g&ddlZddlmZddlmZmZmZmZddlmZ ddl m Z ddl m Z GddZdZhd Z d d Z dd Zdd Zy)N) _pandas_api)CodecTable concat_tablesschema)_feather) FeatherErrorc.eZdZdZddZddZdZd dZy) FeatherDataseta Encapsulates details of reading a list of Feather files. Parameters ---------- path_or_paths : List[str] A list of file names validate_schema : bool, default True Check that individual file schemas are all the same / compatible c ||_||_yN)pathsvalidate_schema)self path_or_pathsrs D/var/www/openai/venv/lib/python3.12/site-packages/pyarrow/feather.py__init__zFeatherDataset.__init__)s" .NcFt|jd|}|g|_|j|_|jddD]H}t||}|jr|j |||jj |Jt|jS)a, Read multiple feather files as a single pyarrow.Table Parameters ---------- columns : List[str] Names of columns to read from the file Returns ------- pyarrow.Table Content of the file as a table (of columns) rcolumnsN) read_tabler_tablesrrvalidate_schemasappendr)rr_filpathtables rrzFeatherDataset.read_table-s$**Q-9v kk JJqrNDtW5E##%%dE2 LL   & # T\\**rc|jj|js0tdj||j|jy)Nz-Schema in {!s} was different. {!s} vs {!s})requals ValueErrorformat)rpiecers rrzFeatherDataset.validate_schemasFsH{{!!%,,/2$fUDKK%*\\34 40rcF|j|j|S)a Read multiple Parquet files as a single pandas DataFrame Parameters ---------- columns : List[str] Names of columns to read from the file use_threads : bool, default True Use multiple threads when converting to pandas Returns ------- pandas.DataFrame Content of the file as a pandas DataFrame (of columns) r) use_threadsr to_pandas)rrr&s r read_pandaszFeatherDataset.read_pandasMs* w/99#:% %r)Tr )NT)__name__ __module__ __qualname____doc__rrrr)rrr r s /+24%rr c|jdk(ry|jtjtjfvrt dj |t dj |t|j)NrzqColumn '{}' exceeds 2GB maximum capacity of a Feather binary column. This restriction may be lifted in the futurezkColumn '{}' of type {} was chunked on conversion to Arrow and cannot be currently written to Feather format) num_chunkstypeextbinarystringr"r#str)namecols rcheck_chunked_overflowr8ass ~~ xxCJJL#**,//006t > > **0&s388}*EG Gr>lz4zstd uncompressedctjrDtjr4t|tjj r|j }tj|rp|dk(rd}n|dk(rd}n tdtj||}|dk(r;t|jjD]\}} ||} t| | n|}|dk(rYt|j tt#|j kDr td| td|Ntd |t%j&d rd }n)|'|t(vrtd j+|t( t-j.|||||| y#t0$rBt|t2r0 t5j6|#t4j8$rYwxYwwxYw)a Write a pandas.DataFrame to Feather format. Parameters ---------- df : pandas.DataFrame or pyarrow.Table Data to write out as Feather format. dest : str Local destination path. compression : string, default None Can be one of {"zstd", "lz4", "uncompressed"}. The default of None uses LZ4 for V2 files if it is available, otherwise uncompressed. compression_level : int, default None Use a compression level particular to the chosen compressor. If None use the default compression level chunksize : int, default None For V2 files, the internal maximum size of Arrow RecordBatch chunks when writing the Arrow IPC file format. None means use the default, which is currently 64K version : int, default 2 Feather file version. Version 2 is the current. Version 1 is the more limited legacy format rFNz%Version value should either be 1 or 2)preserve_indexz'cannot serialize duplicate column namesz2Feather V1 files do not support compression optionz0Feather V1 files do not support chunksize option lz4_framer9z1compression="{}" not supported, must be one of {}) compressioncompression_level chunksizeversion)r have_pandas has_sparse isinstancepdSparseDataFrameto_dense is_data_framer"r from_pandas enumeraternamesr8len column_namessetr is_available_FEATHER_SUPPORTED_CODECSr#r write_feather Exceptionr5osremoveerror) dfdestr@rArBrCr>rir6r7s rrSrSss2  " "2{~~==>B  $ a<"N \!NDE E!!"^D a<$U\\%7%784Ah&tS19!| u!! "SU-?-?)@%A AFG G  "&' '  &' '  5#5#5k#BK%!::))/ 0I*KL L ud 1B)2G E  dC   $ 88    s0>FG%4G  G% G G%G  G%c Dt||||jdd|i|S)a Read a pandas.DataFrame from Feather format. To read as pyarrow.Table use feather.read_table. Parameters ---------- source : str file path, or file-like object You can use MemoryMappedFile as source, for explicitly use memory map. columns : sequence, optional Only read a specific set of columns. If not provided, all columns are read. use_threads : bool, default True Whether to parallelize reading using multiple threads. If false the restriction is used in the conversion to Pandas as well as in the reading from Feather format. memory_map : boolean, default False Use memory mapping when opening file on disk, when source is a str. **kwargs Additional keyword arguments passed on to `pyarrow.Table.to_pandas`. Returns ------- df : pandas.DataFrame The contents of the Feather file as a pandas.DataFrame )r memory_mapr&r&r.r')sourcerr&r\kwargss r read_featherr_s;6 +JJ !!* N7B NFL NOrctj|||}||jS|Dcgc] }t|}}t t d|r|j |}n\t t d|r|j|}n4|Dcgc]}|j} }tdj|| |jdkr|Stt||k(r|S|j|Scc}wcc}w)a Read a pyarrow.Table from Feather format Parameters ---------- source : str file path, or file-like object You can use MemoryMappedFile as source, for explicitly use memory map. columns : sequence, optional Only read a specific set of columns. If not provided, all columns are read. memory_map : boolean, default False Use memory mapping when opening file on disk, when source is a str use_threads : bool, default True Whether to parallelize reading using multiple threads. Returns ------- table : pyarrow.Table The contents of the Feather file as a pyarrow.Table )use_memory_mapr&c|tk(Sr )intts rzread_table..scrc|tk(Sr )r5rds rrfzread_table..s18rzAQZZ>5):;= = ~~ G  ( ||G$$'8 ?s DD)NNNr=)NTF)NFT)rUpyarrow.pandas_compatr pyarrow.librrrrlibr2pyarrowrpyarrow._featherr r r8rRrSr_rr.rrr{s`& -00)A%A%H G<AE*+Qh48!O@.%r