This is an enhancement request.I have just started using Origin 5.0.
I am using it to analyze, chiefly to do graphs,
on already existing data sets.
The data in these existing data sets already has names for things
such as columns. Unfortunately, these names do not meet Origin's
naming conventions.
The names are things like "ctr16.sim_num_insn/ctr16.br_mispred".
(Note the double quotes. More on that below.)
Let's list the Origin column name rules these break:
a) names are more than 11 characters
b) names contain special characters such as _, ., /
etc.
Going in and fixing these names is really, really, annoying.
It makes the data extremely hard to read.
You might ask 'where in the world did such funky names come from?'
You might be saying 'This guy must be computer illiterate. No programming
language allows names like that.'
Well, SQL, the database query language, does. SQL allows ANY string
of characters at all to be used as a name, so long as it is placed within
double quotes - a so-called delimited identifier.
And guess what: my data comes from an SQL database.
I can't tell you how LIBERATING it is to be able to use completely arbitrary
names. For example, you can simply generate names on the fly. No need
to create funky acronyms. It allows you to mix names from any language
- English, Japanese, COBOL (paragraph-name), C (procedure_Name),
LISP (defun-name).
Arbitrary names are a huge increase in user convenience. I suggest that
you consider implementing the delimited identifier in your symbol table
package.
It's really easy - I added it to my data manipulation tools as soon as I
learned
of it - and it helps.
And, lest I have not said so yet: it is really, REALLY, annoying to have to go
back to typing 11 character alphanumeric names with no punctuation.
Imagine, if you will, the pain of having to go back to MS DOS's 8.3 character
filenames, once you have used UNIX or Windows 95 or Windows NT long file names.
Now make it even worse...