eh%PdZddlmZddlZddlmZddlZddlm Z ddl m Z ddl m Z ddlmZmZmZmZmZerdd lmZmZmZdd lmZmZdd lmZmZmZmZd Z dd Z!dde df ddZ"de f ddZ#de df ddZ$de df ddZ%y)z" data hash pandas / numpy objects ) annotationsN) TYPE_CHECKING)hash_object_array) is_list_like)CategoricalDtype) ABCDataFrameABCExtensionArrayABCIndex ABCMultiIndex ABCSeries)HashableIterableIterator) ArrayLikenpt) DataFrameIndex MultiIndexSeries0123456789123456c  t|}t j |g|}tjd}tj|tjdz}d}t|D]4\}}||z }||z}||z}|tjd|z|zz }|}6|dz|k(sJd|tjdz }|S#t$r(tjgtjcYSwxYw) z Parameters ---------- arrays : Iterator[np.ndarray] num_items : int Returns ------- np.ndarray[uint64] Should be the same as CPython's tupleobject.c )dtypeiCBixV4riXBzFed in wrong num_itemsi|) next StopIterationnparrayuint64 itertoolschain zeros_like enumerate) arrays num_itemsfirstmultoutlast_iia inverse_is V/var/www/skyplay_api_hub/venv/lib/python3.12/site-packages/pandas/core/util/hashing.pycombine_hash_arraysr-/s-V __eWf -F 99W D -- 8!4 4C F&!1M  q t   %)+i788  A: "<$<< "299U C J! -xx")),,-s C.DDTutf8cfddlm}tttr|t ddStt r7tjjdd}||dd}|Sttrttjjdd}|r1fdd D}tj|g|} t| d }||jdd}|Sttrfd j!D} t#j$} |r2fd d D} | d z } tj| | } d| D} t| | }||jdd}|St'dt))a> Return a data hash of the Index/Series/DataFrame. Parameters ---------- obj : Index, Series, or DataFrame index : bool, default True Include the index in the hash (if Series/DataFrame). encoding : str, default 'utf8' Encoding for data & key when strings. hash_key : str, default _default_hash_key Hash_key for string key to encode. categorize : bool, default True Whether to first categorize object arrays before hashing. This is more efficient when the array contains duplicate values. Returns ------- Series of uint64, same length as the object Examples -------- >>> pd.util.hash_pandas_object(pd.Series([1, 2, 3])) 0 14639053686158035780 1 3869563279212530728 2 393322362522515241 dtype: uint64 r)rrF)rcopyr0)indexrr0c3fK|](}tjdj*ywF)r2encodinghash_key categorizeNhash_pandas_objectr2_values.0_r7r5r6objs r, z%hash_pandas_object..s? #II%%)  '  .1Nc3TK|]\}}t|j!ywrA) hash_arrayr:)r<r=seriesr7r5r6s r,r?z%hash_pandas_object..s, 6 v~~x: F s%(c3fK|](}tjdj*ywr4r8r;s r,r?z%hash_pandas_object..s? $#II%%)  '  $r@rc3 K|]}|ywrA)r<xs r,r?z%hash_pandas_object..s)Aa)s zUnexpected type for hashing )pandasr_default_hash_key isinstancer hash_tuplesr rDr:astyper rr r-r2ritemslencolumns TypeErrortype)r>r2r5r6r7rhser index_iterr#hashesr$index_hash_generator_hashess` ``` r,r9r9SsF$#}%k#x:(QVWW C " s{{Hh C J J 5 K Qc>d Ja C # s{{Hh C J J 5 K     J__aS*5F#FA.AQciixeD< J9 C &  YY[  $  $  $  NI oof.BCG))F  2QciixeD J6tCykBCCc t|s tdddlm}m}t |t s|j|}n|}t|jDcgc]9}|j|j|t|j|d;}}fd|D}t|t|} | Scc}w)a Hash an MultiIndex / listlike-of-tuples efficiently. Parameters ---------- vals : MultiIndex or listlike-of-tuples encoding : str, default 'utf8' hash_key : str, default _default_hash_key Returns ------- ndarray[np.uint64] of hashed values z'must be convertible to a list-of-tuplesr) CategoricalrF categoriesorderedc3FK|]}|jdyw)Fr5r6r7N)_hash_pandas_object)r<catr5r6s r,r?zhash_tuples..s,  HQVWs!)rrRrJr\rrLr from_tuplesrangenlevels _simple_newcodesrlevelsr-rP) valsr5r6r\rmilevelcat_valsrWrTs `` r,rMrMs$  ABB dM * #Z # #D ) 2::&    HHUO  %(8% H HF FCM2A Hs>Cc t|ds tdt|tr|j |||St|t j s"tdt|jdt||||S)a Given a 1d array, return an array of deterministic integers. Parameters ---------- vals : ndarray or ExtensionArray encoding : str, default 'utf8' Encoding for data & key when strings. hash_key : str, default _default_hash_key Hash_key for string key to encode. categorize : bool, default True Whether to first categorize object arrays before hashing. This is more efficient when the array contains duplicate values. Returns ------- ndarray[np.uint64, ndim=1] Hashed values, same length as the vals. Examples -------- >>> pd.util.hash_array(np.array([1, 2, 3])) array([ 6238072747940578789, 15839785061582574730, 2185194620014831856], dtype=uint64) rzmust pass a ndarray-likeraz6hash_array requires np.ndarray or ExtensionArray, not z!. Use hash_pandas_object instead.) hasattrrRrLr rbrndarrayrS__name__ _hash_ndarray)rjr5r6r7s r,rDrDs> 4 !233$)*''Z(   dBJJ ' DDz""##D F  x: >>rZcf|j}tj|tjr8t |j |||}t |j |||}|d|zzS|tk(r|jd}n"t|jtjtjfr#|jdjdd}nt|jtjrG|jdkr8|jd|jjjd}n`|rPdd lm}m}m} | |d \} } t)|| d }|j+| |} | j-||d S t/|||}||d z z}|tj6dz}||dz z}|tj6dz}||dz z}|S#t0$r6t/|jt2jt4||}YwxYw)z! See hash_array.__doc__. u8i8Fr1ur)r\r factorize)sortr]rale9z lb&&& )rr issubdtype complex128rrrealimagboolrN issubclassrS datetime64 timedelta64viewnumberitemsizerJr\rryrrgrbrrRstrobjectr) rjr5r6r7r hash_real hash_imagr\rryrhr^rcs r,rrrrs JJE }}UBMM*!$))XxL !$))XxL 2 >)) }{{4  EJJ ? @yy%%d%7 EJJ *u~~/Byy1TZZ00123::4@    !*$U ; E:$j0A5QE))%7C**!H+  $T8X>D DBJDBII( ))DDBJDBII( ))DDBJD K $ C ''/8D s G11zIndex | DataFrame | Seriesr2rr5rr6z str | Noner7rrr)rjz+MultiIndex | Iterable[tuple[Hashable, ...]]r5rr6rrr) rjrr5rr6rr7rrr) rjz np.ndarrayr5rr6rr7rrr)&__doc__ __future__rrtypingrnumpyrpandas._libs.hashingrpandas.core.dtypes.commonrpandas.core.dtypes.dtypesrpandas.core.dtypes.genericrr r r r collections.abcr rrpandas._typingrrrJrrrrrKr-r9rMrDrrrHrZr,rsq# 226  '! !-0!!L, a #a aa a  a  aL%/ 5/ / /  / h% .? .?.?.? .?  .?f% 9 999 9  9rZ