Configuration

class pypdf.Configuration(maximum_declared_stream_length: int = 75000000, array_based_stream_maximum_output_length: int = 75000000, jbig2_maximum_output_length: int = 75000000, lzw_maximum_output_length: int = 75000000, run_length_maximum_output_length: int = 75000000, zlib_maximum_output_length: int = 75000000, zlib_maximum_recovery_input_length: int = 5000000, flate_maximum_columns: int = 250000, flate_maximum_row_length: int = 4000000, image_maximum_buffer_size: int = 75000000, xmp_maximum_input_length: int = 5000000, xmp_maximum_element_count: int = 100000, outline_maximum_entries: int = 100000, outline_maximum_depth: int = 100, page_tree_maximum_entries: int = 100000, page_tree_maximum_depth: int = 100, xform_maximum_invocations_per_extraction: int = 5000, jbig2dec_binary: str | None = <factory>, page_merge_box: ~typing.Literal['cropbox', 'trimbox'] = 'cropbox', disable_legacy_handling: bool = False)[source]

Bases: object

Configuration used while processing PDF files.

These limits mostly protect against excessive resource consumption caused by malformed or malicious PDF files.

Note

Avoid relying on specific implementation details of this class. The main public API consists of the values being considered read-only for each instance with the overwrite method to create a new instance with the specified values changed.

maximum_declared_stream_length: int = 75000000: The maximum allowed declared /Length value for streams.

array_based_stream_maximum_output_length: int = 75000000: The maximum allowed output length for array-based streams.

jbig2_maximum_output_length: int = 75000000: The maximum allowed number of uncompressed bytes during decompression when using the /JBIG2Decode filter (JBIG2 images).

lzw_maximum_output_length: int = 75000000: The maximum allowed number of uncompressed bytes during decompression when using the /LZWDecode filter (LZW compression).

run_length_maximum_output_length: int = 75000000: The maximum allowed number of uncompressed bytes during decompression when using the /RunLengthDecode filter (run-length compression).

zlib_maximum_output_length: int = 75000000: The maximum allowed number of uncompressed bytes during decompression when using the /FlateDecode filter (zlib compression).

zlib_maximum_recovery_input_length: int = 5000000: The maximum allowed number of bytes to attempt the recovery with when using the /FlateDecode filter (zlib compression).

flate_maximum_columns: int = 250000: The maximum allowed number of columns when using the /FlateDecode filter.

flate_maximum_row_length: int = 4000000: The maximum allowed row length when using the /FlateDecode filter.

image_maximum_buffer_size: int = 75000000: The maximum allowed number of bytes to allocate for images.

xmp_maximum_input_length: int = 5000000: The maximum allowed actual decompressed stream length in bytes for XMP data.

xmp_maximum_element_count: int = 100000: The maximum allowed number of elements for XMP data.

outline_maximum_entries: int = 100000: The maximum allowed number of outline entries.

outline_maximum_depth: int = 100: The maximum allowed depth for outline entries.

page_tree_maximum_entries: int = 100000: The maximum allowed number of entries in the page tree.

page_tree_maximum_depth: int = 100: The maximum allowed depth of the page tree.

xform_maximum_invocations_per_extraction: int = 5000: The maximum allowed number of /XObject forms per page during text extraction.

jbig2dec_binary: str | None

The location of the jbig2dec binary.

This either is a string holding the path to the binary, or None if the binary could not be found or the binary should not be called.

page_merge_box: Literal['cropbox', 'trimbox'] = 'cropbox': The page box to use during merge.

disable_legacy_handling: bool = False

Temporary configuration value for applications which do not rely on old overwrites and thus are able to skip legacy handling.

This disables looking for changes of the old constants during each initialization of the reader class. Please note that while this reduces the required overhead, it will disable all related deprecation warnings/errors. Setting this flag while legacy constants are intentionally being modified is unsupported/incorrect.

After the deprecation period, this configuration value will be removed with its own deprecation period as well.

with_overwrites(**kwargs: Any) → Configuration[source]

Creates a new configuration with the specified values overwritten.

Parameters:: **kwargs – Configuration field names and their new values.
Returns:: A new configuration with the specified values overwritten.

pypdf.apply_configuration(configuration: Configuration | None = None, **overwrites: Any) → Iterator[Configuration][source]

Temporarily overwrite the configuration.

Parameters:

configuration – A direct configuration instance to use.
**overwrites – Configuration parameters to overwrite in the current or given configuration.

Returns:

The new configuration.

pypdf.get_configuration() → Configuration[source]

Get the current configuration.

Returns:: The current configuration.

pypdf.overwrite_configuration(configuration: Configuration | None = None, **overwrites: Any) → Configuration[source]

Overwrite the configuration for the current execution context.

Parameters:

configuration – A direct configuration instance to use.
**overwrites – Configuration parameters to overwrite in the current or given configuration.

Returns:

The new configuration.