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

List:       koffice-devel
Subject:    Re: META in KOffice document
From:       Ariya Hidayat <ariya () tf ! itb ! ac ! id>
Date:       2002-05-17 20:43:45
[Download RAW message or body]


> - editing info and not deleting the meta data is quite hard to do, you have
> to have extra data structures that match the data structure of the file.
> Thereby keeping a double householding of data.

Hmm, I don't really get into this. Say I have a dBASE file with 2 fields:
name (as text) and age (as number). All I want to do is preserve this info
(there are 2 fields, one is name, the other is age, and so on) in KSpread.
Later, when I do some deleting/inserting/editing and then save the file,
maybe to the same DBF file, the export filter can compose a DBF file whose
structure is somehow described in that metainfo. So practically we have to
keep the structure anyway.

> - its impossible to keep the 2 data types in sync; when you edit one the
> other will be invalid.  Plus when you store invisible data it is
> non-transparent _why_ one conversion would succeed and another does not
> succeed. (app says: "Your meta data is missing"; user: "But its the exact
> same document as the other one that did succeed!")

The dBASE meta info is used to construct database fields definition. If no
meta info is found, then guess the structure (as Nicolas suggested, in
other reply of this thread) or ask the user. MS Excel uses the former
approach. It simply creates a structure which fits for the data.

Note that the goal is to make KSpread able to deal with simple database
file, only for editing records and such. IMHO it's no good to _create_ a
database in a spreadsheet application like KSpread or even Excel.

Regards,

Ariya

_______________________________________________
koffice-devel mailing list
koffice-devel@mail.kde.org
http://mail.kde.org/mailman/listinfo/koffice-devel
[prev in list] [next in list] [prev in thread] [next in thread]