Login | Register
My pages Projects Community openCollabNet

Discussions > users > Re: Invalid or incomplete multibyte or wide character

fsvs
Discussion topic

Hide all messages in topic

All messages in topic

Re: Invalid or incomplete multibyte or wide character

Author pmarek
Full name P.Marek
Date 2007-06-27 08:20:49 PDT
Message On Mittwoch, 27. Juni 2007 Worley, Chris B wrote:
> Again, I apologize. It works fine in these cases:
Ok, NP.

> Having made these extra revisions, how do I get rid of them?
- Either just ignore them, or
- dump/load the repository, or
- if it's a fsfs repository, and these unwanted revisions are at the end, just
  change the "current" file, and delete the corresponding "revs/"
  and "rev-props/" files.


Regards,

Phil


--
Versioning your /etc, /home or even your whole installation?
             Try fsvs (fsvs.tigris.org)!

RE: Re: Invalid or incomplete multibyte or wide character

Author "Worley, Chris B" <chris dot b dot worley at intel dot com>
Full name "Worley, Chris B" <chris dot b dot worley at intel dot com>
Date 2007-06-27 07:26:31 PDT
Message >-----Original Message-----
>From: Philipp Marek [mailto:philipp at marek dot priv dot at]
>Sent: Wednesday, June 27, 2007 12:30 AM
>To: users at fsvs dot tigris dot org
>Cc: Worley, Chris B
>Subject: Re: Invalid or incomplete multibyte or wide character
>
>On Dienstag, 26. Juni 2007 Worley, Chris B wrote:
>> Sorry, I had tried both ways, with "-m" and with the editor.
>>
>> On the output I showed, it was with the editor, and not the "-m"
switch,
>> and I entered nothing in that case (just exited the editor) and did
not
>> have output redirected.
>>
>> The "-m" commit switch is working properly. Sorry for the confusion.
>>
>> The "-D..." switch still aborts.
>Hmmmm ... it doesn't for me.
>What is your exact command line, and what does FSVS print?

Again, I apologize. It works fine in these cases:

1) With -m:

# fsvs ci -d -D hlp___do_convert -m "test"
07:21:54.673 hlp___do_convert[helper.c:138] before iconv from=test
07:21:54.673 hlp___do_convert[helper.c:145] after iconv to=test ret=0
Committing to file:///tmp/images
committed revision 7 on 2007-06-27T14:21:54.679312Z as root

2) Without -m, if I enter something and save in vi.

It only gives the error if I exit vi w/o saving:

An error occurred at 07:22:23.495: Success (0)
  in ci__work [commit.c:760]: mmap commit message
(/tmp/commit-tmp.87zhiU, 0 bytes)

...I'm sure this is expected behavior.

Having made these extra revisions, how do I get rid of them?
>
>
>Regards,
>
>Phil
>
>--
>Versioning your /etc, /home or even your whole installation?
> Try fsvs (fsvs.tigris.org)!
>
>----------------​--------------------​--------------------​-------------
>To unsubscribe, e-mail: users-unsubscribe@fs​vs.tigris.org
>For additional commands, e-mail: users-help at fsvs dot tigris dot org

Re: Invalid or incomplete multibyte or wide character

Author pmarek
Full name P.Marek
Date 2007-06-26 23:29:58 PDT
Message On Dienstag, 26. Juni 2007 Worley, Chris B wrote:
> Sorry, I had tried both ways, with "-m" and with the editor.
>
> On the output I showed, it was with the editor, and not the "-m" switch,
> and I entered nothing in that case (just exited the editor) and did not
> have output redirected.
>
> The "-m" commit switch is working properly. Sorry for the confusion.
>
> The "-D..." switch still aborts.
Hmmmm ... it doesn't for me.
What is your exact command line, and what does FSVS print?


Regards,

Phil

--
Versioning your /etc, /home or even your whole installation?
             Try fsvs (fsvs.tigris.org)!

Re: Invalid or incomplete multibyte or wide character

Author pmarek
Full name P.Marek
Date 2007-06-26 23:27:30 PDT
Message > >Please note that the shell expands the "*" - so you'll have a list of
> >currently existing files that will get ignored, not *every* possibly
> >existing file in the future!
>
> Yes, I actually did the commands in the opposite order, having figured
> out the command line expansion after trying the unquoted version. There
> was no way to "unignore" all the expansions.
Well, use
    fsvs ignore dump > /tmp/file
    vi /tmp/file
    fsvs ignore load < /tmp/file
for such things.

> >There's no init anymore. Where did you see it?
>
> In the "doc/USAGE" file.
Got it, thank you.


Regards,

Phil

--
Versioning your /etc, /home or even your whole installation?
             Try fsvs (fsvs.tigris.org)!

RE: Re: Invalid or incomplete multibyte or wide character

Author "Worley, Chris B" <chris dot b dot worley at intel dot com>
Full name "Worley, Chris B" <chris dot b dot worley at intel dot com>
Date 2007-06-26 13:30:41 PDT
Message Sorry, I had tried both ways, with "-m" and with the editor.

On the output I showed, it was with the editor, and not the "-m" switch,
and I entered nothing in that case (just exited the editor) and did not
have output redirected.

The "-m" commit switch is working properly. Sorry for the confusion.

The "-D..." switch still aborts.

>-----Original Message-----
>From: Philipp Marek [mailto:philipp at marek dot priv dot at]
>Sent: Tuesday, June 26, 2007 2:24 PM
>To: users at fsvs dot tigris dot org
>Cc: Worley, Chris B
>Subject: Re: Invalid or incomplete multibyte or wide character
>
>On Dienstag, 26. Juni 2007 Worley, Chris B wrote:
>> I had both a -m, and didn't redirect the output on this run. The
editor
>> was called, I exited having entered nothing.
>You had a "-m message", and the editor was called??
>Then I don't understand...
>
>Could you do a commit, then touch a single file, and send me the output
of
> fsvs ci -d -m test
>if the editor was called again?
>
>
>Regards,
>
>Phil
>
>
>--
>Versioning your /etc, /home or even your whole installation?
> Try fsvs (fsvs.tigris.org)!
>
>----------------​--------------------​--------------------​-------------
>To unsubscribe, e-mail: users-unsubscribe@fs​vs.tigris.org
>For additional commands, e-mail: users-help at fsvs dot tigris dot org

RE: Re: Invalid or incomplete multibyte or wide character

Author "Worley, Chris B" <chris dot b dot worley at intel dot com>
Full name "Worley, Chris B" <chris dot b dot worley at intel dot com>
Date 2007-06-26 13:24:53 PDT
Message >-----Original Message-----
>From: Philipp Marek [mailto:philipp at marek dot priv dot at]
>Sent: Tuesday, June 26, 2007 2:15 PM
>To: users at fsvs dot tigris dot org
>Cc: Worley, Chris B
>Subject: Re: Invalid or incomplete multibyte or wide character
>
>On Dienstag, 26. Juni 2007 Worley, Chris B wrote:
>> >> It is strange that
>> >> even when "ignored" the file names show up during "commit".
>> >
>> >They shouldn't. Are you sure that you ignored them correctly?
>>
>> I'm never sure I'm doing anything correctly ;)
>>
>> I used both:
>>
>> # fsvs ignore "./roots/*/var/log/*"
>> # fsvs ignore ./roots/2/var/log/*
>Please note that the shell expands the "*" - so you'll have a list of
>currently existing files that will get ignored, not *every* possibly
existing
>file in the future!

Yes, I actually did the commands in the opposite order, having figured
out the command line expansion after trying the unquoted version. There
was no way to "unignore" all the expansions.

>
>> ...and my dump looks like:
>>
>> # fsvs ignore dump
>> ...
>> ./roots/2/var/log/messages
>> ...
>> ./roots/*/var/log/*
>>
>> ...and my commit looks like:
>>
>> # fsvs commit -m "test"
>> ...
>> .mC. 13786 ./roots/2/var/log/messages
>> ...
>I think your problem is that they're already registered for versioning,
so
>they'll not be ignored anymore.
>
>Take a look at "fsvs unversion".

Got it. Thanks.
>
>
>> Note that I never did an "fsvs init"... the README " How is it used?"
>> section didn't contain one... I didn't see the "init" command until I
>> read the "USAGE".
>There's no init anymore. Where did you see it?

In the "doc/USAGE" file.

Thanks again,

Chris
>
>
>Regards,
>
>Phil
>
>
>--
>Versioning your /etc, /home or even your whole installation?
> Try fsvs (fsvs.tigris.org)!
>
>----------------​--------------------​--------------------​-------------
>To unsubscribe, e-mail: users-unsubscribe@fs​vs.tigris.org
>For additional commands, e-mail: users-help at fsvs dot tigris dot org

Re: Invalid or incomplete multibyte or wide character

Author pmarek
Full name P.Marek
Date 2007-06-26 13:23:40 PDT
Message On Dienstag, 26. Juni 2007 Worley, Chris B wrote:
> I had both a -m, and didn't redirect the output on this run. The editor
> was called, I exited having entered nothing.
You had a "-m message", and the editor was called??
Then I don't understand...

Could you do a commit, then touch a single file, and send me the output of
    fsvs ci -d -m test
if the editor was called again?


Regards,

Phil


--
Versioning your /etc, /home or even your whole installation?
             Try fsvs (fsvs.tigris.org)!

RE: Re: Invalid or incomplete multibyte or wide character

Author "Worley, Chris B" <chris dot b dot worley at intel dot com>
Full name "Worley, Chris B" <chris dot b dot worley at intel dot com>
Date 2007-06-26 13:21:23 PDT
Message >-----Original Message-----
>From: Philipp Marek [mailto:philipp at marek dot priv dot at]
>Sent: Tuesday, June 26, 2007 2:20 PM
>To: users at fsvs dot tigris dot org
>Cc: Worley, Chris B
>Subject: Re: Invalid or incomplete multibyte or wide character
>
>On Dienstag, 26. Juni 2007 Worley, Chris B wrote:
>> # fsvs ci -d -D hlp___do_convert
>> An error occurred at 12:27:25.333: Success (0)
>> in ci__work [commit.c:760]: mmap commit message
>> (/tmp/commit-tmp.NCbeZ7, 0 bytes)
>I think I know what the problem is.
>You redirected the output to a file, but had no commit message given
via "-m",
>right?
>
>Then FSVS tries to call an editor (like svn does), but without a valid
STDOUT
>(because it's redirected) that won't work ...

I had both a -m, and didn't redirect the output on this run. The editor
was called, I exited having entered nothing.

Chris
>
>
>Regards,
>
>Phil
>
>
>--
>Versioning your /etc, /home or even your whole installation?
> Try fsvs (fsvs.tigris.org)!
>
>----------------​--------------------​--------------------​-------------
>To unsubscribe, e-mail: users-unsubscribe@fs​vs.tigris.org
>For additional commands, e-mail: users-help at fsvs dot tigris dot org

Re: Invalid or incomplete multibyte or wide character

Author pmarek
Full name P.Marek
Date 2007-06-26 13:19:33 PDT
Message On Dienstag, 26. Juni 2007 Worley, Chris B wrote:
> # fsvs ci -d -D hlp___do_convert
> An error occurred at 12:27:25.333: Success (0)
> in ci__work [commit.c:760]: mmap commit message
> (/tmp/commit-tmp.NCbeZ7, 0 bytes)
I think I know what the problem is.
You redirected the output to a file, but had no commit message given via "-m",
right?

Then FSVS tries to call an editor (like svn does), but without a valid STDOUT
(because it's redirected) that won't work ...


Regards,

Phil


--
Versioning your /etc, /home or even your whole installation?
             Try fsvs (fsvs.tigris.org)!

Re: Invalid or incomplete multibyte or wide character

Author pmarek
Full name P.Marek
Date 2007-06-26 13:14:59 PDT
Message On Dienstag, 26. Juni 2007 Worley, Chris B wrote:
> >> It is strange that
> >> even when "ignored" the file names show up during "commit".
> >
> >They shouldn't. Are you sure that you ignored them correctly?
>
> I'm never sure I'm doing anything correctly ;)
>
> I used both:
>
> # fsvs ignore "./roots/*/var/log/*"
> # fsvs ignore ./roots/2/var/log/*
Please note that the shell expands the "*" - so you'll have a list of
currently existing files that will get ignored, not *every* possibly existing
file in the future!

> ...and my dump looks like:
>
> # fsvs ignore dump
> ...
> ./roots/2/var/log/messages
> ...
> ./roots/*/var/log/*
>
> ...and my commit looks like:
>
> # fsvs commit -m "test"
> ...
> .mC. 13786 ./roots/2/var/log/messages
> ...
I think your problem is that they're already registered for versioning, so
they'll not be ignored anymore.

Take a look at "fsvs unversion".


> Note that I never did an "fsvs init"... the README " How is it used?"
> section didn't contain one... I didn't see the "init" command until I
> read the "USAGE".
There's no init anymore. Where did you see it?


Regards,

Phil


--
Versioning your /etc, /home or even your whole installation?
             Try fsvs (fsvs.tigris.org)!

RE: Invalid or incomplete multibyte or wide character

Author "Worley, Chris B" <chris dot b dot worley at intel dot com>
Full name "Worley, Chris B" <chris dot b dot worley at intel dot com>
Date 2007-06-26 13:01:19 PDT
Message >-----Original Message-----
>From: Philipp Marek [mailto:philipp at marek dot priv dot at]
>Sent: Tuesday, June 26, 2007 1:48 PM
>To: users at fsvs dot tigris dot org
>Cc: Worley, Chris B
>Subject: Re: Invalid or incomplete multibyte or wide character
>
>Hello Chris!
>
>
>On Dienstag, 26. Juni 2007 Worley, Chris B wrote:
>> I had the "multibyte" error a few times in my commit. The
work-around
>> is to "fsvs ignore" those files (or delete them).
>Yes, but that's really only a work-around. That should Just Work(TM).
>Please see my other mail.

Yes, I'll fix the locale settings. Thanks for the fix.

>
>> It is strange that
>> even when "ignored" the file names show up during "commit".
>They shouldn't. Are you sure that you ignored them correctly?

I'm never sure I'm doing anything correctly ;)

I used both:

# fsvs ignore "./roots/*/var/log/*"
# fsvs ignore ./roots/2/var/log/*

...and my dump looks like:

# fsvs ignore dump
...
./roots/2/var/log/messages
...
./roots/*/var/log/*

...and my commit looks like:

# fsvs commit -m "test"
...
.mC. 13786 ./roots/2/var/log/messages
...

Note that I never did an "fsvs init"... the README " How is it used?"
section didn't contain one... I didn't see the "init" command until I
read the "USAGE".

Thanks,

Chris
P.S. thanks for the explanation of the diff issue. I just won't worry
about the error output.
P.P.S. Lines wrapped by Outlook. Oulook su***.
>
>> I also had an error in "fsvs diff" of binary files:
>>
>> [root@challenger challanger.x86_64]# fsvs diff
>> ./roots/2/var/lib/rpm/Packages
>> diff -u ./roots/2/var/lib/rp​m/Packages.r2
>> ./roots/2/var/lib/rp​m/Packages.local
>> Binary files ./roots/2/var/lib/rpm/Packages Rev. 2 (Fri
Jun
>> 8 09:54:56 2007) and ./roots/2/var/lib/rpm/Packages Local
>> version (Wed Jun 13 16:10:50 2007) differ
>I hope that the lines are just wrapped because of the mail, and not in
the
>console output?
>
>> An error occurred: No such file or directory (2)
>> in df__do_diff: Child 16391 gave an exit status 2
>Well, diff has documented that "... exit status of 0 means no
differences were
>found, 1 means some differences were found, and 2 means trouble."
>
>I didn't know that binary files mean trouble; I'd diffed some of them
>with "-a" very successfully.
>
>Hmmm ... What to do?
>I don't like to ignore the error code 2, as that could mean *real*
trouble,
>too.
>The other way would be to check the files ourself for binaryness, which
is not
>the Right Thing ... as more or less the same check is duplicated in
diff, and
>we'd possibly had to read the whole files.
>
>I wrote a mail to the GNU diff mailing list
>(http://lists.gnu.org​/archive/html/bug-gn​u-utils/2007-06/thre​ads.html),
let's
>take a look what they say.
>
>Perhaps the easiest way (for now!) is to write a small shell script
that
>simply contains something like
> #!/bin/sh
> diff "$@"
> exit 0
>and give its path via FSVS_DIFF to fsvs. That allows to ignore such
errors.
>
>
>I'll keep track of this issue; would you like to be notified if there's
any
>news or do you check yourself?
>
>
>Regards,
>
>Phil
>
>
>--
>Versioning your /etc, /home or even your whole installation?
> Try fsvs (fsvs.tigris.org)!

Re: Invalid or incomplete multibyte or wide character

Author pmarek
Full name P.Marek
Date 2007-06-26 12:47:47 PDT
Message Hello Chris!


On Dienstag, 26. Juni 2007 Worley, Chris B wrote:
> I had the "multibyte" error a few times in my commit. The work-around
> is to "fsvs ignore" those files (or delete them).
Yes, but that's really only a work-around. That should Just Work(TM).
Please see my other mail.

> It is strange that
> even when "ignored" the file names show up during "commit".
They shouldn't. Are you sure that you ignored them correctly?

> I also had an error in "fsvs diff" of binary files:
>
> [root@challenger challanger.x86_64]# fsvs diff
> ./roots/2/var/lib/rpm/Packages
> diff -u ./roots/2/var/lib/rp​m/Packages.r2
> ./roots/2/var/lib/rp​m/Packages.local
> Binary files ./roots/2/var/lib/rpm/Packages Rev. 2 (Fri Jun
> 8 09:54:56 2007) and ./roots/2/var/lib/rpm/Packages Local
> version (Wed Jun 13 16:10:50 2007) differ
I hope that the lines are just wrapped because of the mail, and not in the
console output?

> An error occurred: No such file or directory (2)
> in df__do_diff: Child 16391 gave an exit status 2
Well, diff has documented that "... exit status of 0 means no differences were
found, 1 means some differences were found, and 2 means trouble."

I didn't know that binary files mean trouble; I'd diffed some of them
with "-a" very successfully.

Hmmm ... What to do?
I don't like to ignore the error code 2, as that could mean *real* trouble,
too.
The other way would be to check the files ourself for binaryness, which is not
the Right Thing ... as more or less the same check is duplicated in diff, and
we'd possibly had to read the whole files.

I wrote a mail to the GNU diff mailing list
(http://lists.gnu.org​/archive/html/bug-gn​u-utils/2007-06/thre​ads.html), let's
take a look what they say.

Perhaps the easiest way (for now!) is to write a small shell script that
simply contains something like
    #!/bin/sh
    diff "$@"
    exit 0
and give its path via FSVS_DIFF to fsvs. That allows to ignore such errors.


I'll keep track of this issue; would you like to be notified if there's any
news or do you check yourself?


Regards,

Phil


--
Versioning your /etc, /home or even your whole installation?
             Try fsvs (fsvs.tigris.org)!

Re: Invalid or incomplete multibyte or wide character

Author pmarek
Full name P.Marek
Date 2007-06-26 12:44:35 PDT
Message Hello Chris!

On Dienstag, 26. Juni 2007 Worley, Chris B wrote:
> Thanks for responding.
>
> I only use subversion for fsvs, and only locally on this machine.
Ok.

> Here's the output from the commands:
>
> # fsvs ci -d -D hlp___do_convert
> An error occurred at 12:27:25.333: Success (0)
> in ci__work [commit.c:760]: mmap commit message
> (/tmp/commit-tmp.NCbeZ7, 0 bytes)
That's strange, I'll have a look.

> # locale
> LANG=
> LC_CTYPE="POSIX"
> LC_NUMERIC="POSIX"
> LC_TIME="POSIX"
> LC_COLLATE="POSIX"
> LC_MONETARY="POSIX"
> LC_MESSAGES="POSIX"
> LC_PAPER="POSIX"
> LC_NAME="POSIX"
> LC_ADDRESS="POSIX"
> LC_TELEPHONE="POSIX"
> LC_MEASUREMENT="POSIX"
> LC_IDENTIFICATION="POSIX"
> LC_ALL=
Well, that's all I really needed.

You have files with non-7bit-ASCII characters in the name
> An error occurred: Invalid or incomplete multibyte or wide character (84)
>  in hlp___do_convert: Conversion of string failed. Next bytes are
> \xE6\x73\x6D\x69
but there's no locale defined that would tell FSVS how to get that character
into UTF-8.

Could you try using a matching locale?


Regards,

Phil


--
Versioning your /etc, /home or even your whole installation?
             Try fsvs (fsvs.tigris.org)!

RE: Re: Invalid or incomplete multibyte or wide character

Author "Worley, Chris B" <chris dot b dot worley at intel dot com>
Full name "Worley, Chris B" <chris dot b dot worley at intel dot com>
Date 2007-06-26 12:38:25 PDT
Message Thanks for responding.

I only use subversion for fsvs, and only locally on this machine.

Here's the output from the commands:

# fsvs ci -d -D hlp___do_convert
An error occurred at 12:27:25.333: Success (0)
  in ci__work [commit.c:760]: mmap commit message
(/tmp/commit-tmp.NCbeZ7, 0 bytes)
# locale
LANG=
LC_CTYPE="POSIX"
LC_NUMERIC="POSIX"
LC_TIME="POSIX"
LC_COLLATE="POSIX"
LC_MONETARY="POSIX"
LC_MESSAGES="POSIX"
LC_PAPER="POSIX"
LC_NAME="POSIX"
LC_ADDRESS="POSIX"
LC_TELEPHONE="POSIX"
LC_MEASUREMENT="POSIX"
LC_IDENTIFICATION="POSIX"
LC_ALL=

Without the "-D hlp___do_convert", there is no abort... but the output
is long (7.5M lines).

Here's the tail of the long output:

12:33:35.739 waa__output_tree[waa.c:1120] new subdir 8519848 #1
12:33:35.739 waa__output_tree[waa.c:1033] finished subdir
12:33:35.739 waa___find_position[waa.c:822] at 1=9209101 - 3=9209104 -
6=9209107
12:33:35.739 waa___find_position[waa.c:822] at 1=9209101 - 2=9209102 -
3=9209104
12:33:35.739 waa___find_position[waa.c:822] at 1=9209101 - 1=9209101 -
2=9209102
12:33:35.739 waa___find_position[waa.c:839] believing in 1 9209101
12:33:35.739 waa__output_tree[waa.c:1120] new subdir 8536329 #1
12:33:35.739 waa__output_tree[waa.c:1033] finished subdir
12:33:35.739 waa__output_tree[waa.c:1033] finished subdir
12:33:35.739 waa__output_tree[waa.c:1033] finished subdir
12:33:35.739 waa___find_position[waa.c:783] short path taken for 0<1
12:33:35.739 waa__output_tree[waa.c:1120] new subdir 17648 #0
12:33:35.739 waa__output_tree[waa.c:1033] finished subdir
12:33:35.739 waa___find_position[waa.c:783] short path taken for 0<1
12:33:35.739 waa__output_tree[waa.c:1120] new subdir 7392144 #0
12:33:35.739 waa___find_position[waa.c:783] short path taken for 0<1
12:33:35.739 waa___find_position[waa.c:783] short path taken for 0<1
12:33:35.739 waa___find_position[waa.c:783] short path taken for 0<1
12:33:35.739 waa___find_position[waa.c:783] short path taken for 0<1
12:33:35.739 waa___find_position[waa.c:783] short path taken for 0<1
12:33:35.739 waa__output_tree[waa.c:1033] finished subdir
12:33:35.739 waa__output_tree[waa.c:1033] finished subdir
12:33:35.739 waa___find_position[waa.c:783] short path taken for 0<1
12:33:35.739 waa__output_tree[waa.c:1120] new subdir 9209103 #0
12:33:35.739 waa__output_tree[waa.c:1033] finished subdir
12:33:35.739 waa___find_position[waa.c:783] short path taken for 0<1
12:33:35.739 waa__output_tree[waa.c:1120] new subdir 5996810 #0
12:33:35.739 waa___find_position[waa.c:783] short path taken for 0<1
12:33:35.739 waa___find_position[waa.c:783] short path taken for 0<1
12:33:35.739 waa___find_position[waa.c:783] short path taken for 0<1
12:33:35.740 waa___find_position[waa.c:783] short path taken for 0<1
12:33:35.740 waa___find_position[waa.c:783] short path taken for 0<1
12:33:35.740 waa___find_position[waa.c:783] short path taken for 0<1
12:33:35.740 waa___find_position[waa.c:783] short path taken for 0<1
12:33:35.740 waa___find_position[waa.c:783] short path taken for 0<1
12:33:35.740 waa___find_position[waa.c:783] short path taken for 0<1
12:33:35.740 waa___find_position[waa.c:783] short path taken for 0<1
12:33:35.740 waa___find_position[waa.c:783] short path taken for 0<1
12:33:35.740 waa__output_tree[waa.c:1033] finished subdir
12:33:35.740 waa__output_tree[waa.c:1033] finished subdir
12:33:35.740 waa___find_position[waa.c:783] short path taken for 0<1
12:33:35.740 waa__output_tree[waa.c:1120] new subdir 6013238 #0
12:33:35.740 waa___find_position[waa.c:783] short path taken for 0<1
12:33:35.740 waa___find_position[waa.c:783] short path taken for 0<1
12:33:35.740 waa___find_position[waa.c:783] short path taken for 0<1
12:33:35.740 waa__output_tree[waa.c:1033] finished subdir
12:33:35.740 waa__output_tree[waa.c:1033] finished subdir
12:33:35.740 waa___find_position[waa.c:783] short path taken for 0<1
12:33:35.740 waa__output_tree[waa.c:1120] new subdir 6029607 #0
12:33:35.740 waa___find_position[waa.c:783] short path taken for 0<1
12:33:35.740 waa___find_position[waa.c:783] short path taken for 0<1
12:33:35.740 waa___find_position[waa.c:783] short path taken for 0<1
12:33:35.740 waa___find_position[waa.c:783] short path taken for 0<1
12:33:35.740 waa___find_position[waa.c:783] short path taken for 0<1
12:33:35.740 waa___find_position[waa.c:783] short path taken for 0<1
12:33:35.740 waa___find_position[waa.c:783] short path taken for 0<1
12:33:35.740 waa___find_position[waa.c:783] short path taken for 0<1
12:33:35.740 waa___find_position[waa.c:783] short path taken for 0<1
12:33:35.740 waa___find_position[waa.c:783] short path taken for 0<1
12:33:35.740 waa___find_position[waa.c:783] short path taken for 0<1
12:33:35.740 waa___find_position[waa.c:783] short path taken for 0<1
12:33:35.740 waa___find_position[waa.c:783] short path taken for 0<1
12:33:35.740 waa___find_position[waa.c:783] short path taken for 0<1
12:33:35.740 waa___find_position[waa.c:783] short path taken for 0<1
12:33:35.740 waa___find_position[waa.c:783] short path taken for 0<1
12:33:35.740 waa___find_position[waa.c:783] short path taken for 0<1
12:33:35.740 waa___find_position[waa.c:783] short path taken for 0<1
12:33:35.740 waa___find_position[waa.c:783] short path taken for 0<1
12:33:35.740 waa___find_position[waa.c:783] short path taken for 0<1
12:33:35.740 waa___find_position[waa.c:783] short path taken for 0<1
12:33:35.740 waa___find_position[waa.c:783] short path taken for 0<1
12:33:35.740 waa___find_position[waa.c:783] short path taken for 0<1
12:33:35.740 waa__output_tree[waa.c:1033] finished subdir
12:33:35.740 waa__output_tree[waa.c:1033] finished subdir
12:33:35.740 waa__output_tree[waa.c:1033] finished subdir
12:33:35.740 waa__close[waa.c:504] filehandle 3 should be
/var/spool/fsvs/26/6​b/91bfd604466862fab7​939acdc330/dir
12:33:35.742 url___set_internal_n​ums[url.c:573] inum for
file:///tmp/images is 1
12:33:35.742 waa__get_waa_directo​ry[waa.c:278] path is
/hpc/tftpboot/images​/challanger.x86_64
12:33:35.742 waa__get_waa_directo​ry[waa.c:323] md5 of
/hpc/tftpboot/images​/challanger.x86_64
12:33:35.742 waa__get_waa_directo​ry[waa.c:369] returning
/etc/fsvs/266b91bfd6​04466862fab7939acdc3​30/
12:33:35.742 waa__open[waa.c:429] tmp for target
/etc/fsvs/266b91bfd6​04466862fab7939acdc3​30/Urls is
/etc/fsvs/266b91bfd6​04466862fab7939acdc3​30/Urls.tmp
12:33:35.742 waa__open[waa.c:445] got fh 3
12:33:35.742 url__output_list[url.c:801] writing line 1 4
T:-1,N:,P:0,file:///tmp/images
12:33:35.742 waa__close[waa.c:504] filehandle 3 should be
/etc/fsvs/266b91bfd6​04466862fab7939acdc3​30/Urls
12:33:35.742 wa__summary[warnings.c:229] 0# meta-mtime: 0x
12:33:35.742 wa__summary[warnings.c:229] 1# meta-user: 0x
12:33:35.742 wa__summary[warnings.c:229] 2# meta-group: 0x
12:33:35.742 wa__summary[warnings.c:229] 3# meta-umask: 0x
12:33:35.742 wa__summary[warnings.c:229] 4# entry-not-found: 0x
12:33:35.742 wa__summary[warnings.c:229] 5# fsentry-not-found: 0x
12:33:35.742 wa__summary[warnings.c:229] 6# no-urllist: 0x
12:33:35.742 wa__summary[warnings.c:229] 7# charset-invalid: 0x
12:33:35.742 wa__summary[warnings.c:229] 8# chmod-eperm: 0x
12:33:35.742 wa__summary[warnings.c:229] 9# chmod-other: 0x
12:33:35.742 wa__summary[warnings.c:229] 10# chown-eperm: 0x
12:33:35.742 wa__summary[warnings.c:229] 11# chown-other: 0x
12:33:35.742 wa__summary[warnings.c:229] 12# overlayed-entries: 0x
12:33:35.742 wa__summary[warnings.c:229] 13# propname-reserved: 0x
12:33:35.742 wa__summary[warnings.c:229] 14# _test-warning: 1x
12:33:35.742 main[fsvs.c:869] memory stats: 0x558000 to 0x136a000, 14408
KB


Note that I deleted all the offending files, so the error is gone.

>-----Original Message-----
>From: Philipp Marek [mailto:philipp at marek dot priv dot at]
>Sent: Tuesday, June 26, 2007 1:23 PM
>To: users at fsvs dot tigris dot org
>Cc: Worley, Chris B
>Subject: Re: Invalid or incomplete multibyte or wide character
>
>Hello Chris!
>
>On Montag, 25. Juni 2007 Worley, Chris B wrote:
>> Doing a commit, I get this error, any idea why or how to work-around?
>> ...
>> N... dir ./roots/3/usr/share/​doc/man-pages-da-0.1​.1
>> N... 241 ./roots/3/usr/share/​doc/man-pages-da-0.1​.1/AUTHORS
>> N... 753 ./roots/3/usr/share/​doc/man-pages-da-0.1​.1/ChangeLog
>>
>>
>> An error occurred: Invalid or incomplete multibyte or wide character
>> (84)
>> in hlp___do_convert: Conversion of string failed. Next bytes are
>> \xE6\x73\x6D\x69
>This error means that FSVS tried to send invalid some byte sequence to
the
>locale -> UTF8 converter.
>
>Could you please post the output of
> locale
>and (if it's not too much information that's leaked)
> fsvs ci -d
>If that includes too much information, please try
> fsvs ci -d -D hlp___do_convert
>maybe that's not so critical.
>
>Do you use subversion on this machine? Does it have problems with these
files
>too?
>
>
>Regards,
>
>Phil
>
>
>--
>Versioning your /etc, /home or even your whole installation?
> Try fsvs (fsvs.tigris.org)!
>
>----------------​--------------------​--------------------​-------------
>To unsubscribe, e-mail: users-unsubscribe@fs​vs.tigris.org
>For additional commands, e-mail: users-help at fsvs dot tigris dot org

Re: Invalid or incomplete multibyte or wide character

Author pmarek
Full name P.Marek
Date 2007-06-26 12:22:44 PDT
Message Hello Chris!

On Montag, 25. Juni 2007 Worley, Chris B wrote:
> Doing a commit, I get this error, any idea why or how to work-around?
> ...
> N... dir ./roots/3/usr/share/​doc/man-pages-da-0.1​.1
> N... 241 ./roots/3/usr/share/​doc/man-pages-da-0.1​.1/AUTHORS
> N... 753 ./roots/3/usr/share/​doc/man-pages-da-0.1​.1/ChangeLog
>
>
> An error occurred: Invalid or incomplete multibyte or wide character
> (84)
> in hlp___do_convert: Conversion of string failed. Next bytes are
> \xE6\x73\x6D\x69
This error means that FSVS tried to send invalid some byte sequence to the
locale -> UTF8 converter.

Could you please post the output of
    locale
and (if it's not too much information that's leaked)
    fsvs ci -d
If that includes too much information, please try
    fsvs ci -d -D hlp___do_convert
maybe that's not so critical.

Do you use subversion on this machine? Does it have problems with these files
too?


Regards,

Phil


--
Versioning your /etc, /home or even your whole installation?
             Try fsvs (fsvs.tigris.org)!

RE: Invalid or incomplete multibyte or wide character

Author "Worley, Chris B" <chris dot b dot worley at intel dot com>
Full name "Worley, Chris B" <chris dot b dot worley at intel dot com>
Date 2007-06-26 10:17:49 PDT
Message I had the "multibyte" error a few times in my commit. The work-around
is to "fsvs ignore" those files (or delete them). It is strange that
even when "ignored" the file names show up during "commit".

I also had an error in "fsvs diff" of binary files:

[root@challenger challanger.x86_64]# fsvs diff
./roots/2/var/lib/rpm/Packages
diff -u ./roots/2/var/lib/rp​m/Packages.r2
./roots/2/var/lib/rp​m/Packages.local
Binary files ./roots/2/var/lib/rpm/Packages Rev. 2 (Fri Jun
8 09:54:56 2007) and ./roots/2/var/lib/rpm/Packages Local
version (Wed Jun 13 16:10:50 2007) differ


An error occurred: No such file or directory (2)
  in df__do_diff: Child 16391 gave an exit status 2

>-----Original Message-----
>From: Worley, Chris B [mailto:chris.b.worl​ey at intel dot com]
>Sent: Monday, June 25, 2007 2:30 PM
>To: users at fsvs dot tigris dot org
>Subject: Invalid or incomplete multibyte or wide character
>
>Doing a commit, I get this error, any idea why or how to work-around?
>
>
>...
>N... dir ./roots/3/usr/share/​doc/man-pages-da-0.1​.1
>N... 241 ./roots/3/usr/share/​doc/man-pages-da-0.1​.1/AUTHORS
>N... 753 ./roots/3/usr/share/​doc/man-pages-da-0.1​.1/ChangeLog
>
>
>An error occurred: Invalid or incomplete multibyte or wide character
>(84)
> in hlp___do_convert: Conversion of string failed. Next bytes are
>\xE6\x73\x6D\x69
>
>----------------​--------------------​--------------------​-------------
>To unsubscribe, e-mail: users-unsubscribe@fs​vs.tigris.org
>For additional commands, e-mail: users-help at fsvs dot tigris dot org

Invalid or incomplete multibyte or wide character

Author "Worley, Chris B" <chris dot b dot worley at intel dot com>
Full name "Worley, Chris B" <chris dot b dot worley at intel dot com>
Date 2007-06-25 13:30:16 PDT
Message Doing a commit, I get this error, any idea why or how to work-around?


...
N... dir ./roots/3/usr/share/​doc/man-pages-da-0.1​.1
N... 241 ./roots/3/usr/share/​doc/man-pages-da-0.1​.1/AUTHORS
N... 753 ./roots/3/usr/share/​doc/man-pages-da-0.1​.1/ChangeLog


An error occurred: Invalid or incomplete multibyte or wide character
(84)
  in hlp___do_convert: Conversion of string failed. Next bytes are
\xE6\x73\x6D\x69
Messages per page: