William W. Austin schreef:
On 2006-04-05 01:10:43, Jan Bassez wrote:
Daniel Kasak schreef:
Jan Bassez wrote:
Is there a way to sort numeric values in a normal text view without
converting them the numbers with preceding zero's?
(SNIP)
I would split the numeric and strings parts. ie have a numeric
field, and then an extra field to store your 'modifier' or whatever.
Then sort by numeric_field, string_field
That's an option but I see that more as a workaround.
I also have my classical records in a Base catalog. Many Classical
pieces have Opus number, so I have a field for that. It contains this
content:
op. 14
op. 45
op. 125
and it has the same sorting problem.
One could say leave the 'op.'
but pieces of Bach have BWV instead of op., also you have WoO
(Without opus) and KV & P for other composers.
Splitting them up is an idea I dont like too much....
Isn't it possible for some kind of combinable field String/numeric
field.
or a sorting option wich looks for numeric values in a string field,
and threats them as such?
..in future versions?
Well well, what a solution,
ehm...
I get the point of it, but wouldn't know how to start this.
I'm in windows, so I need 'The' free unix tools....
uhm...where, or what should I look for?
then further (see inside text)->
While I don't like it either I'm somewhat in the same boat - my wife
and I both used to be classical musicians (sang opera in the US and in
Europe) and have about 3000 LP's, 2500 tapes (r-to-r and cassette, no
dat's yet), 1000 or so CD's, films, video's, DVD's, etc. ad nauseum.
I put it all into a database about 15 years ago and had to face the
same problem. I started out in informix, and later I moved to MySQL -
I will probably not use base since the effort would not justify the
results.
Essentially I did what you said you didn't want to do - I split the
"opus-designator" and the "opus-number" components into separate
fields and stored them that way. And of course it didn't startout
that way.
If you are on Linux then the conversion is moderately easy - if you
are on windows then you can download the free unix tools to do the
same thing. I will be presumptuous and show the procedure below:
The assumption here is that your current "opus" field is named
something like "OPUS".
1) write an ascii report which fakes an "unload" procedure,
delimiting the fields with "|" characters so that each record give you
something that looks like:
Ok, what's that ascii report...
Is there a way to export the database to a simple ascii text file? Or am
I wrong
How I do that?
1st 2nd
<- field group #1 --------------> time time <- field group #4 ------>
<field 1>|<field 2>|<field 3>|...|OPUS|OPUS|<field N>|<field N+!> ...
I suppose this is how it will look in ascii...
Then use the utility "CUT" to separate out the two occurrences of the
OPUS field.
CUT is probably one of the unix tools I should look for?
With it's manual being 'man cut' (in windows too)
(I would have installed Debian linux if it detected my hard drives, but
there is some new version of the kernel needed for some reason, it
doesn't seem to detect my SATA HD-controller, have ratehr new PC)
Use sed to trim everything to the right of the space in the 1st OPUS
field.
Use sed again to trim everything to the LEFT of the space in the 2nd
OPUS field.
Then put them all back together using paste to create the new lines
which look like:
BWV real
<- field group #1 --------------> op. opus <- field group #4 ---->
etc. num.
<field 1>|<field 2>|<field 3>|...|OPUS|OPUS|<field N>|<field N+!> ...
SED is one of the other tools?
Then create a new database with the original field "OPUS" removed and
2 new fields added, say "OpusDes" (designator) and "OpusNum" (number
itself). Load the "|"-delimited file into the database and you're set.
the "|"-delimited file??
When you display the contents of the database you can set up things to
concatenate the two opus fields - OpusDes + " " + OpusNum - and you're
set.
I can import that ascii file back into a new database??
While it would be nice to have a function such as you describe, the
implementation on the coding side is more than a tad complex ... and
most likely very inefficient in the long run.
I think of it because windows explorer does it allright (with my
directories)
it sorts like this
...
Bach, Johann Sebastian - BWV 70 - Wachet! betet! betet! wachet!
Bach, Johann Sebastian - BWV 565 - Toccata und Fugue d-moll
Bach, Johann Sebastian - BWV 816 - French Suite in G
Bach, Johann Sebastian - BWV 1046-1051 - Brandenburgische Concerten
...
If I remember correctly, when I did this (years ago) I had only about
3000 entries in the database, and it took me about 90 minutes to do
the entire conversion. (As you may have guessed, I'm also a computer
geek now, since I had to stop singing as my day job due to health
reasons.)
Well, I like your solution and want to try, but don't get the whole
thing right now...
Thanks!
Good luck - and I hope my suggestion helps,
- Bill
--william w. austin [EMAIL PROTECTED]
"life is just another phase i'm going through. this time, anyway ..."
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
