j jSrSSKrSSKrSSKrSSKJr SSKJr SSKJ r SSK J r J r J r JrJr S\S\4S jrS \S\4S jrS\4S jrS rSr\S:Xa\"5 gg)a training_ppt.py =============== Training module for PPT/PPTX presentation formats. Handles text extraction from PowerPoint presentations (.ppt, .pptx), semantic chunking, and FAISS embedding creation. - .pptx: Extracted directly using python-pptx - .ppt: Converted to .pptx via LibreOffice, then extracted Shared utilities (semantic chunking, embedding creation, FAISS merging) are imported from training.py to maintain DRY principles. N) Presentation)Document)OllamaEmbeddings) split_text_with_semantic_chunkercount_tokens_in_documentssave_documents_to_txtcreate_and_save_embeddingsmerge_all_faiss file_pathreturnc[U5n/n[URSS9GHup4/nURHn[ US5(aHUR R 5(a)URUR R 55 UR(dMoURRHJnURH7nUR R 5n U (dM&URU 5 M9 ML M UR(adURR(aIURRR R 5n U (aURSU 35 U(dGMjURSUS35 URU5 GM SR!U5$)z Extract all text content from a .pptx file using python-pptx. Iterates through all slides and shapes to capture text from: - Text boxes and titles - Tables within slides - Grouped shapes (recursive extraction) - Notes sections )starttextz[Notes] z --- Slide z --- )r enumerateslidesshapeshasattrrstripappend has_tabletablerowscellshas_notes_slide notes_slidenotes_text_frameextendjoin) r prs text_parts slide_idxslide slide_textsshaperowcell cell_text notes_texts D/var/www/eduai.edurigo.com/question_generate/staging/training_ppt.py_extract_text_from_pptxr,&sT y !CJ%cjj:  \\Euf%%%***:*:*<*<""5::#3#3#56 ;;++C # $(IIOO$5 $9'..y9!*,"  U%6%6%G%G**;;@@FFHJ""Xj\#:; ;    9+T: ;   k *3;6 99Z  ppt_file_pathc X[RRU5=(d Sn[R"SSSSSUU/SSSS 9nUR S :wa%[ S UR S UR35e[RR[RRU55S n[RRXS35n[RRU5(d[ SU35eU$![a [ S 5ef=f)z Convert a legacy .ppt file to .pptx using LibreOffice headless mode. Returns the path to the newly created .pptx file. Raises RuntimeError if LibreOffice conversion fails. . libreofficez --headlessz --convert-topptxz--outdirTx)capture_outputrtimeoutrz)LibreOffice conversion failed (exit code z): zYLibreOffice is not installed or not in PATH. It is required to convert legacy .ppt files..pptxz:LibreOffice conversion produced no output file. Expected: ) ospathdirname subprocessrun returncode RuntimeErrorstderrFileNotFoundErrorsplitextbasenamer exists)r. output_dirresult base_name pptx_paths r+_convert_ppt_to_pptxrGQs./63J J         !;F>) $ $"  %  !   ;   s ADD)cr[RRU5SR5nSnUS:Xa [ U5nO,US:Xa[ U5n[ U5nO[ SUS35eUR5(da[SU35 /U(aJ[RRU5(a%[R"U5 [SU35 $$$[UXS .S 9/n[S US [U5S 35 UU(aJ[RRU5(a%[R"U5 [SU35 $$$!U(aJ[RRU5(a%[R"U5 [SU35 fff=f)a0 Load a PPT or PPTX file and return a list of LangChain Document objects. For .pptx files, text is extracted directly using python-pptx. For .ppt files, the presentation is first converted to .pptx via LibreOffice. Returns: list[Document]: List of LangChain Document objects with page_content set. rNr6z.pptzUnsupported file type: z$. Only .ppt and .pptx are supported.z(Warning: No text content extracted from z%Cleaned up temporary converted file: )sourceformat) page_contentmetadatazSuccessfully loaded z presentation: z characters extracted.) r7r8r@lowerr,rG ValueErrorrprintrBremoverlen)r file_extensionconverted_pptx_pathr documentss r+load_pptrUsWW%%i0399;N#Q W $*95D v %"6y"A *+>?D).)9:56  zz|| >2E#F#F II) * 9:M9NO P$G  !$-H   ">"2/4yk/ 1  277>>2E#F#F II) * 9:M9NO P$G  277>>2E#F#F II) * 9:M9NO P$G sA"E#()E##AF6clSnUH+nURR5nU[U5- nM- U$)zx Count total tokens across all Document objects. Uses whitespace tokenization for consistency with the existing system. r)rKsplitrQ)rT total_tokensdoctokenss r+count_tokens_in_ppt_documentsr[s= L!!'')F #  r-cSnSnSnSUSU3n[R"5n[SU35 [U5nU(d [S5 g[U5n[S U35 [ S S 9n[ XW5n[ X5 [XU5 [X5 [R"5n [S X- S S35 [SU35 g)z/Test the PPT/PPTX training pipeline standalone.ztest_presentation.pptx test_clienttest_refztemp/_z Start Time: zNo documents loaded. Exiting.NzTotal tokens: znomic-embed-text)modelzTraining process took z.2fz secondszTotal tokens processed: ) timerOrUr[rrrr r ) r client_id reference_idrC start_timedocs token_count embeddingssplit_documentsend_times r+mainrjs(IIL 1\N3JJ L %& I D  -./5K N;- ()"(:;J6tHO/6<HI,yy{H "8#8"=X FG $[M 23r-__main__)__doc__r7r:rar2rlangchain_core.documentsrlangchain_ollamartrainingrrrr r strr,rGrUr[rj__name__r-r+rss~   --(!s(!s(!V---h0Q0Qn  "4J zFr-