cygwin + imap cache failures and possible fix
- To: mutt-dev@xxxxxxxx
- Subject: cygwin + imap cache failures and possible fix
- From: "Robbie Gates" <robbie@xxxxxxx>
- Date: Sun, 16 Mar 2008 23:12:47 +1100
- Dkim-signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=domainkey-signature:received:received:message-id:date:from:sender:to:subject:mime-version:content-type:content-transfer-encoding:content-disposition:x-google-sender-auth; bh=EKs3dPbUf7LNiINWik1Aqw4XRWC3ThuyVRZ1NzAsouQ=; b=YoECyE9L4VLibDCi/3tIEHU7k3hpgeaIUIjj/kFMfUqGRy6aUvYa8T4N/2/0wY28RZ6CzfBnXtjY8UuMgpgQto7ZxrhLsZR6Gd/qLCVwessPVw4UFwlRKGw1S+5jg/lrW0B3ObbQ971+VcubvMoCrlrF/XbAWHZmqcItQI6YJUE=
- Domainkey-signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=message-id:date:from:sender:to:subject:mime-version:content-type:content-transfer-encoding:content-disposition:x-google-sender-auth; b=SN+2C1SSa67rE7DFhHLpG23EI7+0ftsLTGUjrh0ivgdpwOhi6QG2+ljcNiCurBpxNknjL3u7OGfBICPl+EmiKD1aMSF2jx4yG+GnXaTchArgkoxwggjXdPB0ra180K3AaHIq+wo+zfXE4J4S5GDvTBj98Pt6B5iOV61FZZaqqGA=
- List-post: <mailto:mutt-dev@mutt.org>
- List-unsubscribe: send mail to majordomo@mutt.org, body only "unsubscribe mutt-dev"
- Sender: owner-mutt-dev@xxxxxxxx
Hi All,
i built mutt 1.5.17 on cygwin, and had trouble getting the header/body
caching to work. A bit of investigation revealed that the culprit is
the inability to use a ':' in a filename on cygwin/win32 coupled with
the cache naming scheme based on the imap url which includes a ':'. As
a quick fix, i patched both bcache_path in bcache.c and
mutt_hcache_per_folder in hcache.c to mutt_sanitize_filename their
cache paths before returning them, and removed ':' from the list of
legal filename characters safe_chars for mutt_sanitize_filename in
lib.c. Things appear to work with these fixes.
i'm not sure how acceptable removing : from the characters considered
safe is, or if there's a way to make this platform dependent.
i've attached a patch generated using diff against the distributed
1.5.17 sources.i left the dprint i used to check what was going on in
also.
Apologies for not using mercurial as advised in devel-notes.txt, but
my attempts to use the win32 all in one versions of mercurial to clone
mutt failed (abort: premature EOF reading chunk (got 33076 bytes,
expected 37378)).
- robbie