9:14 p.m.
Dear all, For example I can't run
open http://www.principlesofeconometrics.com/poe4/data/dat/golf.dat (this file as .gdt is
supplied with Gretl as a data file for Hill Griffiths Lim) But if download it and rename
as golf.dat, gretl open it ok. May be, something like this open filename --as-txt will
be useful? It's only suggestion because I can't do costs-benefits analysis: costs
are C code, and at the benefits side I do not know how many users will need it.
Oleh
12:50 a.m.
On Mon, 9 Nov 2015, oleg_komashko(a)ukr.net wrote:
Dear all, For example I can't run
open http://www.principlesofeconometrics.com/poe4/data/dat/golf.dat
(this file as .gdt is supplied with Gretl as a data file for Hill
Griffiths Lim) But if download it and rename as golf.dat,
Do you mean, rename it as golf.txt?
gretl open it ok. May be, something like this open filename
--as-txt will be useful? It's only suggestion because I can't do
costs-benefits analysis: costs are C code, and at the benefits
side I do not know how many users will need it.
Benefits are small, I think. So far as gretl is concerned, ".dat" is
the suffix for JMulti data files. Renaming as .txt is not too
onerous.
Allin
12:32 p.m.
Do you mean, rename it as golf.txt? exactly > So far as gretl is
concerned, ".dat" is
the suffix for JMulti data files. open
http://www.jmulti.de/download/datasets/e1.dat
works ok In http://www.principlesofeconometrics.com/poe4/data/dat/golf.dat .dat is an
idiosyncratic extension for a text data file
If I understand it correctly, gretl tries to open golf.dat as JMulti data file and fails,
since it is a text file. May be, it will be useful to have some text on renaming
non-standard extensions before opening? Oleh
1:32 p.m.
On Tue, 10 Nov 2015, oleg_komashko(a)ukr.net wrote:
> Do you mean, rename it as golf.txt? exactly > So far as gretl
is concerned, ".dat" is
the suffix for JMulti data files. open http://www.jmulti.de/download/datasets/e1.dat
works ok In http://www.principlesofeconometrics.com/poe4/data/dat/golf.dat .dat is an
idiosyncratic extension for a text data file
If I understand it correctly, gretl tries to open golf.dat as JMulti
data file and fails, since it is a text file. May be, it will be useful
to have some text on renaming non-standard extensions before opening?
I mean, really.
It's ok to make users' lives easy, but up to a point. If someone, for some
reason, rename a Stata file to an XLS extension, do you suppose we should
try and correct for that?
-------------------------------------------------------
Riccardo (Jack) Lucchetti
Dipartimento di Scienze Economiche e Sociali (DiSES)
Università Politecnica delle Marche
(formerly known as Università di Ancona)
r.lucchetti(a)univpm.it
http://www2.econ.univpm.it/servizi/hpp/lucchetti
-------------------------------------------------------
1:47 p.m.
I tested another non-standard extension .raw (used in data sets for Wooldridge) Gretl
opens it OK. So, may be, .dat is a special case because of confusing with JMjmulti files.
So, I am agree, it's to particular to do something
Oleh
10 листопада 2015, 15:33:03, від "Riccardo (Jack) Lucchetti" <
r.lucchetti(a)univpm.it >:
On Tue, 10 Nov 2015, oleg_komashko(a)ukr.net wrote:
> Do you mean, rename it as golf.txt? exactly > So far as gretl
is concerned, ".dat" is
the suffix for JMulti data files. open http://www.jmulti.de/download/datasets/e1.dat
works ok In http://www.principlesofeconometrics.com/poe4/data/dat/golf.dat .dat is an
idiosyncratic extension for a text data file
If I understand it correctly, gretl tries to open golf.dat as JMulti
data file and fails, since it is a text file. May be, it will be useful
to have some text on renaming non-standard extensions before opening?
I mean, really.
It's ok to make users' lives easy, but up to a point. If someone, for some
reason, rename a Stata file to an XLS extension, do you suppose we should
try and correct for that?
-------------------------------------------------------
Riccardo (Jack) Lucchetti
Dipartimento di Scienze Economiche e Sociali (DiSES)
Università Politecnica delle Marche
(formerly known as Università di Ancona)
r.lucchetti(a)univpm.it
http://www2.econ.univpm.it/servizi/hpp/lucchetti
-------------------------------------------------------
_______________________________________________
Gretl-devel mailing list
Gretl-devel(a)lists.wfu.edu
http://lists.wfu.edu/mailman/listinfo/gretl-devel
2:18 p.m.
New subject: open ASCI files: arbitrary extensions
Gretl eats arbitrary extensions for text data files! I've just opened
"401K.booo" My first mistake was as follows: I didn't think an official book
site would'n check known econometric software for data files extensions. I'd say
they are biased against GNU software.
Oleh
-------------------------------------------------------
_______________________________________________
Gretl-devel mailing list
Gretl-devel(a)lists.wfu.edu
http://lists.wfu.edu/mailman/listinfo/gretl-devel
4:06 p.m.
New subject: open ASCI files: arbitrary extensions
On Tue, 10 Nov 2015, oleg_komashko(a)ukr.net wrote:
Gretl eats arbitrary extensions for text data files! I've just
opened "401K.booo" My first mistake was as follows: I didn't think
an official book site would'n check known econometric software for
data files extensions. I'd say they are biased against GNU software.
Acually I wouldn't say anyone is to blame here. The ".dat" extension
has been used by many programs over the years to indicate a data file
of some type, usually a plain text format but sometimes binary. It's
no-one's "property". It just happens that JMulTi uses it for a quite
specific text format, and some years back we wrote an importer for
that format and assigned the ".dat" extension to it.
If there's any blame to assign, it's for using such a non-specific
suffix, presumably on the presumption that nobody will ever want to
open your data other than in your own program. However, gretl is
arguably quilty of the same sort of thing with our generic ".inp"
suffix for script files. It's a legacy thing, from before people
became savvy about mime types and all that.
Allin
3643
days inactive
3644
days old
6 comments
3 participants
participants (3)
-
Allin Cottrell -
oleg_komashko＠ukr.net -
Riccardo (Jack) Lucchetti