f]ddlmZmZmZddlZddlZddlZddlZddlZddl m Z ddl m Z ddl Z e jeZGddeZGddeZGdd eZGd d eZGd d eZGddeZy))divisionprint_functionunicode_literalsN) unicode_strcTeZdZdZddZdZdZdZdZdZ d Z d Z d Z d Z d Zy) SpellCheckerdNc.||_d|_i|_yN) _language_db_backend_cached_queries)self language_dbs 6/usr/lib/python3/dist-packages/Onboard/SpellChecker.py__init__zSpellChecker.__init__(s' !cT|.|jr|jjd|_nh|dk(rt}nt}|jrt |j|k(s2|jr|jj||_|j y)z' Switch spell check backend on the fly Nr)r stophunspell aspell_cmdtypeinvalidate_query_cache)rbackend_classs r set_backendzSpellChecker.set_backend-sz ?}} ""$ DM!|!#== &&0==MM&&( &  ##%rcd}|j|}|jr||jjk(ss|jj|r|jj |d}n9t j djt|j||j|S)NFTz0No matching dictionaries for '{backend}' {dicts}rdicts) _find_matching_dictsr get_active_dict_idsrstart_loggerinfoformatrr)rdict_idssuccessidss r set_dict_idszSpellChecker.set_dict_idsAs''1 ==dmm7799 MM    ##C( O$fT$---@+3%56 ##%rcbg}|D]'}|j|}|s|j|)|Sr )_find_matching_dictappend)rr&resultsdict_idids rr z!SpellChecker._find_matching_dictsQs< #G))'2Br" #rc|j}d}||vr|}|S|jdd}||vr|}|S|jrI|jj|\}}||vr|}|S|jj |}|r||vr|}|S)z Try to match up the given dict_id with the available dictionaries. Look for alternatives if there is no direct match. _-)get_supported_dict_idsreplacer split_lang_idget_main_lang_id)rr.available_dict_idsresultalt_id lang_code country_codelang_ids rr+z SpellChecker._find_matching_dictYs "88: ( (F" __S#.F++ ""*.*;*;*I*I'*R' < 22&F #//@@KG7.@#@!( rcd}g}|jr2|j|}|D]}|dd|kDr||fS|d}|d}||fS)z Return spelling suggestions for word. Multiple result sets may be returned, as the spell checkers may return more than one result for certain tokens, e.g. before and after hyphens. Nrr query_cached)rword caret_offsetspan suggestionsr-r9s rfind_correctionszSpellChecker.find_correctionsvsw ==''-G" !!9Q<,.[  %Qi ay  ! [  rcrg}|jr#|j|}|Dcgc]}|d }}|Scc}w)z Return misspelled spans (begin- to end-index) inside word. Multiple result sets may be returned, as the spell checkers may return more than one result for certain tokens, e.g. before and after hyphens. rr@)rrBspansr-r9s rfind_incorrect_spansz!SpellChecker.find_incorrect_spanssC ==''-G.556VAY5E5 6s 4c|jj|}|t|j}||jk\rc|dz}tj dj ||t|jjd}t||d|_|j|}d|g}||j|<tj|d<|dS) zF Return cached query or ask the backend if necessary. Nz,shrinking query cache from {} to {} entries.c|ddS)Nr?r)xs rz+SpellChecker.query_cached..s1ar)keygrr?) rgetlenMAX_QUERY_CACHE_SIZEr#debugr%sorteditemsdictquerytime)rrBrXsizenew_sizequeriesr-s rrAzSpellChecker.query_cacheds$$((. =t++,Dt00019 L%vdH57!!5!5!;!;!='8:'+GHI,>'?$jj&G'NE).D  &99;aQxrc8|jj|Sr )r rX)rrBs rrXzSpellChecker.querys}}""4((rci|_yr )rrs rrz#SpellChecker.invalidate_query_caches !rc6|jjSr )r r4r_s rr4z#SpellChecker.get_supported_dict_idss}}3355rr )__name__ __module__ __qualname__rSrrr)r r+rFrIrArXrr4rMrrrr%s?" &( :!, 8)"6rrc>eZdZdZd dZd dZdZdZdZdZ d Z y) SCBackendz2 Abstract base class of all spellchecker backends Nc d|_d|_yr ) _active_dicts_prr&s rrzSCBackend.__init__s!rcp||_tjdjt ||y)Nzstarting '{backend}' {dicts}r)rgr#r$r%rris rr"zSCBackend.starts,% 3fT$Zxf@ Brc|jr6tjdjt |d|_yy)Nzstopping '{backend}')r) is_runningr#r$r%rrgr_s rrzSCBackend.stops8 ??  LL/ &d&4 6!%D  rctSr NotImplementedErrorr_s rrlzSCBackend.is_runnings "$$rcg}|jr!|jjd|_|jrt|tdk(rd|zdz}|jd}nd|zdz}|jjj ||jjj  |jjjjd}|j}|s |S|ddd k(r|jd }|d j}t|d dz }|t|dz}|||dg} |djjd } |j| | g|dddk(rc|jd }|d j}t|ddz }|t|dz}|||dg} g} |j| | g@|Sz Query for spelling suggestions. Text may contain one or more words. Each word generates its own list of suggestions. The spell checker backend decides about word boundaries. Nr1^ UTF-8^ r?&:rz, #rKrhpollrencodestdinwriteflushstdoutreadlinedecodestripsplitintrRr, rtextr-linessectionsabeginendrDrEs rrXzSCBackend.query 77477<<>1DG 77DzT"X%TzD({{7+d{U* GGMM   % GGMM   !GGNN++-44W=GGI$#Ra5C< wws|H  ))+A!IME!C!I-C!3!-D"*1+"3"3"5";";D"AKNND+#67Ra5C< wws|H  ))+A!IME!C!I-C!3!-D"$KNND+#67),rct6 Return raw supported dictionary ids. rnr_s rr4z SCBackend.get_supported_dict_ids "##rc|jSz/ Return active dictionary ids. rgr_s rr!zSCBackend.get_active_dict_ids!!!rr ) rarbrc__doc__rr"rrlrXr4r!rMrrreres*<B & %/b$ "rreceZdZdZd dZd fd ZfdZdZejdejejzZ dZ dZd Zed Zd ZxZS) ra Hunspell backend using the C API. Doctests: # known word >>> sp = hunspell(["en_US"]) >>> sp.query("test") [] # unknown word >>> sp = hunspell(["en_US"]) >>> sp.query("jdaskljasd") # doctest: +ELLIPSIS [[... cfd|_tj|||r|j|yyr ) _osk_hunspellrerr"ris rrzhunspell.__init__$s.!4*  JJx  rctt| ||r|j|d\}}tj dj |||rY tj|||_ tj dj |jjyyy#t$r2}tjdt|zd|_ Yd}~yd}~wwxYw)Nrzusing hunspell files '{}', '{}'zdictionary encoding '{}'z#failed to create hunspell backend: )superrr"_search_dict_filesr#r$r%oskHunspellr get_encoding Exceptionerrorr)rr&dicaffe __class__s rr"zhunspell.start*s h#H- ..x{;HC LL:#VC- /.),c3)?D&LL!;"(&););)H)H)J"KM !.MM"G"-a.#12)-D&&.sAB00 C+9(C&&C+chtt| |jrd|_d|_yyr )rrrrlrrgrrs rrz hunspell.stop;s/ h"$ ?? !%D !%D  rc|jduSr )rr_s rrlzhunspell.is_runningAs%%--rz[^-_\s]+cg}|jr|jj|}|D]}|j}|j }|j }|||g} |jj |dk(r7t|jj|} |j|| g|S#t$r|j|ggYwxYw)u Query for spelling suggestions. Text may contain one or more words. Each word generates its own list of suggestions. Word boundaries are chosen to match the behavior of the hunspell command line tool, i.e. split at '-','_' and whitespace. Doctests: # one prediction token, two words for the spell checker >>> sp = hunspell(["en_US"]) >>> q = sp.query("conter_trop") >>> q # doctest: +ELLIPSIS [[[0, 6, 'conter'], [... >>> len(q) 2 # dashes >>> sp = hunspell(["en_US"]) >>> sp.query("ubuntu-system") # doctest: +ELLIPSIS [[[0, 6, 'ubuntu'], ['Ubuntu', 'Kubuntu', 'urbanite', 'urbanity']]] # unrecognized word returns error span with zero choices (# mark) >>> q = sp.query("ἄναρχος") >>> q # doctest: [[[0, 7, 'ἄναρχος'], []]] >>> len(q) 1 # dictionaries come in various encodings, spot-check German >>> sp = hunspell(["de_DE"]) >>> sp.query("Frühstück") [] >>> sp.query("Früstück") # doctest: +ELLIPSIS [[[0, 8, 'Früstück'], ['Frühstück', ... # dictionaries come in various encodings, spot-check Spanish >>> sp = hunspell(["es_ES"]) >>> sp.query("situación") [] >>> sp.query("situción") # doctest: +ELLIPSIS [[[0, 8, 'situción'], ['situación', ... # dictionaries come in various encodings, spot-check Portuguese >>> sp = hunspell(["pt_PT"]) >>> sp.query("questão") [] >>> sp.query("quetão") # doctest: +ELLIPSIS [[[0, 6, 'quetão'], ['questão', ... # dictionaries come in various encodings, spot-check French >>> sp = hunspell(["fr_FR"]) >>> sp.query("garçon") [] >>> sp.query("garçoon") # doctest: +ELLIPSIS [[[0, 7, 'garçoon'], ['garçon', ... # dictionaries come in various encodings, spot-check Italian >>> sp = hunspell(["it_IT"]) >>> sp.query("parlerò") [] >>> sp.query("parllerò") # doctest: +ELLIPSIS [[[0, 8, 'parllerò'], ['parlerò', ... # dictionaries come in various encodings, spot-check Russian >>> sp = hunspell(["ru_RU"]) >>> sp.query("ОБЕДЕННЫЙ") # dinner [] >>> sp.query("ОБЕДЕНЫЙ") # doctest: +ELLIPSIS [[[0, 8, 'ОБЕДЕНЫЙ'], ['ОБЕДЕННЫЙ', ... # dictionaries come in various encodings, spot-check Greek >>> sp = hunspell(["el_GR"]) >>> sp.query("Ωκεανού") [] >>> sp.query("Ωκεαανού") # doctest: +ELLIPSIS [[[0, 8, 'Ωκεαανού'], ['Ωκεανού', ... r) r SPLITWORDSfinditergroupr"rspelllistsuggestr,UnicodeEncodeError) rrr-matchesmatchrBrrrDrEs rrXzhunspell.queryFs^   oo..t4G  /{{}  sD)/))//5:&*4+=+=+E+Ed+K&L k':; /  */NND":. /s%AB>>CCcg}g}|j}|D]d}tjj|dtjjzdz}|j t j |f|D]r}tjjtjj|\}}|jjdrb|j|t|S)aO Return raw supported dictionary ids. They may not all be valid language ids, e.g. en-GB instead of en_GB for myspell dicts. Doctests: # Just check it does anything at all >>> sp = hunspell(["en_US"]) >>> sp.get_supported_dict_ids() # doctest: +ELLIPSIS ['... *rhyph) _get_dict_pathsospathjoinextsepextendglobsplitextbasenamelower startswithr,)rr& filenamespathsrfnr=_exts rr4zhunspell.get_supported_dict_idss $$& ,DdC"''..$85$@AB   TYYr] + ,  )BGG,,RWW-=-=b-ABMGT==?--f5( ) rcv|j}|j||d}|j||d}||fS)z9 Locate dictionary and affix files for the given dict_id rr)r_find_file_in_paths)rr.rrrs rrzhunspell._search_dict_filessA$$&&&ugu=&&ugu=Cxrc|D]c}tjj||tjjz|z}tjj |sa|cSy)z+ Search for a file name in multiple paths. N)rrrrexists)rname extensionrfiles rrzhunspell._find_file_in_pathssS D77<<dRWW^^&;i&GHDww~~d#  rcg}tjj}gd}gd}gd}|jdtjd}|r |j |j ||j |tjd}|r"|j||j ||j ||S)z Return all paths where dictionaries may be located. Path logic taken from the source of the hunspell command line tool. )z/usr/share/hunspellz/usr/share/myspellz/usr/share/myspell/dictsz/Library/Spelling)z.openoffice.org/3/user/wordbookz.openoffice.org2/user/wordbookz .openoffice.org2.0/user/wordbookzLibrary/Spelling) z+/opt/openoffice.org/basis3.0/share/dict/oooz//usr/lib/openoffice.org/basis3.0/share/dict/ooozP/opt/openoffice.org2.4/share/dict/ooo,/usr/lib/openoffice.org2.4/share/dict/ooo,z%/opt/openoffice.org2.3/share/dict/oooz)/usr/lib/openoffice.org2.3/share/dict/oooz%/opt/openoffice.org2.2/share/dict/oooz)/usr/lib/openoffice.org2.2/share/dict/ooozO/opt/openoffice.org2.1/share/dict/ooo,/usr/lib/openoffice.org2.1/share/dict/oooz%/opt/openoffice.org2.0/share/dict/oooz)/usr/lib/openoffice.org2.0/share/dict/ooo.DICPATHHOME)rrpathsepr,getenvrr)rrrLIBDIRS USEROOODIRSOOODIRSdicpathhomes rrzhunspell._get_dict_pathss ''//!  9  S))I&  LLw/ 0 Wyy   LL  LL % W rr )rarbrcrrr"rrlrecompileUNICODEDOTALLrrXr4r staticmethodrr __classcell__rs@rrrsg ! ."& .KBII)=>JcJ:.rrcFeZdZdZdfd ZfdZdZdZdZdZ xZ S) SCBackend_cmdz. Abstract base class of command line backends c:tt| |d|_yr )rrrrh)rr&rs rrzSCBackend_cmd.__init__ s mT+H5rctt| t|j r<|j j |j jd|_yyr )rrrr&rlrh terminatewaitrs rrzSCBackend_cmd.stopsE mT+H5 ??  GG    GGLLNDG rc|jduSr )rhr_s rrlzSCBackend_cmd.is_runnings77d?"rcg}|jr!|jjd|_|jrt|tdk(rd|zdz}|jd}nd|zdz}|jjj ||jjj  |jjjjd}|j}|s |S|ddd k(r|jd }|d j}t|d dz }|t|dz}|||dg} |djjd } |j| | g|dddk(rc|jd }|d j}t|ddz }|t|dz}|||dg} g} |j| | g@|Srqr{rs rrXzSCBackend_cmd.queryrrctrrnr_s rr4z$SCBackend_cmd.get_supported_dict_idsKrrc|jSrrr_s rr!z!SCBackend_cmd.get_active_dict_idsQrrr ) rarbrcrrrrlrXr4r!rrs@rrr s%8#/b$ "rrc2eZdZdZddZdfd ZdZxZS) hunspell_cmda Hunspell backend, using the "hunspell" command. Doctests: # known word >>> sp = hunspell_cmd(["en_US"]) >>> sp.query("test") [] # unknown word >>> sp = hunspell_cmd(["en_US"]) >>> sp.query("jdaskljasd") # doctest: +ELLIPSIS [[... cXtj|||r|j|yyr rerr"ris rrzhunspell_cmd.__init__g&4*  JJx  rc tt| |gd}|r|ddj|gz } t j |tj tj d|_|jjjy#t$r@}tjtddj||d|_Yd}~yd}~wwxYw)N)r-az-irtz-d,Tr~r close_fdsFailed to execute '{}', {} )rrr"r subprocessPopenPIPErhrrOSErrorr#r_formatrr&argsrrs rr"zhunspell_cmd.startls lD'10  T388H-. .D  &&t-7__.8oo157DG GGNN # # %  MM'">HHTNA/ 0DGG AB C6CCc xg}ddg} tj|tjtjtjd}|jd\}}d}|j dj dD]n}|rVd |vr|St jj|}|jjd s|j||jd smd}p |S#t$r:} tjtd d j!|| Yd} ~ |Sd} ~ wwxYw)z Return raw supported dictionary ids. They may not all be valid language ids, e.g. en-GB for myspell dicts. rz-DT)r~rstderrrr1Frtrs/rzAVAILABLE DICTIONARIESrrN)rrr communicaterrrrrrrr,rr#rrr) rr&rpouterrin_dictsrr=rs rr4z#hunspell_cmd.get_supported_dict_ids~s" D! 0  '1(2(2+/ 1A }}R(HCH 7+11$7 $$;!gg..t4G"==?55f= 0??#;<#H $  0 MM'">HHTNA/ 0 0 0s%BC6A C6/C66 D9?/D44D9r rarbrcrrr"r4rrs@rrrXs ! $"rrc2eZdZdZddZdfd ZdZxZS)rz Aspell backend. Doctests: # known word >>> sp = aspell_cmd(["en_US"]) >>> sp.query("test") [] # unknown word >>> sp = aspell_cmd(["en_US"]) >>> sp.query("jdaskljasd") # doctest: +ELLIPSIS [[... cXtj|||r|j|yyr rris rrzaspell_cmd.__init__rrc tt| |ddg}|r|ddj|gz } t j |tj tj d|_|jjjy#t$r@}tjtddj||d|_Yd}~yd}~wwxYw) Naspellrz-lrTrrr)rrr"rrrrrhrrrr#rrrs rr"zaspell_cmd.starts j$%h/$  T388H-. .D  &&t-7__.8oo157DG GGNN # # %  MM'">HHTNA/ 0DGG rc 0g}gd} tj|jdjd}|Dcgc]}|s| c}S#t$r9}t j tddj||Yd}~Pd}~wwxYwcc}w)r)r dumprrtrsrrN) r check_outputrrrr#rrr)rr&rrr/s rr4z!aspell_cmd.get_supported_dict_idss* 0!..t4!'t  &,r,, 0 MM'">HHTNA/ 0 0 0-s"3ABB B/B  Br rrs@rrrs ! $ -rr) __future__rrrrrrYrr Onboard.utilsr Onboard.oskrlogging getLoggerrar#objectrrerrrrrMrrrs*BA  % '  H %Z66Z6zP"P"fqyqhL"IL"^H=HV2-2-r