# Search normalization rules
#
# See basic.tr for formatting options
#
# The rest of search handling is done in the various larger alphabet groups
# e.g latin.tr and cyrillic.tr
# They will have rules called <identifier>_search_cleanup

endline_search_normalize:
# Turns endlines \n, \r and \r\n into a space
U+000A = U+0020
U+000D = U+0020

tab_search_normalize:
# Turns tabs \t into a space
U+0008 = U+0020

specialwords_search_normalize:
U+00AA = "a"  # feminine ordinal indicator (spanish)
U+00BA = "o"  # masculine ordinal indicator (spanish)

punctuation_normalize:
# Turn some symbols which are considered punctuations marks into
# the ASCII punctuation U+002E
U+0021 = U+002E # exclamation mark !
U+003F = U+002E # question mark ?
U+002C = U+002E # comma ,

latin_search_decompose:
U+00B1 = "+-"    # Plus/minus sign
U+00B2 = "2"     # Superscript two
U+00B3 = "3"     # Superscript three
U+00B9 = "1"     # Superscript one
U+00BA = "1"     # Superscript one
U+00BC = "1/4"   # Vulgar fraction one quarter
U+00BD = "1/2"   # Vulgar fraction one half
U+00BE = "3/4"   # Vulgar fraction three quarter