@'hddlZddlZddlZddlZddlmZddlmZddlm Z ddl m Z ddl m Z mZmZddlmZmZmZmZmZmZmZddlmZddlZdd lmZdd lmZmZdd l m!Z!m"Z"m#Z#dd l$m%Z%dd l&m'Z'ddl(m)Z)ddl*m+Z+ddl,m-Z-ddl.m/Z/m0Z0m1Z1m2Z2m3Z3m4Z4m5Z5ee6e7e fZ8dZ9dZ:dZ;dZGdde'Z?y)N)asynccontextmanager)deepcopy)Enum)BufferedIOBase)PathPurePath PurePosixPath)AnyAsyncGeneratorDictListOptionalTupleUnion)urlparse)AbstractFileSystem) asyncio_runrun_jobs)Field PrivateAttrfield_validator)DEFAULT_BASE_URL)BasePydanticReader)get_default_fs)Document) JobResult)SUPPORTED_FILE_TYPES ResultType ParsingModeFailedPageModenest_asyncio_errnest_asyncio_msgmake_api_requestz --- z./api/parsing/job/{job_id}/result/{result_type}z/api/parsing/job/{job_id}z/api/parsing/uploadbase_urlorganization_id project_idreturnc i}|r||d<|r||d<|r=|dzdj|jDcgc] \}}|d|c}}zS|Scc}}w)Nr%r&?&=)joinitems)r$r%r& query_paramskvs f/home/kushmeetdev/Regenta/Chatbot/venv/lib/python3.12/site-packages/llama_cloud_services/parse/base.py build_urlr2-slL*9 &'%/ \"#~ASASAU)VAQCq*)V WWW O*WsA ceZdZdZdZdZdZy)BackoffPatternzBackoff pattern for polling.constantlinear exponentialN)__name__ __module__ __qualname____doc__CONSTANTLINEAR EXPONENTIALr1r4r4<s&H FKr@r4c(eZdZUdZedddZeed<eedZ eed <ed d Z e eed <ed d Z e eed<eddZ eed<eej dZeed<eddZeed<ed dZe ej*ed<eddZeed<eddZeed<eddd d!"Zeed#<eej6d$Zeed%<edd&Zeed'<edd(Zeed)<edd*Zeed+<ed,d-Z e eed.<ed,d/Z!e eed0<ed,d1Z"e eed2<ed d3Z#e eed4<ed,d5Z$e eed6<ed,d7Z%e eed8<ed d9Z&e eed:<ed d;Z'e eed<<ed d=Z(e eed><ed d?Z)e eed@<ed dAZ*e eedB<ed dCZ+e eedD<ed dEZ,e e-edF<ed dGZ.e e-edH<ed dIZ/e e-edJ<ed dKZ0e e-edL<ed,dMZ1e eedN<ed,dOZ2e eedP<ed,dQZ3e eedR<ed,dSZ4e eedT<ed,dUZ5e eedV<ed,dWZ6e eedX<ed,dYZ7e eedZ<ed,d[Z8e eed\<ed,d]Z9e eed^<ed,d_Z:e eed`<ed,daZ;e eedb<ed,dcZe eedh<ed,diZ?e eedj<ed dkZ@e eedl<ed,dmZAe eedn<ed doZBe e-edp<ed dqZCe e-edr<edsdtZDe eedu<ed dvZEe eedw<ed dxZFe eedy<ed,dzZGe eed{<ed d|ZHe eed}<ed d~ZIe eed<ed,dZJe eed<ed dZKe e-ed<ed dZLe eed<ed dZMe eed<ed dZNe eed<ed dZOe ePeQefed<ed,dZRe eed<ed dZSe eed<ed,dZTe eed<ed dZUe eVed<ed dZWe eed<ed dZXe eed<ed,dZYe eed<ed,dZZe eed<ed,dZ[e eed<ed,dZ\e eed<ed,dZ]e eed<ed,dZ^e eed<ed,dZ_e eed<ed dZ`e eed<ed dZae eed<ed dZbe eed<ed dZce eed<ed,dZde eed<ed dZee eed<ed dZfe eed<ed dZge eed<ed dZhe eed<ed dZie eed<ed dZje eed<ed dZke eed<ed dZle eed<ed dZme eed<ed,d¬Zne eed<ed dĬZoe eed<ed,dƬZpe eed<eddȬZqe eed<ed,dʬZre eed<esdddͫetdedefdЄZuesd ddͫetdedefdфZvewd d,ҫZxePej*d fed<eydej*fdԄZze{de|ej*d ffdՄZ}de~defdׄZde~defd؄Z dde~de ede edefd܄Zde-de-fdބZ dded%ed+edeeeffdZ dde~de ede ed%e edeeeeefff dZ dde~de ede ed+edeef dZ ddePee~e~fde ede edeefdZ ddePee~e~fde ede edeefdZ ddePee~e~fde ede edePeddffdZ ddePee~e~fde ede edePeddffdZ dde~de edeefdZ ddePee~e~fde edeefdZ ddePee~e~fde edeefdZ ddePee~e~fde edeefdZdeedededeefdZdeededeefdZdeededeefdZdeededeefdZdeededeefdZdeededeefdZdeededeefdZdeedeefdZy ) LlamaParsezA smart-parser for files.z#The API key for the LlamaParse API.T)default descriptionvalidate_defaultapi_keyz&The base URL of the Llama Parsing API.)rDrEr$Nz+The organization ID for the LlamaParse API.r%z&The project ID for the LlamaParse API.r&z8The interval in seconds to check if the parsing is done.check_intervalzcControls the backoff pattern when retrying failed requests: 'constant', 'linear', or 'exponential'.backoff_patternzNMaximum interval in seconds between polling attempts when checking job status.max_check_intervalz2A custom HTTPX client to use for sending requests. custom_clientz?Whether or not to ignore and skip errors raised during parsing. ignore_errorsizAThe maximum timeout in seconds to wait for the parsing to finish. max_timeoutrz>The number of workers to use sending API requests for parsing.)rDgtltrE num_workerszThe result type for the parser. result_typez*Show progress when parsing multiple files. show_progressz1Whether to split by page using the page separator split_by_pagez-Whether to print the progress of the parsing.verboseFzNIf set to true, LlamaParse will try to detect long table and adapt the output.adaptive_long_tablez:Annotate links found in the document to extract their URL.annotate_linkszIf set to true, the parser will automatically select the best mode to extract text from documents based on the rules provide. Will use the 'accurate' default mode by default and will upgrade page that match the rule to Premium mode. auto_modezA JSON string containing the configuration for the auto mode. If set, the parser will use the provided configuration for the auto mode.auto_mode_configuration_jsonzdIf auto_mode is set to true, the parser will upgrade the page that contain an image to Premium mode."auto_mode_trigger_on_image_in_pagezcIf auto_mode is set to true, the parser will upgrade the page that contain a table to Premium mode."auto_mode_trigger_on_table_in_pagezdIf auto_mode is set to true, the parser will upgrade the page that contain the text to Premium mode.!auto_mode_trigger_on_text_in_pagezdIf auto_mode is set to true, the parser will upgrade the page that match the regexp to Premium mode.#auto_mode_trigger_on_regexp_in_pagezAzure Openai API Versionazure_openai_api_versionzAzure Openai Deployment Nameazure_openai_deployment_namezAzure Openai Endpointazure_openai_endpointzAzure Openai Keyazure_openai_keyzThe bottom margin of the bounding box to use to extract text from documents expressed as a float between 0 and 1 representing the percentage of the page height. bbox_bottomzThe left margin of the bounding box to use to extract text from documents expressed as a float between 0 and 1 representing the percentage of the page width. bbox_leftzThe right margin of the bounding box to use to extract text from documents expressed as a float between 0 and 1 representing the percentage of the page width. bbox_rightzThe top margin of the bounding box to use to extract text from documents expressed as a float between 0 and 1 representing the percentage of the page height.bbox_topzaIf set to true, the parser will output compact markdown table (without trailing spaces in cells).compact_markdown_tablezhParse documents continuously, leading to better results on documents where tables span across two pages.continuous_modezbDisable the OCR on the document. LlamaParse will only extract the copyable text from the document. disable_ocrz]If set to true, the parser will not extract images from the document. Make the parser faster.disable_image_extractionzIf set to true, the document will not be cached. This mean that you will be re-charged it you reprocess them as they will not be cached. do_not_cachezIf set to true, the parser will keep column in the text according to document layout. Reduce reconstruction accuracy, and LLM's/embedings performances in most case.do_not_unroll_columnszEIf set to true, the parser will extract/tag charts from the document.extract_chartszgIf set to true, the parser will extract the layout information of the document. Cost 1 credit per page.extract_layoutzNote: Non compatible with gpt-4o. If set to true, the parser will use a faster mode to extract text from documents. This mode will skip OCR of images, and table/heading reconstruction. fast_modez2Whether to guess the sheet names of the xlsx file.guess_xlsx_sheet_nameszmIf set to true, when parsing HTML the parser will consider all elements display not element as display block.html_make_all_elements_visiblezgIf set to true, when parsing HTML the parser will remove fixed elements. Useful to hide cookie banners.html_remove_fixed_elementszsIf set to true, when parsing HTML the parser will remove navigation elements. Useful to hide menus, header, footer.html_remove_navigation_elementszX(optional) If set with input_url will use the specified http proxy to download the file. http_proxyznIf set to true, the parser will ignore document elements for layout detection and only rely on a vision model.-ignore_document_elements_for_layout_detectionz@The region of the input S3 bucket if input_s3_path is specified.input_s3_regionzIf set to true, the cache will be ignored and the document re-processes. All document are kept in cache for 48hours after the job was completed to avoid processing the same document twice.invalidate_cachezjThe extra time in seconds to wait for the parsing to finish per page. Get added to job_timeout_in_seconds.*job_timeout_extra_time_per_page_in_secondszThe maximum timeout in seconds to wait for the parsing to finish. Override default timeout of 30 minutes. Minimum is 120 seconds.job_timeout_in_secondsenz"The language of the text to parse.languagezeThe separator to use to split the header of the markdown table into multiple lines. Default is:
)markdown_table_multiline_header_separatorzThe maximum number of pages to extract text from documents. If set to 0 or not set, all pages will be that should be extracted will be extracted (can work in combination with targetPages). max_pagesz\If set to true, the parser will also output a PDF of the document. (except for spreadsheets)output_pdf_of_documentzAn S3 path prefix to store the output of the parsing job. If set, the parser will upload the output to S3. The bucket need to be accessible from the LlamaIndex organization.output_s3_path_prefixzHThe AWS region of the output S3 bucket defined in output_s3_path_prefix.output_s3_regionzFIf set to true, the parser will output tables as HTML in the markdown.output_tables_as_HTMLzThe error tolerance for the number of pages with error in a doc (percentage express as 0-1). If we fail to parse a greater percentage of pages than the tolerance value we fail the job.page_error_tolerancezA templated prefix to add to the beginning of each page. If it contain `{page_number}`, it will be replaced by the page number. page_prefixzA templated page separator to use to split the text. If it contain `{page_number}`,it will be replaced by the next page number. If not set will the default separator '\n---\n' will be used.page_separatorzA templated suffix to add to the beginning of each page. If it contain `{page_number}`, it will be replaced by the page number. page_suffixzBThe parsing mode to use, see ParsingMode enum for possible values parse_modez(Use our best parser mode if set to True. premium_modezThe preset to use for the parser. If set, the parser will use the preset configuration. See LlamaParse documentation for available presets. Preset override most other parameters.presetz1Preserve grid alignment across page in text mode.&preserve_layout_alignment_across_pageszThe mode to use to replace the failed page, see FailedPageMode enum for possible value. If set, the parser will replace the failed page with the specified mode. If not set, the default mode (raw_text) will be used.replace_failed_page_modezYA prefix to add before error message in failed pages. If not set, no prefix will be used.-replace_failed_page_with_error_message_prefixzXA suffix to add after error message in failed pages. If not set, no suffix will be used.-replace_failed_page_with_error_message_suffixzlIf set to true, the parser will ignore diagonal text (when the text rotation in degrees modulo 90 is not 0).skip_diagonal_textzvIf set to true, the parser will extract sub-tables from the spreadsheet when possible (more than one table per sheet).spreadsheet_extract_sub_tableszfIf set to true, the parser will fail if it can't extract text from a document because of a buggy font.strict_mode_buggy_fontzTIf set to true, the parser will fail if it can't extract an image from the document.strict_mode_image_extractionzPIf set to true, the parser will fail if it can't OCR an image from the document.strict_mode_image_ocrzdIf set to true, the parser will fail if it can't reconstruct a table or a heading from the document.strict_mode_reconstructionzYIf set to true, the parser will output structured data based on the provided JSON Schema.structured_outputzA JSON Schema to use to structure the output of the parsing job. If set, the parser will output structured data based on the provided JSON Schema.structured_output_json_schemazThe named JSON Schema to use to structure the output of the parsing job. For convenience / testing, LlamaParse provides a few named JSON Schema that can be used directly. Use 'imFeelingLucky' to let llamaParse dream the schema."structured_output_json_schema_namezPThe system prompt. Replace llamaParse default system prompt, may impact accuracy system_promptz*String to append to default system prompt.system_prompt_appendz8Whether to take screenshot of each page of the document.take_screenshotzThe target pages to extract text from documents. Describe as a comma separated list of page numbers. The first page of the document is page 0 target_pagesz7The user prompt. Replace llamaParse default user prompt user_promptz#The API key for the multimodal API.vendor_multimodal_api_keyz-The model name for the vendor multimodal API.vendor_multimodal_model_namez|j>| d<|j@|j@| d<|jB|jB| d<|jD|jD| d<|jF|jF| d<|jH|jH| d<|jJr|jJ| d<|jLrtOd|jL| d<|jPrtOd|jP| d <|jRr|jR| d!<|jTr|jT| d"<|jVr|jV| d#<|jXr|jX| d$<|jZr|jZ| d%<|j\r|j\| d&<|j^r|j^| d'<|j`r|j`| d(<|jbrtOd)|jb| d*<|jdr|jd| d+<|jfr|jf| d,<|jhr|jh| d-<|jjr|jj| d.<|jl|jl| d/<|jnr|jn| d0<|d}t|| d1<|d}t|| d2<|jp|jp| d3<|jrr|jr| d4<|jtrtOd)|jt| d5<|jv|jv| d6<|jx|jx| d7<|jzr|jz| d8<|j||j|| d9<|j~r|j~| d:<|j|j| d;<|j|j| d<<|jr|j| d=<|j|j| d><|j|j| d?<|j|j| d@<|j|j| dA<|jrtOdB|j| dC<|jr|j| dD<|jr|j| dE<|jr|j| dF<|j|j| dG<|j|jj| dH<|j|j| dI<|j|j| dJ<|jr|j| dK<|jr|j| dL<|jr|j| dM<|jr|j| dN<|jr|j| dO<|jr|j| dP<|jr|j| dQ<|j|j| dR<|j|j| dS<|j|j| dT<|j|j| dU<|jr|j| dV<|j|j| dW<|j|j| dX<|jr|j| dY<|j|j| dZ<|j|j| d[<|j|j| d\<|j|j| d]<|j|j| d^<|jr|j| d_<|j|j| d`< tt|j|j} t|jda| |j|| bd{}|j|jdc||jSS7:#tj$r*}dd|jj}t#||d}~wwxYw#||jwwxYww)eN file_name;file_name must be provided in extra_info when passing bytesrfilerHz8Currently, only the following file types are supported: z Current file type: rbzJfile_input must be either a file path string, file bytes, or buffer objectTfrom_python_packagerYrZr[r\r]r^r_r`rarbrcrdrerfrgrhrizWARNING: complemental_formatting_instruction is deprecated and may be remove in a future release. Use system_prompt, system_prompt_append or user_prompt instead.rzWARNING: content_guideline_instruction is deprecated and may be remove in a future release. Use system_prompt, system_prompt_append or user_prompt instead.rrjrkrlrmrnrorprqzWARNING: formatting_instruction is deprecated and may be remove in a future release. Use system_prompt, system_prompt_append or user_prompt instead.rrrrsrtrurvrw input_url input_s3_pathrxryrrzr{r}rrrrrrrrrzkWARNING: parsing_instruction is deprecated. Use system_prompt, system_prompt_append or user_prompt instead.rrrrrrrrrrrrrrrrrrrrrrrrrrr~rrrPOST)rfilesdataidzFailed to parse the file: )trrrbytesrr mimetypes guess_typerrr rrpathsplitextlowerrrropenbasenamerYrZr[r\r]r^r_r`rarbrcrdrerfrgrhrirprintrrjrkrlrmrnrorprqrrrrsrtrurvrwrxryrrzr{r}rrrrrrrrrrrrrrrvaluerrrrrrrrrrrrrrrrrrrrr~rrrr2JOB_UPLOAD_ROUTEr%r&r#rrOraise_for_statusjsoncloserHTTPStatusErrorresponsetext)rrrrr file_handlerrr mime_typerfile_extrrresperrmsgs r1 _create_jobzLlamaParse._create_jobs  "&"4"4Z"@Jd &*ooj&A t j5."9 :J!> Q#;/I!,,Y7:IiY?@E  "E  &E  S$ x$H IJIww'' 215;;=H33NOcNde**25",,Y7:I'~'B''*d3Kbgg..y9; RSE\  "&* "#  # #*.*B*BD& '   %)%8%8D! " >> $D   , , 8373T3TD/ 0  2 277 4   2 277 4   1 1 =66 3   3 3 ?88 5   ( ( 4/3/L/LD+ ,  , , 8373T3TD/ 0  % % 1,0,F,FD( )  ,'+'<'> % $D  ?? &!%D  == $#}}D   & &-1-H-HD) *  3 3 t  88 5   - - n 594V4VD0 1   &*&:&:D" #   "&"2"2D   ( (/3/L/LD+ ,   #'#4#4D  % %,0,F,FD( )   %)%8%8D! "   %)%8%8D! " >> $D   & & g .2-H-HD) *  & &-1-H-HD) *  . .595X5XD1 2  * *151P1PD- .  / /44 1  ?? &!%D   = =BB ?   E #ID   $E$' $6D !    +&*&:&:D" #  '+'<'> % $D   & &-1-H-HD) *  % % 1,0,F,FD( )  ,'+'<'(>D$ %  - - 9484V4VD0 1  2 2 >77 4     )$($6$6D !  $ $ 0+/+D+DD' (   &*&:&:D" #    (#'#4#4D    '"&"2"2D   + +262R2RD. /  ) ) 5040N0ND, -  , , 8373T3TD/ 0    '"&"2"2D   9 9 E>> ;     (#'#4#4D ??!%D     )$($6$6D ! $,d.B.BDOOTC)$,,TM]M]ejquvvD  ! ! #99;t$ &!!#'w$$ *.s||/@/@.ABCC.c ) *&!!#'sIh&l )A j36j17&j3l 1j33k0%k++k00k33ll current_intervalc@|jtjk(r|S|jtjk(r"t |dzt |j S|jtjk(r"t |dzt |j S|S)zCalculate the next backoff interval based on the backoff pattern. Args: current_interval: The current interval in seconds Returns: The next interval in seconds rH)rJr4r<r=minfloatrLr>)rrs r1_calculate_backoffzLlamaParse._calculate_backoffps   >#:#: :# #  ! !^%:%: :'!+U43J3J-KL L  ! !^%?%? ?'!+U43J3J-KL Lr@job_idc Ktj}d}d}t|j}|j} t j |d{|dz }|j tj|d{} | j| j} | d} | dk(r>|j tj||d{} | jS| dk(r^tj} | |z |jkDrtd ||r|d zdk(rtd d d |j|}n=| j dd}| j dd}d|d| d|d|}t| [7B77#t j"t j$t j&t j(t j*t j,t j.f$rp}|dz }tj} | |z |jkDrtd |||r|d zdk(rtd|dd|j|}Yd}~d}~wwxYww)NrTrH)rstatusSUCCESS)rrUPENDINGz Timeout while parsing the file: .rC)endflush error_codezNo error code found error_messagezNo error message foundzJob ID: z failed with status: z, Error code: z, Error message: z HTTP error: z...)r )timerrIrasynciosleepgetJOB_STATUS_ROUTEformatrrJOB_RESULT_URLrOrrrr ConnectError ReadError WriteErrorConnectTimeout ReadTimeout WriteTimeoutr)rrrUrXstarttries error_countrrr result_jsonr parsed_resultrr r  exception_strrs r1_get_job_resultzLlamaParse._get_job_results  "'(;(;"< 2 Mmm$4555 %zz*:*A*A*A*PQQ'')$kkm $X.Y&*0**&--V-U+%M)--//y())+CU{T%5%55'*J6((STT52:?cr6'+'>'>?O'P$!,?T!UJ$/OO')A%M#6(*?xH''1l2CM?T"$M22%)5Q %,""  $$!!""%% Mq iik;!1!11#:6(CurzQ&se3/"$(#:#:;K#L ) Mst;JF F1F F AF F!F 4J5B F JF F F A.I>A&I94J9I>>JcK|j|||d{}|jrtd|z|j||xs|jj |jd{}||fS7`7 w)z-Create one parse job and wait for the result.rrNz(Started parsing the file under job_id %s)rX)rrXrrrUr)rrrrrUrrs r1 _parse_onezLlamaParse._parse_ones'' jR'PP << }t|tr|nd} td| d||jrgcYd}~S|d}~wwxYww)Load data from the input path.r!NrmetadataError while parsing the file '':) r"rrUrrW _get_sub_docsrrrrrN) rrrrrX_job_idrdocse file_reprs r1 _aload_datazLlamaParse._aload_datas $(OOjR%4%OGV 0 0 6 67'-2D !!))$//  %/ 3%? EUI 29+R@! D!!   sWB7A-A+A A-(B7)A-*B7+A-- B461B/'B4(B7-B//B44B7c \Kt|ttttt fr'|j ||||jd{St|tr|Dcgc]0}|j ||||jxr |j 2}} t||jd|jd{}|Dcgc] }|D]}| c}}Std7cc}w7/cc}}w#t$r'} tt| vrtt| d} ~ wwxYww)r$)rrrXN Parsing filesworkersdescrV:The input file_path must be a string or a list of strings.)rrr rrrr/rXlistrVrrT RuntimeErrorr!r"r) rrrrfjobsresultssublistitemr-s r1 aload_datazLlamaParse.aload_datas@ i#}dE>!R S))jR*  4 (#   ) LLC1C1C-C !D  ( ,,("&"4"4 !-4HHHHHL ;I #s1v-&'788G  slAD, C* D,#5C,D,&C9C1 C9 C3C9 D,,D,1C93C99 D)"D$$D))D,c t|j|||S#t$r'}tt |vrtt |d}~wwxYw)r$rN)rr=r7r!rr"rrrrr-s r1 load_datazLlamaParse.load_datasN ty*LM M 3q6)"#344   A"A  ArcKt|ttttt frt|tt fr|rd|vr t d|d}n t|}|j|||tjjd{\}}t||||j|j|j|jxst St|t"r*g}|D]X}t|tt fr&|rd|vr t d|j%|d?|j%t|Z t'|Dcgc]/}|j|||tjj1c}|j(d|j*d{} t-| D cgc]M\} \}}t||| ||j|j|j|jxst Oc}}} St d7cc}w7}cc}}} w#t.$r'} t0t| vrt/t2| d} ~ wwxYww) az Parse the file and return a JobResult object instead of Document objects. This method is similar to aload_data but returns JobResult objects that provide direct access to the various output formats (text, markdown, json, etc.) Args: file_path: Path to the file to parse. Can be a string, path, bytes, file-like object, or a list of these. extra_info: Additional metadata to include in the result. fs: Optional filesystem to use for reading files. Returns: JobResult object or list of JobResult objects if multiple files were provided rr)rrrUN)rr job_resultrGr$rrzGetting job resultsr2r5)rrr rrrrr"rJSONrrrGr$rr_DEFAULT_SEPARATORr6appendrrTrV enumerater7r!r") rrrrrrrD file_namesr8 job_resultsir-s r1aparsezLlamaParse.aparse)s`* i#}dE>!R S)e^%<=![ %B$U'{3  N '+%&OO11 (7(" FJ #% ||#22H6H  4 (J .a%!89%J)F(Y%%j&=>%%c!f- ." $,"+ '1!(2(=(= (!,,."&"4"4 %  44=[3I  0/FJ%",Q-#- $ !%#||'+':':'P>P  &L A"<    #s1v-&'788G  shBI$H B9I$ H14H#"H1*H(+H1>AH*H1 I$#H1*H11 I!:"II!!I$c t|j|||S#t$r'}tt |vrtt |d}~wwxYw)ay Parse the file and return a JobResult object instead of Document objects. This method is similar to load_data but returns JobResult objects that provide direct access to the various output formats (text, markdown, json, etc.) Args: file_path: Path to the file to parse. Can be a string, path, bytes, file-like object, or a list of these. extra_info: Additional metadata to include in the result. fs: Optional filesystem to use for reading files. Returns: JobResult object or list of JobResult objects if multiple files were provided r?N)rrLr7r!rr"r@s r1parsezLlamaParse.parsesN( t{{9jR{HI I 3q6)"#344  rBcjK |j||tjjd{\}}||d<t |t t fst||d<|gS73#t$r>}t |tr|nd}td|d||jrgcYd}~S|d}~wwxYww)r$)rrUNrrr'r(r)) r"rrErrrrrrrrN)rrrrrr-r.s r1 _aget_jsonzLlamaParse._aget_jsons #'??%&OO11$3$NFF &F8 i%)@A&))n{#8O %/ 3%? EUI 29+R@! D!!   sJB3/A)A'2A)&B3'A)) B021B+#B0$B3)B++B00B3cKt|ttfr|j||d{St|tre|Dcgc]}|j||}} t ||j d|jd{}|Dcgc] }|D]}| c}}Std7cc}w7/cc}}w#t$r'}tt|vrtt|d}~wwxYww)r$)rNr1r2r5) rrrrPr6rrTrVr7r!r"r) rrrr8r9r:r;r<r-s r1 aget_jsonzLlamaParse.aget_jsons i#t -zJJ J  4 (GPQ!DOOA*O=QDQ  ( ,,("&"4"4 !-4HHHHHL 'KQI #s1v-&'788G  si-C6B4C6B6"C6%&C B; CB=%C( C66C6;C=C C3 "C..C33C6c t|j||S#t$r'}tt |vrtt |d}~wwxYw)zParse the input path.N)rrRr7r!rr")rrrr-s r1get_json_resultzLlamaParse.get_json_resultsJ  t~~iDE E 3q6)"#344   A "AA c&|j||S)r$)rT)rrrs r1get_jsonzLlamaParse.get_jsons ##Iz::r@ json_result download_path asset_keyc Ktjj|stj||j} g}|D]R}|d}|dD]A}|j r(t d|jd|dd||||D]} | d} tjj||d| } | jd s| jd s| d z } | | d <|| d<|jd d | d<|d| d<t| d5} |jd|d| } t|d| |jd {}|j| j!|j"d d d |j%| DU|S7S#1swY'xYw#t&$r*}t d|d||j(rgcYd }~S|d }~wwxYww)z:Download assets (images or charts) from the parsed result.rpagesz> z for page pagez: name-z.pngz.jpgrrNoriginal_file_path page_numberwb/api/parsing/job/z/result/image/GETrzError while downloading z from the parsed result:)rrexistsmakedirsrrXr capitalizer,endswithrrr$r#rOrwritecontentrGrrN)rrXrYrZrassetsrrr]asset asset_name asset_pathr8 asset_urlrr-s r1 aget_assetszLlamaParse.aget_assetss ww~~m, KK &) F% -)"7O-D|| !5!5!7 8 4<.PRSWXaSbRcd"&i-%*6] &(WW\\)fXQzl+C& *226:#-#6#6v#> *f 4 (2f *0h6 XLSXz.xlsxrrNr`rbrcz/result/raw/xlsxrdrzError while downloading xlsx:)rrrerfrrXrr,rrr$r#rOrrirjrGrrN) rrXrYr xlsx_listrr xlsx_pathxlsxr8xlsx_urlresr-s r1 aget_xlsxzLlamaParse.aget_xlsxVsl ww~~m, KK & I% ')<<(OGGLL6(%8HI (V !'X-3ZZ T-J)*)T*)a==/):6(BRS!1x9I9I!C((*GGCKK()  &- '.   ))  11 5!!   sgAE3A3E7/D5&D3'/D5E2E33D55D> :E E0 E+#E0$E3)E++E00E3c t|j||S#t$r'}tt |vrtt |d}~wwxYw)rN)rrr7r!rr"rus r1get_xlsxzLlamaParse.get_xlsxsJ t~~k=IJ J 3q6)"#344  rUr,cg}|jxst}|D]U}|jj|}|D]3}t |t |j }|j|5W|S)z$Split docs into pages, by separator.r%)rrFrsplitrrr&rG)rr,sub_docs separatordoc doc_chunks doc_chunksub_docs r1r*zLlamaParse._get_sub_docssv''=+=  )C 2J' ) ""%cll3(  ) )r@)NN)F)NNN)NNF)N)r8r9r:r;rrGr__annotations__rr$r%rr&rIintr4r=rJrLrMrrrNboolrOrTrTXTrUrVrWrXrYrZr[r\r]r^r_r`rarbrcrdrerrfrgrhrirjrkrlrmrnrorprqrrrsrtrurvrwrxryrzr{r}r~rrrrrrrrrrrrrrrrr rrrrrrrrrrrrrrrrrrrrrrrrrrrrr classmethodrrrrpropertyrrr r FileInputrrdictrrrr r rrr"r rr/r=rArLrNrPrRrTrWrprvr|r~rrrr*r?r@r1rBrBDs#9GS  <Hc&+A&OXc]!&<!J  NNC ',%%y'O^$d 27"V2M8E--. UM4WK T K $,MK "NM4 GM4"QGT +0d+$&+P&NHTN!&!Ix~38^3 (3-:?z:&:?y:&8=z8%x}:?z:'#/4"</hsm38"@3 (3-,1"9,8C=',"4'hsm$)w$K%"'t"Ix#(u#J!&t!Hhuo.3w.HTN',~'OXd^#(x#K$05s0htn$)_$L(4.-2{-8D>&+[&NHTN&+}&NHTN!&O!Ix~ .3H.HTN6;D6"HTN27}27<J7#Xd^!&n!J EJEE18D>&+V&OXc](-S(htnCHAC./4X/HUO$"FHhsm@E{@-x} %S Ix}.3r.HTN,1D,8C=',^'hsm-2\-8D>-2O-(5/"'V"K#%*X%NHSM"'V"K#5:X5J{C/01$)>$L(4."IFHSM>CG>*HTN:?m:h~6DIoD18C=DInD18C=*/C*6;M6"HTN .3|.HTN 49j4 (4. -2f-8D> 27z2 ).o)x~49i4!8C=9>z9& $)f$M8C=+0@+(3-',N'OXd^#(d#L(3-"'M"K#0590x}38C3 (3-"'R"K# #(F#L(3-:?j:'#49|4!8C=-2q-HSM"'H"J$)Q$M8C=16y1x~*/ I*#38?3$ YXDA   B ZhTB,#,#,C, 0;4e/THeE%%t+,T ** nU5F5F5L&MyT,I$&*+/ T$T$TNT$ ' ( T$ T$l  5 U $>C?M?M(+?M6:?M c3h?MH&*+/%) TN ' (  c]  sDcN" # &&*+/ TN ' (    h D&*+/ 'i)34'TN' ' ( ' h 'X&*+/ i)34 TN  ' (  h $&*+/ ai)34aTNa ' ( a tK +- . aL&*+/ i)34TN ' (  tK +- . :BF"08 d6&*i)34TN d B&* i)34 TN  d "&*;i)34;TN; d ;2:2692FI2 d2h : 69  d  : 69  d d4jdd4jd(:(69( d(TDJstDz $x. T(^ r@rB)@r rrr  contextlibrcopyrenumriorpathlibrrr typingr r r r rrr urllib.parserrfsspecrllama_index.core.async_utilsrr llama_index.core.bridge.pydanticrrrllama_index.core.constantsrllama_index.core.readers.baser"llama_index.core.readers.file.baserllama_index.core.schemar llama_cloud_services.parse.typesr llama_cloud_services.parse.utilsrrrr r!r"r#rrrrFrrrr2r4rBr?r@r1rs *11JJJ! %>PP7<=,6 #un, - B.(  $,SM ?G}   S$ S#Sr@