Content type for word document

Here is an (almost) complete file extensions’s MIME in a JSON format.
You can do these examples: MIME["ppt"], MIME["docx"], etc

{
    "x3d": "application/vnd.hzn-3d-crossword",
    "3gp": "video/3gpp",
    "3g2": "video/3gpp2",
    "mseq": "application/vnd.mseq",
    "pwn": "application/vnd.3m.post-it-notes",
    "plb": "application/vnd.3gpp.pic-bw-large",
    "psb": "application/vnd.3gpp.pic-bw-small",
    "pvb": "application/vnd.3gpp.pic-bw-var",
    "tcap": "application/vnd.3gpp2.tcap",
    "7z": "application/x-7z-compressed",
    "abw": "application/x-abiword",
    "ace": "application/x-ace-compressed",
    "acc": "application/vnd.americandynamics.acc",
    "acu": "application/vnd.acucobol",
    "atc": "application/vnd.acucorp",
    "adp": "audio/adpcm",
    "aab": "application/x-authorware-bin",
    "aam": "application/x-authorware-map",
    "aas": "application/x-authorware-seg",
    "air": "application/vnd.adobe.air-application-installer-package+zip",
    "swf": "application/x-shockwave-flash",
    "fxp": "application/vnd.adobe.fxp",
    "pdf": "application/pdf",
    "ppd": "application/vnd.cups-ppd",
    "dir": "application/x-director",
    "xdp": "application/vnd.adobe.xdp+xml",
    "xfdf": "application/vnd.adobe.xfdf",
    "aac": "audio/x-aac",
    "ahead": "application/vnd.ahead.space",
    "azf": "application/vnd.airzip.filesecure.azf",
    "azs": "application/vnd.airzip.filesecure.azs",
    "azw": "application/vnd.amazon.ebook",
    "ami": "application/vnd.amiga.ami",
    "N/A": "application/andrew-inset",
    "apk": "application/vnd.android.package-archive",
    "cii": "application/vnd.anser-web-certificate-issue-initiation",
    "fti": "application/vnd.anser-web-funds-transfer-initiation",
    "atx": "application/vnd.antix.game-component",
    "dmg": "application/x-apple-diskimage",
    "mpkg": "application/vnd.apple.installer+xml",
    "aw": "application/applixware",
    "mp3": "audio/mpeg",
    "les": "application/vnd.hhe.lesson-player",
    "swi": "application/vnd.aristanetworks.swi",
    "s": "text/x-asm",
    "atomcat": "application/atomcat+xml",
    "atomsvc": "application/atomsvc+xml",
    "atom": "application/atom+xml",
    "ac": "application/pkix-attr-cert",
    "aif": "audio/x-aiff",
    "avi": "video/x-msvideo",
    "aep": "application/vnd.audiograph",
    "dxf": "image/vnd.dxf",
    "dwf": "model/vnd.dwf",
    "par": "text/plain-bas",
    "bcpio": "application/x-bcpio",
    "bin": "application/octet-stream",
    "bmp": "image/bmp",
    "torrent": "application/x-bittorrent",
    "cod": "application/vnd.rim.cod",
    "mpm": "application/vnd.blueice.multipass",
    "bmi": "application/vnd.bmi",
    "sh": "application/x-sh",
    "btif": "image/prs.btif",
    "rep": "application/vnd.businessobjects",
    "bz": "application/x-bzip",
    "bz2": "application/x-bzip2",
    "csh": "application/x-csh",
    "c": "text/x-c",
    "cdxml": "application/vnd.chemdraw+xml",
    "css": "text/css",
    "cdx": "chemical/x-cdx",
    "cml": "chemical/x-cml",
    "csml": "chemical/x-csml",
    "cdbcmsg": "application/vnd.contact.cmsg",
    "cla": "application/vnd.claymore",
    "c4g": "application/vnd.clonk.c4group",
    "sub": "image/vnd.dvb.subtitle",
    "cdmia": "application/cdmi-capability",
    "cdmic": "application/cdmi-container",
    "cdmid": "application/cdmi-domain",
    "cdmio": "application/cdmi-object",
    "cdmiq": "application/cdmi-queue",
    "c11amc": "application/vnd.cluetrust.cartomobile-config",
    "c11amz": "application/vnd.cluetrust.cartomobile-config-pkg",
    "ras": "image/x-cmu-raster",
    "dae": "model/vnd.collada+xml",
    "csv": "text/csv",
    "cpt": "application/mac-compactpro",
    "wmlc": "application/vnd.wap.wmlc",
    "cgm": "image/cgm",
    "ice": "x-conference/x-cooltalk",
    "cmx": "image/x-cmx",
    "xar": "application/vnd.xara",
    "cmc": "application/vnd.cosmocaller",
    "cpio": "application/x-cpio",
    "clkx": "application/vnd.crick.clicker",
    "clkk": "application/vnd.crick.clicker.keyboard",
    "clkp": "application/vnd.crick.clicker.palette",
    "clkt": "application/vnd.crick.clicker.template",
    "clkw": "application/vnd.crick.clicker.wordbank",
    "wbs": "application/vnd.criticaltools.wbs+xml",
    "cryptonote": "application/vnd.rig.cryptonote",
    "cif": "chemical/x-cif",
    "cmdf": "chemical/x-cmdf",
    "cu": "application/cu-seeme",
    "cww": "application/prs.cww",
    "curl": "text/vnd.curl",
    "dcurl": "text/vnd.curl.dcurl",
    "mcurl": "text/vnd.curl.mcurl",
    "scurl": "text/vnd.curl.scurl",
    "car": "application/vnd.curl.car",
    "pcurl": "application/vnd.curl.pcurl",
    "cmp": "application/vnd.yellowriver-custom-menu",
    "dssc": "application/dssc+der",
    "xdssc": "application/dssc+xml",
    "deb": "application/x-debian-package",
    "uva": "audio/vnd.dece.audio",
    "uvi": "image/vnd.dece.graphic",
    "uvh": "video/vnd.dece.hd",
    "uvm": "video/vnd.dece.mobile",
    "uvu": "video/vnd.uvvu.mp4",
    "uvp": "video/vnd.dece.pd",
    "uvs": "video/vnd.dece.sd",
    "uvv": "video/vnd.dece.video",
    "dvi": "application/x-dvi",
    "seed": "application/vnd.fdsn.seed",
    "dtb": "application/x-dtbook+xml",
    "res": "application/x-dtbresource+xml",
    "ait": "application/vnd.dvb.ait",
    "svc": "application/vnd.dvb.service",
    "eol": "audio/vnd.digital-winds",
    "djvu": "image/vnd.djvu",
    "dtd": "application/xml-dtd",
    "mlp": "application/vnd.dolby.mlp",
    "wad": "application/x-doom",
    "dpg": "application/vnd.dpgraph",
    "dra": "audio/vnd.dra",
    "dfac": "application/vnd.dreamfactory",
    "dts": "audio/vnd.dts",
    "dtshd": "audio/vnd.dts.hd",
    "dwg": "image/vnd.dwg",
    "geo": "application/vnd.dynageo",
    "es": "application/ecmascript",
    "mag": "application/vnd.ecowin.chart",
    "mmr": "image/vnd.fujixerox.edmics-mmr",
    "rlc": "image/vnd.fujixerox.edmics-rlc",
    "exi": "application/exi",
    "mgz": "application/vnd.proteus.magazine",
    "epub": "application/epub+zip",
    "eml": "message/rfc822",
    "nml": "application/vnd.enliven",
    "xpr": "application/vnd.is-xpr",
    "xif": "image/vnd.xiff",
    "xfdl": "application/vnd.xfdl",
    "emma": "application/emma+xml",
    "ez2": "application/vnd.ezpix-album",
    "ez3": "application/vnd.ezpix-package",
    "fst": "image/vnd.fst",
    "fvt": "video/vnd.fvt",
    "fbs": "image/vnd.fastbidsheet",
    "fe_launch": "application/vnd.denovo.fcselayout-link",
    "f4v": "video/x-f4v",
    "flv": "video/x-flv",
    "fpx": "image/vnd.fpx",
    "npx": "image/vnd.net-fpx",
    "flx": "text/vnd.fmi.flexstor",
    "fli": "video/x-fli",
    "ftc": "application/vnd.fluxtime.clip",
    "fdf": "application/vnd.fdf",
    "f": "text/x-fortran",
    "mif": "application/vnd.mif",
    "fm": "application/vnd.framemaker",
    "fh": "image/x-freehand",
    "fsc": "application/vnd.fsc.weblaunch",
    "fnc": "application/vnd.frogans.fnc",
    "ltf": "application/vnd.frogans.ltf",
    "ddd": "application/vnd.fujixerox.ddd",
    "xdw": "application/vnd.fujixerox.docuworks",
    "xbd": "application/vnd.fujixerox.docuworks.binder",
    "oas": "application/vnd.fujitsu.oasys",
    "oa2": "application/vnd.fujitsu.oasys2",
    "oa3": "application/vnd.fujitsu.oasys3",
    "fg5": "application/vnd.fujitsu.oasysgp",
    "bh2": "application/vnd.fujitsu.oasysprs",
    "spl": "application/x-futuresplash",
    "fzs": "application/vnd.fuzzysheet",
    "g3": "image/g3fax",
    "gmx": "application/vnd.gmx",
    "gtw": "model/vnd.gtw",
    "txd": "application/vnd.genomatix.tuxedo",
    "ggb": "application/vnd.geogebra.file",
    "ggt": "application/vnd.geogebra.tool",
    "gdl": "model/vnd.gdl",
    "gex": "application/vnd.geometry-explorer",
    "gxt": "application/vnd.geonext",
    "g2w": "application/vnd.geoplan",
    "g3w": "application/vnd.geospace",
    "gsf": "application/x-font-ghostscript",
    "bdf": "application/x-font-bdf",
    "gtar": "application/x-gtar",
    "texinfo": "application/x-texinfo",
    "gnumeric": "application/x-gnumeric",
    "kml": "application/vnd.google-earth.kml+xml",
    "kmz": "application/vnd.google-earth.kmz",
    "gqf": "application/vnd.grafeq",
    "gif": "image/gif",
    "gv": "text/vnd.graphviz",
    "gac": "application/vnd.groove-account",
    "ghf": "application/vnd.groove-help",
    "gim": "application/vnd.groove-identity-message",
    "grv": "application/vnd.groove-injector",
    "gtm": "application/vnd.groove-tool-message",
    "tpl": "application/vnd.groove-tool-template",
    "vcg": "application/vnd.groove-vcard",
    "h261": "video/h261",
    "h263": "video/h263",
    "h264": "video/h264",
    "hpid": "application/vnd.hp-hpid",
    "hps": "application/vnd.hp-hps",
    "hdf": "application/x-hdf",
    "rip": "audio/vnd.rip",
    "hbci": "application/vnd.hbci",
    "jlt": "application/vnd.hp-jlyt",
    "pcl": "application/vnd.hp-pcl",
    "hpgl": "application/vnd.hp-hpgl",
    "hvs": "application/vnd.yamaha.hv-script",
    "hvd": "application/vnd.yamaha.hv-dic",
    "hvp": "application/vnd.yamaha.hv-voice",
    "sfd-hdstx": "application/vnd.hydrostatix.sof-data",
    "stk": "application/hyperstudio",
    "hal": "application/vnd.hal+xml",
    "html": "text/html",
    "irm": "application/vnd.ibm.rights-management",
    "sc": "application/vnd.ibm.secure-container",
    "ics": "text/calendar",
    "icc": "application/vnd.iccprofile",
    "ico": "image/x-icon",
    "igl": "application/vnd.igloader",
    "ief": "image/ief",
    "ivp": "application/vnd.immervision-ivp",
    "ivu": "application/vnd.immervision-ivu",
    "rif": "application/reginfo+xml",
    "3dml": "text/vnd.in3d.3dml",
    "spot": "text/vnd.in3d.spot",
    "igs": "model/iges",
    "i2g": "application/vnd.intergeo",
    "cdy": "application/vnd.cinderella",
    "xpw": "application/vnd.intercon.formnet",
    "fcs": "application/vnd.isac.fcs",
    "ipfix": "application/ipfix",
    "cer": "application/pkix-cert",
    "pki": "application/pkixcmp",
    "crl": "application/pkix-crl",
    "pkipath": "application/pkix-pkipath",
    "igm": "application/vnd.insors.igm",
    "rcprofile": "application/vnd.ipunplugged.rcprofile",
    "irp": "application/vnd.irepository.package+xml",
    "jad": "text/vnd.sun.j2me.app-descriptor",
    "jar": "application/java-archive",
    "class": "application/java-vm",
    "jnlp": "application/x-java-jnlp-file",
    "ser": "application/java-serialized-object",
    "java": "text/x-java-source,java",
    "js": "application/javascript",
    "json": "application/json",
    "joda": "application/vnd.joost.joda-archive",
    "jpm": "video/jpm",
    "jpeg": "image/x-citrix-jpeg",
    "jpg": "image/x-citrix-jpeg",
    "pjpeg": "image/pjpeg",
    "jpgv": "video/jpeg",
    "ktz": "application/vnd.kahootz",
    "mmd": "application/vnd.chipnuts.karaoke-mmd",
    "karbon": "application/vnd.kde.karbon",
    "chrt": "application/vnd.kde.kchart",
    "kfo": "application/vnd.kde.kformula",
    "flw": "application/vnd.kde.kivio",
    "kon": "application/vnd.kde.kontour",
    "kpr": "application/vnd.kde.kpresenter",
    "ksp": "application/vnd.kde.kspread",
    "kwd": "application/vnd.kde.kword",
    "htke": "application/vnd.kenameaapp",
    "kia": "application/vnd.kidspiration",
    "kne": "application/vnd.kinar",
    "sse": "application/vnd.kodak-descriptor",
    "lasxml": "application/vnd.las.las+xml",
    "latex": "application/x-latex",
    "lbd": "application/vnd.llamagraphics.life-balance.desktop",
    "lbe": "application/vnd.llamagraphics.life-balance.exchange+xml",
    "jam": "application/vnd.jam",
    "123": "application/vnd.lotus-1-2-3",
    "apr": "application/vnd.lotus-approach",
    "pre": "application/vnd.lotus-freelance",
    "nsf": "application/vnd.lotus-notes",
    "org": "application/vnd.lotus-organizer",
    "scm": "application/vnd.lotus-screencam",
    "lwp": "application/vnd.lotus-wordpro",
    "lvp": "audio/vnd.lucent.voice",
    "m3u": "audio/x-mpegurl",
    "m4v": "video/x-m4v",
    "hqx": "application/mac-binhex40",
    "portpkg": "application/vnd.macports.portpkg",
    "mgp": "application/vnd.osgeo.mapguide.package",
    "mrc": "application/marc",
    "mrcx": "application/marcxml+xml",
    "mxf": "application/mxf",
    "nbp": "application/vnd.wolfram.player",
    "ma": "application/mathematica",
    "mathml": "application/mathml+xml",
    "mbox": "application/mbox",
    "mc1": "application/vnd.medcalcdata",
    "mscml": "application/mediaservercontrol+xml",
    "cdkey": "application/vnd.mediastation.cdkey",
    "mwf": "application/vnd.mfer",
    "mfm": "application/vnd.mfmp",
    "msh": "model/mesh",
    "mads": "application/mads+xml",
    "mets": "application/mets+xml",
    "mods": "application/mods+xml",
    "meta4": "application/metalink4+xml",
    "mcd": "application/vnd.mcd",
    "flo": "application/vnd.micrografx.flo",
    "igx": "application/vnd.micrografx.igx",
    "es3": "application/vnd.eszigno3+xml",
    "mdb": "application/x-msaccess",
    "asf": "video/x-ms-asf",
    "exe": "application/x-msdownload",
    "cil": "application/vnd.ms-artgalry",
    "cab": "application/vnd.ms-cab-compressed",
    "ims": "application/vnd.ms-ims",
    "application": "application/x-ms-application",
    "clp": "application/x-msclip",
    "mdi": "image/vnd.ms-modi",
    "eot": "application/vnd.ms-fontobject",
    "xls": "application/vnd.ms-excel",
    "xlam": "application/vnd.ms-excel.addin.macroenabled.12",
    "xlsb": "application/vnd.ms-excel.sheet.binary.macroenabled.12",
    "xltm": "application/vnd.ms-excel.template.macroenabled.12",
    "xlsm": "application/vnd.ms-excel.sheet.macroenabled.12",
    "chm": "application/vnd.ms-htmlhelp",
    "crd": "application/x-mscardfile",
    "lrm": "application/vnd.ms-lrm",
    "mvb": "application/x-msmediaview",
    "mny": "application/x-msmoney",
    "pptx": "application/vnd.openxmlformats-officedocument.presentationml.presentation",
    "sldx": "application/vnd.openxmlformats-officedocument.presentationml.slide",
    "ppsx": "application/vnd.openxmlformats-officedocument.presentationml.slideshow",
    "potx": "application/vnd.openxmlformats-officedocument.presentationml.template",
    "xlsx": "application/vnd.openxmlformats-officedocument.spreadsheetml.sheet",
    "xltx": "application/vnd.openxmlformats-officedocument.spreadsheetml.template",
    "docx": "application/vnd.openxmlformats-officedocument.wordprocessingml.document",
    "dotx": "application/vnd.openxmlformats-officedocument.wordprocessingml.template",
    "obd": "application/x-msbinder",
    "thmx": "application/vnd.ms-officetheme",
    "onetoc": "application/onenote",
    "pya": "audio/vnd.ms-playready.media.pya",
    "pyv": "video/vnd.ms-playready.media.pyv",
    "ppt": "application/vnd.ms-powerpoint",
    "ppam": "application/vnd.ms-powerpoint.addin.macroenabled.12",
    "sldm": "application/vnd.ms-powerpoint.slide.macroenabled.12",
    "pptm": "application/vnd.ms-powerpoint.presentation.macroenabled.12",
    "ppsm": "application/vnd.ms-powerpoint.slideshow.macroenabled.12",
    "potm": "application/vnd.ms-powerpoint.template.macroenabled.12",
    "mpp": "application/vnd.ms-project",
    "pub": "application/x-mspublisher",
    "scd": "application/x-msschedule",
    "xap": "application/x-silverlight-app",
    "stl": "application/vnd.ms-pki.stl",
    "cat": "application/vnd.ms-pki.seccat",
    "vsd": "application/vnd.visio",
    "vsdx": "application/vnd.visio2013",
    "wm": "video/x-ms-wm",
    "wma": "audio/x-ms-wma",
    "wax": "audio/x-ms-wax",
    "wmx": "video/x-ms-wmx",
    "wmd": "application/x-ms-wmd",
    "wpl": "application/vnd.ms-wpl",
    "wmz": "application/x-ms-wmz",
    "wmv": "video/x-ms-wmv",
    "wvx": "video/x-ms-wvx",
    "wmf": "application/x-msmetafile",
    "trm": "application/x-msterminal",
    "doc": "application/msword",
    "docm": "application/vnd.ms-word.document.macroenabled.12",
    "dotm": "application/vnd.ms-word.template.macroenabled.12",
    "wri": "application/x-mswrite",
    "wps": "application/vnd.ms-works",
    "xbap": "application/x-ms-xbap",
    "xps": "application/vnd.ms-xpsdocument",
    "mid": "audio/midi",
    "mpy": "application/vnd.ibm.minipay",
    "afp": "application/vnd.ibm.modcap",
    "rms": "application/vnd.jcp.javame.midlet-rms",
    "tmo": "application/vnd.tmobile-livetv",
    "prc": "application/x-mobipocket-ebook",
    "mbk": "application/vnd.mobius.mbk",
    "dis": "application/vnd.mobius.dis",
    "plc": "application/vnd.mobius.plc",
    "mqy": "application/vnd.mobius.mqy",
    "msl": "application/vnd.mobius.msl",
    "txf": "application/vnd.mobius.txf",
    "daf": "application/vnd.mobius.daf",
    "fly": "text/vnd.fly",
    "mpc": "application/vnd.mophun.certificate",
    "mpn": "application/vnd.mophun.application",
    "mj2": "video/mj2",
    "mpga": "audio/mpeg",
    "mxu": "video/vnd.mpegurl",
    "mpeg": "video/mpeg",
    "m21": "application/mp21",
    "mp4a": "audio/mp4",
    "mp4": "video/mp4",
    "m3u8": "application/vnd.apple.mpegurl",
    "mus": "application/vnd.musician",
    "msty": "application/vnd.muvee.style",
    "mxml": "application/xv+xml",
    "ngdat": "application/vnd.nokia.n-gage.data",
    "n-gage": "application/vnd.nokia.n-gage.symbian.install",
    "ncx": "application/x-dtbncx+xml",
    "nc": "application/x-netcdf",
    "nlu": "application/vnd.neurolanguage.nlu",
    "dna": "application/vnd.dna",
    "nnd": "application/vnd.noblenet-directory",
    "nns": "application/vnd.noblenet-sealer",
    "nnw": "application/vnd.noblenet-web",
    "rpst": "application/vnd.nokia.radio-preset",
    "rpss": "application/vnd.nokia.radio-presets",
    "n3": "text/n3",
    "edm": "application/vnd.novadigm.edm",
    "edx": "application/vnd.novadigm.edx",
    "ext": "application/vnd.novadigm.ext",
    "gph": "application/vnd.flographit",
    "ecelp4800": "audio/vnd.nuera.ecelp4800",
    "ecelp7470": "audio/vnd.nuera.ecelp7470",
    "ecelp9600": "audio/vnd.nuera.ecelp9600",
    "oda": "application/oda",
    "ogx": "application/ogg",
    "oga": "audio/ogg",
    "ogv": "video/ogg",
    "dd2": "application/vnd.oma.dd2+xml",
    "oth": "application/vnd.oasis.opendocument.text-web",
    "opf": "application/oebps-package+xml",
    "qbo": "application/vnd.intu.qbo",
    "oxt": "application/vnd.openofficeorg.extension",
    "osf": "application/vnd.yamaha.openscoreformat",
    "weba": "audio/webm",
    "webm": "video/webm",
    "odc": "application/vnd.oasis.opendocument.chart",
    "otc": "application/vnd.oasis.opendocument.chart-template",
    "odb": "application/vnd.oasis.opendocument.database",
    "odf": "application/vnd.oasis.opendocument.formula",
    "odft": "application/vnd.oasis.opendocument.formula-template",
    "odg": "application/vnd.oasis.opendocument.graphics",
    "otg": "application/vnd.oasis.opendocument.graphics-template",
    "odi": "application/vnd.oasis.opendocument.image",
    "oti": "application/vnd.oasis.opendocument.image-template",
    "odp": "application/vnd.oasis.opendocument.presentation",
    "otp": "application/vnd.oasis.opendocument.presentation-template",
    "ods": "application/vnd.oasis.opendocument.spreadsheet",
    "ots": "application/vnd.oasis.opendocument.spreadsheet-template",
    "odt": "application/vnd.oasis.opendocument.text",
    "odm": "application/vnd.oasis.opendocument.text-master",
    "ott": "application/vnd.oasis.opendocument.text-template",
    "ktx": "image/ktx",
    "sxc": "application/vnd.sun.xml.calc",
    "stc": "application/vnd.sun.xml.calc.template",
    "sxd": "application/vnd.sun.xml.draw",
    "std": "application/vnd.sun.xml.draw.template",
    "sxi": "application/vnd.sun.xml.impress",
    "sti": "application/vnd.sun.xml.impress.template",
    "sxm": "application/vnd.sun.xml.math",
    "sxw": "application/vnd.sun.xml.writer",
    "sxg": "application/vnd.sun.xml.writer.global",
    "stw": "application/vnd.sun.xml.writer.template",
    "otf": "application/x-font-otf",
    "osfpvg": "application/vnd.yamaha.openscoreformat.osfpvg+xml",
    "dp": "application/vnd.osgi.dp",
    "pdb": "application/vnd.palm",
    "p": "text/x-pascal",
    "paw": "application/vnd.pawaafile",
    "pclxl": "application/vnd.hp-pclxl",
    "efif": "application/vnd.picsel",
    "pcx": "image/x-pcx",
    "psd": "image/vnd.adobe.photoshop",
    "prf": "application/pics-rules",
    "pic": "image/x-pict",
    "chat": "application/x-chat",
    "p10": "application/pkcs10",
    "p12": "application/x-pkcs12",
    "p7m": "application/pkcs7-mime",
    "p7s": "application/pkcs7-signature",
    "p7r": "application/x-pkcs7-certreqresp",
    "p7b": "application/x-pkcs7-certificates",
    "p8": "application/pkcs8",
    "plf": "application/vnd.pocketlearn",
    "pnm": "image/x-portable-anymap",
    "pbm": "image/x-portable-bitmap",
    "pcf": "application/x-font-pcf",
    "pfr": "application/font-tdpfr",
    "pgn": "application/x-chess-pgn",
    "pgm": "image/x-portable-graymap",
    "png": "image/x-png",
    "ppm": "image/x-portable-pixmap",
    "pskcxml": "application/pskc+xml",
    "pml": "application/vnd.ctc-posml",
    "ai": "application/postscript",
    "pfa": "application/x-font-type1",
    "pbd": "application/vnd.powerbuilder6",
    "pgp": "application/pgp-signature",
    "box": "application/vnd.previewsystems.box",
    "ptid": "application/vnd.pvi.ptid1",
    "pls": "application/pls+xml",
    "str": "application/vnd.pg.format",
    "ei6": "application/vnd.pg.osasli",
    "dsc": "text/prs.lines.tag",
    "psf": "application/x-font-linux-psf",
    "qps": "application/vnd.publishare-delta-tree",
    "wg": "application/vnd.pmi.widget",
    "qxd": "application/vnd.quark.quarkxpress",
    "esf": "application/vnd.epson.esf",
    "msf": "application/vnd.epson.msf",
    "ssf": "application/vnd.epson.ssf",
    "qam": "application/vnd.epson.quickanime",
    "qfx": "application/vnd.intu.qfx",
    "qt": "video/quicktime",
    "rar": "application/x-rar-compressed",
    "ram": "audio/x-pn-realaudio",
    "rmp": "audio/x-pn-realaudio-plugin",
    "rsd": "application/rsd+xml",
    "rm": "application/vnd.rn-realmedia",
    "bed": "application/vnd.realvnc.bed",
    "mxl": "application/vnd.recordare.musicxml",
    "musicxml": "application/vnd.recordare.musicxml+xml",
    "rnc": "application/relax-ng-compact-syntax",
    "rdz": "application/vnd.data-vision.rdz",
    "rdf": "application/rdf+xml",
    "rp9": "application/vnd.cloanto.rp9",
    "jisp": "application/vnd.jisp",
    "rtf": "application/rtf",
    "rtx": "text/richtext",
    "link66": "application/vnd.route66.link66+xml",
    "rss": "application/rss+xml",
    "shf": "application/shf+xml",
    "st": "application/vnd.sailingtracker.track",
    "svg": "image/svg+xml",
    "sus": "application/vnd.sus-calendar",
    "sru": "application/sru+xml",
    "setpay": "application/set-payment-initiation",
    "setreg": "application/set-registration-initiation",
    "sema": "application/vnd.sema",
    "semd": "application/vnd.semd",
    "semf": "application/vnd.semf",
    "see": "application/vnd.seemail",
    "snf": "application/x-font-snf",
    "spq": "application/scvp-vp-request",
    "spp": "application/scvp-vp-response",
    "scq": "application/scvp-cv-request",
    "scs": "application/scvp-cv-response",
    "sdp": "application/sdp",
    "etx": "text/x-setext",
    "movie": "video/x-sgi-movie",
    "ifm": "application/vnd.shana.informed.formdata",
    "itp": "application/vnd.shana.informed.formtemplate",
    "iif": "application/vnd.shana.informed.interchange",
    "ipk": "application/vnd.shana.informed.package",
    "tfi": "application/thraud+xml",
    "shar": "application/x-shar",
    "rgb": "image/x-rgb",
    "slt": "application/vnd.epson.salt",
    "aso": "application/vnd.accpac.simply.aso",
    "imp": "application/vnd.accpac.simply.imp",
    "twd": "application/vnd.simtech-mindmapper",
    "csp": "application/vnd.commonspace",
    "saf": "application/vnd.yamaha.smaf-audio",
    "mmf": "application/vnd.smaf",
    "spf": "application/vnd.yamaha.smaf-phrase",
    "teacher": "application/vnd.smart.teacher",
    "svd": "application/vnd.svd",
    "rq": "application/sparql-query",
    "srx": "application/sparql-results+xml",
    "gram": "application/srgs",
    "grxml": "application/srgs+xml",
    "ssml": "application/ssml+xml",
    "skp": "application/vnd.koan",
    "sgml": "text/sgml",
    "sdc": "application/vnd.stardivision.calc",
    "sda": "application/vnd.stardivision.draw",
    "sdd": "application/vnd.stardivision.impress",
    "smf": "application/vnd.stardivision.math",
    "sdw": "application/vnd.stardivision.writer",
    "sgl": "application/vnd.stardivision.writer-global",
    "sm": "application/vnd.stepmania.stepchart",
    "sit": "application/x-stuffit",
    "sitx": "application/x-stuffitx",
    "sdkm": "application/vnd.solent.sdkm+xml",
    "xo": "application/vnd.olpc-sugar",
    "au": "audio/basic",
    "wqd": "application/vnd.wqd",
    "sis": "application/vnd.symbian.install",
    "smi": "application/smil+xml",
    "xsm": "application/vnd.syncml+xml",
    "bdm": "application/vnd.syncml.dm+wbxml",
    "xdm": "application/vnd.syncml.dm+xml",
    "sv4cpio": "application/x-sv4cpio",
    "sv4crc": "application/x-sv4crc",
    "sbml": "application/sbml+xml",
    "tsv": "text/tab-separated-values",
    "tiff": "image/tiff",
    "tao": "application/vnd.tao.intent-module-archive",
    "tar": "application/x-tar",
    "tcl": "application/x-tcl",
    "tex": "application/x-tex",
    "tfm": "application/x-tex-tfm",
    "tei": "application/tei+xml",
    "txt": "text/plain",
    "dxp": "application/vnd.spotfire.dxp",
    "sfs": "application/vnd.spotfire.sfs",
    "tsd": "application/timestamped-data",
    "tpt": "application/vnd.trid.tpt",
    "mxs": "application/vnd.triscape.mxs",
    "t": "text/troff",
    "tra": "application/vnd.trueapp",
    "ttf": "application/x-font-ttf",
    "ttl": "text/turtle",
    "umj": "application/vnd.umajin",
    "uoml": "application/vnd.uoml+xml",
    "unityweb": "application/vnd.unity",
    "ufd": "application/vnd.ufdl",
    "uri": "text/uri-list",
    "utz": "application/vnd.uiq.theme",
    "ustar": "application/x-ustar",
    "uu": "text/x-uuencode",
    "vcs": "text/x-vcalendar",
    "vcf": "text/x-vcard",
    "vcd": "application/x-cdlink",
    "vsf": "application/vnd.vsf",
    "wrl": "model/vrml",
    "vcx": "application/vnd.vcx",
    "mts": "model/vnd.mts",
    "vtu": "model/vnd.vtu",
    "vis": "application/vnd.visionary",
    "viv": "video/vnd.vivo",
    "ccxml": "application/ccxml+xml,",
    "vxml": "application/voicexml+xml",
    "src": "application/x-wais-source",
    "wbxml": "application/vnd.wap.wbxml",
    "wbmp": "image/vnd.wap.wbmp",
    "wav": "audio/x-wav",
    "davmount": "application/davmount+xml",
    "woff": "application/x-font-woff",
    "wspolicy": "application/wspolicy+xml",
    "webp": "image/webp",
    "wtb": "application/vnd.webturbo",
    "wgt": "application/widget",
    "hlp": "application/winhlp",
    "wml": "text/vnd.wap.wml",
    "wmls": "text/vnd.wap.wmlscript",
    "wmlsc": "application/vnd.wap.wmlscriptc",
    "wpd": "application/vnd.wordperfect",
    "stf": "application/vnd.wt.stf",
    "wsdl": "application/wsdl+xml",
    "xbm": "image/x-xbitmap",
    "xpm": "image/x-xpixmap",
    "xwd": "image/x-xwindowdump",
    "der": "application/x-x509-ca-cert",
    "fig": "application/x-xfig",
    "xhtml": "application/xhtml+xml",
    "xml": "application/xml",
    "xdf": "application/xcap-diff+xml",
    "xenc": "application/xenc+xml",
    "xer": "application/patch-ops-error+xml",
    "rl": "application/resource-lists+xml",
    "rs": "application/rls-services+xml",
    "rld": "application/resource-lists-diff+xml",
    "xslt": "application/xslt+xml",
    "xop": "application/xop+xml",
    "xpi": "application/x-xpinstall",
    "xspf": "application/xspf+xml",
    "xul": "application/vnd.mozilla.xul+xml",
    "xyz": "chemical/x-xyz",
    "yaml": "text/yaml",
    "yang": "application/yang",
    "yin": "application/yin+xml",
    "zir": "application/vnd.zul",
    "zip": "application/zip",
    "zmm": "application/vnd.handheld-entertainment+xml",
    "zaz": "application/vnd.zzazz.deck+xml"
}

Edit:

According to @Octo Poulos suggestions:

• Separated jpeg and jpg MIME into two different keys
• updated the MIME of mp4 format to video/mp4

This article describes MIME Types and the corresponding file extension of the Microsoft Office documents. It will very be useful for document analysis, and you can easily define ContentType for the Microsoft Office associated documents in ASP.NET applications. Also you can use this details to customize MIME types in IIS server configuration.

 For Microsoft Office Excel, you can define content type like this example.
 Aspx page
<%response.ContentType=“application/vnd.ms-excel”%>

 C#
  Response.ContentType = “application/vnd.ms-excel”;

The following table lists the MIME types and file extensions that are associated  with the Microsoft Office documents.

Extension	MIME Type
.doc	application/msword
.dot	application/msword
.docx	application/vnd.openxmlformats-officedocument.wordprocessingml.document
.dotx	application/vnd.openxmlformats-officedocument.wordprocessingml.template
.docm	application/vnd.ms-word.document.macroEnabled.12
.dotm	application/vnd.ms-word.template.macroEnabled.12
.xls	application/vnd.ms-excel
.xlt	application/vnd.ms-excel
.xla	application/vnd.ms-excel
.xlsx	application/vnd.openxmlformats-officedocument.spreadsheetml.sheet
.xltx	application/vnd.openxmlformats-officedocument.spreadsheetml.template
.xlsm	application/vnd.ms-excel.sheet.macroEnabled.12
.xltm	application/vnd.ms-excel.template.macroEnabled.12
.xlam	application/vnd.ms-excel.addin.macroEnabled.12
.xlsb	application/vnd.ms-excel.sheet.binary.macroEnabled.12
.ppt	application/vnd.ms-powerpoint
.pot	application/vnd.ms-powerpoint
.pps	application/vnd.ms-powerpoint
.ppa	application/vnd.ms-powerpoint
.pptx	application/vnd.openxmlformats-officedocument.presentationml.presentation
.potx	application/vnd.openxmlformats-officedocument.presentationml.template
.ppsx	application/vnd.openxmlformats-officedocument.presentationml.slideshow
.ppam	application/vnd.ms-powerpoint.addin.macroEnabled.12
.pptm	application/vnd.ms-powerpoint.presentation.macroEnabled.12
.potm	application/vnd.ms-powerpoint.template.macroEnabled.12
.ppsm	application/vnd.ms-powerpoint.slideshow.macroEnabled.12

Thanks,
Morgan
Software Developer

This is a list of the MIME Types of all well known and lesser known file formats and kinds of documents from the Microsoft Office family.

You can find the MIME Types in the table ordered by the corresponding application Microsoft Word, Excel, Power Point and Access as well as the extension of the file.

Microsoft Word

Extension	MIME-Type
DOC	application/msword
DOCX	application/vnd.openxmlformats-officedocument.wordprocessingml.document
DOT	application/msword
DOTX	application/vnd.openxmlformats-officedocument.wordprocessingml.template
DOCM	application/vnd.ms-word.document.macroEnabled.12
DOTM	application/vnd.ms-word.template.macroEnabled.12
DOCM	application/vnd.ms-word.document.macroEnabled.12
DOTM	application/vnd.ms-word.template.macroEnabled.12
WORD	application/msword
W6W	application/msword

Microsoft Excel

Extension	MIME-Type
XLS	application/msexcel
XLSX	application/vnd.openxmlformats-officedocument.spreadsheetml.sheet
XLT	application/msexcel
XLTX	application/vnd.openxmlformats-officedocument.spreadsheetml.template
XLA	application/msexcel
XLW	application/msexcel
XLSM	application/vnd.ms-excel.sheet.macroEnabled.12
XLSB	application/vnd.ms-excel.sheet.binary.macroEnabled.12
XLTM	application/vnd.ms-excel.template.macroEnabled.12
XLAM	application/vnd.ms-excel.addin.macroEnabled.12

Microsoft PowerPoint

Extension	MIME-Type
PPT	application/mspowerpoint
PPTX	application/vnd.openxmlformats-officedocument.presentationml.presentation
POT	application/mspowerpoint
POTX	application/vnd.openxmlformats-officedocument.presentationml.template
PPS	application/mspowerpoint
PPSX	application/vnd.openxmlformats-officedocument.presentationml.slideshow
PPA	application/mspowerpoint
PPAM	application/vnd.ms-powerpoint.addin.macroEnabled.12
PPTM	application/vnd.ms-powerpoint.presentation.macroEnabled.12
PPSM	application/vnd.ms-powerpoint.slideshow.macroEnabled.12
POTM	application/vnd.ms-powerpoint.template.macroEnabled.12

Microsoft Access

Extension	MIME-Type
MDB	application/msaccess
ACCDA	application/msaccess
ACCDB	application/msaccess
ACCDE	application/msaccess
ACCDR	application/msaccess
ACCDT	application/msaccess
ADE	application/msaccess
ADP	application/msaccess
ADN	application/msaccess
MDE	application/msaccess
MDF	application/msaccess
MDN	application/msaccess
MDT	application/msaccess
MDW	application/msaccess

Other

Extension	MIME-Type
CALCX	application/vnd.ms-office.calx
HLP (Microsoft Windows Help)	application/x-winhelp
MPP (Microsoft Project)	application/msproject
THMX	application/vnd.ms-officetheme
WRI (Microsoft Write)	application/mswrite

I hope, I could help you with this list. If I should have forgotten something, just write a comment.

Как я разбирал docx с помощью XSLT

Задача обработки документов в формате docx, а также таблиц xlsx и презентаций pptx является весьма нетривиальной. В этой статье расскажу как научиться парсить, создавать и обрабатывать такие документы используя только XSLT и ZIP архиватор.

Зачем?

docx — самый популярный формат документов, поэтому задача отдавать информацию пользователю в этом формате всегда может возникнуть. Один из вариантов решения этой проблемы — использование готовой библиотеки, может не подходить по ряду причин:

• библиотеки может просто не существовать
• в проекте не нужен ещё один чёрный ящик
• ограничения библиотеки по платформам и т.п.
• проблемы с лицензированием
• скорость работы

Поэтому в этой статье будем использовать только самые базовые инструменты для работы с docx документом.

Структура docx

Для начала разоберёмся с тем, что собой представляет docx документ. docx это zip архив который физически содержит 2 типа файлов:

• xml файлы с расширениями xml и rels
• медиа файлы (изображения и т.п.)

А логически — 3 вида элементов:

• Типы (Content Types) — список типов медиа файлов (например png) встречающихся в документе и типов частей документов (например документ, верхний колонтитул).
• Части (Parts) — отдельные части документа, для нашего документа это document.xml, сюда входят как xml документы так и медиа файлы.
• Связи (Relationships) идентифицируют части документа для ссылок (например связь между разделом документа и колонтитулом), а также тут определены внешние части (например гиперссылки).

Они подробно описаны в стандарте ECMA-376: Office Open XML File Formats, основная часть которого — PDF документ на 5000 страниц, и ещё 2000 страниц бонусного контента.

Минимальный docx

Простейший docx после распаковки выглядит следующим образом

Давайте посмотрим из чего он состоит.

[Content_Types].xml

Находится в корне документа и перечисляет MIME типы содержимого документа:

<Types xmlns="http://schemas.openxmlformats.org/package/2006/content-types">
    <Default Extension="rels" ContentType="application/vnd.openxmlformats-package.relationships+xml"/>
    <Default Extension="xml" ContentType="application/xml"/>
    <Override PartName="/word/document.xml"
              ContentType="application/vnd.openxmlformats-officedocument.wordprocessingml.document.main+xml"/>
</Types>

_rels/.rels

Главный список связей документа. В данном случае определена всего одна связь — сопоставление с идентификатором rId1 и файлом word/document.xml — основным телом документа.

<Relationships xmlns="http://schemas.openxmlformats.org/package/2006/relationships">
    <Relationship 
        Id="rId1" 
        Type="http://schemas.openxmlformats.org/officeDocument/2006/relationships/officeDocument"
        Target="word/document.xml"/>
</Relationships>

word/document.xml

Основное содержимое документа.

<w:document xmlns:wpc="http://schemas.microsoft.com/office/word/2010/wordprocessingCanvas"
            xmlns:mc="http://schemas.openxmlformats.org/markup-compatibility/2006"
            xmlns:o="urn:schemas-microsoft-com:office:office"
            xmlns:r="http://schemas.openxmlformats.org/officeDocument/2006/relationships"
            xmlns:m="http://schemas.openxmlformats.org/officeDocument/2006/math"
            xmlns:v="urn:schemas-microsoft-com:vml"
            xmlns:wp14="http://schemas.microsoft.com/office/word/2010/wordprocessingDrawing"
            xmlns:wp="http://schemas.openxmlformats.org/drawingml/2006/wordprocessingDrawing"
            xmlns:w10="urn:schemas-microsoft-com:office:word"
            xmlns:w="http://schemas.openxmlformats.org/wordprocessingml/2006/main"
            xmlns:w14="http://schemas.microsoft.com/office/word/2010/wordml"
            xmlns:wpg="http://schemas.microsoft.com/office/word/2010/wordprocessingGroup"
            xmlns:wpi="http://schemas.microsoft.com/office/word/2010/wordprocessingInk"
            xmlns:wne="http://schemas.microsoft.com/office/word/2006/wordml"
            xmlns:wps="http://schemas.microsoft.com/office/word/2010/wordprocessingShape"
            mc:Ignorable="w14 wp14">
    <w:body>
        <w:p w:rsidR="005F670F" w:rsidRDefault="005F79F5">
            <w:r>
                <w:t>Test</w:t>
            </w:r>
            <w:bookmarkStart w:id="0" w:name="_GoBack"/>
            <w:bookmarkEnd w:id="0"/>
        </w:p>
        <w:sectPr w:rsidR="005F670F">
            <w:pgSz w:w="12240" w:h="15840"/>
            <w:pgMar w:top="1440" w:right="1440" w:bottom="1440" w:left="1440" 
                     w:header="720" w:footer="720" w:gutter="0"/>
            <w:cols w:space="720"/>
            <w:docGrid w:linePitch="360"/>
        </w:sectPr>
    </w:body>
</w:document>

Здесь:

• <w:document> — сам документ
• <w:body> — тело документа
• <w:p> — параграф
• <w:r> — run (фрагмент) текста
• <w:t> — сам текст
• <w:sectPr> — описание страницы

Если открыть этот документ в текстовом редакторе, то увидим документ из одного слова Test.

word/_rels/document.xml.rels

Здесь содержится список связей части word/document.xml. Название файла связей создаётся из названия части документа к которой он относится и добавления к нему расширения rels. Папка с файлом связей называется _rels и находится на том же уровне, что и часть к которой он относится. Так как связей в word/document.xml никаких нет то и в файле пусто:

<Relationships xmlns="http://schemas.openxmlformats.org/package/2006/relationships">
</Relationships>

Даже если связей нет, этот файл должен существовать.

docx и Microsoft Word

docx созданный с помощью Microsoft Word, да в принципе и с помощью любого другого редактора имеет несколько дополнительных файлов.

Вот что в них содержится:

• docProps/core.xml — основные метаданные документа согласно Open Packaging Conventions и Dublin Core [1], [2].
• docProps/app.xml — общая информация о документе: количество страниц, слов, символов, название приложения в котором был создан документ и т.п.
• word/settings.xml — настройки относящиеся к текущему документу.
• word/styles.xml — стили применимые к документу. Отделяют данные от представления.
• word/webSettings.xml — настройки отображения HTML частей документа и настройки того, как конвертировать документ в HTML.
• word/fontTable.xml — список шрифтов используемых в документе.
• word/theme1.xml — тема (состоит из цветовой схемы, шрифтов и форматирования).

В сложных документах частей может быть гораздо больше.

Реверс-инжиниринг docx

Итак, первоначальная задача — узнать как какой-либо фрагмент документа хранится в xml, чтобы потом создавать (или парсить) подобные документы самостоятельно. Для этого нам понадобятся:

• Архиватор zip
• Библиотека для форматирования XML (Word выдаёт XML без отступов, одной строкой)
• Средство для просмотра diff между файлами, я буду использовать git и TortoiseGit

Инструменты

• Под Windows: zip, unzip, libxml2, git, TortoiseGit
• Под Linux: apt-get install zip unzip libxml2 libxml2-utils git

Также понадобятся скрипты для автоматического (раз)архивирования и форматирования XML.
Использование под Windows:

• unpack file dir — распаковывает документ file в папку dir и форматирует xml
• pack dir file — запаковывает папку dir в документ file

Использование под Linux аналогично, только ./unpack.sh вместо unpack, а pack становится ./pack.

Использование

Поиск изменений происходит следующим образом:

1. Создаём пустой docx файл в редакторе.
2. Распаковываем его с помощью unpack в новую папку.
3. Коммитим новую папку.
4. Добавляем в файл из п. 1. изучаемый элемент (гиперссылку, таблицу и т.д.).
5. Распаковываем изменённый файл в уже существующую папку.
6. Изучаем diff, убирая ненужные изменения (перестановки связей, порядок пространств имён и т.п.).
7. Запаковываем папку и проверяем что получившийся файл открывается.
8. Коммитим изменённую папку.

Пример 1. Выделение текста жирным

Посмотрим на практике, как найти тег который определяет форматирование текста жирным шрифтом.

1. Создаём документ bold.docx с обычным (не жирным) текстом Test.
2. Распаковываем его: unpack bold.docx bold.
3. Коммитим результат.
4. Выделяем текст Test жирным.
5. Распаковываем unpack bold.docx bold.
6. Изначально diff выглядел следующим образом:

Рассмотрим его подробно:

docProps/app.xml

@@ -1,9 +1,9 @@
-  <TotalTime>0</TotalTime>
+  <TotalTime>1</TotalTime>

Изменение времени нам не нужно.

docProps/core.xml

@@ -4,9 +4,9 @@
-  <cp:revision>1</cp:revision>
+  <cp:revision>2</cp:revision>
   <dcterms:created xsi:type="dcterms:W3CDTF">2017-02-07T19:37:00Z</dcterms:created>
-  <dcterms:modified xsi:type="dcterms:W3CDTF">2017-02-07T19:37:00Z</dcterms:modified>
+  <dcterms:modified xsi:type="dcterms:W3CDTF">2017-02-08T10:01:00Z</dcterms:modified>

Изменение версии документа и даты модификации нас также не интересует.

word/document.xml

«`diff
@@ -1,24 +1,26 @@

—
+

+

—
+

+

Test

—
+
«`

Изменения в w:rsidR не интересны — это внутренняя информация для Microsoft Word. Ключевое изменение тут

в параграфе с Test. Видимо элемент <w:b/> и делает текст жирным. Оставляем это изменение и отменяем остальные.

word/settings.xml

@@ -1,8 +1,9 @@
+  <w:proofState w:spelling="clean"/>
@@ -17,10 +18,11 @@
+    <w:rsid w:val="00F752CF"/>

Также не содержит ничего относящегося к жирному тексту. Отменяем.

7 Запаковываем папку с 1м изменением (добавлением <w:b/>) и проверяем что документ открывается и показывает то, что ожидалось.
8 Коммитим изменение.

Пример 2. Нижний колонтитул

Теперь разберём пример посложнее — добавление нижнего колонтитула.
Вот первоначальный коммит. Добавляем нижний колонтитул с текстом 123 и распаковываем документ. Такой diff получается первоначально:

Сразу же исключаем изменения в docProps/app.xml и docProps/core.xml — там тоже самое, что и в первом примере.

[Content_Types].xml

@@ -4,10 +4,13 @@
   <Default Extension="xml" ContentType="application/xml"/>
   <Override PartName="/word/document.xml" ContentType="application/vnd.openxmlformats-officedocument.wordprocessingml.document.main+xml"/>
+  <Override PartName="/word/footnotes.xml" ContentType="application/vnd.openxmlformats-officedocument.wordprocessingml.footnotes+xml"/>
+  <Override PartName="/word/endnotes.xml" ContentType="application/vnd.openxmlformats-officedocument.wordprocessingml.endnotes+xml"/>
+  <Override PartName="/word/footer1.xml" ContentType="application/vnd.openxmlformats-officedocument.wordprocessingml.footer+xml"/>

footer явно выглядит как то, что нам нужно, но что делать с footnotes и endnotes? Являются ли они обязательными при добавлении нижнего колонтитула или их создали заодно? Ответить на этот вопрос не всегда просто, вот основные пути:

• Посмотреть, связаны ли изменения друг с другом
• Экспериментировать
• Ну а если совсем не понятно что происходит:

Идём пока что дальше.

word/_rels/document.xml.rels

Изначально diff выглядит вот так:

«`diff
@@ -1,8 +1,11 @@

+

+

—
—
+
+
+

«`

Видно, что часть изменений связана с тем, что Word изменил порядок связей, уберём их:
«`diff
@@ -3,6 +3,9 @@
+
+
+
«`
Опять появляются footer, footnotes, endnotes. Все они связаны с основным документом, перейдём к нему:

word/document.xml

@@ -15,10 +15,11 @@
       </w:r>
       <w:bookmarkStart w:id="0" w:name="_GoBack"/>
       <w:bookmarkEnd w:id="0"/>
     </w:p>
     <w:sectPr w:rsidR="0076695C" w:rsidRPr="00290C70">
+      <w:footerReference w:type="default" r:id="rId6"/>
       <w:pgSz w:w="11906" w:h="16838"/>
       <w:pgMar w:top="1134" w:right="850" w:bottom="1134" w:left="1701" w:header="708" w:footer="708" w:gutter="0"/>
       <w:cols w:space="708"/>
       <w:docGrid w:linePitch="360"/>
     </w:sectPr>

Редкий случай когда есть только нужные изменения. Видна явная ссылка на footer из sectPr. А так как ссылок в документе на footnotes и endnotes нет, то можно предположить что они нам не понадобятся.

word/settings.xml

@@ -1,19 +1,30 @@
 <?xml version="1.0" encoding="UTF-8" standalone="yes"?>
 <w:settings xmlns:mc="http://schemas.openxmlformats.org/markup-compatibility/2006" xmlns:o="urn:schemas-microsoft-com:office:office" xmlns:r="http://schemas.openxmlformats.org/officeDocument/2006/relationships" xmlns:m="http://schemas.openxmlformats.org/officeDocument/2006/math" xmlns:v="urn:schemas-microsoft-com:vml" xmlns:w10="urn:schemas-microsoft-com:office:word" xmlns:w="http://schemas.openxmlformats.org/wordprocessingml/2006/main" xmlns:w14="http://schemas.microsoft.com/office/word/2010/wordml" xmlns:w15="http://schemas.microsoft.com/office/word/2012/wordml" xmlns:sl="http://schemas.openxmlformats.org/schemaLibrary/2006/main" mc:Ignorable="w14 w15">
   <w:zoom w:percent="100"/>
+  <w:proofState w:spelling="clean"/>
   <w:defaultTabStop w:val="708"/>
   <w:characterSpacingControl w:val="doNotCompress"/>
+  <w:footnotePr>
+    <w:footnote w:id="-1"/>
+    <w:footnote w:id="0"/>
+  </w:footnotePr>
+  <w:endnotePr>
+    <w:endnote w:id="-1"/>
+    <w:endnote w:id="0"/>
+  </w:endnotePr>
   <w:compat>
     <w:compatSetting w:name="compatibilityMode" w:uri="http://schemas.microsoft.com/office/word" w:val="15"/>
     <w:compatSetting w:name="overrideTableStyleFontSizeAndJustification" w:uri="http://schemas.microsoft.com/office/word" w:val="1"/>
     <w:compatSetting w:name="enableOpenTypeFeatures" w:uri="http://schemas.microsoft.com/office/word" w:val="1"/>
     <w:compatSetting w:name="doNotFlipMirrorIndents" w:uri="http://schemas.microsoft.com/office/word" w:val="1"/>
     <w:compatSetting w:name="differentiateMultirowTableHeaders" w:uri="http://schemas.microsoft.com/office/word" w:val="1"/>
   </w:compat>
   <w:rsids>
     <w:rsidRoot w:val="00290C70"/>
+    <w:rsid w:val="000A7B7B"/>
+    <w:rsid w:val="001B0DE6"/>

А вот и появились ссылки на footnotes, endnotes добавляющие их в документ.

word/styles.xml

«`diff
@@ -480,6 +480,50 @@

+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+

«`

Изменения в стилях нас интересуют только если мы ищем как поменять стиль. В данном случае это изменение можно убрать.

word/footer1.xml

Посмотрим теперь собственно на сам нижний колонтитул (часть пространств имён опущена для читабельности, но в документе они должны быть):

<w:ftr xmlns:w="http://schemas.openxmlformats.org/wordprocessingml/2006/main">
  <w:p w:rsidR="000A7B7B" w:rsidRDefault="000A7B7B">
    <w:pPr>
      <w:pStyle w:val="a6"/>
    </w:pPr>
    <w:r>
      <w:t>123</w:t>
    </w:r>
  </w:p>
</w:ftr>

Тут виден текст 123. Единственное, что надо исправить — убрать ссылку на <w:pStyle w:val="a6"/>.

В результате анализа всех изменений делаем следующие предположения:

• footnotes и endnotes не нужны
• В [Content_Types].xml надо добавить footer
• В word/_rels/document.xml.rels надо добавить ссылку на footer
• В word/document.xml в тег <w:sectPr> надо добавить <w:footerReference>

Уменьшаем diff до этого набора изменений:

Затем запаковываем документ и открываем его.
Если всё сделано правильно, то документ откроется и в нём будет нижний колонтитул с текстом 123. А вот и итоговый коммит.

Таким образом процесс поиска изменений сводится к поиску минимального набора изменений, достаточного для достижения заданного результата.

Практика

Найдя интересующее нас изменение, логично перейти к следующему этапу, это может быть что-либо из:

• Создания docx
• Парсинг docx
• Преобразования docx

Тут нам потребуются знания XSLT и XPath.

Давайте напишем достаточно простое преобразование — замену или добавление нижнего колонтитула в существующий документ. Писать я буду на языке Caché ObjectScript, но даже если вы не знаете — не беда. В основном будем вызовать XSLT и архиватор. Ничего более. Итак, приступим.

Алгоритм

Алгоритм выглядит следующим образом:

1. Распаковываем документ
2. Добавляем наш нижний колонтитул
3. Прописываем ссылку на него в [Content_Types].xml и word/_rels/document.xml.rels
4. В word/document.xml в тег <w:sectPr> добавляем тег <w:footerReference> или заменяем в нём ссылку на наш нижний колонтитул.
5. Запаковываем документ

Приступим.

Распаковка

В Caché ObjectScript есть возможность выполнять команды ОС с помощью функции $zf(-1, oscommand). Вызовем unzip для распаковки документа с помощью обёртки над $zf(-1):

/// Используя %3 (unzip) распаковать файл %1 в папку %2
Parameter UNZIP = "%3 %1 -d %2";

/// Распаковать архив source в папку targetDir
ClassMethod executeUnzip(source, targetDir) As %Status
{
    set timeout = 100
    set cmd = $$$FormatText(..#UNZIP, source, targetDir, ..getUnzip())
    return ..execute(cmd, timeout)
}

Создаём файл нижнего колонтитула

На вход поступает текст нижнего колонтитула, запишем его в файл in.xml:

В XSLT (файл — footer.xsl) будем создавать нижний колонтитул с текстом из тега xml (часть пространств имён опущена, вот полный список):

<xsl:stylesheet 
  xmlns:xsl="http://www.w3.org/1999/XSL/Transform" 
  xmlns="http://schemas.openxmlformats.org/package/2006/relationships" version="1.0">
    <xsl:output method="xml" omit-xml-declaration="no" indent="yes" standalone="yes"/>
    <xsl:template match="/">

        <w:ftr xmlns:w="http://schemas.openxmlformats.org/wordprocessingml/2006/main">
            <w:p>
                <w:r>
                    <w:rPr>
                        <w:lang w:val="en-US"/>
                    </w:rPr>
                    <w:t>
                        <xsl:value-of select="//xml/text()"/>
                    </w:t>
                </w:r>
            </w:p>
        </w:ftr>
    </xsl:template>
</xsl:stylesheet>

Теперь вызовем XSLT преобразователь:

do ##class(%XML.XSLT.Transformer).TransformFile("in.xml", "footer.xsl", footer0.xml")    

В результате получится файл нижнего колонтитула footer0.xml:

<w:ftr xmlns:w="http://schemas.openxmlformats.org/wordprocessingml/2006/main">
    <w:p>
        <w:r>
            <w:rPr>
                <w:lang w:val="en-US"/>
            </w:rPr>
            <w:t>TEST</w:t>
        </w:r>
    </w:p>
</w:ftr>

Добавляем ссылку на колонтитул в список связей основного документа

Сссылки с идентификатором rId0 как правило не существует. Впрочем можно использовать XPath для получения идентификатора которого точно не существует.
Добавляем ссылку на footer0.xml c идентификатором rId0 в word/_rels/document.xml.rels:

«`xml

<xsl:template match="/*">
    <xsl:copy>
        <xsl:copy-of select="$new"/>
        <xsl:copy-of select="@* | node()"/>
    </xsl:copy>
</xsl:template>

</xsl:stylesheet>

</spoiler>

#### Прописываем ссылки в документе

Далее надо в каждый тег `<w:sectPr>` добавить тег `<w:footerReference>` или заменить в нём ссылку на наш нижний колонтитул. [Оказалось](https://msdn.microsoft.com/en-us/library/documentformat.openxml.wordprocessing.footerreference(v=office.14).aspx), что у каждого тега `<w:sectPr>` может быть 3 тега `<w:footerReference>` - для первой страницы, четных страниц и всего остального:
<spoiler title="XSLT">
```xml
<xsl:stylesheet xmlns:xsl="http://www.w3.org/1999/XSL/Transform" 
xmlns:r="http://schemas.openxmlformats.org/officeDocument/2006/relationships"
xmlns:w="http://schemas.openxmlformats.org/wordprocessingml/2006/main"
version="1.0">
    <xsl:output method="xml" omit-xml-declaration="yes" indent="yes" />
    <xsl:template match="//@* | //node()">
        <xsl:copy>
            <xsl:apply-templates select="@*"/>
            <xsl:apply-templates select="node()"/>
        </xsl:copy>
    </xsl:template>
    <xsl:template match="//w:sectPr">
        <xsl:element name="{name()}" namespace="{namespace-uri()}">
            <xsl:copy-of select="./namespace::*"/>
            <xsl:apply-templates select="@*"/>
            <xsl:copy-of select="./*[local-name() != 'footerReference']"/>
            <w:footerReference w:type="default" r:id="rId0"/>
            <w:footerReference w:type="first" r:id="rId0"/>
            <w:footerReference w:type="even" r:id="rId0"/>
        </xsl:element>
    </xsl:template>
</xsl:stylesheet>

Добавляем колонтитул в [Content_Types].xml

Добавляем в [Content_Types].xml информацию о том, что /word/footer0.xml имеет тип application/vnd.openxmlformats-officedocument.wordprocessingml.footer+xml:

«`xml

<xsl:template match="/*">
    <xsl:copy>
        <xsl:copy-of select="@* | node()"/> 
        <xsl:copy-of select="$new"/>
    </xsl:copy>
</xsl:template>

</xsl:stylesheet>

</spoiler>


#### В результате

Весь код [опубликован](https://github.com/intersystems-ru/Converter/blob/master/Converter/Footer.cls.xml). Работает он так:
```cos
do ##class(Converter.Footer).modifyFooter("in.docx", "out.docx", "TEST")

Где:

• in.docx — исходный документ
• out.docx — выходящий документ
• TEST — текст, который добавляется в нижний колонтитул

Выводы

Используя только XSLT и ZIP можно успешно работать с документами docx, таблицами xlsx и презентациями pptx.

Открытые вопросы

1. Изначально хотел использовать 7z вместо zip/unzip т..к. это одна утилита и она более распространена на Windows. Однако я столкнулся с такой проблемой, что документы запакованные 7z под Linux не открываются в Microsoft Office. Я попробовал достаточно много вариантов вызова, однако положительного результата добиться не удалось.
2. Ищу XSD со схемами ECMA-376 версии 5 и комментариями. XSD версии 5 без комментариев доступен к загрузке на сайте ECMA, но без комментариев в нём сложно разобраться. XSD версии 2 с комментариями доступен к загрузке.

Ссылки

• ECMA-376
• Описание docx
• Подробная статья про docx
• Репозиторий со скриптами
• Репозиторий с преобразователем нижнего колонтитула

With approximately one billion people using Microsoft Office, the DOCX format is the most popular de facto standard for exchanging document files between offices. Its closest competitor — the ODT format — is only supported by Open/LibreOffice and some open source products, making it far from standard. The PDF format is not a competitor because PDFs can’t be edited and they don’t contain a full document structure, so they can only take limited local changes like watermarks, signatures, and the like. This is why most business documents are created in the DOCX format; there’s no good alternative to replace it.

While DOCX is a complex format, you may want to parse it manually for simpler tasks such as indexing, converting to TXT and making other small modifications. I’d like to give you enough information on DOCX internals so you don’t have to reference the ECMA specifications, a massive 5,000 page manual.

The best way to understand the format is to create a simple one-word document with MSWord and observe how editing the document changes the underlying XML. You’ll face some cases where the DOCX doesn’t format properly in MS Word and you don’t know why, or come across instances when it’s not evident how to generate the desired formatting. Seeing and understanding exactly what’s going on in the XML will help that.

I worked for about a year on a collaborative DOCX editor, CollabOffice, and I want to share some of that knowledge with the developer community. In this article I will explain the DOCX file structure, summarising information that is scattered over the internet. This article is an intermediary between the huge, complex ECMA specification and the simple internet tutorials currently available. You can find the files that accompany this article in the toptal-docx project on my github account.

A Simple DOCX file

A DOCX file is a ZIP archive of XML files. If you create a new, empty Microsoft Word document, write a single word ‘Test’ inside and unzip it contents, you will see the following file structure:

Even though we’ve created a simple document, the save process in Microsoft Word has generated default themes, document properties, font tables, and so on, in XML format.

All the files inside a DOCX are XML files, even those with the «.rels» extension.

To start, let us remove the unused stuff and focus on document.xml, which contains the main text elements. When you delete a file, make sure you have deleted all the relationship references to it from other the xml files. Here is a code-diff example on how I’ve cleared dependencies to app.xml and core.xml. If you have any unresolved/missing references, MSWord will consider the file broken.

Here’s the structure of our simplified, minimal DOCX document (and here’s the project on github):

Let’s break it down by file from here, from the top:

_rels/.rels

This defines the reference that tells MS Word where to look for the document contents. In this case, it references word/document.xml:

<?xml version="1.0" encoding="UTF-8" standalone="yes"?>
<Relationships xmlns="http://schemas.openxmlformats.org/package/2006/relationships">
   <Relationship Id="rId1" Type="http://schemas.openxmlformats.org/officeDocument/2006/relationships/officeDocument"
                 Target="word/document.xml"/>
</Relationships>

_rels/document.xml.rels

This file defines references to resources, such as images, embedded in the document content. Our simple document has no embedded resources, so the relationship tag is empty:

<?xml version="1.0" encoding="UTF-8" standalone="yes"?>
<Relationships xmlns="http://schemas.openxmlformats.org/package/2006/relationships">
</Relationships>

[Content_Types].xml

[Content_Types].xml contains information about the types of media inside the document. Since we only have text content, it’s pretty simple:

<?xml version="1.0" encoding="UTF-8" standalone="yes"?>
<Types xmlns="http://schemas.openxmlformats.org/package/2006/content-types">
   <Default Extension="rels" ContentType="application/vnd.openxmlformats-package.relationships+xml"/>
   <Default Extension="xml" ContentType="application/xml"/>
   <Override PartName="/word/document.xml"
             ContentType="application/vnd.openxmlformats-officedocument.wordprocessingml.document.main+xml"/>
</Types>

document.xml

Finally, here is the main XML with the document’s text content. I have removed some of namespace declarations for clarity, but you can find the full version of the file in the github project. In that file you’ll find that some of the namespace references in the document are unused, but you shouldn’t delete them because MS Word needs them.

Here’s our simplified example:

<w:document>
   <w:body>
       <w:p w:rsidR="005F670F" w:rsidRDefault="005F79F5">
           <w:r><w:t>Test</w:t></w:r>
       </w:p>
       <w:sectPr w:rsidR="005F670F">
           <w:pgSz w:w="12240" w:h="15840"/>
           <w:pgMar w:top="1440" w:right="1440" w:bottom="1440" w:left="1440" w:header="720" w:footer="720"
                    w:gutter="0"/>
           <w:cols w:space="720"/>
           <w:docGrid w:linePitch="360"/>
       </w:sectPr>
   </w:body>
</w:document>

The main node <w:document> represents the document itself, <w:body> contains paragraphs, and nested within <w:body> are page dimensions defined by <w:sectPr>.

<w:rsidR> is an attribute that you can ignore; it’s used by MS Word internals.

Let’s take a look at a more complex document with three paragraphs. I have highlighted the XML with the same colors on the screenshot from Microsoft Word, so you can see the correlation:

<w:p w:rsidR="0081206C" w:rsidRDefault="00E10CAE">
<w:r> <w:t xml:space="preserve">This is our example first paragraph. It's default is left aligned, and now I'd like to introduce</w:t> </w:r>
    <w:r> <w:rPr>
    <w:rFonts w:ascii="Arial" w:hAnsi="Arial" w:cs="Arial"/>
    <w:color w:val="000000"/>
    </w:rPr> <w:t>some bold</w:t>
    </w:r>
    <w:r> <w:rPr>
    <w:rFonts w:ascii="Arial" w:hAnsi="Arial" w:cs="Arial"/>
    <w:b/> <w:color w:val="000000"/>
    </w:rPr> <w:t xml:space="preserve"> text</w:t>
    </w:r>
    <w:r>  <w:rPr> <w:rFonts w:ascii="Arial"
w:hAnsi="Arial" w:cs="Arial"/> <w:color w:val="000000"/> </w:rPr>
    <w:t xml:space="preserve">, </w:t>
    </w:r>
    <w:proofErr w:type="gramStart"/>
    <w:r> <w:t xml:space="preserve">and also change the</w:t> </w:r>
    <w:r w:rsidRPr="00E10CAE">  <w:rPr><w:rFonts w:ascii="Impact" w:hAnsi="Impact"/>
    </w:rPr>  <w:t>font style</w:t> </w:r>
    <w:r>
    <w:rPr>  <w:rFonts w:ascii="Impact" w:hAnsi="Impact"/>  </w:rPr>  <w:t xml:space="preserve"> </w:t>
    </w:r>
    <w:r> <w:t>to 'Impact'.</w:t></w:r>
    </w:p>
    <w:p w:rsidR="00E10CAE" w:rsidRDefault="00E10CAE"> <w:r> <w:t>This is new paragraph.</w:t> </w:r></w:p>
    <w:p w:rsidR="00E10CAE" w:rsidRPr="00E10CAE" w:rsidRDefault="00E10CAE">
    <w:r> <w:t>This is one more paragraph, a bit longer.</w:t> </w:r>
    </w:p>

Paragraph Structure

A simple document consists of paragraphs, a paragraph consists of runs (a series of text with the same font, color, etc), and runs consist of characters (such as <w:t>).<w:t> tags may have several characters inside, and there might be a few in the same run.

Again, we can ignore <w:rsidR>.

Text properties

Basic text properties are font, size, color, style, and so on. There are about 40 tags that specify text appearance. As you can see in our three paragraph example, each run has its own properties inside <w:rPr>, specifying <w:color>, <w:rFonts> and boldness <w:b>.

An important thing to note is that properties make a distinction between the two groups of characters, normal and complex script (Arabic, for instance), and that the properties have a different tag depending on which type of character it’s affecting.

Most normal script property tags have a matching complex script tag with an added “C” specifying the property is for complex scripts. For example: <w:i> (italic) becomes <w:iCs>, and the bold tag for normal script, <w:b>, becomes <w:bCs> for complex script.

Styles

There’s an entire toolbar in Microsoft Word dedicated to styles: normal, no spacing, heading 1, heading 2, title, and so on. These styles are stored in /word/styles.xml (note: in the first step in our simple example, we removed this XML from DOCX. Make a new DOCX to see this).

Once you have text defined as a style, you will find reference to this style inside the paragraph properties tag, <w:pPr>. Here’s an example where I’ve defined my text with the style Heading 1:

<w:p>
   <w:pPr>
       <w:pStyle w:val="Heading1"/>
   </w:pPr>
   <w:r>
       <w:t>My heading 1</w:t>
   </w:r>
</w:p>

and here is the style itself from styles.xml:

<w:style w:type="paragraph" w:styleId="Heading1">
   <w:name w:val="heading 1"/>
   <w:basedOn w:val="Normal"/>
   <w:next w:val="Normal"/>
   <w:link w:val="Heading1Char"/>
   <w:uiPriority w:val="9"/>
   <w:qFormat/>
   <w:rsid w:val="002F7F18"/>
   <w:pPr>
       <w:keepNext/>
       <w:keepLines/>
       <w:spacing w:before="480" w:after="0"/>
       <w:outlineLvl w:val="0"/>
   </w:pPr>
   <w:rPr>
       <w:rFonts w:asciiTheme="majorHAnsi" w:eastAsiaTheme="majorEastAsia" w:hAnsiTheme="majorHAnsi"
                 w:cstheme="majorBidi"/>
       <w:b/>
       <w:bCs/>
       <w:color w:val="365F91" w:themeColor="accent1" w:themeShade="BF"/>
       <w:sz w:val="28"/>
       <w:szCs w:val="28"/>
   </w:rPr>
</w:style>

The <w:style/w:rPr/w:b> xpath specifies that the font is bold, and <w:style/w:rPr/w:color> indicates the font color. <w:basedOn> instructs MSWord to use “Normal” style for any missing properties.

Property Inheritance

Text properties are inherited. A run has its own properties (w:p/w:r/w:rPr/*), but it also inherits properties from paragraph (w:r/w:pPr/*), and both can reference style properties from the /word/styles.xml.

<w:r>
 <w:rPr>
   <w:rStyle w:val="DefaultParagraphFont"/>
   <w:sz w:val="16"/>
 </w:rPr>
 <w:tab/>
</w:r>

Paragraphs and runs start with default properties: w:styles/w:docDefaults/w:rPrDefault/*
and w:styles/w:docDefaults/w:pPrDefault/*. To get the end result of a character’s properties you should:

1. Use default run/paragraph properties
2. Append run/paragraph style properties
3. Append local run/paragraph properties
4. Append result run properties over paragraph properties

When I say “append” B to A, I mean to iterate through all B properties and override all A’s properties, leaving all non-intersecting properties as-is.

One more place where default properties may be located is in the <w:style> tag with w:type="paragraph" and w:default="1". Note, that characters themselves inside a run never have a default style, so <w:style w:type="character" w:default="1"> doesn’t actually affect any text.

1554402290400-dbb29eef3ba6035df7ad726dfc99b2af.png)

Characters in a run can inherit from its paragraph and both can inherit from styles.xml.

Toggle properties

Some of the properties are “toggle” properties, such as <w:b> (bold) or <w:i> (italic); these attributes behave like an XOR operator.

This means if the parent style is bold and a child run is bold, the result will be regular, non-bold text.

You have to do lots of testing and reverse-engineering to handle toggle attributes correctly. Take a look at paragraph 17.7.3 of ECMA-376 Open XML specification to get the formal, detailed rules for toggle properties/

Toggle properties are the most complex for a layouter to handle correctly.

Fonts

Fonts follow the same common rules as other text attributes, but font property default values are specified in a separate theme file, referenced under word/_rels/document.xml.rels like this:

<Relationship Id="rId7" Type="http://schemas.openxmlformats.org/officeDocument/2006/relationships/theme" Target="theme/theme1.xml"/>

Based on the above reference, the default font name will be found in word/theme/themes1.xml, inside a <a:theme> tag, a:themeElements/a:fontScheme/a:majorFont or a:minorFont tag.

The default font size is 10 unless the w:docDefaults/w:rPrDefault tag is missing, then it is size 11.

Text alignment

Text alignment is specified by a <w:jc> tag with four w:val modes available: "left", "center", "right" and "both".

"left" is the default mode; text is started at the left of paragraph rectangle (usually the page width). (This paragraph is aligned to the left, which is standard.)

"center" mode, predictably, centers all characters inside the page width. (Again, this paragraph exemplifies centered alignment.)

In "right" mode, paragraph text is aligned to the right margin. (Notice how this text is aligned to the right side.)

"both" mode puts extra spacing between words so that lines get wider and occupy the full paragraph width, with the exception of the last line which is left aligned. (This paragraph is a demonstration of that.)

Images

DOCX supports two sorts of images: inline and floating.

Inline images appear inside a paragraph along with the other characters, <w:drawing> is used instead of using <w:t> (text). You can find image ID with the following xpath syntax:

w:drawing/wp:inline/a:graphic/a:graphicData/pic:pic/pic:blipFill/a:blip/@r:embed

The image ID is used to look up the filename in the word/_rels/document.xml.rels file, and it should point to gif/jpeg file inside word/media subfolder. (See the github project’s word/_rels/document.xml.rels file, where you can see the image ID.)

Floating images are placed relative to paragraphs with text flowing around them. (Here’s th github project sample document with a floating image.)

Floating images use <wp:anchor> instead of <w:drawing>, so if you delete any text inside <w:p>, be careful with the anchors if you don’t want the images removed.

MS Word’s image options refer to image alignment as «text wrapping mode».

Tables

XML tags for tables are similar to HTML table markup– is the same as <table>, matches with <tr>, etc.

<w:tbl>, the table itself, has table properties <w:tblPr>, and each column property is presented by <w:gridCol> inside <w:tblGrid>. Rows follow one by one as <w:tr> tags and each row should have same number of columns as specified in <w:tblGrid>:

<w:tbl>
 <w:tblPr>
   <w:tblW w:w="5000" w:type="pct" />
 </w:tblPr>
 <w:tblGrid><w:gridCol/><w:gridCol/></w:tblGrid>
 <w:tr>
   <w:tc><w:p><w:r><w:t>left</w:t></w:r></w:p></w:tc>
   <w:tc><w:p><w:r><w:t>right</w:t></w:r></w:p></w:tc>
 </w:tr>
</w:tbl>

Width for table columns can be specified in the <w:tblW> tag, but if you don’t define it MS Word will use its internal algorithms to find the optimal width of columns for the smallest effective table size.

Units

Many XML attributes inside DOCX specify sizes or distances. While they’re integers inside the XML, they all have different units so some conversion is necessary. The topic is a complicated one, so I’d recommend this article by Lars Corneliussen on units in DOCX files. The table he presents is useful, though with a small misprint: inches should be pt/72, not pt*72.

Here’s a cheat sheet:

Tips for Implementing a Layouter

If you want to convert a DOCX file (to PDF, for instance), draw it on canvas, or count number of pages, you’ll have to implement a layouter. A layouter is an algorithm for calculating character positions from a DOCX file.

This is a complex task if you need 100 percent fidelity rendering. The amount of time needed to implement a good layouter is measured in man-years, but if you only need a simple, limited one, it can be done relatively quickly.

A layouter fills a parent rectangle, which is usually a rectangle of the page. It add words from a run one by one. When the current line overflows, it starts a new one. If the paragraph is too high for the parent rectangle, it’s wrapped to the next page.

Here are some important things to keep in mind if you decide to implement a layouter:

• The layouter should take care about text alignment and text floating over images
• It should be capable of handling nested objects, such as nested tables
• If you want to provide full support for such images, you’ll have to implement a layouter with at least two passes, the first step collects floating images’ positions and the second fills empty space with text characters.
• Be aware of indentations and spacings. Each paragraph has spacing before and after, and these numbers are specified by the w:spacing tag. Vertical spacing is specified by w:after and w:before tags. Note that line spacing is specified by w:line, but this is not the size of the line as one may expect. To get the size of the line, take the current font height, multiply by w:line and divide by 12.
• DOCX files contain no information about pagination. You won’t find the number of pages in the document unless you calculate how much space you need for each line to ascertain the number of pages. If you need to find exact coordinates of each character on the page, be sure to take into account all spacings, indentations and sizes.
• If you implement a full-featured DOCX layouter that handles tables, note the special cases when tables span multiple pages. A cell which causes a page overflow also affects other cells.
• Creating an optimal algorithm for calculating a table columns’ width is a challenging math problem and word processors and layouters usually use some suboptimal implementations. I propose using the algorithm from W3C HTML table documentation as a first approximation. I haven’t found a description of the algorithm used by MS Word, and Microsoft has fine-tuned the algorithm over time so different versions of Word may lay out tables slightly differently.

If something is unclear: reverse-engineer the XML!

When it’s not obvious how this or that XML tag works inside MS Word, there are two main approaches to figuring it out:

• Create the desired content step-by-step. Start with a simple docx file. Save each step to its own file, as in 1.docx, 2.docx, for example. Unzip each of them and use a visual diff tool for folder comparison to see which tags appear after your changes. (For a commercial option, try Araxis Merge, or for a free option, WinMerge.)

• If you generate a DOCX file that MS Word doesn’t like, work backwards. Simplify your XML step by step. At some point you will learn which change MS Word found incorrect.

DOCX is quite complex, isn’t it?

It is complex, and Microsoft’s license forbids using MS Word on the server side for processing DOCX– this is pretty standard for commercial products. Microsoft has, however, provided the XSLT file to handle most DOCX tags, but it won’t give you 100 percent or even 99 percent fidelity. Processes such as text wrapping over images are not supported, but you will be able to support the majority of documents. (If you don’t need complexity, consider using Markdown as an alternative.)

If you have a sufficient budget (there is no free DOCX rendering engine), you may want to use commercial products such as Aspose or docx4j. The most popular free solution is LibreOffice for converting between DOCX and other formats, including PDF. Unfortunately, LibreOffice contains many small bugs during conversion, and since it’s a sophisticated, open-source C++ product, it’s slow and difficult to fix fidelity issues.

Alternatively, if you find DOCX layouting too complicated to implement yourself, you can also convert it to HTML and use a browser to render it. You can also consider one of Toptal’s freelance XML developers.

DOCX Resources for further reading

• ECMA DOCX specification
• OpenXML library for DOCX manipulation from C#. It doesn’t contain information on layouting or rendering code, but offers a class hierarchy matching each possible XML node in DOCX.
• You can always search or ask on stackoverflow with keywords like docx4j, OpenXML and docx; there are people in the community who are knowledgeable.

MIME (Multipurpose Internet Mail Extensions) is an internet standard that is used to identify the types of content found in various files. These types can include applications, sounds, video, text, and many others.

MIME Types Explained

In a way, they are much like the file extensions you’re familiar with on your desktop or laptop. For example, the .doc extension that is used for Word documents, the .exe for executable windows files, and .xls that is found on Excel files are all file extensions you are undoubtedly familiar with for the files on your computer.

MIME types are defined in HTML by the type attribute on links, objects, and script and style tags.

In this article, we will list the various MIME types for applications, sounds, images, mail messages, text files, video files, and virtual world files. You can use this article as a handy catalog of all of these files should you need these MIME types in the future.

The MIME type for HTML is:

text/html

Applications and Their MIME Types

Here’s a list of applications, their MIME types, and their file extensions.

Sound Files and Their MIME Types

Here’s a list of sound files, their MIME types, and their file extensions.

Image Files and Their MIME Types

Here’s a list of image files, their MIME types, and their file extensions.

Mail Message Files and Their MIME Types

Here’s a list of mail message files, their MIME types, and their file extensions.

Text Files and Their MIME Types

Here’s a list of text files, their MIME types, and their file extensions.

Video Files and Their MIME Types

Here’s a list of video files, their MIME types, and their file extensions.

Virtual World Files and Their MIME Types

Here’s a list of virtual world files, their MIME types, and their file extensions.