=wgBAdZddlmZddlmZmZddlmZmZm Z m Z ddl m Z ddl m Z mZddl mZmZmZmZGdd eZd ZGd d eZGd deZGddeZGddeZGddeZGddeZy)z The classes in this module encode and decode posting information for a field. The field format essentially determines what information is stored about each occurance of a term. ) defaultdict) unstoppedentoken) iteritemsdumpsloadsb emptybytes) _INT_SIZE _FLOAT_SIZE) pack_uint unpack_uint pack_float unpack_floatc\eZdZdZdZdZeeZddZ dZ dZ dZ d Z d Zd Zd Zy )FormataAbstract base class representing a storage format for a field or vector. Format objects are responsible for writing and reading the low-level representation of a field. It controls what kind/level of information to store about the indexed fields. T field_boostc ||_||_y)z :param field_boost: A constant boost factor to scale to the score of all queries matching terms in this field. Nroptionsselfrrs E/var/www/horilla/myenv/lib/python3.12/site-packages/whoosh/formats.py__init__zFormat.__init__8s ' cp|xr3|j|juxr|j|jk(SN) __class____dict__)rothers r__eq__z Format.__eq__As44NNeoo54MMU^^3 5rcN|jjd|jdS)Nz(boost=))r!__name__rrs r__repr__zFormat.__repr__Fs!%!8!8$:J:JKKrc:|jdkry|jSNr) posting_sizer(s rfixed_value_sizezFormat.fixed_value_sizeIs   q    rc t)aTakes the text value to be indexed and yields a series of ("tokentext", frequency, weight, valuestring) tuples, where frequency is the number of times "tokentext" appeared in the value, weight is the weight (a float usually equal to frequency in the absence of per-term boosts) and valuestring is encoded field-specific posting value for the token. For example, in a Frequency format, the value string would be the same as frequency; in a Positions format, the value string would encode a list of token positions at which "tokentext" occured. :param value: The unicode text to index. :param analyzer: The analyzer to use to process the text. )NotImplementedError)rvalueanalyzerkwargss r word_valueszFormat.word_valuesNs "!rc t|d|zS)zReturns True if this format supports interpreting its posting value as 'name' (e.g. "frequency" or "positions"). decode_)hasattrrnames rsupportszFormat.supports^stY-..rc t|d|zS)zReturns the bound method for interpreting value as 'name', where 'name' is for example "frequency" or "positions". This object must have a corresponding Format.decode_() method. r5)getattrr7s rdecoderzFormat.decoderds tY-..rc0|j||S)zInterprets the encoded value string as 'astype', where 'astype' is for example "frequency" or "positions". This object must have a corresponding decode_() method. )r<)rastype valuestrings r decode_aszFormat.decode_asks $t||F#K00rN?)r' __module__ __qualname____doc__r,textualdictfloat __inittypes__rr$r)r-r3r9r<r@rrrr-sF LGU+M5 L! " / /1rrct|ttfrt|fi|}t |S||fi|}t |Sr ) isinstancetuplelistrr)r0r1r2gens rtokensrPysF%%'e&v& S>u'' S>rcLeZdZdZdZeeZd dZdZ dZ dZ dZ d Z y ) ExistenceaOnly indexes whether a given term occurred in a given document; it does not store frequencies or positions. This is useful for fields that should be searchable but not scorable, such as file path. Supports: frequency, weight (always reports frequency = 1). rrc ||_||_yr rrs rrzExistence.__init__s& rc l|jtdt|||D}fd|DS)Nc34K|]}|jywr )text).0ts r z(Existence.word_values..sFaffFsc32K|]}|dtfywNr )rWwfbs rrYz(Existence.word_values..s81Ar:&8s)rsetrP)rr0r1r2wordsetr^s @rr3zExistence.word_valuess2   FfUHf&EFF888rctSr r )rr0s rencodezExistence.encodercyNr\rJrr?s rdecode_frequencyzExistence.decode_frequencysrc|jSr rrfs r decode_weightzExistence.decode_weightsrctSr r rvss rcombinezExistence.combinercrNrA)r'rCrDrEr,rGrHrIrr3rbrgrirmrJrrrRrRs6LU+M9  rrRcHeZdZdZeZeeeZ d dZ dZ dZ dZ dZy) FrequencyzUStores frequency information for each posting. Supports: frequency, weight. )r boost_as_freqc Dt|tsJ||_||_y)z :param field_boost: A constant boost factor to scale to the score of all queries matching terms in this field. N)rLrHrr)rrrprs rrzFrequency.__init__s#+u---& rc 4 |jd}tt}tt d|d<t |||D]?}|dz }||j xxdz cc< |j xx|j z cc<A fdt|D}|S)NrTboostsr\c3NK|]\}}|||zt|fywr )r)rWr]freqr^weightss rrYz(Frequency.word_values..s.$wq$4b)D/:$s"%)rrintrHrPrVboostr) rr0r1r2lengthfreqsrXwvsr^rvs @@rr3zFrequency.word_valuess   C e$xx0 'A aKF !&&MQ M AFFOqww &O ' $E"$ rct|dSr+)rrfs rrgzFrequency.decode_frequencys;'**rc<t|d}||jzSr+)rr)rr?rus rrizFrequency.decode_weights";'*d&&&&rc>ttfd|DS)Nc3@K|]}j|ywr ) decode_value)rWvrs rrYz$Frequency.combine..s>aT..q1>s)rsumrks` rrmzFrequency.combines>2>>??rN)rBF)r'rCrDrEr r,rGrHboolrIrr3rgrirmrJrrroros4 LU$?M  +'@rroc:eZdZdZdZdZdZdZdZdZ dZ y ) PositionszStores position information in each posting, to allow phrase searching and "near" queries. Supports: frequency, weight, positions, position_boosts (always reports position boost = 1.0). c+K|j}tt}tt}d|d<d|d<t |||D]K}||j j |j||j xx|jz cc<Mt|D]-\}} |j| }|t| |||z|f/yw)NT positionsrs) rrrNrHrPrVappendposrxrrblen) rr0r1r2r^posesrvrXr]poslists rr3zPositions.word_valuess   D!e$"{xx0 'A !&&M  ' AFFOqww &O '$E* |j||jzSr )rgrrfs rrizPositions.decode_weights$$[1D4D4DDDrcN|j|Dcgc]}|dfc}Scc}wre)r)rr?rs rdecode_position_boostsz Positions.decode_position_boostss%$($9$9+$FGSaGGGs "ct}|D]"}|j|j|$|jt |Sr )r_updaterrbsorted)rrlsrs rrmzPositions.combinesB E /A HHT**1- . /{{6!9%%rN) r'rCrDrEr3rbrrgrirrmrJrrrrs, <9 7EH&rrc.eZdZdZdZdZdZdZdZy) CharacterszStores token position and character start and end information for each posting. Supports: frequency, weight, positions, position_boosts (always reports position boost = 1.0), characters. c+K|j}tt}tt}d|d<d|d<d|d<t |||D]b}||j j |j|j|jf||j xx|jz cc<dt|D]-\}} |j| }|t| |||z|f/ywNTrcharsrs)rrrNrHrPrVrr startcharendcharrxrrbr) rr0r1r2r^seenrvrXr]rs rr3zCharacters.word_valuess   4 e$"{wxx0 'A L   QYY ? @ AFFOqww &O '$D/ "Y.0 1GH   V%fa(888rc|jtds|tdz }t|td}d}d}g}|D].}|d|z}|d|z}|d|z}|j |||f0|S)Nrrr\rr)rr?rrr posns_charsrrs rdecode_characterszCharacters.decode_characters-s##AcF+ 1S6 !Kk)*-.  ?DAw)HQ')I1g )G   )W= >  ? rc|jtds|tdz }t|td}d}g}|D]}|d|z}|j ||Srrrr?rrposnsrs rrzCharacters.decode_positions;sm##AcF+ 1S6 !Kk)*-. #DAw)H LL " # rc<i}|D]K}|j|D]5\}}}||vr$||\}}t||t||f||</||f||<7Mt|j Dcgc]}|||d||df} }|j | Scc}w)Nrr\)rminmaxrkeysrb) rrlrrrscecold_scold_ecrs rrmzCharacters.combineFs  &A#55a8 & R!8%(VNFF!"fos2v?AcF "XAcF  & &9?qvvx8HI#qvay!C&),II{{5!!Js.BN) r'rCrDrEr3rbrrrmrJrrrr s < 9   "rrc4eZdZdZdZdZdZdZdZdZ y) PositionBoostszA format that stores positions and per-position boost information in each posting. Supports: frequency, weight, positions, position_boosts. c+zK|j}tt}d|d<d|d<t|||D]:}|j}|j }||j j||f<t|D]:\} } |j| }| t| td| D|z|f<yw)NTrrsc3&K|] }|d ywr[rJ)rWps rrYz-PositionBoosts.word_values..gs%:qad%:s) rrrNrPrrxrVrrrbrr) rr0r1r2r^rrXrrxr]rs rr3zPositionBoosts.word_valuesZs   4 "{xx0 .A%%CGGE L  e - . "$ HHAuKK&Ec%j#%:E%:":R"?G G HsB9B;cg}d}d}|D]"\}}||z }|j||z |f|}$tt|t|zt |dzSr)rrrrr)rrrr summedboostrrxs rrbzPositionBoosts.encodeiss  JC 5 K LL#*e, -D #e*% ;(??q/" #rc|jtds|tdz }t|ttzd}d}g}|D] }|d|z}|j ||df"|S)Nrrr\rr rr r r)rr?rr posns_boostsrs rrz%PositionBoosts.decode_position_booststs##AcF+ 1S6 !Kk)k"9":;<  5DAw)H   47 3 4 5rc|jtds|tdz }t|ttzd}d}g}|D]}|d|z}|j ||Srrrs rrzPositionBoosts.decode_positionsss##AcF+ 1S6 !Kk)k"9":;< #DAw)H LL " # rc`t|tttzd}||jzSr+)rr r r)rrrs rrizPositionBoosts.decode_weights."1Yy;/F#GHK T----rctt}|D](}|j|D]\}}||xx|z cc<*|jt |j Sr )rrHrrbritems)rrlrrrrxs rrmzPositionBoosts.combinesb   A"99!< U#%  {{6!''),--rN) r'rCrDrEr3rbrrrirmrJrrrrSs& H #  ..rrc:eZdZdZdZdZdZdZdZdZ dZ y ) CharacterBoostszA format that stores positions, character start and end, and per-position boost information in each posting. Supports: frequency, weight, positions, position_boosts, characters, character_boosts. c+pKtt}d|d<d|d<d|d<t|||D]L}||jj |j |j |j|jfNt|D]*\}}|j|\}}|t|||f,ywr) rrNrPrVrrrrrxrrbr) rr0r1r2rrXr]rrs rr3zCharacterBoosts.word_valuess4 "{wxx0 JA L   QYY H I J"$ 6HAu!%U!3 E;c%j+u5 5 6sB4B6c|j}g}d}d}d}|D].\}}} } |j||z ||z | |z | f|}| }|| z }0tt|t ||zzt |dz|fSr)rrrrrr) rrr^rrrrrrrrxs rrbzCharacterBoosts.encodes    .3 ! *CGU LL#-X)=!I-u6 7GH 5 K  !3u:&K"4D)EE?#%02 2rc|jtds|tdz }t|ttzd}d}d}g}|D]2}||dz}||dz}||dz}|j ||||df4|S)Nrrr\rr)rr?rrrposn_char_boostsrrs rdecode_character_boostsz'CharacterBoosts.decode_character_boostss##AcF+ 1S6 !Kk)k"9":;< MD$q')H$q')I$q')G  # #Xy'47$K L  M  rcP|j|Dcgc]}|d c}Scc}wr+r)rr?items rrz CharacterBoosts.decode_positionss%$($@$@$MNDQNNNs #c l|j|Dcgc] \}}}}|||fc}}}}Scc}}}}wr r)rr?rrr_s rrz!CharacterBoosts.decode_characterss>// <>>.Hc9gqi)> >>s. c d|j|Dcgc] \}}}}||f c}}}Scc}}}wr r)rr?rrrxs rrz&CharacterBoosts.decode_position_boostss;// <>>!1aEe > >>s+c Zi}|D]R}|j|D]<\}}}}||vr)||\}} } t||t|| | |zf||<5|||f||<>Tt|j Dcgc]\}\}}}||||f} }}}}|j | dScc}}}}wr+)rrrrrrb) rrlrrrrrrxrr old_boostrs rrmzCharacterBoosts.combines  -A&*&B&B1&E -"RU!803A-FFI!"fos2v'%/1AcF!"e_AcF  - -1779%''*>#B#r2u%''{{5!!$$'s8B% N) r'rCrDrEr3rbrrrrrmrJrrrrs+ 62"  O>> %rrN)rE collectionsrwhoosh.analysisrr whoosh.compatrrrr whoosh.systemr r r rrrrobjectrrPrRrorrrrrJrrrs8 $.44$0JJ C1VC1XB,@,@^6&6&rF"F"R@.Y@.FK%jK%r