o i@sddlZddlZddlmZddlmZddlmZddlZ ddl m Z ddl m Z mZddlmZmZmZmZmZmZddlmZdd lmZdd lmZmZeeZGd d d Ze d GdddeZ!dS)N)Sequence)Any) ParserCreate)make_tool_call_id)ChatCompletionRequestChatCompletionToolsParam)DeltaFunctionCall DeltaMessage DeltaToolCallExtractedToolCallInformation FunctionCallToolCall) init_logger) TokenizerLike) ToolParserToolParserManagerc@seZdZdZddZddZdedefdd Zd edefd d Z de fd dZ dedefddZ dedefddZ dede fddZdede fddZdede fddZdedeedBeffdd Zd!edefd"d#Zdedefd$d%Zd&efd'd(ZdKd)edBfd*d+Zd,ed-eeeffd.d/Zd0efd1d2Zd,efd3d4Zd5d6Zd7eedBfd8d9Zd,ed-eeefdedBfd:d;Z d,ed-eeefdedBfdd?Z"d@edefdAdBZ#dCed@ede$fdDdEZ%dFe$d@edefdGdHZ&dIdJZ'dS)LStreamingXMLToolCallParserze Simplified streaming XML tool call parser Supports streaming input, parsing, and output cCs6|d|_d|_d|_d|_d|_d|_d|_dS)N z )reset_streaming_statetoolstool_call_start_tokentool_call_end_tokenfunction_start_tokenfunction_end_tokenparameter_start_tokenparameter_end_tokenselfr![/home/ubuntu/vllm_env/lib/python3.10/site-packages/vllm/tool_parsers/step3p5_tool_parser.py__init__(s z#StreamingXMLToolCallParser.__init__cCsg|_d|_d|_d|_d|_d|_i|_d|_d|_d|_ d|_ d|_ d|_ d|_ d|_d|_d|_d|_d|_d|_d|_t|_|dS)zReset streaming parsing staterNF)deltastool_call_indexcurrent_call_idlast_completed_call_idcurrent_function_namecurrent_function_open parameterscurrent_param_namecurrent_param_valuecurrent_param_value_convertedcurrent_param_is_firstshould_emit_end_newlinestart_quote_emittedstreaming_bufferlast_processed_postext_content_buffer_pre_inside_parameter_pre_param_buffer_pre_current_param_namedefer_current_parameterdeferred_param_raw_valuerparser setup_parserrr!r!r"r4s. z0StreamingXMLToolCallParser.reset_streaming_state xml_chunkreturnc st|j}|j}|j}|j|7_|}d|dur)|j|kr(|j|kr(|n|jdur8|j|dkr8|j|rzZ|j|d}durh|j|vrhtfdd|D}|sh|jr`| d|j rh| ddur|j |vrtfdd|D}|s|jr| d|j r| d| dWnt y} z t d | WYd} ~ nd} ~ ww||} | S|jr|jd krt|jd } || d |_| Sdur|j|vs|j |vr|jr| d|j|vr|j r| d|j |vr| d||} | Stdd S) a Parse single streaming XML chunk and return Delta response This is the actual streaming interface that receives chunks one by one and maintains internal state Args: xml_chunk: Single XML chunk string Returns: DeltaMessage: Contains delta information generated by this chunk, returns empty response if no complete elements Nc3.|]}|jotfdd|jDVqdS)c3s:|]}|jo|jkot|jjto|jjdvVqdS))}{}N)functionid isinstance argumentsstr.0tcfallback_call_idr!r" s   UStreamingXMLToolCallParser.parse_single_streaming_chunks...N tool_callsanyrHtdrJr!r"rL   zKStreamingXMLToolCallParser.parse_single_streaming_chunks.. parameterrBc3r?)c3s6|]}|jdko|jo|jjdko|jkVqdS)rBr$N)typerBrErCrGrJr!r"rLs   rMNrNrQrJr!r"rLrS tool_callzError with fallback parsing: %srcontentr$)lenr%r'r&r2_process_complete_xml_elementsrrPr, _end_elementr)r Exceptionloggerwarning$_merge_new_deltas_to_single_responser4r _emit_delta) r r<initial_delta_count entry_call_identry_tool_call_indexfound_elements new_deltashas_function_closehas_toolcall_closee result_delta text_deltar!rJr"parse_single_streaming_chunksVs                         z8StreamingXMLToolCallParser.parse_single_streaming_chunkstextcCs2dddddd}|D] \}}|||}q |S)z Escape XML special characters Args: text: Original text Returns: Escaped text z&z<z>z"z')&<>"')itemsreplace)r rl xml_escapescharescaper!r!r"_escape_xml_special_charss z4StreamingXMLToolCallParser._escape_xml_special_charsc Cs|d}|jt|jkr||j\}}|dur |S||r"||_qzv||}|ds6|drL|jdkrL|j rLt |j d}| |d|_ |dr|jdkr|j r|j r|jrf|d|jrn|d t dddt|jd |j d tddd d gd }| |||j|dd}Wnty}z td|WYd}~nd}~ww||_|jt|jks |S)z Process complete XML elements in buffer Returns: bool: Whether complete elements were found and processed FNrznamerEindexrCrUrB)rolerXreasoning_contentrOTz#Error when parsing XML elements: %s)r3rYr2_find_next_complete_element_should_skip_element_preprocess_xml_chunkstrip startswithr&r4r r`r'r)r,r[r*r r!_reset_xml_parser_after_tool_callr:Parser\r]r^)r found_anyelementend_pospreprocessed_elementrj final_deltarhr!r!r"rZsnA             Fz9StreamingXMLToolCallParser._process_complete_xml_elementschunkc Cs|||}dD]}d|d}||vrq||}||d}|d}|dt|}|dkrA|dks8||krA||d|vrAq||t|d}tddt|Dt|} |d| } | sbq| |drt| t|d d} | r| ds| dd vs| dd} | r| ds| dd vr| sq|d kr|| sq|d|d|d| d|d| dd }q|S) a Fallback: fix incomplete ) Examples: , Also handles missing = cases: -> , -> Only fixes tags that pass validation (parameter exists in tool definition) )rTrBrn=Nrocss |] \}}|dvr|VqdS))  rnNr!)rHichr!r!r"rLhszJStreamingXMLToolCallParser._fix_incomplete_tag_in_chunk..r>)-_rT) #_fix_missing_equals_in_function_tagfindrYnext enumeraterisalnum_validate_parameter_namers) r rtag_typepattern start_idx after_taggt_poslt_posrXrtag_namer!r!r"_fix_incomplete_tag_in_chunkFsP           (z7StreamingXMLToolCallParser._fix_incomplete_tag_in_chunkc Csd|vr|Sd}t||}|r2|d}|r2||r2|d}d|d}|||d}|Sd}t||}|r^|d}|r^||r^|d}d|d}|||d}|S|S)a Fix missing = in function tags: or Examples: -> -> Only fixes if function name exists in tool definition rz(r>rroz%)researchgroupr_validate_function_namers) r rpattern1match1 func_nameoriginalfixedpattern2match2r!r!r"rs*       z>StreamingXMLToolCallParser._fix_missing_equals_in_function_tagrcCsV|jsdS|jD] }t|dr(|jdkr(t|dr(t|jdr(|jj|kr(dSqdS)z1Check if function name exists in tool definitionsFrUrBryT)rhasattrrUrBry)r rtoolr!r!r"rs    z2StreamingXMLToolCallParser._validate_function_name param_namecCs|jr|jsdS|jD]?}t|drJ|jdkrJt|drJt|jdrJ|jj|jkrJt|jds2dS|jj}t|trG| d|}||vSdSq dS)z?Check if parameter exists in current function's tool definitionTrUrBryr+ properties) rr)rrUrBryr+rDdictget)r rrparamsrr!r!r"rs*        z3StreamingXMLToolCallParser._validate_parameter_namercCs\||js||js||jrdS|jdur$|r$|j|7_dS|jdur+dS| S)z Determine whether an element should be skipped Args: element: Element to evaluate Returns: bool: True means should skip, False means should process FNT)rrrrr'r4)r rr!r!r"rs    z/StreamingXMLToolCallParser._should_skip_element start_posNcCs|j|d}|s d|fS|dr|dod|ddv}|do-d|ddv}|s2|r[|r6dnd }d |d}||}|d kr[|d|t|}|||t|fS|dd } |dd } | d kr| d kr| | kr}|d| || fS|d| d || d fS| d kr|d| || fS| d kr|d| d || d fS|jdur|d dt|krd|fS|ds|ddt|krd|fS||t|fSd|fS|d} | d kr|d| } | || fS|} | |t| fS)a Find next complete XML element from specified position Args: start_pos: Position to start searching Returns: (Complete element string, element end position), returns (None, start_pos) if no complete element found NrnrrorrrrTrBzr)r2rsplitrrYr')r rbufferis_incomplete_paramis_incomplete_funcr closing_tag closing_poscomplete_elementtag_endtag_end2 next_tag_pos text_content remainingr!r!r"r~sP           z6StreamingXMLToolCallParser._find_next_complete_element initial_countc Cs&t|j|kr tddS|j|d}t|dkr|dSg}d}|D]b}|jr-||j7}|jr|jD]Q}d}|D] }|j|jkrE|}nq9|r|jr|jrX|jjrX|jj|j_|jrw|jjdurw|jjdurkd|j_|jj} |jj| 7_|j r~|j |_ q3| |q3q#t|r||dSd|dS)a Merge newly generated deltas from this processing into a single DeltaMessage Args: initial_count: Delta count before processing Returns: Merged DeltaMessage containing all newly generated delta information NrWr>rr$)rXrO) rYr%r rXrOrCrBryrErUappend) r rremerged_tool_callsmerged_contentdeltarV existing_callexistingnew_argsr!r!r"r_RsP          z?StreamingXMLToolCallParser._merge_new_deltas_to_single_responsec Csd}||js||jrd}||js'||js'|ds'td|r)d}||js5||jr7d}|j dusF|dsF|drK| |}t dd |}t d d |}|}|j r|d r|j }d|_||_d|_ d |_ d|_||}|d S|j d kr|jr||jnd}|dv}|dvp|dp|d} d|vpd|vpd|v} d} | rd} n |r| rd|vrd} | sd|_ ||S|j |7_ d S|drtd|} | r| d|_d|_ d |_ |S|s||}|S)z Preprocess XML chunk, handle non-standard formats, and escape special characters Args: chunk: Original XML chunk Returns: Processed XML chunk FTz ]+)>zz]+)>zrr$string)object)arrayarrsequencerlist[{(rqzr>)rrrrrrmatchrrr'rsubr5r6r8r9r7rw_get_param_typer) r r is_tool_call processedoriginal_chunk body_text safe_text param_typeis_object_typeis_complex_typehas_container_hint need_defermr!r!r"rs                z0StreamingXMLToolCallParser._preprocess_xml_chunkrcCs|j|dS)z&Emit Delta response (streaming output)N)r%r)r rr!r!r"r`sz&StreamingXMLToolCallParser._emit_delta incoming_tagcCsL|jr|d|dvr|jr|d|dkr"|jr$|ddSdSdS)aBefore starting to process new elements, if there are unclosed tags from before, automatically complete their endings to the parser. - If there are unclosed parameters, it's equivalent to feeding `` - When about to start a new function or tool_call, if there are unclosed functions, complete ``. - When about to start a new tool_call, if there are unclosed tool_calls, complete ``. rT)rBrVrBrVN)r,r[r)r')r rr!r!r"$_auto_close_open_parameter_if_neededs  z?StreamingXMLToolCallParser._auto_close_open_parameter_if_neededryattrsc Cs|dkrdS|dkr"|di|_t|_d|_|jd7_dS|ds+|dkrd|js4|di|d|||}||_ d|_ |rbt t |jd|jdt |ddd gd }||dSdS|d sm|d kr|d |||}||_d|_d|_d |_|r|jsd |d}t t |jd|jdt d|dd gd }||d|_dSd|d}t t |jd|jdt d|dd gd }||d |_dSdSdS)zHandle XML start element eventsrootNrVTr>rBr$rxrzrOrTFz{"z": z, ")rr+rr'r/r&r_start_element_extract_function_namer)r*r r rr`_extract_parameter_namer,r-r.r1)r ryr function_namerr json_start json_continuer!r!r"r4s           z)StreamingXMLToolCallParser._start_elementdatac Cs~|r|jr|jr+|}|jrd|}d|_|dr"d|_|dd}|j|7_dS||j}|js?|dr?|dd}|dvrb|jsbtt |j d|j dt dd d d gd }| |d|_|sfdS|}|jrrd|}d|_|drd|_|dd}|j|7_||j|}|||}|t|jd}||_tt |j d|j dt d|d d gd }| |dSdSdS) z Handle XML character data eventsrFTNrr>rrFrlvarcharruenumrBrprxrzr)r,r8r0endswithr-rrr1r r r&r'rr`_convert_param_value_convert_for_json_streamingrYr.) r r original_datar quote_deltaconverted_value output_data delta_datarr!r!r" _char_datasn            z%StreamingXMLToolCallParser._char_datac Cs|dkrdS|ds|dks|dkr|jr||ds#|dkr|jr|j}|j}|jr|jr5|jn|}d}d}z|jrD|d}n|}t|}t j |dd}Wnt yet j |dd}|}Ynwt t |jd |jdtd|d d gd }||d|_||j|<d|_d |_d |_d|_d|_d |_dS||} ||| } | dvr|s|jst t |jd |jdtddd d gd }||nt t |jd |jdtddd d gd }||d|_| |j|<d|_d |_d |_d|_dS|ds|dkrB|jr!t t |jd |jdtddd d gd }||nt t |jd |jdtddd d gd }||d|_d|_dS|dkr|jrY|jrT|d|dt t |jd |jdtdd d d gd }|||jrt |jd} || |dSdS)zHandle XML end element eventsrNrBrVrTrF ensure_asciir>rxrzrr$rz""rpr@rArW)rr,rr-r8r9r0ast literal_evaljsondumpsr\r r r&r'rr`r+r.r1rrr*r)r[r4rr) r ryr param_valueraw_text parsed_valueoutput_arguments raw_for_parserrrrjr!r!r"r[s                          z'StreamingXMLToolCallParser._end_elementcCs*d|j_|j|j_|j|j_|j|j_dS)z Set up XML parser event handlersTN)r: buffer_textrStartElementHandlerr[EndElementHandlerrCharacterDataHandlerrr!r!r"r;s  z'StreamingXMLToolCallParser.setup_parserrcCs ||_dS)z"Set tool configuration informationN)r)r rr!r!r" set_toolss z$StreamingXMLToolCallParser.set_toolscCL|r d|vr |dSd|vr$|dd}t|dkr$|ddkr$|dSdS)z*Extract function name from various formatsryrr>rrBNrrYr ryrpartsr!r!r"r  z1StreamingXMLToolCallParser._extract_function_namecCr)z+Extract parameter name from various formatsryrr>rrrTNr r r!r!r"rr z2StreamingXMLToolCallParser._extract_parameter_namecCs |jr|jsdS|jD]x}t|drt|drt|jdsq |jdkr|jj|jkrt|jds3dS|jj}t|trad|vra|d}||vr^t||tr^| t || ddSdSt|tr||vr||}t|tr| t | ddSdSq dS)zGet parameter type based on tool configuration, defaults to string Args: param_name: Parameter name Returns: Parameter type rrUrBryr+r) rr)rrBrUryr+rDrrepair_param_typerFr)r rrrr param_configr!r!r"rsD         z*StreamingXMLToolCallParser._get_param_typercCsz|dvs9|ds9|ds9|ds9|ds9|ds9|ds9|ds9|d vs9|d vs9|d s9|d r;|Sd S)zRepair unknown parameter types by treating them as string Args: param_type: Parameter type Returns: Repaired parameter type rintuintlongshortunsignednumfloatbooleanboolbinary)rrrrrrr)r)r rr!r!r"r s. z,StreamingXMLToolCallParser.repair_param_typerc Cs|dkrdS|}|dvr|S|ds-|ds-|ds-|ds-|drEzt|WSttfyDtd |Y|Sw|d sO|d rvzt|}|t|d kr_|WSt|WSttfyutd |Y|Sw|dvr|}|dkS|S)zConvert value based on parameter type Args: param_value: Parameter value param_type: Parameter type Returns: Converted value nullNrrrrrrzr)rrrDrF)r rrr!r!r"r!s  z6StreamingXMLToolCallParser._convert_for_json_streamingcCst|_||jr|j|_d|_d|_d|_i|_d|_d|_ d|_ d|_ d|_ d|_ d|_d|_d|_d|_d|_d|_dS)z Each tool_call is treated as a separate XML document, so we need to reset the parser after each tool_call. NFr$)rr:r;r'r(r)r*r+r,r-r.r/r0r1r4r5r6r7r8r9rr!r!r"r9s( zr$ryrEr)r:rr)r*rrrkrOr rXrBryrr rCrUr rEr{rY)r r.r/resultrOrV tool_indexr!r!r"extract_tool_callshs`         z$Step3p5ToolParser.extract_tool_calls previous_text current_text delta_textprevious_token_idscurrent_token_idsdelta_token_idsNc Cst|s|jg|_g|_|r|j|j|s?|r?||jj||jj}|dkr2|jj dks8|jj s=|r=t ddSdS|j |} | r| j r| j D]j} | j r| jdurZ| jnt|jd} t|j| krx|jdddt|j| ksht|j| kr|jdt|j| ks| j jr| j j|j| d<| j jdur|j| d| j j7<|j| | j j7<qM| S)Nrr$rWr>rxryrE)r:rr)r*rrcountrrr&r rkrOrBr{rYrryrE) r r4r5r6r7r8r9r/ open_callsr1rVr2r!r!r"extract_tool_calls_streamings\            z.Step3p5ToolParser.extract_tool_calls_streamingcCsdS)zE Skip the remaining_call calculation in serving_chat Fr!rr!r!r"2parser_should_check_for_unstreamed_tool_arg_tokensszDStep3p5ToolParser.parser_should_check_for_unstreamed_tool_arg_tokens)r r!r"rr#rFrr r3rrr r<rr= __classcell__r!r!r-r"r&Zs6  B Hr&)"rrcollections.abcrtypingrxml.parsers.expatrregexrvllm.entrypoints.chat_utilsr0vllm.entrypoints.openai.chat_completion.protocolrr'vllm.entrypoints.openai.engine.protocolrr r r r r vllm.loggerrvllm.tokenizersr&vllm.tool_parsers.abstract_tool_parserrrr r]rregister_moduler&r!r!r!r"s4       B