Doc/lib/libmimetypes.tex

   1 \section{\module{mimetypes} ---
   2          Map filenames to MIME types}
   3
   4 \declaremodule{standard}{mimetypes}
   5 \modulesynopsis{Mapping of filename extensions to MIME types.}
   6 \sectionauthor{Fred L. Drake, Jr.}{fdrake@acm.org}
   7
   8
   9 \indexii{MIME}{content type}
  10
  11 The \module{mimetypes} module converts between a filename or URL and
  12 the MIME type associated with the filename extension.  Conversions are
  13 provided from filename to MIME type and from MIME type to filename
  14 extension; encodings are not supported for the latter conversion.
  15
  16 The module provides one class and a number of convenience functions.
  17 The functions are the normal interface to this module, but some
  18 applications may be interested in the class as well.
  19
  20 The functions described below provide the primary interface for this
  21 module.  If the module has not been initialized, they will call
  22 \function{init()} if they rely on the information \function{init()}
  23 sets up.
  24
  25
  26 \begin{funcdesc}{guess_type}{filename\optional{, strict}}
  27 Guess the type of a file based on its filename or URL, given by
  28 \var{filename}.  The return value is a tuple \code{(\var{type},
  29 \var{encoding})} where \var{type} is \code{None} if the type can't be
  30 guessed (missing or unknown suffix) or a string of the form
  31 \code{'\var{type}/\var{subtype}'}, usable for a MIME
  32 \mailheader{content-type} header\indexii{MIME}{headers}.
  33
  34 \var{encoding} is \code{None} for no encoding or the name of the
  35 program used to encode (e.g. \program{compress} or \program{gzip}).
  36 The encoding is suitable for use as a \mailheader{Content-Encoding}
  37 header, \emph{not} as a \mailheader{Content-Transfer-Encoding} header.
  38 The mappings are table driven.  Encoding suffixes are case sensitive;
  39 type suffixes are first tried case sensitively, then case
  40 insensitively.
  41
  42 Optional \var{strict} is a flag specifying whether the list of known
  43 MIME types is limited to only the official types \ulink{registered
  44 with IANA}{http://www.isi.edu/in-notes/iana/assignments/media-types}
  45 are recognized.  When \var{strict} is true (the default), only the
  46 IANA types are supported; when \var{strict} is false, some additional
  47 non-standard but commonly used MIME types are also recognized.
  48 \end{funcdesc}
  49
  50 \begin{funcdesc}{guess_all_extensions}{type\optional{, strict}}
  51 Guess the extensions for a file based on its MIME type, given by
  52 \var{type}.
  53 The return value is a list of strings giving all possible filename extensions,
  54 including the leading dot (\character{.}).  The extensions are not guaranteed
  55 to have been associated with any particular data stream, but would be mapped
  56 to the MIME type \var{type} by \function{guess_type()}.
  57
  58 Optional \var{strict} has the same meaning as with the
  59 \function{guess_type()} function.
  60 \end{funcdesc}
  61
  62
  63 \begin{funcdesc}{guess_extension}{type\optional{, strict}}
  64 Guess the extension for a file based on its MIME type, given by
  65 \var{type}.
  66 The return value is a string giving a filename extension, including the
  67 leading dot (\character{.}).  The extension is not guaranteed to have been
  68 associated with any particular data stream, but would be mapped to the
  69 MIME type \var{type} by \function{guess_type()}.  If no extension can
  70 be guessed for \var{type}, \code{None} is returned.
  71
  72 Optional \var{strict} has the same meaning as with the
  73 \function{guess_type()} function.
  74 \end{funcdesc}
  75
  76
  77 Some additional functions and data items are available for controlling
  78 the behavior of the module.
  79
  80
  81 \begin{funcdesc}{init}{\optional{files}}
  82 Initialize the internal data structures.  If given, \var{files} must
  83 be a sequence of file names which should be used to augment the
  84 default type map.  If omitted, the file names to use are taken from
  85 \constant{knownfiles}.  Each file named in \var{files} or
  86 \constant{knownfiles} takes precedence over those named before it.
  87 Calling \function{init()} repeatedly is allowed.
  88 \end{funcdesc}
  89
  90 \begin{funcdesc}{read_mime_types}{filename}
  91 Load the type map given in the file \var{filename}, if it exists.  The
  92 type map is returned as a dictionary mapping filename extensions,
  93 including the leading dot (\character{.}), to strings of the form
  94 \code{'\var{type}/\var{subtype}'}.  If the file \var{filename} does
  95 not exist or cannot be read, \code{None} is returned.
  96 \end{funcdesc}
  97
  98
  99 \begin{funcdesc}{add_type}{type, ext\optional{, strict}}
 100 Add a mapping from the mimetype \var{type} to the extension \var{ext}.
 101 When the extension is already known, the new type will replace the old
 102 one. When the type is already known the extension will be added
 103 to the list of known extensions.
 104
 105 When \var{strict} is the mapping will added to the official
 106 MIME types, otherwise to the non-standard ones.
 107 \end{funcdesc}
 108
 109
 110 \begin{datadesc}{inited}
 111 Flag indicating whether or not the global data structures have been
 112 initialized.  This is set to true by \function{init()}.
 113 \end{datadesc}
 114
 115 \begin{datadesc}{knownfiles}
 116 List of type map file names commonly installed.  These files are
 117 typically named \file{mime.types} and are installed in different
 118 locations by different packages.\index{file!mime.types}
 119 \end{datadesc}
 120
 121 \begin{datadesc}{suffix_map}
 122 Dictionary mapping suffixes to suffixes.  This is used to allow
 123 recognition of encoded files for which the encoding and the type are
 124 indicated by the same extension.  For example, the \file{.tgz}
 125 extension is mapped to \file{.tar.gz} to allow the encoding and type
 126 to be recognized separately.
 127 \end{datadesc}
 128
 129 \begin{datadesc}{encodings_map}
 130 Dictionary mapping filename extensions to encoding types.
 131 \end{datadesc}
 132
 133 \begin{datadesc}{types_map}
 134 Dictionary mapping filename extensions to MIME types.
 135 \end{datadesc}
 136
 137 \begin{datadesc}{common_types}
 138 Dictionary mapping filename extensions to non-standard, but commonly
 139 found MIME types.
 140 \end{datadesc}
 141
 142
 143 The \class{MimeTypes} class may be useful for applications which may
 144 want more than one MIME-type database:
 145
 146 \begin{classdesc}{MimeTypes}{\optional{filenames}}
 147   This class represents a MIME-types database.  By default, it
 148   provides access to the same database as the rest of this module.
 149   The initial database is a copy of that provided by the module, and
 150   may be extended by loading additional \file{mime.types}-style files
 151   into the database using the \method{read()} or \method{readfp()}
 152   methods.  The mapping dictionaries may also be cleared before
 153   loading additional data if the default data is not desired.
 154
 155   The optional \var{filenames} parameter can be used to cause
 156   additional files to be loaded ``on top'' of the default database.
 157
 158   \versionadded{2.2}
 159 \end{classdesc}
 160
 161 An example usage of the module:
 162
 163 \begin{verbatim}
 164 >>> import mimetypes
 165 >>> mimetypes.init()
 166 >>> mimetypes.knownfiles
 167 ['/etc/mime.types', '/etc/httpd/mime.types', ... ]
 168 >>> mimetypes.suffix_map['.tgz']
 169 '.tar.gz'
 170 >>> mimetypes.encodings_map['.gz']
 171 'gzip'
 172 >>> mimetypes.types_map['.tgz']
 173 'application/x-tar-gz'
 174 \end{verbatim}
 175
 176 \subsection{MimeTypes Objects \label{mimetypes-objects}}
 177
 178 \class{MimeTypes} instances provide an interface which is very like
 179 that of the \refmodule{mimetypes} module.
 180
 181 \begin{datadesc}{suffix_map}
 182   Dictionary mapping suffixes to suffixes.  This is used to allow
 183   recognition of encoded files for which the encoding and the type are
 184   indicated by the same extension.  For example, the \file{.tgz}
 185   extension is mapped to \file{.tar.gz} to allow the encoding and type
 186   to be recognized separately.  This is initially a copy of the global
 187   \code{suffix_map} defined in the module.
 188 \end{datadesc}
 189
 190 \begin{datadesc}{encodings_map}
 191   Dictionary mapping filename extensions to encoding types.  This is
 192   initially a copy of the global \code{encodings_map} defined in the
 193   module.
 194 \end{datadesc}
 195
 196 \begin{datadesc}{types_map}
 197   Dictionary mapping filename extensions to MIME types.  This is
 198   initially a copy of the global \code{types_map} defined in the
 199   module.
 200 \end{datadesc}
 201
 202 \begin{datadesc}{common_types}
 203   Dictionary mapping filename extensions to non-standard, but commonly
 204   found MIME types.  This is initially a copy of the global
 205   \code{common_types} defined in the module.
 206 \end{datadesc}
 207
 208 \begin{methoddesc}{guess_extension}{type\optional{, strict}}
 209   Similar to the \function{guess_extension()} function, using the
 210   tables stored as part of the object.
 211 \end{methoddesc}
 212
 213 \begin{methoddesc}{guess_type}{url\optional{, strict}}
 214   Similar to the \function{guess_type()} function, using the tables
 215   stored as part of the object.
 216 \end{methoddesc}
 217
 218 \begin{methoddesc}{read}{path}
 219   Load MIME information from a file named \var{path}.  This uses
 220   \method{readfp()} to parse the file.
 221 \end{methoddesc}
 222
 223 \begin{methoddesc}{readfp}{file}
 224   Load MIME type information from an open file.  The file must have
 225   the format of the standard \file{mime.types} files.
 226 \end{methoddesc}