"Fossies" - the Fresh Open Source Software Archive

Member "recoll-1.26.3/sampleconf/fields" (17 Sep 2018, 6238 Bytes) of package /linux/privat/recoll-1.26.3.tar.gz:


As a special service "Fossies" has tried to format the requested text file into HTML format (style: standard) with prefixed line numbers. Alternatively you can here view or download the uninterpreted source code file.

    1 # (C) 2007-2011 J.F.Dockes
    2 # License: GPL V2
    3 #
    4 # Field names configuration. This defines how one may search ie for:
    5 #   author:Hemingway
    6 #
    7 # Important: 
    8 #   - the field names MUST be all lowercase alphabetic ascii here. They can
    9 #     be anycased in the documents.
   10 
   11 [prefixes]
   12 
   13 #####################################################
   14 # This section defines what prefix the terms inside named fields will be
   15 # indexed with (in addition to prefix-less indexing for general search)
   16 # ALL prefixes MUST be all ASCII UPPERCASE (NO DIGITS)
   17 # 
   18 # The field names should be the canonic ones, not the aliases defined in
   19 # the following section. Don't change those which are predefined here, 
   20 # quite a few are hard-coded in the c++. But you can add more (for new
   21 # fields emitted by filters).
   22 #
   23 # Fields can have two relevance boost factors defined, such as in:
   24 # caption = S ; wdfinc=10
   25 #  and/or
   26 # caption = S ; boost = 10
   27 # The first line would boost the xapian "within document frequency" of
   28 # caption terms by a factor of 10 at indexing time. The second one 
   29 # (NOT CURRENTLY IMPLEMENTED) would automatically boost the weight of a
   30 # caption-based field query (ie: caption:mytitle or title:mytitle) at query
   31 # time.
   32 #
   33 # The pfxonly attribute can also be set on entries to express that terms
   34 # from the field should be indexed only with a prefix (in general, field
   35 # terms are indexed both with and without a prefix).
   36 
   37 # The following ones are probably hard-coded in the c code, can't change at
   38 # all.
   39 # Also reserved: F(parentid), Q(uniqueid)
   40 author = A
   41 xapdate = D
   42 keywords= K
   43 xapyearmon = M
   44 title = S ; wdfinc = 10
   45 mtype = T
   46 ext = XE; noterms = 1
   47 rclmd5 = XM
   48 dir = XP ; noterms = 1
   49 abstract = XS
   50 filename = XSFN ; noterms = 1
   51 containerfilename = XCFN ; pfxonly = 1 ; noterms = 1
   52 rclUnsplitFN = XSFS
   53 xapyear = Y
   54 recipient = XTO
   55 
   56 # Extension examples. These are actually used by default by Recoll, you can
   57 # add your own to search for fields produced by the filters and not handled
   58 # by default. 
   59 # Some values are internally reserved by recoll: 
   60 #   XE (file ext), XP (for path elements), XSFN, XSFS, XXST, XXND, XXPG
   61 rclbes = XB ; noterms = 1
   62 # Using XX was not a good idea. 
   63 #
   64 # I hereby commit to not using XY for Recoll:
   65 # *** USE XY for beginning your local prefixes *** ie:
   66 # myfield = XYMYPREF
   67 
   68 [values]
   69 ###########
   70 ## Fields which will be stored in Xapian values, authorizing range query
   71 ## processing.
   72 # Entries are specified as 'fieldname = valueslot;[px=val1;py=val2...]'.
   73 # Xapian value slots are 32 bits numbers. Numbers below 1000 are reserved
   74 # by Recoll or Xapian. Numbers above are available for user configuration
   75 # Values have types, which can be 'int' or 'string' at the moment. ints have
   76 # an additional 'len' attributes, which specifies the padding size used for
   77 # sorting (leading zeroes: all xapian sorting is text-based). 10 is fine
   78 # for an unsigned 32 bits integer.
   79 # myfield = 1001; type=int; len = 10
   80 # mystrfield = 1002; type = string
   81 
   82 [stored]
   83 ############################
   84 # Some fields are stored in the document data record inside the index and
   85 # can be displayed in result lists. There is no necessity that stored fields
   86 # should be indexed (have a prefix in the preceding section). Example: "url"
   87 #
   88 # Some fields are stored by default, don't add them here:
   89 #    caption, mimetype, url
   90 # Only canonical names should be used here, not aliases.
   91 # "rclaptg" is used for viewer specialization (depending on local config)
   92 # "rclbes" defines the backend type (ie normal fs, firefox cache). Should
   93 #   probably be hardcoded, don't remove it
   94 abstract=
   95 author=
   96 filename=
   97 keywords=
   98 rclaptg=
   99 rclbes=
  100 recipient=
  101 
  102 [aliases]
  103 ##########################
  104 # This section defines field names aliases or synonyms. Any right hand side
  105 # value will be turned into the lhs canonic name before further treatment
  106 #
  107 # The left-hand values in the recoll distribution file are well known and
  108 # must match names used in the c++ code, or even the index data
  109 # record. They can't change! But you can add others.
  110 #
  111 # Filters should only add canonic names to the meta array when indexing,
  112 # not aliases.
  113 abstract = summary dc:summary description xesam:description
  114 author = creator dc:creator xesam:author xesam:creator from
  115 title = caption title dc:title subject
  116 # catg = dc:type contentCategory
  117 dbytes = size xesam:size
  118 dmtime = date dc:date dc:datemodified datemodified contentmodified \
  119        xesam:contentmodified
  120 ext = fileextension xesam:fileextension
  121 # Don't add subject to keywords aliases, it's better to keep it for email
  122 keywords = keyword xesam:keyword tag tags dc:subject xesam:subject \
  123 	 dc:description
  124 mtype = mime mimetype xesam:mimetype contenttype xesam:contenttype dc:format
  125 recipient = to xesam:recipient
  126 url = dc:identifier xesam:url
  127 
  128 ##################
  129 # The queryaliases section defines aliases which are used exclusively at
  130 # query time: there is no risk to pick up a random field from a document
  131 # (e.g. an HTML meta field) and index it. 
  132 [queryaliases]
  133 filename = fn
  134 containerfilename = cfn
  135 
  136 [xattrtofields]
  137 ######################
  138 # Processing for extended file attributes.
  139 # By default, attributes are processed as document fields (after removing
  140 # the 'user' prefix from the name on Linux).
  141 # You can enter name translations as "xattrname = fieldname". Case matters.
  142 # Entering an empty translation will disable use of the attribute.
  143 # The values from the extended attributes will extend, not replace, the
  144 # data found from equivalent fields inside the document. 
  145 
  146 # As an example, the following would map a quite plausible "tags" extended
  147 # attribute into the "keywords" field.
  148 tags = keywords
  149 
  150 # Proposed or standard freedesktop.org extended attributes
  151 xdg.tags = keywords
  152 xdg.comment = abstract
  153 
  154 # Some standard fields are not to be indexed
  155 mime_type =
  156 charset =
  157 
  158 ########################
  159 # Sections reserved for specific filters follow
  160 #
  161 
  162 ##########################
  163 # Mail filter section. You can specify mail headers to be indexed 
  164 # in addition to the standard ones: (To:, Cc:, From:, Subject:, Date, 
  165 # Message-Id), along with the field name to be used. For this to be useful, 
  166 # the field name should also be listed in the [prefixes] and possibly the
  167 # [stored] sections.
  168 #
  169 # [mail]
  170 # x-my-tag = mymailtag