Hello,
I use PostgreSQL 7.0.2 on linux.
The base was set with initdb -E UNICODE.
I have many Strings with accents (french language).
Some of them aren't supported by queries or pg_dump:
WORKING EXAMPLE:
DB=# select * from element_attribute where java_lang_string like 'Scholtè_s';
doc_id | tag_id | java_lang_string | type | java_lang_integer
--------+--------+------------------+------+-------------------
(0
rows)
NOT WORKING EXAMPLE (psql still waiting for characters to be entered):
DB=# select * from element_attribute where java_lang_string like 'Scholtès';
DB'#
ERRORS IN FILE GENERATED BY PG_DUMP (AS A RESULT THE WHOLE TABLE WON'T BE
RESTORED):
2220 3 Faure java.lang.String \N
2221 3 Rosières java.lang.String \N
2222 3 Rosières java.lang.String \N
2223 3 Rosières java.lang.String \N
2224 3 Rosières java.lang.String \N
2225 3 Rosières java.lang.String \N
2226 3 Rosières java.lang.String \N
2227 3 Scholtès
2228 3 Scholtès
2229 3 Scholtès
2230 3 Scholtès
2231 3 Scholtès
2232 3 Scholtès
2233 3 Scholtès
2234 3 Scholtès
2235 3 Whirlpool java.lang.String \N
As you can see, the string ended by 'ès' is quite lethal for Postgre.
The success of pg_dump depends on database content !
It seems that any string ending by an accent followed by less than 2 characters
is a problem.
Is it a known problem?
Is it a database configuration problem or a real bug in Postgre?
This problem is on my production database. I would appreciate If you could give
me an answer soon on that point.
Thanks in advance.
Samuel Minne
