tweaked the parser to support cjk languages

This commit is contained in:
SJ
2013-11-26 11:43:21 +01:00
parent c0301ceaca
commit 1e44042b82
5 changed files with 12 additions and 2 deletions

View File

@ -103,6 +103,13 @@ archive_only_mydomains=0
; minimum word length in mail body to index
min_word_len=1
; whether to enable CJK (=Chinese, Japanese, and Korean) "characters".
; the text piler can see with CJK languages may have extremely long
; sequences without any whitespace. To prevent the parser to drop
; these very long sequences, enable (1) this feature. By default it's
; disabled (0).
enable_cjk=0
; if piler detects this line in the mail header, then it will assume
; the message is a spam. You should include your own antispam solution's
; specific line.