3 VmWi@s@dZddlmZmZddlZddlZddlZddlZddlZddl Z ddl Z ddl Z ddl Z ddlmZmZddlmZmZmZmZddlmZmZmZddlmZdd lmZd d ZGd d d e jZ e j!Z"Gddde#Z$Gddde#Z%d$ddZ&ddZ'ddZ(ddZ)ddZ*ddZ+ddZ,d d!Z-d"d#Z.dS)%aW weasyprint.pdf -------------- Post-process the PDF files created by cairo and add metadata such as hyperlinks and bookmarks. Rather than trying to parse any valid PDF, we make some assumptions that hold for cairo in order to simplify the code: * All newlines are '\n', not '\r' or '\r\n' * Except for number 0 (which is always free) there is no "free" object. * Most white space separators are made of a single 0x20 space. * Indirect dictionary objects do not contain '>>' at the start of a line except to mark the end of the object, followed by 'endobj'. (In other words, '>>' markers for sub-dictionaries are indented.) * The Page Tree is flat: all kids of the root page node are page objects, not page tree nodes. However the code uses a lot of assert statements so that if an assumptions is not true anymore, the code should (hopefully) fail with an exception rather than silently behave incorrectly. :copyright: Copyright 2011-2014 Simon Sapin and contributors, see AUTHORS. :license: BSD, see LICENSE for details. )divisionunicode_literalsN)VERSION_STRING Attachment)xrange iteritemsizipunquote) iri_to_uriurlsplitURLFetchingError) W3C_DATE_RE)LOGGERcCs&t|tr|jd}|jddddS)zlEscape parentheses and backslashes in ``value``. ``value`` must be unicode, or latin1 bytestring. latin1z\(z\)z\\)()\) isinstancebytesdecode translate)valuerN/var/www/html/enquirykeeper_venv/lib/python3.6/site-packages/weasyprint/pdf.py pdf_escape4s  rcs,eZdZdZfddZfddZZS) PDFFormatterzLike str.format except: * Results are byte strings * The new !P conversion flags encodes a PDF string. (UTF-16 BE with a BOM, then backslash-escape parentheses.) Except for fields marked !P, everything should be ASCII-only. cs<|dkr&djtd|jdjdStt|j||SdS)NPz({0})uz utf-16-ber)formatrencodersuperr convert_field)selfr conversion) __class__rrr!IszPDFFormatter.convert_fieldcstt|j|||}|jdS)Nr)r rvformatr)r" format_stringargskwargsresult)r$rrr%SszPDFFormatter.vformat)__name__ __module__ __qualname____doc__r!r% __classcell__rr)r$rr?s  rc@s@eZdZddZddZiZddZddZd d Zd d Z d S) PDFDictionarycCs||_||_dS)N) object_number byte_string)r"r0r1rrr__init__\szPDFDictionary.__init__cCs|jjt|j|jfS)N)r$r*reprr0r1)r"rrr__repr__`szPDFDictionary.__repr__cCsF|jj||f}|s4tjtd||}||j||f<|j|jjdS)Nz/{0} {1}r) _re_cachegetrecompile pdf_formatsearchr1group)r"keyZvalue_reZregexrrr get_valuefs zPDFDictionary.get_valuecCs|jddjdS)z9 :returns: the value for the /Type key. ZTypez/(\w+)ascii)r=r)r"rrrget_typemszPDFDictionary.get_typecCs$t|j|d}t|||j|S)zRead the value for `key` and follow the reference, assuming it is an indirect dictionary object. :return: a new PDFDictionary instance. z (\d+) 0 R)intr=type read_object)r"r<pdf_filer0rrrget_indirect_dictuszPDFDictionary.get_indirect_dictcsP|j|djd}|j}|j s(tt||jfddtt|DS)zRead the value for `key` and follow the references, assuming it is an array of indirect dictionary objects. :return: a list of new PDFDictionary instance. z \[(.+?)\]s 0 Rcsg|]}||qSrr).0n)class_readrr sz9PDFDictionary.get_indirect_dict_array..) r=splitpopstripAssertionErrorrArBmapr@)r"r<rCpartstrailr)rGrHrget_indirect_dict_arrays z%PDFDictionary.get_indirect_dict_arrayN) r*r+r,r2r4r5r=r?rDrQrrrrr/[s r/c@sfeZdZdZejdejZddZddZ ddZ d d Z d d Z d dZ ddZddZddZdS)PDFFilezc :param fileobj: A seekable binary file-like object for a PDF generated by cairo. s% trailer (.+) startxref (\d+) %%EOF $c Csf|jdtj|jj|jj\}}td|}t|}|j|t |}|dksVt t |}|j \}}|dksvt t|}t |}|dkst dg}xDt d|D]6}t |}|dddkst |j t|ddqW||_||_|jd|} |jd |} | jd |} | jd |} td d | Ds2t ||_| |_| |_| |_| |_d|_i|_g|_dS)Nsxref 0s0000000000 65535 f r s 00000 n ZInfoZRootZPagesZKidscss|]}|jdkVqdS)ZPageN)r?)rEprrr sz#PDFFile.__init__..Fi8)seekosSEEK_END trailer_rer:rHgroupsr/r@nextrMrJrappendfileobjobjects_offsetsrDrQall startxrefinfocatalog page_treepagesfinishedoverwritten_objects_offsetsnew_objects_offsets) r"r_trailerrblineZ first_objectZ total_objectsr`r0rcrdrerfrrrr2sB          zPDFFile.__init__cCs|j}|j|j|t|}|jds,tt|dd|ksDtg}xB|D]:}|dkr~t|dksjt|jddj|S|j|qNWdS) z :param object_number: An integer N so that 1 <= N < len(self.objects_offsets) :returns: The object content as a byte string. s 0 obj Ns>> sendobj s>>i) r_rXr`r]endswithrMr@r^join)r"r0r_rkZ object_linesrrrrBs   zPDFFile.read_objectcCs|j|||j|<dS)aWrite the new content for an existing object at the end of the file. :param object_number: An integer N so that 1 <= N < len(self.objects_offsets) :param byte_string: The new object content as a byte string. N) _write_objectrh)r"r0r1rrroverwrite_objects zPDFFile.overwrite_objectcCs4|jjdst|j|j|jdd|ddS)zaOverwrite a dictionary object after adding content inside the << >> delimiters. s>>Ns >>)r1rnrMrqr0)r" dictionaryZ new_contentrrr extend_dictszPDFFile.extend_dictcCst|jt|jS)zKReturn the object number that would be used by write_new_object(). )lenr`ri)r"rrrnext_object_numberszPDFFile.next_object_numbercCs |j}|jj|j|||S)zWrite a new object at the end of the file. :param byte_string: The object content as a byte string. :return: The new object number. )rwrir^rp)r"r1r0rrrwrite_new_objects zPDFFile.write_new_objectc Cs|j\}}d|_|dx&t|jD]\}}|td||q&W|jrt|j}|td|t|jx(t|j|dD]\}}|td|qxW|td|j |j j |j j |j |dd S) z Write the cross-reference table and the trailer for the new and overwritten objects. This makes `fileobj` a valid (updated) PDF file. Tsxref z{0} 1 {1:010} 00000 n z{0} {1} )startz{0:010} 00000 n zftrailer << /Size {size} /Root {root} 0 R /Info {info} 0 R /Prev {prev} >> startxref {startxref} %%EOF )sizerootrcprevrbN)_start_writingrgrrhr9rirvr` enumeraterwrdr0rcrb)r"Z new_startxrefwriter0offsetZfirst_new_objectrrrfinishs*  zPDFFile.finishcCs.|j\}}|td||||d|S)Nz {0} 0 obj s endobj )r}r9)r"r0r1rrrrrrp+s  zPDFFile._write_objectcCs.|j s t|j}|jdtj|j|jfS)Nr)rgrMr_rXrYrZtellr)r"r_rrrr}2s zPDFFile._start_writingN)r*r+r,r-r7r8DOTALLr[r2rBrqrurwrxrrpr}rrrrrRs /  #rRccsBx<|D]4\}}}|||fVxt||dD] }|Vq,WqWdS)Nr)flatten_bookmarks) bookmarksdepthlabeltargetchildrenr)rrrr9s rc sfdd|jD}g}xt|j|D]\}}g}x|D]\}} } |dkrl| \} } } | f|| j| | } | \}}}}|j||\}}|j||\}}||||||f} |j|| | fq:W|j|q(Wddi}g}|g}|g}xtt|j|dD]\}\}} }| \} } } | f|| j| | } ddddd||d|| d}|t |dkrp|||dd <n,|||d <|||d <||d=||d=x&t |D]}||dd7<qW|||dd <|j||j||j|qW|||fS) ayChange metadata into data structures closer to the PDF objects. In particular, convert from WeasyPrint units (CSS pixels from the top-left corner) to PDF units (points from the bottom-left corner.) :param scale: PDF points per CSS pixels. Defaults to 0.75, but is affected by `zoom` in :meth:`weasyprint.document.Document.write_pdf`. cs$g|]}tj |jdqS))ZxxyyZy0)cairoZMatrixheight)rEpage)scalerrrIMsz$prepare_metadata..internalCountrrN)rFirstLastPrevNextParentrrrrrr) rfr Z resolve_linksZtransform_pointZtransform_distancer^r~rZmake_bookmark_treervrange)documentbookmark_root_idrZmatriceslinks page_linksZmatrixZnew_page_links link_typer rectangleZ target_pageZtarget_xZtarget_yZrect_xZrect_ywidthr bookmark_rootZ bookmark_listZlast_id_by_depthZ last_by_depthZ bookmark_idrrbookmarkir)rrprepare_metadata@sT                rcsJ|j}|d}|d}|d}|d}|j\}}|td||td||||dd} d} tj} tj} xNtfd d d D]8} | t| 7} | j | | j | }| t|7} ||qW| j tj }| t|7} |||d |d |j j||jtd| |jtd| j|jtd| |j|ksFt|S)a Write a file like object as ``/EmbeddedFile``, compressing it with deflate. In fact, this method writes multiple PDF objects to include length, compressed length and MD5 checksum. :return: the object number of the compressed file stream object rrrz {0} 0 obj zm<< /Type /EmbeddedFile /Length {0} 0 R /Filter /FlateDecode /Params << /CheckSum {1} 0 R /Size {2} 0 R >> >> sstream rcs jdS)Ni)rHr)filerrsz/_write_compressed_file_object..rms endstream sendobj z{0}z<{0}>)rwr}r9hashlibmd5zlib compressobjiterrvupdatecompressflushZ_FINISHrir^rx hexdigestrM)pdfrr0Zexpected_next_object_numberZ length_numberZ md5_numberZuncompressed_length_numberrrZuncompressed_lengthZcompressed_lengthrrdata compressedr)rr_write_compressed_file_objects@           rcCsd}|r|jd}|r|S|rLt|}|jdkrL|jjdd}|dkrLd}|dkrd}|r|jd}|dkrtd }qtj|pd }nd }d |}n>tjd d krt |}t |t s|j d}|j d}nt |}|S)z Derives a filename from a fetched resource. This is either the filename returned by the URL fetcher, the last URL path component or a synthetic name if the URL has no path Nfilenamer/r mime_typez text/plainz.txtz.bin attachmentrrrzutf-8)r6r schemepathrJ mimetypesZguess_extensionsys version_infor rrrr)urlr)rrJ extensionrrrr_get_filename_from_results6        rcCs~g}x*|D]"}t|||}|dk r |j|q Wt|dkr@dSdg}x|D]}|jtd|qLW|jd|jdj|S)z Writes attachments as embedded files (document attachments). :return: the object number of the name dictionary or :obj:`None` Nrs << /Names [z (attachment{0}) {0} 0 R s ] >>rm)_write_pdf_attachmentr^rvr9rxro)r attachments url_fetcherZ file_spec_idsrZ file_spec_idcontentfsrrr_write_pdf_embedded_filess      rc Csytt|tr$|\}}t|||d}nt|ts:t||d}|j,\}}}}t|tr^tj|}t||}WdQRXWn,tk r} zt j d| dSd} ~ XnXt |d} |j t d| ||jpdS)z Writes an attachment to the PDF stream :return: the object number of the ``/Filespec`` object or :obj:`None` if the attachment couldn't be read. )rr description)ZguessrNzFailed to load attachment: %szF<< /Type /Filespec /F () /UF {0!P} /EF << /F {1} 0 R >> /Desc {2!P} >>r)rtuplersourcerioBytesIOrr rwarningrrxr9r) rrrrrZ source_typer_Zfile_stream_idexcrrrrrs(       rcCsNi}xD|D]<}x6|D].\}}}|dkr||krt||df|||<qWq W|S)z Write all annotation attachments to the PDF file. :return: a dictionary that maps URLs to PDF object numbers, which can be :obj:`None` if the resource failed to load. rN)r)rrr annot_filesrrrrrrr_write_pdf_annotation_files=s rcCst|}|j}t|||\}} } | r|jtd|d|d|dx| D]} td| dg} | d\} }}| jtd|j| j||| dr| jtd | dx*d3D]"}| |r| jtd || |qW| jd|jdj| qLWt ||j |pg|}| s|dk rVd}| r0|td|7}|dk rH|td|7}|j |j |t || |}x*t|j| D]\}}g}x|D]\}}}td4|g} |dks||dkr| jd|dkr| jtd5|n| jtdtt|nD||dk s t|jtd6|}| jd| jtd|||| jd|j|jdj| qW|rr|j |tddjdd|DqrWtd tg}x8d:D]0\}}t||}|dk r|jtd'||qWx>d=D]6\}}t||}|dk r|jtd'|d,j|qWx>d@D]6\}}tt|||}|dk r|jtd1||qW|jd2|j|jjdj||jdS)Az:Append to a seekable file-like object to add PDF metadata.z=<< /Type /Outlines /Count {0} /First {1} 0 R /Last {2} 0 R >>rrrz<< /Title {0!P} rrz@/A << /Type /Action /S /GoTo /D [{0} 0 R /XYZ {1:f} {2:f} 0] >> z /Count {0} rrrz /{0} {1} 0 R s>>rmNz) /Outlines {0} 0 R /PageMode /UseOutlinesz$ /Names << /EmbeddedFiles {0} 0 R >>@<< /Type /Annot /Rect [{0:f} {1:f} {2:f} {3:f}] /Border [0 0 0] rs/Subtype /Link r> z*/A << /Type /Action /S /URI /URI ({0}) >> ^<< /Type /XObject /Subtype /Form /BBox [{0:f} {1:f} {2:f} {3:f}] /Length 0 >> stream endstreams/Subtype /FileAttachment z&/T () /FS {0} 0 R /AP << /N {1} 0 R >>z /Annots [{0}] css|]}dj|VqdS)z{0} 0 RN)r)rErFrrrrWsz%write_pdf_metadata..z<< /Producer {0!P} titleTitlerSubject generatorCreatorz /{0} {1!P}authorsAuthorkeywordsKeywordsz, created CreationDatemodifiedModDatez /{0} (D:{1})s >>)rrrrr)r)r)rrrrrrr)rrrrrrr)rrrrrr)rr)rRrwrrxr9r^rfr0rorrrurdrziprr rMrgetattrw3c_date_to_pdfrqrcr)rr_rmetadatarrrrrrrrrZpage_numZpos_xZpos_yr<Zembedded_files_idparamsrrr annotationsrrrZlink_aprcattrrrrrwrite_pdf_metadataOs                     rcCs|dkr dStj|}|dkr0tjd||dS|j}|d|dpHd|dpTd|dp`d|dpld|d pxd}|dr|dst|d s|d 7}|d r|d jdst|dst|d|d |df7}n|d7}|S)z YYYYMMDDHHmmSSOHH'mm' NzInvalid %s date: %ryearmonthrdayhourminutesecondZ00Ztz_hour+-Z tz_minutez%s'%s'Z)rr)rmatchrr groupdictrM startswith)string attr_namerr\Zpdf_daterrrrs& 6  r)r)/r- __future__rrrrrrYr7rrrZ cairocffirrrrcompatrrr r Zurlsr r r htmlrloggerrr Formatterrrr9objectr/rRrrrrrrrrrrrrrs<   6) D;>&v