j!ddlZddlZddlZddlmZmZddlmZddlm Z ddl m Z ddl m Z mZmZdZd efd Zd efd Zed k(reyy)N)AnyDict)dedent) entry_points) __version__) MarkItDown StreamInfoDocumentConverterResultc tjddtjtdj }|j ddddt d |j d d d |j ddd|j ddd|j ddd|j}|j dddd|j ddddd !|j d"d#td$%|j d&td'%|j d(td)%|j d*td+%|j d,d-dd.|j d/dd0|j d1dd2|j d3d45|j}|j}|E|j j}t|d6kDr|jd7sd7|z}nd}|j}|C|j }t|d6kDr#|jd8d9k7rt!d:|nd}|j"}|A|j }t|d6kDr! t%j&|j(}nd}d}|||t-|||<}|j.rt1d=t3t5d>?}t|d6k(rt1d@t1dAn9|D])}t1dB|j(dCdD|j6dE+t1dFt9j:d6|j<rR|j> t!dGn|j@ t!dHtC|jD|j>I} n|jFr|jH t!dJn|j@ t!dKdL|jHi} |jJ|jJ| dM<|jLd9dNl'm(} |jLjSdOD cgc]1} | j s| j j3} } g}| D]} |jU| ||| dQ<tCdUdR|jDi| } ntC|jDS} |j@7| jYt8jZj\||j^T}n(| ja|j@||j^T}tc||y#t*$rt!d;|YwxYwcc} w#tV$rt!dP|YwxYw)VNz)Convert various file formats to markdown. markitdowna SYNTAX: markitdown If FILENAME is empty, markitdown reads from stdin. EXAMPLE: markitdown example.pdf OR cat example.pdf | markitdown OR markitdown < example.pdf OR to save to a file use markitdown example.pdf -o example.md OR markitdown example.pdf > example.md ) descriptionprogformatter_classusagez-vz --versionversionz %(prog)s z show the version number and exit)actionrhelpz-oz--outputz?Output file name. If not provided, output is written to stdout.)rz-xz --extensionzHProvide a hint about the file extension (e.g., when reading from stdin).z-mz --mime-typez*Provide a hint about the file's MIME type.z-cz --charsetz5Provide a hint about the file's charset (e.g, UTF-8).z-dz--use-docintel store_truezyUse Document Intelligence to extract text instead of offline conversion. Requires a valid Document Intelligence Endpoint.)rrz--use-cuz--use-content-understandinguse_cuzHUse Azure Content Understanding to extract text. Requires --cu-endpoint.)rdestrz-ez --endpointzHDocument Intelligence Endpoint. Required if using Document Intelligence.)typerz --cu-endpointz;Content Understanding Endpoint. Required if using --use-cu.z --cu-analyzerzOContent Understanding analyzer ID. If not specified, auto-selects by file type.z--cu-file-typeszComma-separated list of file types to route to Content Understanding (e.g., pdf,jpeg,mp4). If omitted, all supported types are routed.z-pz --use-pluginszTUse 3rd-party plugins to convert files. Use --list-plugins to see installed plugins.z--list-pluginsz^List installed 3rd-party plugins. Plugins are loaded when using the -p or --use-plugin option.z--keep-data-urisz_Keep data URIs (like base64-encoded images) in the output. By default, data URIs are truncated.filename?)nargsr./rzInvalid MIME type: zInvalid charset: ) extensionmimetypecharsetz(Installed MarkItDown 3rd-party Plugins: zmarkitdown.plugin)groupz# * No 3rd-party plugins installed.zI Find plugins by searching for the hashtag #markitdown-plugin on GitHub. z * z<16z (package: )zC Use the -p (or --use-plugins) option to enable 3rd-party plugins. zLDocument Intelligence Endpoint is required when using Document Intelligence.z6Filename is required when using Document Intelligence.)enable_pluginsdocintel_endpointzOContent Understanding Endpoint (--cu-endpoint) is required when using --use-cu.z6Filename is required when using Content Understanding. cu_endpointcu_analyzer_id)ContentUnderstandingFileType,zUnknown file type: cu_file_typesr#)r#) stream_infokeep_data_uris)2argparseArgumentParserRawDescriptionHelpFormatterrstrip add_argumentradd_mutually_exclusive_groupstr parse_argsrlowerlen startswith mime_typecount_exit_with_errorr codecslookupname LookupErrorr list_pluginsprintlistrvaluesysexit use_docintelendpointrr use_pluginsrr% cu_analyzerr) convertersr'splitappend ValueErrorconvert_streamstdinbufferr+convert_handle_output)parser cloud_groupargsextension_hintmime_type_hint charset_hintr*plugin_entry_points entry_pointr cu_kwargsr't type_namescu_typesr=results T/root/aria/tools/markitdown-venv/lib/python3.12/site-packages/markitdown/__main__.pymainr`s  $ $?  <<  6 %'? FD  K=) /    N    W    9    D 557K I %  W     W    J    ^   V    c   m   n   #.    D^^N!'--/557 ~  "!,,S1!$~!5!N^^N!'--/ ~  "##C(A- #6~6F!GH!N<QQRST U W     == ^ ]] " U V++t}}      # a ]] " U V 4++%     '*.*:*:I& '    ) @,0+=+=+C+CC+H&'AGGI !JH" CCOO$@$FG C *2Io &Mt/?/?M9M t/?/?@  }}** II  #..+  ## MM{4CVCV$ 4 o E #4\N!CD Ex"C$':4&%ABCs09UU# U#U(U U (VVr^cx|jr=t|jdd5}|j|jdddyt |jj t jjdjt jjy#1swYyxYw)zHandle output to stdout or filewzutf-8)encodingNreplace)errors) outputopenwritemarkdownr@encoderCstdoutrcdecode)rTr^fs r_rQrQs {{ $++sW 5 % GGFOO $ % %  OO " "3::#6#6y " I P P ##   % %s B00B9messagecDt|tjdy)Nr)r@rCrD)rns r_r:r:s 'NHHQK__main__)r-rCr;typingrrtextwraprimportlib.metadatar __about__r _markitdownr r r r`rQr3r:__name__r,rpr_rxsV +"HHt!n  !8  c  zFrp