[libreoffice-users] Re: Does LO Writer converts (internally) any format other than .odt to .odt when it wants to open such files?

On 08/07/2013 01:22 AM, Nino Novak wrote:
> Am 06.08.2013 21:55, schrieb Sina Momken:
>
> > ... does LO Writer
> > converts files to .odt when it tries to open them or it can directly
> > work with formats other than .odt like .doc and .docx?
>
> I think no software "works" directly with a file format. Files are
> "frozen data streams", they are static. Any data bits from the file get
> "imported" into the software which stores them into its internal
> variables, arrays, objects, and more, then it works with them and after
> that stores the result again in the file.
>
> So there is always some kind of import "translation" of data when a
> software "opens" (i.e. reads) the file (resp. export translation when
> writing it).
>
> However, in the "native" format (as is odf for LibreOffice), this
> translation mostly goes 1:1 as the file format is chosen to support the
> software's needs best (or vice versa).
>
> In contrast, for other (i.e. non-native) formats, the software first has
> to convert the "foreign" data format into what it understands. This is
> done by import resp. export filters, which map the software's needs to
> file format givens and thus allows the software to "understand" foreign
> formats.
I exactly meant what you said above in my question. Of course a file is
not understandable by a software and it imports its info into its data
structures. A software has a 1:1 relationship between its internal set
of data structures and its native file format. But is it necessary that
an alien file format be converted (filtered) to the current set of data
structures in the software? Maybe or maybe not. If the software has only
one set of data structures and only one system to process them then the
external format must be converted. But if the software has 2 different
systems for storing data in memory (RAM) and analyzing them then there
can be no need for conversion, because the second system has a 1:1
relation with stored data in the file.
I wanted to know which of these 2 possibilities is true for LO and found
out that LO has only one set of data structures which goes 1:1 with .odt
format but not with .doc format.

> Does this answer your question? Or did you mean something completely
> diffferent? Why did you ask this question?
Yeah you realized my question right. I asked this question because I
wanted to know whether the process of
doc -> odt -> odt' -> doc'
is running under LO. Because in this case it's faster, more efficient
and much more stable to work directly on .odt files, which reduces the
redundant processes of conversion (filtering).

1.odt -> edit -> 2.odt
is faster and much more stable than
1.doc -> internal DS of 1.odt -> edit -> internal DS of 2.odt -> 2.doc

Regards,
   Sina Momken

> Nino



-- 
To unsubscribe e-mail to: users+unsubscribe@global.libreoffice.org
Problems? http://www.libreoffice.org/get-help/mailing-lists/how-to-unsubscribe/
Posting guidelines + more: http://wiki.documentfoundation.org/Netiquette
List archive: http://listarchives.libreoffice.org/global/users/
All messages sent to this list will be publicly archived and cannot be deleted