o mi@sdZddlmZddlmZmZddlmZddlm Z m Z m Z m Z m Z ddlmZddlmZdgZ ddddZd S)a( This file provides a single function `serialize_in_chunks()` which can serialize a Graph into a number of NT files with a maximum number of triples or maximum file size. There is an option to preserve any prefixes declared for the original graph in the first file, which will be a Turtle file. ) annotations) ExitStackcontextmanager)Path) TYPE_CHECKINGBinaryIO GeneratorOptionalTuple)Graph)_nt_rowserialize_in_chunks'NchunkFgr max_triplesintmax_file_size_kb Optional[int]file_name_stemstr output_dirOptional[Path]write_prefixesboolreturnNonec s&durtstdtdfdd }dd d }|rFttdddd}|||Wdn1sAwYd} t} |dur|d} |rXdnd} t| dD]N\} }t | d}t || krtd|d|dd|| dkr| || \}}d} n| t || kr| d7} | || \}}d} | ||7} qanOt |}||kr|jtdddn/|rdnd} t| dD]!\} }| |dkr| || \}}| d7} |t | dq WddSWddS1s wYdS)a:Serializes a given Graph into a series of n-triples with a given length. Args: g: The graph to serialize. max_file_size_kb: Maximum size per NT file in kB (1,000 bytes) Equivalent to ~6,000 triples, depending on Literal sizes. max_triples: Maximum size per NT file in triples Equivalent to lines in file. If both this parameter and max_file_size_kb are set, max_file_size_kb will be used. file_name_stem: Prefix of each file name. e.g. "chunk" = chunk_000001.nt, chunk_000002.nt... output_dir: The directory you want the files to be written to. write_prefixes: The first file created is a Turtle file containing original graph prefixes. See `../test/test_tools/test_chunk_serializer.py` for examples of this in use. Nz>If you specify an output_dir, it must actually be a directory!file_norr,Generator[Tuple[Path, BinaryIO], None, None]c3sntr dus Jtdt|dd}t|d}||fVWddS1s0wYdS)N_z.ntwb)rrrzfillopen)rfpfhrrW/home/kim/smarthome/.venv/lib/python3.10/site-packages/rdflib/tools/chunk_serializer.py_start_new_file;s "  "z,serialize_in_chunks.._start_new_filerr rcSsBg}|jD]\}}|d|d|dqdt|dS)NzPREFIX z: <> )Znamespace_manager namespacesappendjoinsorted)rZpreskvr'r'r(_serialize_prefixesDsz0serialize_in_chunks.._serialize_prefixesz _000000.ttlwzutf-8)encodingri)NNNzcannot write triple z as it's serialized size of z exceeds max_file_size_kb = z_all.ntnt)Z destinationformat)rrrr)rr rr)rcwdis_dir ValueErrorrr#writer enumerateZtriplesr encodelen enter_context serialize)rrrrrrr)r2r%Z bytes_writtenZxstackZ max_file_sizeritZ row_bytesr$ZfhbZ graph_lengthr'r&r(r sh     $)rNrNF)rr rrrrrrrrrrrr)__doc__ __future__r contextlibrrpathlibrtypingrrrr r Z rdflib.graphr Zrdflib.plugins.serializers.ntr __all__r r'r'r'r(s