# There is a publically editable copy of this file at
# http://www.mediawiki.org/wiki/AntiSpoof/Equivalence_sets

# This is the input file for generateEquivset.php
# The format is:
#
#    <hexadecimal codepoint> <character> => [<hexadecimal codepoint>] <character>
#
# If the codepoint is given, it must match the character, or else a warning
# will be issued and the line will be ignored.
#
# The effect of such a line is to conflate the two identified character, i.e.
# to put them in the same set. If two sets share a member, then they will be
# merged into a single larger set.
#
# We have attempted to include the following types of equivalence:
#    * Case folding. Although letters of different cases are often visually
#      distinct, they can easily be confused by people who are familiar with
#      the alphabet. Two words with a different case may be read as the same
#      word. This is a popular technique for impersonation.
#
#    * Visually similar characters. Cross-script pairs are included, but these
#      tend to produce false conflations within scripts, and so should be
#      avoided. The software implements a blanket restriction against cross-
#      script strings, which makes cross-script pairs mostly redundant.
#
#    * Chinese Simplified/Traditional pairs.
#
# The list is based on one by Neil Harris, which was derived by unknown methods.
# That list also contained transliteration pairs, which we considered excessive
# and have attempted to remove. For example, the latin E and H were considered
# equivalent, because the latin transliteration of the cyrillic "Н" (which
# looks like latin H) is "E".