[prev in list] [next in list] [prev in thread] [next in thread] 

List:       netatalk-devel
Subject:    Re: [Netatalk-devel] Office 2008 Word Type Codes
From:       Scott Sipe <css () cap-press ! com>
Date:       2008-06-20 0:28:39
Message-ID: BBE10A2B-FD59-4489-B7A5-22BAEAC2EF6D () cap-press ! com
[Download RAW message or body]

On Jun 19, 2008, at 6:12 PM, Thomas Kaiser wrote:

> Scott Sipe wrote:
>
>> You're saying that once Word 2008 (or whatever program) opens the  
>> file via
>> netatalk, it rewrites the type code in the metadata, even if the file
>> contents were not changed?
>
> No. It's completely independent from anything a Mac program does or  
> not.
>
> A file might have mac metadata like file type and creator code  
> attached to
> it (when saved via Netatalk/AFP) or not (when stored by other  
> means). This
> sort of metadata will be stored AppleDouble encoded in an afpd- 
> internal file
> (normally residing in the .AppleDouble subfolder with identical name  
> --
> unfortunately there exists an ugly hack called "adouble:osx" which  
> stores
> the AD file in a similar way like MacOS X' VFS layer when accessing  
> flat
> filesystems like SMB or NFS --> ._$filename)
>
> The first time an apfd process accesses a file without AppleDouble  
> metadata
> a lookup in AppleVolumes.system happens and in case there exists a  
> mapping
> for the file's suffix and a specific type/creator combination the  
> latter
> will be automagically assigned to the file in question (an AppleDouble
> encoded file inside the .AppleDouble subfolder will be written  
> containing
> this sort of information).
>
> When this sort of metadata is present (read as: the appropriate AD  
> encoded
> file exists) any sort of mapping table/algorithm is ignored (so the
> aforementioned process is unique unless one removes the file's type/ 
> creator
> allocation again). The same applies to files directly stored via AFP  
> since
> type/creator information is already handed over from the client to the
> server.

Ok, that was my understanding--perhaps I didn't explain properly in my  
original post. My question to which you said "no" was whether  
disconnecting from a netatalk share and reconnecting (starting a new  
instance of afpd right?) was all it took for a change in the  
AppleVolumes.system to take effect. Sorry if I was unclear or didn't  
present enough information.

As a side question, are there are any serverside tools written that  
parse AppleDouble files and spit out the contents in human readable  
form (lists type/number of forks etc)? (for instance to look up type/ 
creator code) Might be nice to have (I might fiddle around a little if  
nothing exists) and slightly easier than just opening the binary file  
to look for type/creator codes.


>> FWIW, we've been running Samba and Netatalk side by side for--I don't
>> know about 10+ years--with minimal problems.


> From my experience this is the usual gap between user's and  
> administrator's
> experience :-(

Well, our office is < 20 users and I'm one of them, so hopefully I  
would be the first to run into any problems. Actually that's exactly  
what happened with this Office 2008 SP1 problem,I was the first to  
upgrade and get hit by microsoft's regression. But anyway..

Also thanks for the link to the wiki site--I didn't know it was  
editable by the public. I will add an entry tomorrow.

Thanks,
Scott

-------------------------------------------------------------------------
Check out the new SourceForge.net Marketplace.
It's the best place to buy or sell services for
just about anything Open Source.
http://sourceforge.net/services/buy/index.php
_______________________________________________
Netatalk-devel mailing list
Netatalk-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/netatalk-devel
[prev in list] [next in list] [prev in thread] [next in thread]