database/postgres
1
0
Fork 0
mirror of https://github.com/postgres/postgres.git synced 2025-07-30 11:03:19 +03:00
Code Activity

Add Greek characters to unaccent.rules.

Browse Source 
Author: Tasos Maschalidis
Reviewed-by: Michael Paquier, Tom Lane
Discussion: https://postgr.es/m/153495048900.1368.11566580687623014380%40wrigleys.postgresql.org
Discussion: https://postgr.es/m/VI1PR01MB38537EBD529FE5EE3FE9A5FEB5370%40VI1PR01MB3853.eurprd01.prod.exchangelabs.com
This commit is contained in:
Thomas Munro
2018-09-02 07:12:24 +12:00
parent ec74369931
commit 5e8d670c31
2 changed files with 236 additions and 4 deletions
Download Patch File Download Diff File Expand all files Collapse all files

19
contrib/unaccent/generate_unaccent_rules.py
View File

@ -29,6 +29,15 @@ import argparse
import sys
import xml.etree.ElementTree as ET
# The ranges of Unicode characters that we consider to be "plain letters".
# For now we are being conservative by including only Latin and Greek. This
# could be extended in future based on feedback from people with relevant
# language knowledge.
PLAIN_LETTER_RANGES = ((ord('a'), ord('z')), # Latin lower case
(ord('A'), ord('Z')), # Latin upper case
(0x03b1, 0x03c9), # GREEK SMALL LETTER ALPHA, GREEK SMALL LETTER OMEGA
(0x0391, 0x03a9)) # GREEK CAPITAL LETTER ALPHA, GREEK CAPITAL LETTER OMEGA
def print_record(codepoint, letter):
print (unichr(codepoint) + "\t" + letter).encode("UTF-8")
@ -39,9 +48,11 @@ class Codepoint:
self.combining_ids = combining_ids
def is_plain_letter(codepoint):
"""Return true if codepoint represents a plain ASCII letter."""
return (codepoint.id >= ord('a') and codepoint.id <= ord('z')) or \
(codepoint.id >= ord('A') and codepoint.id <= ord('Z'))
"""Return true if codepoint represents a "plain letter"."""
for begin, end in PLAIN_LETTER_RANGES:
if codepoint.id >= begin and codepoint.id <= end:
return True
return False
def is_mark(codepoint):
"""Returns true for diacritical marks (combining codepoints)."""
@ -184,7 +195,7 @@ def main(args):
len(codepoint.combining_ids) > 1:
if is_letter_with_marks(codepoint, table):
charactersSet.add((codepoint.id,
chr(get_plain_letter(codepoint, table).id)))
unichr(get_plain_letter(codepoint, table).id)))
elif args.noLigaturesExpansion is False and is_ligature(codepoint, table):
charactersSet.add((codepoint.id,
"".join(unichr(combining_codepoint.id)