# HG changeset patch
# User Goffi <goffi@goffi.org>
# Date 1285668006 -28800
# Node ID 8603fcb8615ad6f49b14027f02ec3b309a4e7a83
# Parent  da8353f26f4b4d1d41124f1d6bc8d24d7cd8e733
--no-unicode-fix commented, and README update

diff -r da8353f26f4b -r 8603fcb8615a README
--- a/README	Tue Sep 28 17:25:28 2010 +0800
+++ b/README	Tue Sep 28 18:00:06 2010 +0800
@@ -22,12 +22,12 @@
 
 
 ** WTF ? **
-gcp is a file copier, loosely inspired from cp, but with high level functionnalities like:
+gcp is a file copier, loosely inspired from cp, but with high level functionalities like:
 - progression indicator
-- gcp continue coping even when there is an issue: he just skip the file with problem, and go on
-- journalization: gcp write what he is doing, this allow to now which files where effectively copied
+- gcp continue copying even when there is an issue: he just skip the file with problem, and go on
+- journalization: gcp write what he is doing, this allow to know which files where effectively copied
 - fixing names to be compatible with the target filesystem (e.g. removing incompatible chars like "?" or "*" on vfat)
-- if you launch a copy when an other is already running, the files are added to the first queue, this avoir you hard drive to move its head all the time
+- if you launch a copy when an other is already running, the files are added to the first queue, this avoid your hard drive to move its head all the time
 - files saving: you can keep track of files you have copied, and re-copy them later (useful when, for example, you always copy some free music to all your friends).
 - gcp will be approximately option-compatible with cp (approximately because the behaviour is not exactly the same, see below)
 
@@ -39,18 +39,31 @@
 Please note that the behaviour is not exactly the same as cp, even if gcp want to be option-compatible. Mainly, the destination filenames can be changed (by default, can be deactivated).
 gcp doesn't implement yet all the options from cp, but it's planed.
 
+** journalizaion **
+The journal is planed to be used by gcp itself, buts remains human-readable. It is located in ~/.gcp/journal
+
+3 states are used:
+- OK means the file is copied and all operation were successful
+- PARTIAL means the file is copied, but something went wrong (e.g. changing the permissions of the file)
+- FAILED: the file is *not* copied
+
+after the state, a list of things which went wront are show, separated by ", "
+
 ** What's next ? **
 
-Several improvment are already planed:
+Several improvment are already planed
 - copy queue management (moving copy order)
 - advanced console interface
 - notification (xmpp and maybe mail) when a long copy is finished
-- retry for files which where not correctly copied
+- retry for files which were not correctly copied
 - badly encoded unicode filenames fix
+- file copy integrity check
 
 ... and other are with a "maybe"
 - graphic interface
 - desktop (Kde, Gnome, XFCE, ...) integration
+- distant copy (ftp)
+- basic server mode, for copying files on network without the need of nfs or other heavy stuff
 
 ** Credits **
 
diff -r da8353f26f4b -r 8603fcb8615a gcp
--- a/gcp	Tue Sep 28 17:25:28 2010 +0800
+++ b/gcp	Tue Sep 28 18:00:06 2010 +0800
@@ -498,8 +498,8 @@
         parser.add_option("--preserve", action="store", default='mode,ownership,timestamps',
                     help=_("preserve  the  specified  attributes"))
         
-        parser.add_option("--no-unicode-fix", action="store_false", dest='unicode_fix', default=True,
-                    help=_("don't fixe name encoding errors")) #TODO
+        #parser.add_option("--no-unicode-fix", action="store_false", dest='unicode_fix', default=True,
+        #            help=_("don't fixe name encoding errors")) #TODO
 
         parser.add_option("--no-fs-fix", action="store_false", dest='fs_fix', default=True,
                     help=_("don't fixe filesystem name incompatibily"))