Skip to content

Automate browser based workflows with AI

[Oncall] Preserve PDF suffix for slash-date invoice downloads (#6082)

Marc Kelechavabeaf8749c25d
diff --git a/skyvern/webeye/actions/handler.py b/skyvern/webeye/actions/handler.py
index f056a0142..909f10810 100644
--- a/skyvern/webeye/actions/handler.py
+++ b/skyvern/webeye/actions/handler.py
@@ -117,6 +117,7 @@ from skyvern.webeye.cdp_download_interceptor import (
MAX_FILE_SIZE_BYTES,
extract_filename,
is_download_response,
+ normalize_download_filename,
)
from skyvern.webeye.scraper.scraped_page import (
CleanupElementTreeFunc,
@@ -556,7 +557,7 @@ class ScopedXhrDownloadCapture:
if not self._is_xhr_download(headers, response.status):
return
raw_filename = extract_filename({"content-disposition": headers.get("content-disposition", "")}, "")
- filename = Path(raw_filename).name if raw_filename else ""
+ filename = normalize_download_filename(raw_filename, headers.get("content-type", ""))
if not filename or filename in self._saved:
return
content_length = headers.get("content-length", "")
diff --git a/skyvern/webeye/cdp_download_interceptor.py b/skyvern/webeye/cdp_download_interceptor.py
index 5e28977e2..dd45b4d7f 100644
--- a/skyvern/webeye/cdp_download_interceptor.py
+++ b/skyvern/webeye/cdp_download_interceptor.py
@@ -102,6 +102,14 @@ DOWNLOAD_MIME_TYPES = frozenset(
}
)
+DOWNLOAD_EXTENSION_BY_MIME_TYPE = {
+ "application/pdf": ".pdf",
+}
+
+_FILENAME_PATH_SEPARATOR_RE = re.compile(r"[\\/]+")
+_FILENAME_CONTROL_CHAR_RE = re.compile(r"[\x00-\x1f\x7f]")
+_WINDOWS_DRIVE_RE = re.compile(r"^[A-Za-z]:[\\/]")
+
def _parse_headers(raw_headers: list[dict[str, str]]) -> dict[str, str]:
"""Convert CDP header list [{name, value}] to a lowercase-keyed dict (last value wins)."""
@@ -122,6 +130,42 @@ def _parse_content_length(headers: dict[str, str]) -> int | None:
return None
+def _normalized_content_type(content_type: str) -> str:
+ return content_type.split(";")[0].strip().lower()
+
+
+def _download_extension_for_content_type(content_type: str) -> str:
+ return DOWNLOAD_EXTENSION_BY_MIME_TYPE.get(_normalized_content_type(content_type), "")
+
+
+def normalize_download_filename(filename: str, content_type: str = "") -> str:
+ """Sanitize a server-provided filename and add a trusted extension when missing."""
+ filename = unquote(filename).strip()
+ filename = _FILENAME_CONTROL_CHAR_RE.sub("", filename)
+ if not filename:
+ return ""
+
+ path_segments = [segment for segment in _FILENAME_PATH_SEPARATOR_RE.split(filename) if segment]
+ has_path_traversal = (
+ filename.startswith(("/", "\\"))
+ or bool(_WINDOWS_DRIVE_RE.match(filename))
+ or any(segment == ".." for segment in path_segments)
+ )
+ if has_path_traversal:
+ filename = next((segment for segment in reversed(path_segments) if segment not in {".", ".."}), "")
+ else:
+ filename = _FILENAME_PATH_SEPARATOR_RE.sub("_", filename)
+
+ filename = filename.strip(" .")
+ if not filename or Path(filename).suffix:
+ return filename
+
+ extension = _download_extension_for_content_type(content_type)
+ if extension:
+ return f"{filename}{extension}"
+ return filename
+
+
def is_download_response(headers: dict[str, str], status_code: int, resource_type: str = "") -> bool:
"""
Determine if a response is a file download.
@@ -142,7 +186,7 @@ def is_download_response(headers: dict[str, str], status_code: int, resource_typ
return False
content_disposition = headers.get("content-disposition", "")
- content_type = headers.get("content-type", "").split(";")[0].strip().lower()
+ content_type = _normalized_content_type(headers.get("content-type", ""))
if content_type in NON_DOWNLOAD_CONTENT_TYPES:
return False
@@ -237,7 +281,7 @@ class CDPDownloadInterceptor:
self._output_dir.mkdir(parents=True, exist_ok=True)
LOG.info("CDP download interceptor download dir set", download_dir=download_dir)
- def _resolve_save_path(self, filename: str = "") -> tuple[Path, str]:
+ def _resolve_save_path(self, filename: str = "", content_type: str = "") -> tuple[Path, str]:
"""Generate a unique save path under _output_dir.
Sanitizes the filename (path traversal prevention), falls back to a UUID-based
@@ -250,10 +294,9 @@ class CDPDownloadInterceptor:
self._output_dir.mkdir(parents=True, exist_ok=True)
self._download_index += 1
- # Sanitize to prevent path traversal (e.g. "../../etc/evil")
- filename = Path(filename).name
+ filename = normalize_download_filename(filename, content_type)
if not filename:
- filename = f"download_{uuid.uuid4().hex[:8]}"
+ filename = f"download_{uuid.uuid4().hex[:8]}{_download_extension_for_content_type(content_type)}"
save_path = self._output_dir / filename
# TODO: implement proper filename dedup (e.g., content hash or UUID suffix)
@@ -389,11 +432,10 @@ class CDPDownloadInterceptor:
LOG.warning("No output_dir set, skipping direct download", url=url)
return
- save_path, filename = self._resolve_save_path(suggested_filename)
-
t0 = time.monotonic()
data: bytes | None = None
method = ""
+ content_type = ""
# Try Playwright's APIRequestContext which shares the BrowserContext's cookies.
# We use the BrowserContext (not a Page) so this survives individual page closes.
@@ -402,6 +444,7 @@ class CDPDownloadInterceptor:
response = await self._browser_context.request.get(url)
if response.ok:
data = await response.body()
+ content_type = response.headers.get("content-type", "")
method = "playwright_api"
else:
LOG.debug(
@@ -418,11 +461,11 @@ class CDPDownloadInterceptor:
req = urllib.request.Request(url, headers={"User-Agent": "Mozilla/5.0"})
ssl_ctx = ssl.create_default_context()
- def _fetch() -> bytes:
+ def _fetch() -> tuple[bytes, str]:
with urllib.request.urlopen(req, context=ssl_ctx) as resp:
- return resp.read()
+ return resp.read(), resp.headers.get("content-type", "")
- data = await asyncio.to_thread(_fetch)
+ data, content_type = await asyncio.to_thread(_fetch)
method = "urllib"
except Exception as e:
LOG.error("Direct HTTP download failed", url=url, error=str(e), exc_info=True)
@@ -441,6 +484,8 @@ class CDPDownloadInterceptor:
)
return
+ save_path, filename = self._resolve_save_path(suggested_filename, content_type)
+
with open(save_path, "wb") as f:
f.write(data)
@@ -640,9 +685,9 @@ class CDPDownloadInterceptor:
return
content_length = _parse_content_length(headers)
- content_type = headers.get("content-type", "").split(";")[0].strip()
+ content_type = _normalized_content_type(headers.get("content-type", ""))
raw_filename = extract_filename(headers, url)
- save_path, filename = self._resolve_save_path(raw_filename)
+ save_path, filename = self._resolve_save_path(raw_filename, content_type)
LOG.info(
"CDP download detected",
diff --git a/tests/unit/test_cdp_download_interceptor.py b/tests/unit/test_cdp_download_interceptor.py
index b88f1cfc7..ee7a08e0a 100644
--- a/tests/unit/test_cdp_download_interceptor.py
+++ b/tests/unit/test_cdp_download_interceptor.py
@@ -256,6 +256,13 @@ class TestResolveSavePath:
assert len(filename) > len("download_")
assert save_path == tmp_path / filename
+ def test_empty_filename_gets_pdf_uuid_fallback(self, tmp_path: Path) -> None:
+ interceptor = self._make_interceptor(tmp_path)
+ save_path, filename = interceptor._resolve_save_path("", "application/pdf")
+ assert filename.startswith("download_")
+ assert filename.endswith(".pdf")
+ assert save_path == tmp_path / filename
+
def test_default_param_empty_string(self, tmp_path: Path) -> None:
"""Calling without arguments should also trigger fallback."""
interceptor = self._make_interceptor(tmp_path)
@@ -269,6 +276,25 @@ class TestResolveSavePath:
assert filename == "evil"
assert save_path == tmp_path / "evil"
+ def test_header_date_separators_are_filename_chars(self, tmp_path: Path) -> None:
+ """Invoice-style slashes in Content-Disposition should not collapse to the last date segment."""
+ interceptor = self._make_interceptor(tmp_path)
+ save_path, filename = interceptor._resolve_save_path("invoice_5/19/2026", "application/pdf")
+ assert filename == "invoice_5_19_2026.pdf"
+ assert save_path == tmp_path / "invoice_5_19_2026.pdf"
+
+ def test_missing_extension_uses_pdf_content_type(self, tmp_path: Path) -> None:
+ interceptor = self._make_interceptor(tmp_path)
+ save_path, filename = interceptor._resolve_save_path("2026", "application/pdf; charset=utf-8")
+ assert filename == "2026.pdf"
+ assert save_path == tmp_path / "2026.pdf"
+
+ def test_existing_pdf_extension_not_duplicated(self, tmp_path: Path) -> None:
+ interceptor = self._make_interceptor(tmp_path)
+ save_path, filename = interceptor._resolve_save_path("invoice_5/19/2026.pdf", "application/pdf")
+ assert filename == "invoice_5_19_2026.pdf"
+ assert save_path == tmp_path / "invoice_5_19_2026.pdf"
+
def test_increments_download_index(self, tmp_path: Path) -> None:
interceptor = self._make_interceptor(tmp_path)
assert interceptor._download_index == 0