# tests/test_pdf_stream.py — Regression tests for BUG-001 (PDF stream HTTP 500)
#
# BUG-001: opening a PDF whose filename contains accented characters
# (e.g. "Bière blonde envoyé par Desja.pdf") returned HTTP 500 because the raw
# Unicode name was injected into the Content-Disposition header, producing an
# invalid (non-ASCII) header value. /pdf/info worked because it sets no
# filename header.
#
# Fix: RFC 5987 header encoding (ASCII `filename` fallback + `filename*=UTF-8''…`).
import os
from pathlib import Path

# Minimal valid PDF so pypdf/stream treats the bytes as a real PDF.
MIN_PDF = (
    b"%PDF-1.4\n"
    b"1 0 obj<</Type/Catalog/Pages 2 0 R>>endobj\n"
    b"2 0 obj<</Type/Pages/Kids[3 0 R]/Count 1>>endobj\n"
    b"3 0 obj<</Type/Page/Parent 2 0 R/MediaBox[0 0 612 792]>>endobj\n"
    b"xref\n0 4\n0000000000 65535 f \n"
    b"trailer<</Size 4/Root 1 0 R>>\nstartxref\n0\n%%EOF\n"
)


def _create_pdf_with_unicode_name():
    vault = Path(os.environ["VAULT_1_PATH"])
    sub = vault / "98_Boite_Outils" / "98.2_Attachments"
    sub.mkdir(parents=True, exist_ok=True)
    pdf = sub / "Bière blonde envoyé par Desja.pdf"
    pdf.write_bytes(MIN_PDF)
    return "98_Boite_Outils/98.2_Attachments/Bière blonde envoyé par Desja.pdf"


class TestPdfStreamUnicodeFilename:
    def test_full_stream_ok(self, client):
        rel = _create_pdf_with_unicode_name()
        resp = client.get("/api/file/TestVault/pdf/stream", params={"path": rel})
        assert resp.status_code == 200, resp.text
        assert resp.headers["content-type"] == "application/pdf"
        cd = resp.headers.get("content-disposition", "")
        # Content-Disposition must be RFC 5987 encoded (ASCII-safe)
        assert "filename*=UTF-8''" in cd, cd
        # UTF-8 percent-encoded accented name present (é -> %C3%A8)
        assert "%C3%A8re%20blonde" in cd

    def test_range_request_ok(self, client):
        rel = _create_pdf_with_unicode_name()
        resp = client.get(
            "/api/file/TestVault/pdf/stream",
            params={"path": rel},
            headers={"Range": "bytes=0-99"},
        )
        assert resp.status_code == 206
        assert resp.headers.get("content-range", "").startswith("bytes 0-99/")

    def test_disposition_ascii_safe(self, client):
        rel = _create_pdf_with_unicode_name()
        resp = client.get("/api/file/TestVault/pdf/stream", params={"path": rel})
        cd = resp.headers.get("content-disposition", "")
        # Every header field value must be ASCII-encodable (no UnicodeEncodeError)
        cd.encode("ascii")
        # ASCII fallback filename present
        assert 'filename="' in cd