a jKk@sddlZddlZddlZddlZddlZgdZedZedZ edZ edZ edZ edZ ed Zed Zed Zed Zed ZGdddeZGdddeZGdddZGdddeZGdddeZGdddeZGdddeZGdddeZddZd d!Zd"d#Zd?d$d%Zd&d'Z d(d)Z!d*d+Z"d,d-Z#d.d/Z$Gd0d1d1Z%d2d3Z&d4d5Z'd6d7Z(d8d9Z)d:d;Z*dZ/e*e/dS)@N)MetaDataloadarff ArffErrorParseArffErrorz^\s*@z^%z^\s+$z^\s*@\S*z^@[Dd][Aa][Tt][Aa]z*^@[Rr][Ee][Ll][Aa][Tt][Ii][Oo][Nn]\s*(\S*)z2^\s*@[Aa][Tt][Tt][Rr][Ii][Bb][Uu][Tt][Ee]\s*(..*$)z{(.+)}z%[Dd][Aa][Tt][Ee]\s+[\"']?(.+?)[\"']?$z'(..+)'\s+(..+$)z(\S+)\s+(..+$)c@s eZdZdS)rN__name__ __module__ __qualname__r r =E:\soccer\ml_env\Lib\site-packages\scipy/io/arff/_arffread.pyr6src@s eZdZdS)rNrr r r r r:src@s4eZdZdZddZeddZddZdd ZdS) AttributeNcCs||_d|_tj|_dSN)namerangenpobject_dtypeselfrr r r __init__EszAttribute.__init__cCsdS)zj Parse the attribute line if it knows how. Returns the parsed attribute, or None. Nr clsr attr_stringr r r parse_attributeJszAttribute.parse_attributecCsdS)- Parse a value of this type. Nr rdata_strr r r parse_dataRszAttribute.parse_datacCs|jd|jS)r,)r type_namerr r r __str__XszAttribute.__str__) rrr rr classmethodrrr!r r r r r As  r csHeZdZdZfddZeddZeddZdd Z d d Z Z S) NominalAttributeZnominalcs6t|||_||_tjtdd|Df|_dS)Ncss|]}t|VqdSr )len.0ir r r gz,NominalAttribute.__init__..)superrvaluesrrbytes_maxr)rrr+ __class__r r rcs zNominalAttribute.__init__cCs4t|}|r(t|d\}}t|StddS)a8Given a string containing a nominal type, returns a tuple of the possible values. A nominal type is defined as something framed between braces ({}). Parameters ---------- atrv : str Nominal type definition Returns ------- poss_vals : tuple possible values Examples -------- >>> from scipy.io.arff._arffread import NominalAttribute >>> NominalAttribute._get_nom_val("{floup, bouga, fl, ratata}") ('floup', 'bouga', 'fl', 'ratata') z(This does not look like a nominal stringN) r_nominalmatchsplit_data_linegrouptuple ValueError)atrvmattrs_r r r _get_nom_valis  zNominalAttribute._get_nom_valcCs(|ddkr ||}|||SdSdS)z Parse the attribute line if it knows how. Returns the parsed attribute, or None. For nominal attributes, the attribute string would be like '{, , }'. r{N)r;)rrrr+r r r rs   z NominalAttribute.parse_attributecCs:||jvr|S|dkr|Stt|dt|jdS)r?z value not in N)r+r6strrr r r rs  zNominalAttribute.parse_datacCsN|jd}tt|jdD]}||j|d7}q||jd7}|d7}|S)Nz,{r0r})rrr$r+rmsgr'r r r r!s  zNominalAttribute.__str__) rrr rr staticmethodr;r"rrr! __classcell__r r r.r r#_s    r#cs8eZdZfddZeddZddZddZZS) NumericAttributecst|d|_tj|_dS)Nnumeric)r*rrrZfloat64rrr.r r rs zNumericAttribute.__init__cCsX|}|dtddksH|dtddksH|dtddkrP||SdSdS)z Parse the attribute line if it knows how. Returns the parsed attribute, or None. For numeric attributes, the attribute string would be like 'numeric' or 'int' or 'real'. NrFintreallowerstripr$rr r r rs z NumericAttribute.parse_attributecCsd|vrtjSt|SdS)a Parse a value of this type. Parameters ---------- data_str : str string to convert Returns ------- f : float where float can be nan Examples -------- >>> from scipy.io.arff._arffread import NumericAttribute >>> atr = NumericAttribute('atr') >>> atr.parse_data('1') 1.0 >>> atr.parse_data('1\n') 1.0 >>> atr.parse_data('?\n') nan r=N)rnanfloatrr r r rszNumericAttribute.parse_datacCs<|jd|jd}t|t|t|t||fSNg?r0sizerZnanminZnanmaxmeanstd)rdatanbfacr r r _basic_statsszNumericAttribute._basic_stats) rrr rr"rrrUrDr r r.r rEs   rEcs(eZdZfddZeddZZS)StringAttributecst|d|_dS)Nstring)r*rrrr.r r rs zStringAttribute.__init__cCs0|}|dtddkr(||SdSdS)z Parse the attribute line if it knows how. Returns the parsed attribute, or None. For string attributes, the attribute string would be like 'string'. NrWrIrr r r rs zStringAttribute.parse_attribute)rrr rr"rrDr r r.r rVs rVcsHeZdZfddZeddZeddZddZfd d Z Z S) DateAttributecs8t|||_||_d|_||_td|j|_dS)Ndater) r*r date_format datetime_unitrrr datetime64r)rrrZr[r.r r rs  zDateAttribute.__init__cCst|}|r|d}d}d|vr:|dd}d}n|dd}d}d|vrb|dd}d }d |vrz|d d }d }d |vr|d d}d}d|vr|dd}d}d|vr|dd}d}d|vsd|vrtd|durtd||fStddS)Nr0Zyyyyz%YYyyz%yZMMz%mMddz%dDZHHz%Hhmmz%Mr8ssz%SszZz6Date type attributes with time zone not supported, yetz"Invalid or unsupported date formatzInvalid or no date format)r_dater2r4rKreplacer6)r7r8patternr[r r r _get_date_format s<        zDateAttribute._get_date_formatcCsB|}|dtddkr:||\}}||||SdSdS) Parse the attribute line if it knows how. Returns the parsed attribute, or None. For date attributes, the attribute string would be like 'date '. NrY)rJrKr$rk)rrrattr_string_lowerrZr[r r r r1s  zDateAttribute.parse_attributecCsT|dd}|dkr*td|jStj||j}t|d|jSdS)r'"r=ZNaTzdatetime64[%s]N)rKrr\r[datetimestrptimerZZastype)rrZdate_strdtr r r rCs zDateAttribute.parse_datacstd|jS)Nr)r*r!rZr r.r r r!OszDateAttribute.__str__) rrr rrCrkr"rrr!rDr r r.r rXs  &  rXcs<eZdZfddZeddZddZfddZZS) RelationalAttributecs*t|d|_tj|_g|_d|_dS)N relational)r*rrrrr attributesdialectrr.r r rUs  zRelationalAttribute.__init__cCs0|}|dtddkr(||SdSdS)rlNrtrI)rrrrmr r r r\s z#RelationalAttribute.parse_attributecs|tttj}|d}g}|dD]4}t|j\_| t fdd|Dq.t |ddjDS)Nzunicode-escape cs g|]}j||qSr )rurr%rowrr r yr)z2RelationalAttribute.parse_data..cSsg|]}|j|jfqSr rrr&ar r r rz|r)) listrr$ruencodedecodesplitr3rvappendr5rarray)rrelemsescaped_stringZ row_tuplesrawr rxr rmszRelationalAttribute.parse_datacs$tdddd|jDS)Nz css|]}t|VqdSr )r>r|r r r r(r)z.RelationalAttribute.__str__..)r*r!joinrur r.r r r!~s zRelationalAttribute.__str__) rrr rr"rrr!rDr r r.r rsSs   rscCsDtttttf}|D] }|||}|dur|Sqtd|dS)Nzunknown attribute %s)r#rErXrVrsrr)rrZ attr_classesrattrr r r to_attributes  rcCs8ttdd}|dur4td}|jdkt_tj}|S)zL Checks if the bug https://bugs.python.org/issue30157 is unpatched. has_bugNz3, 'a'rn)getattrcsv_sniffer_has_bug_last_fieldcsvSniffersniff quotecharr)rrvr r r rs   rcCs4tr0d}dd|dfD]*}t|tjtjB}||}|rqDq||krPdS|j}t|dksfJ|d}|dd} || } |d d} || } |d d} t|| } td t | d | d t | d| d t | d| dt | dtj} t| |}| |_ | |vr$| |_ ||_ | |_dS)zT Workaround for the bug https://bugs.python.org/issue30157 if is unpatched. zG(?P[^\w\n"\'])(?P ?)(?P["\']).*?(?P=quote)(?:$|\n)zI(?P[^\w\n"\'])(?P ?)(?P["\']).*?(?P=quote)(?P=delim)zG(?:^|\n)(?P["\']).*?(?P=quote)(?P[^\w\n"\'])(?P ?)z-(?:^|\n)(?P["\']).*?(?P=quote)(?:$|\n)Nr0rquotedelimspacez((z)|^)\W*z[^z\n]*z\W*((z)|$))rrecompileDOTALL MULTILINEfindall groupindexr$boolescapesearchr delimiter doublequoteskipinitialspace) sniff_linerv delimitersZ right_regexrestrregexpmatchesrr8nrrr dq_regexprr r r %workaround_csv_sniffer_bug_last_fields>     J rcsd}tttdjdddkr6dd}tfdd|Ds`|d7}|durtj ||d}t |||d t t g|}||fS) Nz, r?rwc3s|]}|vVqdSr r )r&dliner r r(r)z"split_data_line..r)r)rrvr) rfield_size_limitrGctypesc_ulongvaluerKanyrrrnextreader)rrvrrryr rr r3s   r3cCs|}t|}|rj|d}t|r@t|\}}t|}qvt|r`t|\}}t|}qvt dn t d|t ||}| dkrt |||}||fS)aParse a raw string in header (e.g., starts by @attribute). Given a raw string attribute, try to get the name and type of the attribute. Constraints: * The first line must start with @attribute (case insensitive, and space like characters before @attribute are allowed) * Works also if the attribute is spread on multilines. * Works if empty lines or comments are in between Parameters ---------- attribute : str the attribute string. Returns ------- name : str name of the attribute value : str value of the attribute next : str next line to be parsed Examples -------- If attribute is a string defined in python as r"floupi real", will return floupi as name, and real as value. >>> from scipy.io.arff._arffread import tokenize_attribute >>> iterable = iter([0] * 10) # dummy iterator >>> tokenize_attribute(iterable, r"@attribute floupi real") ('floupi', 'real', 0) If attribute is r"'floupi 2' real", will return 'floupi 2' as name, and real as value. >>> tokenize_attribute(iterable, r" @attribute 'floupi 2' real ") ('floupi 2', 'real', 0) r0zmulti line not supported yetzFirst line unparsable: %srt) rK r_attributer2r4 r_comattrvaltokenize_single_commar r_wcomattrvaltokenize_single_wcommar6rrJread_relational_attribute)iterable attributeZsattrZmattrr7rtypeZ next_itemr r r tokenize_attributes *             rc Cstt|}|r`z |d}|d}Wqlty\}ztd|WYd}~qld}~00n td|||fSNr0rz Error while tokenizing attributez Error while tokenizing single %s)rr2r4rK IndexErrorr6valr8rrer r r r<s " rc Cstt|}|r`z |d}|d}Wqlty\}ztd|WYd}~qld}~00n td|||fSr)rr2r4rKrr6rr r r rKs " rcCsztd|jd}||snt|}|rdt|}|rVt||\}}|j|qlt d|qt |}qt |}|S)z4Read the nested attributes of a relational attributez^@[Ee][Nn][Dd]\s*z\s*$Error parsing line %s) rrrr2 r_headerlinerrrurr6r)ofileZrelational_attributer'Zr_end_relationalr8isattrrr r r rZs     rcCst|}t|rt|}qd}g}t|st|}|rt|}|rdt||\}}||qt|}|r~| d}n t d|t|}q$t|}q$||fS)z&Read the header of the iterable ofile.Nr0r) r r_commentr2 r_datametarrrr r_relationr4r6)rr'Zrelationrur8rrZisrelr r r read_headerps&           rc@s@eZdZdZddZddZddZdd Zd d Zd d Z dS)raSmall container to keep useful information on a ARFF dataset. Knows about attributes names and types. Examples -------- :: data, meta = loadarff('iris.arff') # This will print the attributes names of the iris.arff dataset for i in meta: print(i) # This works too meta.names() # Getting attribute type types = meta.types() Methods ------- names types Notes ----- Also maintains the list of attributes in order, i.e., doing for i in meta, where meta is an instance of MetaData, will return the different attribute names in the order they were defined. cCs||_dd|D|_dS)NcSsi|] }|j|qSr )rr|r r r r)z%MetaData.__init__..)r _attributes)rrelrr r r rszMetaData.__init__cCsjd}|d|j7}|jD]L}|d|d|j|j7}|j|jr\|dt|j|j7}|d7}q|S)Nz Dataset: %s  z 's type is z , range is %srw)rrrrr>rAr r r __repr__s   zMetaData.__repr__cCs t|jSr )iterrr r r r __iter__szMetaData.__iter__cCs|j|}|j|jfSr )rrr)rkeyrr r r __getitem__s zMetaData.__getitem__cCs t|jS)zReturn the list of attribute names. Returns ------- attrnames : list of str The attribute names. )r~rr r r r namesszMetaData.namescsfddjD}|S)zReturn the list of attribute types. Returns ------- attr_types : list of str The attribute types. csg|]}j|jqSr )rr)r&rr r r rzsz"MetaData.types..)r)rZ attr_typesr r r typess zMetaData.typesN) rrr __doc__rrrrrrr r r r rs  rcCsLt|dr|}nt|}zt|W||ur2|Sn||urF|0dS)a Read an arff file. The data is returned as a record array, which can be accessed much like a dictionary of NumPy arrays. For example, if one of the attributes is called 'pressure', then its first 10 data points can be accessed from the ``data`` record array like so: ``data['pressure'][0:10]`` Parameters ---------- f : file-like or str File-like object to read from, or filename to open. Returns ------- data : record array The data of the arff file, accessible by attribute names. meta : `MetaData` Contains information about the arff file such as name and type of attributes, the relation (name of the dataset), etc. Raises ------ ParseArffError This is raised if the given file is not ARFF-formatted. NotImplementedError The ARFF file has an attribute which is not supported yet. Notes ----- This function should be able to read most arff files. Not implemented functionality include: * date type attributes * string type attributes It can read files with numeric and nominal attributes. It cannot read files with sparse data ({} in the file). However, this function can read files with missing data (? in the file), representing the data points as NaNs. Examples -------- >>> from scipy.io import arff >>> from io import StringIO >>> content = """ ... @relation foo ... @attribute width numeric ... @attribute height numeric ... @attribute color {red,green,blue,yellow,black} ... @data ... 5.0,3.25,blue ... 4.5,3.75,green ... 3.0,4.00,red ... """ >>> f = StringIO(content) >>> data, meta = arff.loadarff(f) >>> data array([(5.0, 3.25, 'blue'), (4.5, 3.75, 'green'), (3.0, 4.0, 'red')], dtype=[('width', '>> meta Dataset: foo width's type is numeric height's type is numeric color's type is nominal, range is ('red', 'green', 'blue', 'yellow', 'black') readN)hasattropen _loadarffclose)frr r r rsF rc szt|\}Wn:tyJ}z"dt|}t||WYd}~n d}~00d}D]}t|trTd}qTt|}|r~tdtd fdd }t ||}t |dd D}||fS) Nz'Error while parsing header, error was: FTz*String attributes not supported yet, sorryrc3s\tt}d}|D]B}t|st|r.qt||\}tfdd|DVqdS)Ncsg|]}||qSr )rr%)rryr r rzcr)z0_loadarff..generator..)r~rrr2r_emptyr3r5)Zrow_iterrrrvrrni)ryr generatorIs z_loadarff..generatorcSsg|]}|j|jfqSr r{r|r r r rzgr)z_loadarff..)r) rr6r>r isinstancerVrNotImplementedErrorr$r~rr) rrrrBhasstrr}metarrSr rr r*s"     rcCs<|jd|jd}t|t|t|t||fSrNrO)rSrTr r r basic_statsnsrc Csr|j}|dks|dks|dkrbt|\}}}}t|d|d|dd|dd|dd|d n tt|dS)NrFrHintegerrr)rrprintr>)rtprSrminr-rQrRr r r print_attributess 4rcCsFt|\}}tt|jt|j|D]}t|||||q(dSr )rrr$rrPr)filenamerSrr'r r r test_weka|s   rF__main__r0)N)0rrpnumpyrrr__all__rZr_metarrrrrrr1rhrrOSErrorrrr r#rErVrXrsrrrr3rrrrrrrrrrrZ__test__rsysargvrr r r r sX           K>S3 2 #EJQD