+d%ddlmZ ddlZeejZn #e$rdZYnwxYwddlmZdZdZ dZ dZ d Z d d Z d ZdS) )VersionN)make_path_posixc|std|dr |dd}|stdt|S)Nzz/_get_parts_of_format_string..BsDDDrRDDDDr) endswithlen enumeraterformatsplitappendr Exception) resolved_string literal_texts format_specs_textbitsi literal_textbitr s r _get_parts_of_format_stringr(sG E DRB)5>>-2C#D#D)/]2.////0%crc* #CRC( $]33??< 2  ?5( r "Z"a"abn"o"oqqq\155JC ! C   !V ?  ?3LQ4GHH  E!K-0111kll+ ? ? ?L))A--?"7 Q"H"HKKKm |m 4555KK{lmm 4555EE),q1u*=>>> ?  E 4yy3DD\DDDEEEABBx Ks?DB F('F(c ddlm}|}d||D}d|D}|D]E}t||D] \}}|||!F|S)aH Reverse the string method format for a list of strings. Given format_string and resolved_strings, for each resolved string find arguments that would give ``format_string.format(**arguments) == resolved_string``. Each item in the output corresponds to a new column with the key setting the name and the values representing a mapping from list of resolved_strings to the related value. Parameters ---------- format_strings : str Format template string as used with str.format method resolved_strings : list List of strings with same pattern as format_string but with fields filled out. Returns ------- args : dict Dict of the form ``{field: [value_0, ..., value_n], ...}`` where values are in the same order as resolved_strings, so: ``format_sting.format(**{f: v[0] for f, v in args.items()}) == resolved_strings[0]`` Examples -------- >>> paths = ['data_2014_01_03.csv', 'data_2014_02_03.csv', 'data_2015_12_03.csv'] >>> reverse_formats('data_{year}_{month}_{day}.csv', paths) {'year': ['2014', '2014', '2015'], 'month': ['01', '02', '12'], 'day': ['03', '03', '03']} >>> reverse_formats('data_{year:d}_{month:d}_{day:d}.csv', paths) {'year': [2014, 2014, 2015], 'month': [1, 2, 12], 'day': [3, 3, 3]} >>> reverse_formats('data_{date:%Y_%m_%d}.csv', paths) {'date': [datetime.datetime(2014, 1, 3, 0, 0), datetime.datetime(2014, 2, 3, 0, 0), datetime.datetime(2015, 12, 3, 0, 0)]} >>> reverse_formats('{state:2}{zip:5}', ['PA19104', 'PA19143', 'MA02534']) {'state': ['PA', 'PA', 'MA'], 'zip': ['19104', '19143', '02534']} See also -------- str.format : method that this reverses reverse_format : method for reversing just one string using a pattern r Formatterc.g|]}|d |dS)rr)rr%s r rz#reverse_formats..}s%BBBAQqTB1Q4BBBrci|]}|gSrr)r field_names r z#reverse_formats..s 9 9 9zJ 9 9 9r)stringr+parsereverse_formatitemsr) format_stringresolved_stringsr+fmt field_namesargsr fieldvalues r reverse_formatsr;Gsb!     )++CCB=!9!9BBBK : 9[ 9 9 9D+&&*=/JJPPRR & &LE5 K  u % % % % & Krcddlm}ddlm}|}i}t|}t ||\}}}} t |siSt| D]/\} } | r(td || 0t|}t|||} tt ||D]\} \} }| r | dr | | | ||| <n|dtdvrt| | || <n|dtdvrt| | || <nO|ddkr$t| | d dd z || <n|| | ||| <#t"$r| | || <Y wxYw|S) a Reverse the string method format. Given format_string and resolved_string, find arguments that would give ``format_string.format(**arguments) == resolved_string`` Parameters ---------- format_string : str Format template string as used with str.format method resolved_string : str String with same pattern as format_string but with fields filled out. Returns ------- args : dict Dict of the form {field_name: value} such that ``format_string.(**args) == resolved_string`` Examples -------- >>> reverse_format('data_{year}_{month}_{day}.csv', 'data_2014_01_03.csv') {'year': '2014', 'month': '01', 'day': '03'} >>> reverse_format('data_{year:d}_{month:d}_{day:d}.csv', 'data_2014_01_03.csv') {'year': 2014, 'month': 1, 'day': 3} >>> reverse_format('data_{date:%Y_%m_%d}.csv', 'data_2016_10_01.csv') {'date': datetime.datetime(2016, 10, 1, 0, 0)} >>> reverse_format('{state:2}{zip:5}', 'PA19104') {'state': 'PA', 'zip': '19104'} See also -------- str.format : method that this reverses reverse_formats : method for reversing a list of strings using one pattern r)datetimer*z$Conversion not allowed. Found on {}.%rbcdoxXeEfFgGnNd)r=r0r+rzipr1anyrrrr( startswithstrptimelistr float format_fieldr)r4r r=r+r6r8r!r7r" conversionsr% conversionr$r.r s r r2r2sOL"!!!!!      )++C D$M22M=@=AYAY>> path_to_glob('{year}/{month}/{day}.csv') '*/*/*.csv' >>> path_to_glob('data/{year:4}{month:02}{day:02}.csv') 'data/*.csv' >>> path_to_glob('data/*.csv') 'data/*.csv' rr*rN*)r0r+ isinstancestrr1)pathr+r6globprev_field_namer&r._s r path_to_globrSs4!     )++C dC   DO*-))D//%%& j!Q   < ?  CKD$ Krcddlm}t|tsdS||}|rK|d}|r4t d|D}||d}|S)a Remove source information from path when using chaching Returns None if path is not str Parameters ---------- path : str Path to data optionally containing format_strings metadata : dict, optional Extra arguments to the class, contains any cache information Returns ------- pattern : str Pattern style path stripped of everything to the left of cache regex. r)strip_protocolNcachec3rK|]2}|ddk|dV3dS)argkeyurlpathregexN)get)rcs r z"path_to_pattern..!s@WWA!%%//Y:VWwWWWWWWrr) fsspec.corerUrMrNr[nextr)rOmetadatarUpatternrVrZs r path_to_patternrbs$+***** dC nT""G/ W%%  /WWWWWWWEmmE**2.G Nrcddlmddlmmdfdt dDS)Nr)choice) ascii_lettersdigitsrc,g|]}zSrr)rnrerdrfs r rz!unique_string..*s(EEEqFF=6122EEEr)randomrdr0rerfjoinrange)rerdrfs@@@r unique_stringrm&sb,,,,,,,, 77EEEEEEE!HHEEE F FFrr)packaging.versionrdask __version__ DASK_VERSION ImportErrorutilsrrr(r;r2rSrbrmrrr rts&%%%%%KKK74+,,LLLLL######'''T>>>BNNNb***Z@GGGGGs ''