiv|geschlossen|leer|umgezogen|entfallen|"
        r"abgelöst|veraltet|überholt|scharf|grün|rot|tabu|kanonisch|"
        r"zurückgestellt|verifiziert|bereinigt)\b", re.I)),
    ("version", re.compile(r"\bv?\d+\.\d+\.\d+\b|\b\d+\.\d+\.\d+-[\w.]+\b")),
    ("path-claim", re.compile(r"`[^`]*(?:/[^`]*|\.(?:md|ya?ml|json|py|ts|toml|crt))`")),
    ("date-claim", re.compile(r"\b(seit|Stand|bis|am|ab)\s+\d{4}-\d{2}-\d{2}\b", re.I)),
    ("mirror-topology", re.compile(
        r"\b(Mirror|Spiegel|gespiegelt|Gitea|rohana|Push-Mirror|kanonisch|Flux-Source)\b",
        re.I)),
    ("issue-ref", re.compile(r"(?:#\d+\b|/-/issues/\d+|\b[A-Z]{3,8}-\d{2}\b)")),
]

FENCE_RE = re.compile(r"^\s*(```|~~~)")


def main():
    rows = []
    docs = [f for f in tracked_files("management", MGMT_REPO)
            if any(f.endswith(s) for s in DOC_SUFFIXES)]
    for src in sorted(docs):
        iso, _, _ = last_commit("management", MGMT_REPO, src)
        text = (MGMT_REPO / src).read_text(encoding="utf-8")
        in_fence = False
        for lineno, line in enumerate(text.splitlines(), start=1):
            if FENCE_RE.match(line):
                in_fence = not in_fence
                continue
            stripped = line.strip()
            if len(stripped) < 12:
                continue
            hits = [name for name, rx in TRIGGERS if rx.search(stripped)]
            if not hits:
                continue
            rows.append([
                src, lineno, ";".join(hits), "yes" if in_fence else "no",
                iso, days_since(iso), cell(stripped)[:400],
            ])

    rows.sort(key=lambda r: (r[0], r[1]))
    write_tsv(
        "claims.tsv",
        ["path", "line", "triggers", "in_code_block", "doc_last_commit_date",
         "doc_days_since_change", "claim_text"],
        rows,
    )


if __name__ == "__main__":
    main()