Based on a previous thread, I gave FasterCSV a try and I’ve been very
impressed with the performance increases. However, something about the
syntax for writing to a file bothers me. Is there any way to create a
“writer object” to write to a file? For example, this is how I’m
currently using FasterCSV:
FCSV.open(my_file,“w”) do |out|
out << my_data
…
end
What I’d like to do is something along the lines of:
out = FCSV.open(my_file,“w”)
out << my_data
…
out.close
Is this functionality present? If not, is this a design decision? I
guess it just bothers me to have to wrap a block around all my code that
has to do with writing to a file.
Well, irb is my friend and I should have used it before asking.
FasterCSV indeed does have this functionality and it works as expected.
I had an error with this functionality before posting but it must have
been my mistake.
What I’d like to do is something along the lines of:
out = FCSV.open(my_file,“w”)
out << my_data
…
out.close
Is this functionality present?
Sure. Should work just fine. (File a bug if it doesn’t!)
Just FYI though, the first example is very idiomatic Ruby. We would
always rather let the language clean up after us than to have to
remember to do it ourselves. That’s why you see this construct all
over Ruby even with normal File writes.
What I’d like to do is something along the lines of:
out = FCSV.open(my_file,“w”)
out << my_data
…
out.close
Is this functionality present?
Sure. Should work just fine. (File a bug if it doesn’t!)
Just FYI though, the first example is very idiomatic Ruby. We would
always rather let the language clean up after us than to have to
remember to do it ourselves. That’s why you see this construct all
over Ruby even with normal File writes.
James Edward G. II
James -
Was hoping you’d chime in. Thanks for the explanation, this is what I
expected. I supposed I just have the beat the Java that’s remaining in
me by getting used to this type of code. I can certainly see the benefit
of clean up being done for you.
I don’t use FCSV but I’d be wouldn’t be surprised if it didn’t have a
close method (although it may - just have a look at the code.) The
pattern of associating a block with the opening/initialization of a
resource such as a file, connection, etc. is one of Ruby’s major
strengths and if the designer of the entity being opened has done their
job right, the entity is (usually) closed for you when the block
terminates. This pattern leads to fewer defects such as leaked file
handles and such. So, I suppose at the end of the day this is a matter
of personal taste but I think most Ruby enthusiasts would prefer the
block to the explicit close. And I do believe this is an example of
the “initialization is acquisition” design pattern, if memory serves
me, so its not as if this is some arcane practice, for what ever my 2
cents are worth
Ken
This forum is not affiliated to the Ruby language, Ruby on Rails framework, nor any Ruby applications discussed here.
