octave-bug-tracker
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[Octave-bug-tracker] [bug #51512] [octave forge] (io) Missing or wrong t


From: Markus Mützel
Subject: [Octave-bug-tracker] [bug #51512] [octave forge] (io) Missing or wrong types when using xlsread with OCT interface
Date: Sat, 7 Oct 2017 09:21:24 -0400 (EDT)
User-agent: Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:57.0) Gecko/20100101 Firefox/57.0

Follow-up Comment #23, bug #51512 (project octave):

Philip,

attached are example files for formulas spanning several cell. The first one
has formulas spanning several rows in a column. The second one has formulas
spanning several columns in one row. For the third file, I let a formula span
several rows in one column and then expanded the area to the right (over
multiple columns). The actual formulas seem to be given in the "header cell"
of the span with a reference to the scope to which they apply.
To get the formula of each respective cell, one would have to take the
relative position of the cell to the "header cell" into account.

Wrt which version would be best, I think it comes down to personal taste. I
didn't notice much difference (i.e. none at all) when it comes to
performance.
Thinking of the additional logic for the formula output, it might be easier to
write and maintain if you arranged distinct cases in reasonable groups.
Formulas could then be read in dedicated regexps depending on whether the
value or the formula itself should be returned. That might also save some time
that was unnecessarily spent reading the values if a user requested the
formulas.
But it's probably best for you to decide as the maintainer.

After having written the above, I found this benchmark [1] from which I don't
know whether it is still valid. However, the following suggests that it might
be better performance-wise to use loads of dedicated regexps:
"Several BT-based algorithms such as onig and PCRE are 10X slower than using
two regex separately, although this is not true for all BT-based algorithms."
Since we are using PCRE, did you ever try to get rid of all "|" in the regular
expressions?

[1] http://lh3lh3.users.sourceforge.net/reb.shtml

(file #42076, file #42077, file #42078)
    _______________________________________________________

Additional Item Attachment:

File name: Excel2016_row_span.xlsx        Size:9 KB
File name: Excel2016_column_span.xlsx     Size:8 KB
File name: Excel2016_rows_then_columns.xlsx Size:9 KB


    _______________________________________________________

Reply to this item at:

  <http://savannah.gnu.org/bugs/?51512>

_______________________________________________
  Message sent via/by Savannah
  http://savannah.gnu.org/




reply via email to

[Prev in Thread] Current Thread [Next in Thread]