j5j-~dZddlmZddlZddlZddlZddlmZddlm Z ddl m Z ddl Z ddl mZddlmZdd lmZmZdd lmZdd lmZmZmZmZdd lmZmZmZmZdd l m!Z!ejDe#Z$ddZ% ddZ&ddZ'ddZ(ddZ)ddZ* ddZ+ddZ,ddZ-ddZ. d ddZ/y) z/Validate a work order from API or command line.) annotationsN)Sequence)Path) copyfileobj)DEFAULT_ROTATE_PAGES_THRESHOLD)unpaper) OcrOptionsProcessingMode)OcrmypdfPluginManager) BadArgsErrorInputFileErrorMissingDependencyErrorOutputFileAccessError)is_file_writablerunning_in_dockerrunning_in_snap safe_symlink)check_external_programcVtjdkrtjdyy)NlzYou are running OCRmyPDF in a 32-bit (x86) Python interpreter. This is not supported. 32-bit does not have enough address space to process large files. Please use a 64-bit (x86-64) version of Python.)sysmaxsizelogwarningJ/var/www/html/qr/venv/lib/python3.12/site-packages/ocrmypdf/_validation.pycheck_platformr's$ {{e  > rcddh}|t|jz}|rtddj|d|syt|jt|z }|r)djd|D}d|d }t |y) Nequosdz^The following languages are for Tesseract's internal use and should not be issued explicitly: , z- Remove them from the -l/--language argument. c3 K|]}|ywNr).0langs r z*check_options_languages..DsAtdAs zOOCR engine does not have language data for the following requested languages: a Please install the appropriate language data for your OCR engine. See the online documentation for instructions: https://ocrmypdf.readthedocs.io/en/latest/languages.html Note: most languages are identified by a 3-letter ISO 639-2 Code. For example, English is 'eng', German is 'deu', and Spanish is 'spa'. Simplified Chinese is 'chi_sim' and Traditional Chinese is 'chi_tra'. )set languagesr joinr)optionsocr_engine_languagesDENIED_LANGUAGESblockedmissing_languages lang_textmsgs rcheck_options_languagesr21su~W%6%6!77G 0yy!"#; ;   G--.5I1JJIIA/@AA  &k   %S))!rcH|jdk(rV|jdk(r td|jtjk(r td|jdz|_|j|j k(s|j|jk(r tdy)N-z5--sidecar filename needed when output file is stdout.z?--sidecar filename needed when output file is /dev/null or NUL.z.txtz@--sidecar file must be different from the input and output files)sidecar output_filer osdevnull input_filer+s rcheck_options_sidecarr<Vs$   # %VW W  BJJ .Q "--6',,,7CVCV0V N  1Wrc|jrd|_|jr|js td|jt k7r|j s td|jrtddtjddyy)NTz&--clean is required for --unpaper-argsz7--rotate-pages is required for --rotate-pages-thresholdrz6.1z--clean, --clean-final)programpackageversion_checker need_version required_for) clean_finalclean unpaper_argsr rotate_pages_thresholdr rotate_pagesrrversionr;s rcheck_options_preprocessingrIesx GMMCDD&&*HH$$TUU}}#OO1  rc t|jtjk7ry|j|j|j |j |j|j|j|jd}td|jD}|rtddj|y)zReject options that cannot apply in strip mode. ``--mode strip`` removes the OCR text layer in place without rasterizing or running OCR, so image-processing and OCR-output options have no effect. N)z--deskewz--cleanz --clean-finalz--remove-backgroundz--rotate-pagesz --oversamplez--remove-vectorsz --sidecarc3,K|] \}}|s |ywr$r)r%namevalues rr'z&check_options_strip..sH;4%$Hs z--mode strip removes the OCR text layer without rasterizing or running OCR, so these options have no effect and are not allowed: r!)moder strip_textdeskewrDrCremove_backgroundrG oversampleremove_vectorsr6sorteditemsr r*)r+ incompatibleuseds rcheck_options_striprXys ||~000NN== ,,&88!..**#22__ L H,*<*<*>H HD  Qyy !   rcZtt|t|t|yr$)rrXr<rIr;s r_check_plugin_invariant_optionsrZs  '"(rc|j||j|j|}t||ddlm}||}|j |y)Nr;r)ValidationCoordinator) check_optionsget_ocr_enginer)r2 ocrmypdf._validation_coordinatorr\validate_all_options)r+plugin_managerr,r\ coordinators r_check_plugin_optionsrcsd   1*888ISSG%9:G'7K$$W-rc2t|t||y)aZCheck options for validity and consistency. This function coordinates validation across the entire system: 1. Core validation (platform, files, preprocessing) 2. Plugin external dependency validation 3. Plugin-specific validation (handled by plugin models) 4. Cross-cutting validation (handled by validation coordinator) N)rZrc)r+ras rr]r]s$G,'>2rc"|jdk(rWtjd|dz }t|d5}t t j j|ddd|dfSt|jdrn|jjs tdtjd|dz }t|d5}t |j|ddd|dfS |d z }t|j||tj|jfS#1swY|dfSxYw#1swY|dfSxYw#t$r?}d |j}tr|d z }nt!r|d z }t||d}~wwxYw) Nr5z reading file from standard inputstdinwbreadablez!Input file stream is not readablezreading file from input streamstreamoriginzFile not found - z Docker cannot access your working directory unless you explicitly share it with the Docker container and set up permissions correctly. You may find it easier to use stdin/stdout: docker run -i --rm jbarlow83/ocrmypdf - - output.pdf z Snap applications cannot access files outside of your home directory unless you explicitly allow it. You may find it easier to use stdin/stdout: snap run ocrmypdf - - output.pdf )r:rinfoopenrrrfbufferhasattrrhr rr8fspathFileNotFoundErrorrr)r+ work_foldertarget stream_bufferer1s rcreate_input_filerusS  34w& &$  9=  ((- 8 9w ##Z 0!!**, !DE E 12x' &$  ;= **M : ;x - 8+F ++V 4299W%7%788 8 9w  ;x ! -%g&8&8%9:C "!"!%1 ,+ -s/%D(D7,;E(D47E F:F  Fc8|jdk(r)tjjrtt dt |jdr%|jj s9tdt|jstd|jd|jrmt |jdsV|jdk7rFtt|jjrtd|jdyyyy) Nr5ztOutput was set to stdout '-' but it looks like stdout is connected to a terminal. Please redirect stdout to a file.writablezOutput stream is not writablezOutput file location (z) is not a writable file.zOutput file already exists: z6 To overwrite it, omit the --no-overwrite / -n option.) r7rstdoutisattyr rnrwrr no_overwriterstrexistsr;s rcheck_requested_output_filer}sc! ::     $$j 1""++-'(GH H g11 2#$W%8%8$99R S  ++Z8   3 & W(() * 1 1 3#*7+>+>*?@D D   4 '9 rcR|g} t|jj}t|jj}t j |5}||t |jzz} ddd||z } || zz } | dks|dkryg} hd} | D]5}t||ds| jd|jddd7|jr| jd | j||jjd r| jd |jr| jd | rd dj!| zdz}nd}t"j%d| dd|y#t$rYywxYw#1swY!xYw)Ng?ia>rPrRrCrQFz--_r5z! was issued, causing transcoding.z>--force-ocr (or --mode force) was issued, causing transcoding.pdfaz8PDF/A conversion was enabled. (Try `--output-type pdf`.)zPlugins were used.z#Possible reasons for this include: r"z@No reason for this increase is known. Please report this issue.zThe output file size is z.2fu× larger than the input file. )rstatst_sizerppikepdfrllenpagesgetattrappendreplace force_ocrextend output_type startswithpluginsr*rr)r+r:r7optimize_messages file_overhead page_overhead output_size input_sizepreasonable_overheadratioreasonable_ratioreasons image_preprocarg explanations rreport_output_file_sizers ;',,.66 *%**,44  k "Ka,mc!''l.JJK * $E"j3F&FG$*u"4GM  7C ' NNS[[c*++LM  WX NN$%%%f-QR+,rs 6" $="8:  7g! "* "*/8"* "*J   ( 6). .)>. .& 3*-Z @/3 7 777, 7  7  7 7r