| 1 |
dpavlin |
11 |
################################################### |
| 2 |
|
|
|
| 3 |
|
|
#IndexDir http://www.rot13.org |
| 4 |
|
|
|
| 5 |
|
|
#IndexFile /data/swish/index/rot13 |
| 6 |
|
|
|
| 7 |
|
|
#IndexName "rot13.org" |
| 8 |
|
|
#IndexDescription "Internet web pages" |
| 9 |
|
|
#IndexPointer "http://www.rot13.org/" |
| 10 |
|
|
#IndexAdmin "dpavlin@rot13.org" |
| 11 |
|
|
|
| 12 |
|
|
# don't index files with this suffix (filesys only?) |
| 13 |
|
|
NoContents .gif .xbm .au .mov .mpg ..ps .bin .exe .wmv |
| 14 |
|
|
|
| 15 |
|
|
# MetaNames automatic |
| 16 |
|
|
MetaNames first author |
| 17 |
|
|
|
| 18 |
|
|
# 0 to 3 - 0 is totally silent, 3 is the most verbose. |
| 19 |
|
|
#IndexReport 2 |
| 20 |
|
|
|
| 21 |
|
|
#FollowSymLinks yes |
| 22 |
|
|
|
| 23 |
|
|
#UseStemming no |
| 24 |
|
|
|
| 25 |
|
|
IgnoreTotalWordCountWhenRanking yes |
| 26 |
|
|
# Put yes to ignore the total number of words in the file when calculating |
| 27 |
|
|
# ranking. Often better with merges and small files. Default is no. |
| 28 |
|
|
|
| 29 |
|
|
WordCharacters abcdefghijklmnopqrstuvwxyz\&#;0123456789.@|,-'"[](~!@$%^{}_+?¹©ðÐèÈæƾ® |
| 30 |
|
|
|
| 31 |
|
|
# We allow a period and a dash within words, but strip them |
| 32 |
|
|
# from the beginning or end of a word. This is done after |
| 33 |
|
|
# WordCharacters above is used to split words. |
| 34 |
|
|
|
| 35 |
|
|
IgnoreFirstChar .-"'`([{< |
| 36 |
|
|
IgnoreLastChar .-,!?;"'`)]}> |
| 37 |
|
|
|
| 38 |
|
|
# Finally, resulting words must begin/end with one |
| 39 |
|
|
# of the characters listed here |
| 40 |
|
|
|
| 41 |
|
|
#BeginCharacters abcdefghijklmnopqrstuvwxyz0123456789é |
| 42 |
|
|
#EndCharacters abcdefghijklmnopqrstuvwxyz0123456789é |
| 43 |
|
|
|
| 44 |
|
|
|
| 45 |
|
|
# This automatically omits words that appear too often in the files |
| 46 |
|
|
# (these words are called stopwords). Specify a whole percentage |
| 47 |
|
|
# and a number, such as "80 256". This omits words that occur in |
| 48 |
|
|
# over 80% of the files and appear in over 256 files. Comment out |
| 49 |
|
|
# to turn of auto-stopwording. |
| 50 |
|
|
IgnoreLimit 90 500 |
| 51 |
|
|
|
| 52 |
|
|
# This option allows the user decide if to index the comments in the files |
| 53 |
|
|
# default is 1. Set to 0 if comment indexing is not required. |
| 54 |
|
|
IndexComments 0 |
| 55 |
|
|
|
| 56 |
|
|
#TranslateCharacters string1 string2 |
| 57 |
|
|
# This option allows to index the characters in string1 to be indexed |
| 58 |
|
|
# as the characteres in string2. |
| 59 |
|
|
# This is done after htnl entities are converted |
| 60 |
|
|
# This option is useful in languages like spanish, french, ... |
| 61 |
|
|
# eg: |
| 62 |
|
|
# TranslateCharacters _á -a |
| 63 |
|
|
# This will index a_b as a-b and ámo as amo |
| 64 |
|
|
TranslateCharacters ¹©ðÐèÈæƾ® ssddcccczz |
| 65 |
|
|
|
| 66 |
|
|
################################ |
| 67 |
|
|
# DIRECTIVES for HTTP METHOD ONLY |
| 68 |
|
|
# Comment out if using FILESYSTEM |
| 69 |
|
|
################################## |
| 70 |
|
|
|
| 71 |
|
|
MaxDepth 0 |
| 72 |
|
|
#(default 5) This defines how many links the spider should |
| 73 |
|
|
#follow before stopping. A value of 0 configures the spider to |
| 74 |
|
|
#traverse all links |
| 75 |
|
|
|
| 76 |
|
|
Delay 0 |
| 77 |
|
|
#(default 60) The number of seconds to wait between issuing |
| 78 |
|
|
#requests to a server. |
| 79 |
|
|
|
| 80 |
|
|
SpiderDirectory /data/swish/spider |
| 81 |
|
|
|
| 82 |
|
|
#EquivalentServer http://portal.pliva.hr http://www.pliva.hr |
| 83 |
|
|
#(default nothing) This allows you to deal with |
| 84 |
|
|
#servers that use respond to multiple DNS names. Each line should have |
| 85 |
|
|
#a list of all the method/names that should be considered equivalent. |
| 86 |
|
|
#If you have multiple directives, each one defines its own set of equivalent |
| 87 |
|
|
#servers. |
| 88 |
|
|
|
| 89 |
|
|
# documents which are html files |
| 90 |
|
|
IndexContents HTML .htm .html .php .pl |
| 91 |
|
|
|
| 92 |
|
|
# text documents |
| 93 |
|
|
IndexContents TXT .txt .log .text |
| 94 |
|
|
|
| 95 |
|
|
# default |
| 96 |
|
|
DefaultContents HTML |
| 97 |
|
|
|
| 98 |
|
|
# use 100 chars from html |
| 99 |
|
|
StoreDescription HTML <body> 300 |
| 100 |
|
|
StoreDescription TXT 300 |
| 101 |
|
|
|
| 102 |
|
|
|
| 103 |
|
|
# index pdf |
| 104 |
|
|
IndexContents HTML .pdf |
| 105 |
|
|
FileFilter .pdf pdftotext "-htmlmeta '%p' -" |
| 106 |
|
|
|
| 107 |
|
|
# various other |
| 108 |
|
|
FileFilter .gz gzip "-cd '%p'" |
| 109 |
|
|
FileFilter .bz2 bzip2 "-cd '%p'" |
| 110 |
|
|
|
| 111 |
|
|
# debug |
| 112 |
|
|
ParserWarnLevel 3 |
| 113 |
|
|
IndexReport 2 |
| 114 |
|
|
|
Parent Directory
| 
Revision Log