Skip to content

Add arbitrary io buffer support to reader #234

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
horgh merged 7 commits into from
Aug 11, 2025
+51 −29
Merged

Add arbitrary io buffer support to reader #234

Changes from 1 commit
Commits
Show all changes
7 commits
Select commit Hold shift + click to select a range
fe042ae
add support for reading a database from buffers
seemanne Aug 8, 2025
ffa7554
clarify reading_from_buffer test
seemanne Aug 8, 2025
0fc074b
fix import order
seemanne Aug 8, 2025
5ce419e
refactor init logic to reduce complexity
seemanne Aug 8, 2025
5f6d083
have _load_buffer return filename to avoid setting it as an attr
seemanne Aug 11, 2025
aa22454
remove unused import
seemanne Aug 11, 2025
12dbdb6
add Any back to make CI happy
seemanne Aug 11, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
75 changes: 40 additions & 35 deletions maxminddb/reader.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -64,39 +64,7 @@ def __init__(
a path. This mode implies MODE_MEMORY.

"""
filename: Any
if (mode == MODE_AUTO and mmap) or mode == MODE_MMAP:
with open(database, "rb") as db_file: # type: ignore[arg-type]
self._buffer = mmap.mmap(db_file.fileno(), 0, access=mmap.ACCESS_READ)
self._buffer_size = self._buffer.size()
filename = database
elif mode in (MODE_AUTO, MODE_FILE):
self._buffer = FileBuffer(database) # type: ignore[arg-type]
self._buffer_size = self._buffer.size()
filename = database
elif mode == MODE_MEMORY:
with open(database, "rb") as db_file: # type: ignore[arg-type]
buf = db_file.read()
self._buffer = buf
self._buffer_size = len(buf)
filename = database
elif mode == MODE_FD:
self._buffer = database.read() # type: ignore[union-attr]
self._buffer_size = len(self._buffer) # type: ignore[arg-type]
# io buffers are not guaranteed to have a name attribute
if hasattr(database, "name"):
filename = database.name # type: ignore[union-attr]
else:
filename = f"<{type(database)}>"
else:
msg = (
f"Unsupported open mode ({mode}). Only MODE_AUTO, MODE_FILE, "
"MODE_MEMORY and MODE_FD are supported by the pure Python "
"Reader"
)
raise ValueError(
msg,
)
self._load_buffer(database, mode)

metadata_start = self._buffer.rfind(
self._METADATA_START_MARKER,
Expand All @@ -106,7 +74,7 @@ def __init__(
if metadata_start == -1:
self.close()
msg = (
f"Error opening database file ({filename}). "
f"Error opening database file ({self.filename}). "
"Is this a valid MaxMind DB file?"
)
raise InvalidDatabaseError(
Expand All @@ -118,7 +86,7 @@ def __init__(
(metadata, _) = metadata_decoder.decode(metadata_start)

if not isinstance(metadata, dict):
msg = f"Error reading metadata in database file ({filename})."
msg = f"Error reading metadata in database file ({self.filename})."
raise InvalidDatabaseError(
msg,
)
Expand Down Expand Up @@ -280,6 +248,43 @@ def _resolve_data_pointer(self, pointer: int) -> Record:
(data, _) = self._decoder.decode(resolved)
return data

def _load_buffer(
self, database: AnyStr | int | PathLike | IO, mode: int = MODE_AUTO
) -> None:
self.filename: Any
Copy link
Contributor

@horgh horgh Aug 8, 2025

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Rather than making this a public attribute, what do you think about returning it from this function?

Copy link
Contributor Author

@seemanne seemanne Aug 11, 2025

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Fine for me, changed it

if (mode == MODE_AUTO and mmap) or mode == MODE_MMAP:
with open(database, "rb") as db_file: # type: ignore[arg-type]
self._buffer = mmap.mmap(db_file.fileno(), 0, access=mmap.ACCESS_READ)
self._buffer_size = self._buffer.size()
self.filename = database
elif mode in (MODE_AUTO, MODE_FILE):
self._buffer = FileBuffer(database) # type: ignore[arg-type]
self._buffer_size = self._buffer.size()
self.filename = database
elif mode == MODE_MEMORY:
with open(database, "rb") as db_file: # type: ignore[arg-type]
buf = db_file.read()
self._buffer = buf
self._buffer_size = len(buf)
self.filename = database
elif mode == MODE_FD:
self._buffer = database.read() # type: ignore[union-attr]
self._buffer_size = len(self._buffer) # type: ignore[arg-type]
# io buffers are not guaranteed to have a name attribute
if hasattr(database, "name"):
self.filename = database.name # type: ignore[union-attr]
else:
self.filename = f"<{type(database)}>"
else:
msg = (
f"Unsupported open mode ({mode}). Only MODE_AUTO, MODE_FILE, "
"MODE_MEMORY and MODE_FD are supported by the pure Python "
"Reader"
)
raise ValueError(
msg,
)

def close(self) -> None:
"""Close the MaxMind DB file and returns the resources to the system."""
with contextlib.suppress(AttributeError):
Expand Down
Loading