o ;Di=@s0ddlmZmZmZddlZddlZddlmZmZddl m Z ddl m Z m Z ddlmZddlmZ  d%d d Zee _Gd d d e Zd dZ  d&ddZee _Gddde ZddZd'ddZee _ddZee _Gddde ZddZddZee _d'dd Zee _Gd!d"d"e Zd#d$ZdS)()absolute_importprint_functiondivisionN)next text_type) ArgumentError)Table asindices)TransformError)convertFc Cst|||||||dS)a Add one or more new fields with values captured from an existing field searched via a regular expression. E.g.:: >>> import petl as etl >>> table1 = [['id', 'variable', 'value'], ... ['1', 'A1', '12'], ... ['2', 'A2', '15'], ... ['3', 'B1', '18'], ... ['4', 'C12', '19']] >>> table2 = etl.capture(table1, 'variable', '([A-Z,a-z]+)([0-9]+)', ... ['treat', 'time']) >>> table2 +-----+-------+-------+------+ | id | value | treat | time | +=====+=======+=======+======+ | '1' | '12' | 'A' | '1' | +-----+-------+-------+------+ | '2' | '15' | 'A' | '2' | +-----+-------+-------+------+ | '3' | '18' | 'B' | '1' | +-----+-------+-------+------+ | '4' | '19' | 'C' | '12' | +-----+-------+-------+------+ >>> # using the include_original argument ... table3 = etl.capture(table1, 'variable', '([A-Z,a-z]+)([0-9]+)', ... ['treat', 'time'], ... include_original=True) >>> table3 +-----+----------+-------+-------+------+ | id | variable | value | treat | time | +=====+==========+=======+=======+======+ | '1' | 'A1' | '12' | 'A' | '1' | +-----+----------+-------+-------+------+ | '2' | 'A2' | '15' | 'A' | '2' | +-----+----------+-------+-------+------+ | '3' | 'B1' | '18' | 'B' | '1' | +-----+----------+-------+-------+------+ | '4' | 'C12' | '19' | 'C' | '12' | +-----+----------+-------+-------+------+ By default the field on which the capture is performed is omitted. It can be included using the `include_original` argument. The ``fill`` parameter can be used to provide a list or tuple of values to use if the regular expression does not match. The ``fill`` parameter should contain as many values as there are capturing groups in the regular expression. If ``fill`` is ``None`` (default) then a ``petl.transform.TransformError`` will be raised on the first non-matching value. ) newfieldsinclude_originalflagsfill) CaptureView)tablefieldpatternr r rrrM/var/www/Datamplify/venv/lib/python3.10/site-packages/petl/transform/regex.pycaptures 7rc@s"eZdZ  dddZddZdS) rNFrcC.||_||_||_||_||_||_||_dSN)sourcerrr r rr)selfrrrr r rrrrr__init__R zCaptureView.__init__cC"t|j|j|j|j|j|j|jSr) itercapturerrrr r rrrrrr__iter__\s zCaptureView.__iter__NFrN__name__ __module__ __qualname__rr rrrrrP  rc#s6t|}t||}zt|} Wn tyg} Ynwttt| } t|t r2|t | kr2|n|| vr<| |nt dt| } |sK| ||rR| |t| V|D]?} | } |rft| }n fddt| D}|| }|dur|dur||ntd| |f||t|VqYdS)N1field invalid: must be either field name or indexcg|] \}}|kr|qSrr.0iv field_indexrr zitercapture..z!value %r did not match pattern %r)iterrecompiler StopIterationlistmapr isinstanceintlenindexrremoveextendtuple enumeratesearchr groups)rrrr r rritproghdrfldsouthdrrowvalueout_rowmatchrr-rrbsF           rcCst|||||||S)a? Add one or more new fields with values generated by splitting an existing value around occurrences of a regular expression. E.g.:: >>> import petl as etl >>> table1 = [['id', 'variable', 'value'], ... ['1', 'parad1', '12'], ... ['2', 'parad2', '15'], ... ['3', 'tempd1', '18'], ... ['4', 'tempd2', '19']] >>> table2 = etl.split(table1, 'variable', 'd', ['variable', 'day']) >>> table2 +-----+-------+----------+-----+ | id | value | variable | day | +=====+=======+==========+=====+ | '1' | '12' | 'para' | '1' | +-----+-------+----------+-----+ | '2' | '15' | 'para' | '2' | +-----+-------+----------+-----+ | '3' | '18' | 'temp' | '1' | +-----+-------+----------+-----+ | '4' | '19' | 'temp' | '2' | +-----+-------+----------+-----+ By default the field on which the split is performed is omitted. It can be included using the `include_original` argument. ) SplitView)rrrr r maxsplitrrrrsplits rLc@s"eZdZ  dddZddZdS) rJNFrcCrr)rrrr r rKr)rrrrr r rKrrrrrrzSplitView.__init__cCrr) itersplitrrrr r rKrrrrrr s zSplitView.__iter__NFrrr"rrrrrJr&rJc#s t|}t||}zt|} Wn tyg} Ynwttt| } t|t r6|t | kr6|| }n|| vr@| |nt dt| } |sO| ||rV| |t| V|D]&} | } |rjt| }n fddt| D}||| |t|Vq]dS)Nr'cr(rrr)r-rrr/r0zitersplit..)r1r2r3rr4r5r6rr7r8r9r:rr;r<r=r>rL)rrrr r rKrrArBrCrDrErFrGrHrr-rrMs:          rMcs(t||fdd}t|||S)z Convenience function to convert values under the given field using a regular expression substitution. See also :func:`re.sub`. csj|dS)N)count)sub)r,rOrBreplrrszsub..)r2r3r )rrrrRrOrconvrrQrrPs  rPcOsVt|dkr d}|d}nt|dkr|d}|d}ntdt||fd|i|S)ai Perform a regular expression search, returning rows that match a given pattern, either anywhere in the row or within a specific field. E.g.:: >>> import petl as etl >>> table1 = [['foo', 'bar', 'baz'], ... ['orange', 12, 'oranges are nice fruit'], ... ['mango', 42, 'I like them'], ... ['banana', 74, 'lovely too'], ... ['cucumber', 41, 'better than mango']] >>> # search any field ... table2 = etl.search(table1, '.g.') >>> table2 +------------+-----+--------------------------+ | foo | bar | baz | +============+=====+==========================+ | 'orange' | 12 | 'oranges are nice fruit' | +------------+-----+--------------------------+ | 'mango' | 42 | 'I like them' | +------------+-----+--------------------------+ | 'cucumber' | 41 | 'better than mango' | +------------+-----+--------------------------+ >>> # search a specific field ... table3 = etl.search(table1, 'foo', '.g.') >>> table3 +----------+-----+--------------------------+ | foo | bar | baz | +==========+=====+==========================+ | 'orange' | 12 | 'oranges are nice fruit' | +----------+-----+--------------------------+ | 'mango' | 42 | 'I like them' | +----------+-----+--------------------------+ The complement can be found via :func:`petl.transform.regex.searchcomplement`. Nrz$expected 1 or 2 positional argumentsr)r9r SearchView)rargskwargsrrrrrr?s (   r?c@seZdZdddZddZdS) rWNrFcC"||_||_||_||_||_dSr)rrrr complement)rrrrrr[rrrr2  zSearchView.__init__cCt|j|j|j|j|jSr) itersearchrrrrr[rrrrr 9szSearchView.__iter__)NrFr"rrrrrW0  rWc #st||t|}zt|}Wn tyYdSwttt|}t|V|dur3fdd}n#t ||} t | dkrJ| dfdd}n t j | fdd}|sh|D] } || ret| VqZdS|D] } || sut| VqjdS)Ncstfdd|DS)Nc3|] }t|VqdSrr?rr*r,rBrr J/itersearch....anyrrcrrrSJszitersearch..rUrcst|Srrari)r:rBrrrSOscstfdd|DS)Nc3r`rrarbrcrrrdRrerfrgri)getvalsrBrrrSRs) r2r3r1rr4r5r6rr=r r9operator itemgetter) rrrrr[rArCrDtestindicesrFr)rkr:rBrr^>s:         r^cOst|g|Rddi|S)aB Perform a regular expression search, returning rows that **do not** match a given pattern, either anywhere in the row or within a specific field. E.g.:: >>> import petl as etl >>> table1 = [['foo', 'bar', 'baz'], ... ['orange', 12, 'oranges are nice fruit'], ... ['mango', 42, 'I like them'], ... ['banana', 74, 'lovely too'], ... ['cucumber', 41, 'better than mango']] >>> # search any field ... table2 = etl.searchcomplement(table1, '.g.') >>> table2 +----------+-----+--------------+ | foo | bar | baz | +==========+=====+==============+ | 'banana' | 74 | 'lovely too' | +----------+-----+--------------+ >>> # search a specific field ... table3 = etl.searchcomplement(table1, 'foo', '.g.') >>> table3 +------------+-----+---------------------+ | foo | bar | baz | +============+=====+=====================+ | 'banana' | 74 | 'lovely too' | +------------+-----+---------------------+ | 'cucumber' | 41 | 'better than mango' | +------------+-----+---------------------+ This returns the complement of :func:`petl.transform.regex.search`. r[T)r?)rrXrYrrrsearchcomplement_s$rpcCst|||||S)a+ Split a field into multiple rows using a regular expression. E.g.: >>> import petl as etl >>> table1 = [['name', 'roles'], ... ['Jane Doe', 'president,engineer,tailor,lawyer'], ... ['John Doe', 'rocket scientist,optometrist,chef,knight,sailor']] >>> table2 = etl.splitdown(table1, 'roles', ',') >>> table2.lookall() +------------+--------------------+ | name | roles | +============+====================+ | 'Jane Doe' | 'president' | +------------+--------------------+ | 'Jane Doe' | 'engineer' | +------------+--------------------+ | 'Jane Doe' | 'tailor' | +------------+--------------------+ | 'Jane Doe' | 'lawyer' | +------------+--------------------+ | 'John Doe' | 'rocket scientist' | +------------+--------------------+ | 'John Doe' | 'optometrist' | +------------+--------------------+ | 'John Doe' | 'chef' | +------------+--------------------+ | 'John Doe' | 'knight' | +------------+--------------------+ | 'John Doe' | 'sailor' | +------------+--------------------+ ) SplitDownViewrrrrKrrrr splitdowns"rsc@seZdZdddZddZdS)rqrcCrZrrr)rrrrrKrrrrrr\zSplitDownView.__init__cCr]r) itersplitdownrrrrKrrrrrr szSplitDownView.__iter__Nrrr"rrrrrqr_rqc #st||}t|}zt|}Wn tyYdSwttt|}t|t r5|t |kr5||}n||vr?| |nt dt |V|D]!} || |D]t fddtt |DVqVqJdS)Nr'c3s$|] }|kr n|VqdSrr)r*r+r.rFr,rrrds"z itersplitdown..)r2r3r1rr4r5r6rr7r8r9r:rr=rLrange) rrrrKrrBrArCrDrGrrvrrts,      &rtr!rNru) __future__rrrr2rl petl.compatrr petl.errorsrpetl.util.baserr petl.transform.basicsr petl.transform.conversionsr rrrrLrJrMrPr?rWr^rprsrqrtrrrrs@    >, # & 3!' %