Hello,

I wrote a patch for header search about a word over two lines. Please check it.

For example, following header string contains a word splitted into two mime encoded block. It's normal behavior of MUAs in CJK environment.

Subject: =?ISO-2022-JP?B?Sk1CGyRCJUQlIiE8JUslZSE8JTkhIRsoQkpBTBskQiVBJWMhPCU/GyhC?= =?ISO-2022-JP?B?GyRCITxKWCRHOVQkLxsoQiAbJEI0WjlxGyhCLxskQiVRJWklKhsoQi8bJEIlNSUkJVElcxsoQg==?=

And now, dovecot search command miss to search the word. word : "ABCDE" MIMED : "AB\r\n\tCDE" search(now): "AB CDE" this patch: "ABCDE"

This patch has a bad side effect that in English or other languages, two words is connected.

For example,

This is a sample. matches to "asample".

I think this is not a big problem, but if there is a good solution, please let me know.

best regards,

Kazuo Moriwaka moriwaka@valinux.co.jp

Index: src/lib-mail/message-header-search.c

RCS file: /home/cvs/dovecot/src/lib-mail/message-header-search.c,v retrieving revision 1.13 diff -r1.13 message-header-search.c 149c149 < chr = ' ';

  	chr = '\t';

153c153 < if (chr == '\r' || chr == '\n')

  if (chr == '\r' || chr == '\n' || chr == '\t')