Implementace prevodu HTML na PDF
Sluzba prijme adresu HTML dokumentu nebo HTML v tele requestu a vrati PDF. Navrzena pro dokumenty o stovkach az tisicich stranek. Rendering: - WeasyPrint jako vychozi engine, spravne CSS Paged Media, nizka pametova narocnost, bez JavaScriptu - Chromium pres Playwright pro dokumenty dokreslovane skripty - rezim auto s detekci skriptu a fallbackem pri selhani WeasyPrintu Velke dokumenty: - deleni na casti na strukturalnich hranicich, rez nikdy uvnitr tabulky nebo odstavce - dvoupruchodovy render obsahu se skutecnymi cisly stranek, pozice nadpisu se ctou z kotev hlasenych u kazde stranky - cislovani stranek bud pres CSS countery, nebo pres cislovaci vrstvu nastampovanou na hotove PDF, rozmer stranky se cte z vysledneho souboru - Chromium se restartuje po N jobech, nikdy vsak behem beziciho renderu API: - POST /convert synchronne, POST /jobs asynchronne se sledovanim stavu, stahovanim vysledku, rusenim a volitelnym callbackem - GET /health s overenim dostupnosti obou enginu a stavem fronty - OpenAPI respektuje prefix reverse proxy pres root_path Bezpecnost a provoz: - SSRF kontrola po DNS resolvu, na kazdem presmerovani a u vsech pozadavku prohlizece - nedostupne assety render nezastavi, ale hlasi se v odpovedi i v logu - fronta s omezenym poctem workeru, rozpracovane joby se pri ukonceni oznaci jako failed, nezmizi potichu - strukturovane JSON logovani s job_id - vsechny limity vypnute ve vychozim stavu Dockerfile je dvoufazovy, obsahuje zavislosti WeasyPrintu, Chromium a fonty s ceskou diakritikou. Autentizace zamerne neni implementovana, zpusob predavani neni domluveny. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
This commit is contained in:
JiriUhlir
co-authored by
Claude Opus 5
Claude Opus 5
parent
e3cc8f418b
commit
156289fe2d
@@ -0,0 +1,46 @@
|
||||
"""Shared fixtures.
|
||||
|
||||
Nothing here starts a real browser. Engine level tests skip themselves when the
|
||||
engine is not installed in the current environment.
|
||||
"""
|
||||
|
||||
from __future__ import annotations
|
||||
|
||||
import pytest
|
||||
|
||||
|
||||
def build_document(sections: int, paragraphs_per_section: int = 12) -> str:
|
||||
"""Synthetic document with predictable structure and size."""
|
||||
parts = [
|
||||
"<!DOCTYPE html><html><head><meta charset='utf-8'>",
|
||||
"<style>body{font-family:sans-serif;font-size:11pt}</style>",
|
||||
"</head><body>",
|
||||
]
|
||||
for index in range(sections):
|
||||
parts.append(f"<section id='sekce-{index}'><h1>Kapitola {index + 1}</h1>")
|
||||
for paragraph in range(paragraphs_per_section):
|
||||
parts.append(
|
||||
f"<p>Odstavec {paragraph + 1} kapitoly {index + 1}. "
|
||||
+ ("Text s ceskou diakritikou pro overeni fontu. " * 8)
|
||||
+ "</p>"
|
||||
)
|
||||
parts.append("</section>")
|
||||
parts.append("</body></html>")
|
||||
return "".join(parts)
|
||||
|
||||
|
||||
@pytest.fixture
|
||||
def small_document() -> str:
|
||||
return build_document(sections=3, paragraphs_per_section=2)
|
||||
|
||||
|
||||
@pytest.fixture
|
||||
def large_document() -> str:
|
||||
"""Roughly twelve hundred pages, used for memory and timing measurements."""
|
||||
return build_document(sections=400, paragraphs_per_section=14)
|
||||
|
||||
|
||||
@pytest.fixture
|
||||
def unsplittable_document() -> str:
|
||||
body = "".join(f"<p>Odstavec {index}. " + ("Text. " * 40) + "</p>" for index in range(200))
|
||||
return f"<!DOCTYPE html><html><head><meta charset='utf-8'></head><body>{body}</body></html>"
|
||||
@@ -0,0 +1,69 @@
|
||||
"""HTTP surface of the service."""
|
||||
|
||||
from __future__ import annotations
|
||||
|
||||
import pytest
|
||||
from fastapi.testclient import TestClient
|
||||
|
||||
from app.main import app
|
||||
|
||||
|
||||
@pytest.fixture
|
||||
def client():
|
||||
with TestClient(app) as test_client:
|
||||
yield test_client
|
||||
|
||||
|
||||
def test_health_is_available(client) -> None:
|
||||
response = client.get("/health")
|
||||
|
||||
assert response.status_code == 200
|
||||
body = response.json()
|
||||
assert body["status"] in ("ok", "degraded")
|
||||
assert "engines" in body
|
||||
|
||||
|
||||
def test_version_reports_the_application(client) -> None:
|
||||
response = client.get("/version")
|
||||
|
||||
assert response.status_code == 200
|
||||
assert response.json()["language"] == "python"
|
||||
|
||||
|
||||
def test_openapi_is_generated(client) -> None:
|
||||
response = client.get("/openapi.json")
|
||||
|
||||
assert response.status_code == 200
|
||||
paths = response.json()["paths"]
|
||||
assert "/convert" in paths
|
||||
assert "/jobs" in paths
|
||||
assert "/jobs/{job_id}/result" in paths
|
||||
|
||||
|
||||
def test_source_requires_exactly_one_input(client) -> None:
|
||||
response = client.post(
|
||||
"/convert",
|
||||
json={"source": {"url": "https://example.com/a.html", "html": "<html></html>"}},
|
||||
)
|
||||
|
||||
assert response.status_code == 422
|
||||
|
||||
|
||||
def test_empty_source_is_rejected(client) -> None:
|
||||
response = client.post("/convert", json={"source": {}})
|
||||
|
||||
assert response.status_code == 422
|
||||
|
||||
|
||||
def test_private_address_is_refused(client) -> None:
|
||||
response = client.post("/convert", json={"source": {"url": "http://127.0.0.1:8000/a.html"}})
|
||||
|
||||
assert response.status_code == 400
|
||||
assert response.json()["error_code"] == "blocked_target"
|
||||
|
||||
|
||||
def test_unknown_job_returns_404(client) -> None:
|
||||
response = client.get("/jobs/00000000-0000-0000-0000-000000000000")
|
||||
|
||||
assert response.status_code == 404
|
||||
assert response.json()["error_code"] == "job_not_found"
|
||||
@@ -0,0 +1,42 @@
|
||||
"""Behaviour when an asset cannot be loaded.
|
||||
|
||||
A missing image must not abort the render, but it must never disappear
|
||||
silently either.
|
||||
"""
|
||||
|
||||
from __future__ import annotations
|
||||
|
||||
import pytest
|
||||
|
||||
pytest.importorskip("weasyprint")
|
||||
|
||||
from app.config import Settings # noqa: E402
|
||||
from app.engines.weasy import WeasyPrintEngine # noqa: E402
|
||||
from app.models import ConvertRequest, Source # noqa: E402
|
||||
from app.services.pipeline import ConversionPipeline # noqa: E402
|
||||
|
||||
pytestmark = pytest.mark.asyncio
|
||||
|
||||
HTML_WITH_BLOCKED_IMAGE = """
|
||||
<!DOCTYPE html><html><head><meta charset="utf-8"></head>
|
||||
<body>
|
||||
<h1>Dokument s nedostupnym obrazkem</h1>
|
||||
<p>Text pokracuje i kdyz se obrazek nenacte.</p>
|
||||
<img src="http://127.0.0.1:9/neexistuje.png" alt="chybi">
|
||||
</body></html>
|
||||
"""
|
||||
|
||||
|
||||
async def test_blocked_asset_is_reported_and_render_continues(tmp_path) -> None:
|
||||
pipeline = ConversionPipeline({"weasyprint": WeasyPrintEngine()}, Settings())
|
||||
request = ConvertRequest(
|
||||
source=Source(html=HTML_WITH_BLOCKED_IMAGE),
|
||||
engine="weasyprint",
|
||||
)
|
||||
|
||||
result = await pipeline.run(request, tmp_path)
|
||||
|
||||
assert result.page_count >= 1
|
||||
assert result.path.exists()
|
||||
assert result.missing_assets, "nedostupny asset musi byt hlaseny v odpovedi"
|
||||
assert "127.0.0.1" in result.missing_assets[0].url
|
||||
@@ -0,0 +1,56 @@
|
||||
"""Splitting of the document into chunks."""
|
||||
|
||||
from __future__ import annotations
|
||||
|
||||
from app.pdf.chunker import split_document
|
||||
from app.pdf.document import SourceDocument
|
||||
from tests.conftest import build_document
|
||||
|
||||
|
||||
def test_document_with_sections_is_split() -> None:
|
||||
document = SourceDocument(build_document(sections=40, paragraphs_per_section=10))
|
||||
chunks = split_document(document, pages_per_chunk=5)
|
||||
|
||||
assert len(chunks) > 1
|
||||
for chunk in chunks:
|
||||
assert chunk.startswith("<!DOCTYPE html>")
|
||||
assert "<body" in chunk
|
||||
assert "</html>" in chunk
|
||||
|
||||
|
||||
def test_every_section_appears_exactly_once() -> None:
|
||||
document = SourceDocument(build_document(sections=30, paragraphs_per_section=8))
|
||||
chunks = split_document(document, pages_per_chunk=3)
|
||||
|
||||
joined = "".join(chunks)
|
||||
for index in range(30):
|
||||
assert joined.count(f"id=\"sekce-{index}\"") == 1
|
||||
|
||||
|
||||
def test_head_is_repeated_in_every_chunk() -> None:
|
||||
document = SourceDocument(build_document(sections=20, paragraphs_per_section=10))
|
||||
chunks = split_document(document, pages_per_chunk=2)
|
||||
|
||||
assert len(chunks) > 1
|
||||
for chunk in chunks:
|
||||
assert "font-family:sans-serif" in chunk
|
||||
|
||||
|
||||
def test_document_without_split_points_stays_whole(unsplittable_document: str) -> None:
|
||||
document = SourceDocument(unsplittable_document)
|
||||
chunks = split_document(document, pages_per_chunk=1)
|
||||
|
||||
assert len(chunks) == 1
|
||||
|
||||
|
||||
def test_single_wrapper_is_traversed() -> None:
|
||||
sections = "".join(f"<section><h1>Kapitola {i}</h1><p>Text</p></section>" for i in range(10))
|
||||
html = f"<!DOCTYPE html><html><head></head><body><main class='obal'>{sections}</main></body></html>"
|
||||
|
||||
document = SourceDocument(html)
|
||||
assert document.container.tag == "main"
|
||||
|
||||
chunks = split_document(document, pages_per_chunk=1)
|
||||
assert len(chunks) > 1
|
||||
for chunk in chunks:
|
||||
assert "class=\"obal\"" in chunk
|
||||
@@ -0,0 +1,133 @@
|
||||
"""Job queue behaviour."""
|
||||
|
||||
from __future__ import annotations
|
||||
|
||||
import asyncio
|
||||
from pathlib import Path
|
||||
|
||||
import pytest
|
||||
|
||||
from app.config import Settings
|
||||
from app.errors import LimitExceededError, QueueFullError
|
||||
from app.models import ConvertRequest, Source
|
||||
from app.services.jobs import JobManager
|
||||
from app.services.pipeline import ConversionResult
|
||||
from app.services.storage import Storage
|
||||
|
||||
pytestmark = pytest.mark.asyncio
|
||||
|
||||
|
||||
class StubPipeline:
|
||||
def __init__(self, behaviour="ok", delay=0.0) -> None:
|
||||
self.behaviour = behaviour
|
||||
self.delay = delay
|
||||
|
||||
async def run(self, request, workdir: Path, progress=None) -> ConversionResult:
|
||||
if progress is not None:
|
||||
progress(1, 1, 1, 1)
|
||||
if self.delay:
|
||||
await asyncio.sleep(self.delay)
|
||||
if self.behaviour == "conversion_error":
|
||||
raise LimitExceededError("Prekrocen limit stranek.", {"limit": 10})
|
||||
if self.behaviour == "crash":
|
||||
raise RuntimeError("necekana chyba enginu")
|
||||
|
||||
workdir.mkdir(parents=True, exist_ok=True)
|
||||
produced = workdir / "out.pdf"
|
||||
produced.write_bytes(b"%PDF-1.7\n%fake\n")
|
||||
return ConversionResult(path=produced, page_count=1, engine_used="stub")
|
||||
|
||||
|
||||
def make_manager(tmp_path, behaviour="ok", delay=0.0, **overrides) -> JobManager:
|
||||
settings = Settings()
|
||||
for key, value in overrides.items():
|
||||
object.__setattr__(settings, key, value)
|
||||
object.__setattr__(settings, "storage_dir", str(tmp_path))
|
||||
|
||||
storage = Storage(str(tmp_path))
|
||||
return JobManager(StubPipeline(behaviour, delay), storage, settings)
|
||||
|
||||
|
||||
def simple_request() -> ConvertRequest:
|
||||
return ConvertRequest(source=Source(html="<html><body><p>ahoj</p></body></html>"))
|
||||
|
||||
|
||||
async def test_successful_job_publishes_a_result(tmp_path) -> None:
|
||||
manager = make_manager(tmp_path)
|
||||
await manager.start()
|
||||
try:
|
||||
job = manager.submit(simple_request())
|
||||
await asyncio.wait_for(job.done.wait(), timeout=5)
|
||||
|
||||
assert job.state.status == "done"
|
||||
assert job.state.page_count == 1
|
||||
assert (Path(tmp_path) / job.id / "result.pdf").exists()
|
||||
finally:
|
||||
await manager.stop()
|
||||
|
||||
|
||||
async def test_conversion_error_keeps_its_code(tmp_path) -> None:
|
||||
manager = make_manager(tmp_path, behaviour="conversion_error")
|
||||
await manager.start()
|
||||
try:
|
||||
job = manager.submit(simple_request())
|
||||
await asyncio.wait_for(job.done.wait(), timeout=5)
|
||||
|
||||
assert job.state.status == "failed"
|
||||
assert job.state.error is not None
|
||||
assert job.state.error.error_code == "limit_exceeded"
|
||||
finally:
|
||||
await manager.stop()
|
||||
|
||||
|
||||
async def test_unexpected_error_is_reported_not_swallowed(tmp_path) -> None:
|
||||
manager = make_manager(tmp_path, behaviour="crash")
|
||||
await manager.start()
|
||||
try:
|
||||
job = manager.submit(simple_request())
|
||||
await asyncio.wait_for(job.done.wait(), timeout=5)
|
||||
|
||||
assert job.state.status == "failed"
|
||||
assert job.state.error.error_code == "internal_error"
|
||||
finally:
|
||||
await manager.stop()
|
||||
|
||||
|
||||
async def test_running_job_can_be_cancelled(tmp_path) -> None:
|
||||
manager = make_manager(tmp_path, delay=5.0)
|
||||
await manager.start()
|
||||
try:
|
||||
job = manager.submit(simple_request())
|
||||
await asyncio.sleep(0.2)
|
||||
manager.cancel(job.id)
|
||||
await asyncio.wait_for(job.done.wait(), timeout=5)
|
||||
|
||||
assert job.state.status == "cancelled"
|
||||
assert not (Path(tmp_path) / job.id).exists()
|
||||
finally:
|
||||
await manager.stop()
|
||||
|
||||
|
||||
async def test_full_queue_is_rejected(tmp_path) -> None:
|
||||
manager = make_manager(tmp_path, delay=5.0, queue_max_size=1, workers=1)
|
||||
await manager.start()
|
||||
try:
|
||||
manager.submit(simple_request())
|
||||
await asyncio.sleep(0.1)
|
||||
manager.submit(simple_request())
|
||||
|
||||
with pytest.raises(QueueFullError):
|
||||
manager.submit(simple_request())
|
||||
finally:
|
||||
await manager.stop()
|
||||
|
||||
|
||||
async def test_shutdown_marks_pending_jobs_as_failed(tmp_path) -> None:
|
||||
manager = make_manager(tmp_path, delay=5.0, workers=1)
|
||||
await manager.start()
|
||||
job = manager.submit(simple_request())
|
||||
await asyncio.sleep(0.1)
|
||||
await manager.stop()
|
||||
|
||||
assert job.state.status == "failed"
|
||||
assert job.state.error.error_code == "service_restarted"
|
||||
@@ -0,0 +1,39 @@
|
||||
"""Measurement fixture for a document of roughly twelve hundred pages.
|
||||
|
||||
Marked slow on purpose. It is here to measure memory and wall clock time of the
|
||||
chunked pipeline, not to assert exact numbers.
|
||||
"""
|
||||
|
||||
from __future__ import annotations
|
||||
|
||||
import time
|
||||
|
||||
import pytest
|
||||
|
||||
pytest.importorskip("weasyprint")
|
||||
|
||||
from app.config import Settings # noqa: E402
|
||||
from app.engines.weasy import WeasyPrintEngine # noqa: E402
|
||||
from app.models import ChunkSettings, ConvertRequest, PageNumbers, Source # noqa: E402
|
||||
from app.services.pipeline import ConversionPipeline # noqa: E402
|
||||
|
||||
pytestmark = [pytest.mark.asyncio, pytest.mark.slow]
|
||||
|
||||
|
||||
async def test_large_document_renders_in_chunks(tmp_path, large_document: str) -> None:
|
||||
pipeline = ConversionPipeline({"weasyprint": WeasyPrintEngine()}, Settings())
|
||||
request = ConvertRequest(
|
||||
source=Source(html=large_document),
|
||||
engine="weasyprint",
|
||||
page_numbers=PageNumbers(enabled=True),
|
||||
chunking=ChunkSettings(enabled=True, pages_per_chunk=50),
|
||||
)
|
||||
|
||||
started = time.monotonic()
|
||||
result = await pipeline.run(request, tmp_path)
|
||||
duration = time.monotonic() - started
|
||||
|
||||
print(f"stranek: {result.page_count}, cas: {duration:.1f} s")
|
||||
|
||||
assert result.page_count > 1000
|
||||
assert result.path.exists()
|
||||
@@ -0,0 +1,90 @@
|
||||
"""Page numbering and table of contents of a merged document.
|
||||
|
||||
This is where the chunking approach usually breaks, so the numbers are read back
|
||||
out of the produced PDF instead of being trusted.
|
||||
"""
|
||||
|
||||
from __future__ import annotations
|
||||
|
||||
import re
|
||||
|
||||
import pytest
|
||||
|
||||
pytest.importorskip("weasyprint")
|
||||
pytest.importorskip("pypdf")
|
||||
|
||||
from app.config import Settings # noqa: E402
|
||||
from app.engines.weasy import WeasyPrintEngine # noqa: E402
|
||||
from app.models import ChunkSettings, ConvertRequest, PageNumbers, Source, TocSettings # noqa: E402
|
||||
from app.services.pipeline import ConversionPipeline # noqa: E402
|
||||
from tests.conftest import build_document # noqa: E402
|
||||
|
||||
pytestmark = pytest.mark.asyncio
|
||||
|
||||
|
||||
def page_texts(path) -> list[str]:
|
||||
from pypdf import PdfReader
|
||||
|
||||
with open(path, "rb") as handle:
|
||||
return [page.extract_text() or "" for page in PdfReader(handle).pages]
|
||||
|
||||
|
||||
def make_pipeline() -> ConversionPipeline:
|
||||
return ConversionPipeline({"weasyprint": WeasyPrintEngine()}, Settings())
|
||||
|
||||
|
||||
async def test_page_numbers_are_continuous_across_chunks(tmp_path) -> None:
|
||||
request = ConvertRequest(
|
||||
source=Source(html=build_document(sections=12, paragraphs_per_section=8)),
|
||||
engine="weasyprint",
|
||||
page_numbers=PageNumbers(enabled=True, format="{page} / {pages}"),
|
||||
chunking=ChunkSettings(enabled=True, pages_per_chunk=2),
|
||||
)
|
||||
|
||||
result = await make_pipeline().run(request, tmp_path)
|
||||
texts = page_texts(result.path)
|
||||
|
||||
assert result.page_count == len(texts)
|
||||
assert result.page_count > 3, "dokument musi mit vic stranek, jinak test nic neoveruje"
|
||||
|
||||
for index, text in enumerate(texts, start=1):
|
||||
assert f"{index} / {result.page_count}" in text.replace("\n", " ")
|
||||
|
||||
|
||||
async def test_table_of_contents_points_to_the_real_pages(tmp_path) -> None:
|
||||
request = ConvertRequest(
|
||||
source=Source(html=build_document(sections=10, paragraphs_per_section=8)),
|
||||
engine="weasyprint",
|
||||
toc=TocSettings(enabled=True, depth=1, title="Obsah"),
|
||||
chunking=ChunkSettings(enabled=True, pages_per_chunk=2),
|
||||
)
|
||||
|
||||
result = await make_pipeline().run(request, tmp_path)
|
||||
texts = page_texts(result.path)
|
||||
toc_text = " ".join(texts[:2]).replace("\n", " ")
|
||||
|
||||
for index in range(10):
|
||||
heading = f"Kapitola {index + 1}"
|
||||
match = re.search(re.escape(heading) + r"\s+(\d+)", toc_text)
|
||||
assert match, f"v obsahu chybi polozka {heading}"
|
||||
|
||||
declared_page = int(match.group(1))
|
||||
actual_pages = [
|
||||
number for number, text in enumerate(texts, start=1) if heading in text.replace("\n", " ")
|
||||
]
|
||||
# The first occurrence after the table of contents is the heading itself.
|
||||
assert declared_page in actual_pages, f"{heading} deklaruje stranku {declared_page}"
|
||||
|
||||
|
||||
async def test_unchunked_document_uses_css_counters(tmp_path) -> None:
|
||||
request = ConvertRequest(
|
||||
source=Source(html=build_document(sections=4, paragraphs_per_section=6)),
|
||||
engine="weasyprint",
|
||||
page_numbers=PageNumbers(enabled=True, format="{page} / {pages}"),
|
||||
chunking=ChunkSettings(enabled=False),
|
||||
)
|
||||
|
||||
result = await make_pipeline().run(request, tmp_path)
|
||||
texts = page_texts(result.path)
|
||||
|
||||
assert f"1 / {result.page_count}" in texts[0].replace("\n", " ")
|
||||
@@ -0,0 +1,65 @@
|
||||
"""SSRF protection.
|
||||
|
||||
The important case is a public hostname that resolves to a loopback address.
|
||||
Checking the URL string alone would let it through.
|
||||
"""
|
||||
|
||||
from __future__ import annotations
|
||||
|
||||
import socket
|
||||
|
||||
import pytest
|
||||
|
||||
from app.config import Settings
|
||||
from app.errors import BlockedTargetError
|
||||
from app.services.security import UrlGuard
|
||||
|
||||
|
||||
@pytest.fixture
|
||||
def guard() -> UrlGuard:
|
||||
return UrlGuard(Settings())
|
||||
|
||||
|
||||
def test_literal_loopback_is_blocked(guard: UrlGuard) -> None:
|
||||
with pytest.raises(BlockedTargetError):
|
||||
guard.check("http://127.0.0.1:8000/dokument.html")
|
||||
|
||||
|
||||
def test_private_range_is_blocked(guard: UrlGuard) -> None:
|
||||
with pytest.raises(BlockedTargetError):
|
||||
guard.check("http://192.168.1.10/dokument.html")
|
||||
|
||||
|
||||
def test_link_local_metadata_endpoint_is_blocked(guard: UrlGuard) -> None:
|
||||
with pytest.raises(BlockedTargetError):
|
||||
guard.check("http://169.254.169.254/latest/meta-data/")
|
||||
|
||||
|
||||
def test_hostname_resolving_to_loopback_is_blocked(guard: UrlGuard, monkeypatch) -> None:
|
||||
def fake_getaddrinfo(host, *args, **kwargs): # noqa: ARG001
|
||||
return [(socket.AF_INET, socket.SOCK_STREAM, 6, "", ("127.0.0.1", 80))]
|
||||
|
||||
monkeypatch.setattr(socket, "getaddrinfo", fake_getaddrinfo)
|
||||
|
||||
with pytest.raises(BlockedTargetError):
|
||||
guard.check("http://vlastni-domena.example.com/dokument.html")
|
||||
|
||||
|
||||
def test_file_scheme_is_blocked(guard: UrlGuard) -> None:
|
||||
with pytest.raises(BlockedTargetError):
|
||||
guard.check("file:///etc/passwd")
|
||||
|
||||
|
||||
def test_public_address_passes(guard: UrlGuard, monkeypatch) -> None:
|
||||
def fake_getaddrinfo(host, *args, **kwargs): # noqa: ARG001
|
||||
return [(socket.AF_INET, socket.SOCK_STREAM, 6, "", ("93.184.216.34", 80))]
|
||||
|
||||
monkeypatch.setattr(socket, "getaddrinfo", fake_getaddrinfo)
|
||||
assert guard.check("https://example.com/dokument.html") == "example.com"
|
||||
|
||||
|
||||
def test_allowlisted_host_skips_the_check() -> None:
|
||||
settings = Settings()
|
||||
object.__setattr__(settings, "ssrf_allowed_hosts", ["localhost"])
|
||||
guard = UrlGuard(settings)
|
||||
assert guard.check("http://localhost:9000/dokument.html") == "localhost"
|
||||
@@ -0,0 +1,37 @@
|
||||
"""Building of the page stylesheet."""
|
||||
|
||||
from __future__ import annotations
|
||||
|
||||
from app.models import PageNumbers, PageSettings
|
||||
from app.pdf.styles import build_page_css, css_content_value, page_size_value
|
||||
|
||||
|
||||
def test_named_format_keeps_orientation() -> None:
|
||||
assert page_size_value(PageSettings(format="A4", orientation="landscape")) == "A4 landscape"
|
||||
|
||||
|
||||
def test_explicit_dimensions_are_used_as_is() -> None:
|
||||
assert page_size_value(PageSettings(format="210mm 297mm")) == "210mm 297mm"
|
||||
|
||||
|
||||
def test_content_uses_counters_when_the_total_is_unknown() -> None:
|
||||
assert css_content_value("{page} / {pages}", None) == 'counter(page) " / " counter(pages)'
|
||||
|
||||
|
||||
def test_content_uses_a_literal_when_the_total_is_known() -> None:
|
||||
assert css_content_value("Strana {page} z {pages}", 120) == '"Strana " counter(page) " z " "120"'
|
||||
|
||||
|
||||
def test_page_numbers_are_absent_when_disabled() -> None:
|
||||
css = build_page_css(PageSettings(), PageNumbers(enabled=False))
|
||||
assert "@bottom-center" not in css
|
||||
|
||||
|
||||
def test_page_numbers_land_in_the_requested_box() -> None:
|
||||
css = build_page_css(PageSettings(), PageNumbers(enabled=True, position="top-right"))
|
||||
assert "@top-right" in css
|
||||
|
||||
|
||||
def test_bookmarks_can_be_switched_off() -> None:
|
||||
css = build_page_css(PageSettings(), None, outline=False)
|
||||
assert "bookmark-level: none" in css
|
||||
Reference in New Issue
Block a user