Re: Using tin to access fido(7) nntp geteways
Urs Janßen <[email protected]> Thu, 10 May 2012 11:56:14 +0200
| Newsgroups | gmane.network.tin.user |
|---|---|
| Message-ID | <[email protected]> |
On Wed, May 09, 2012 at 05:12:28PM +0400, Anton Shepelev wrote: > > do you use overview caching or the like? > I am using: > cache_overview_files = ON tin doesn't cache the original data but converts the articles headers/nov data to the local charset are writes its own nov data. There might be a bug in there, the overview caching is rarely used. On a first quick look the code seems to be ok: when articles subject/from/... is read rfc1522_decode() is used to convert the data into the local charset, rfc1522_decode() takes care about undeclared_charset etc.pp. art.c:read_overview() and/or art.c:parse_headers() when nov entries are written rfc1522_encode(..., tinrc.mm_local_charset, ...) is used to ensure the data is propper mime-encoded with a valid charset if post_8bit_header is not set, otherweise the raw 8bit data is written out. art.c:write_overview() ~line 2042 as you use post_8bit_header=ON (= raw data in nov file), art.c:read_overview() ~line 1750 art->subject = hash_str(eat_re(eat_tab(convert_to_printable(rfc1522_decode(ptr), FALSE)), FALSE)); should convert the raw overview subject back to the local charset via rfc2047.c:rfc1522_decode() ~line 255 process_charsets(&c, &max_len, (CURR_GROUP.attribute->undeclared_charset) ? : "US-ASCII", tinrc.mm_local_charset, FALSE); does your local charset differ from the one given in undeclared_charset? > although I am not sure eternal-september NNTP server > supports NOV files... IIRC eternal september runs a recent version of inn, inn supports nov files for about 20 years now. urs -- "Only whimps use tape backup: _real_ men just upload their important stuff on ftp, and let the rest of the world mirror it ;)" - Linus