Guidance for Testing Apps that require large Data Sets

Rails
1

I’m currently trying to build through TDD an app that scrapes daily
statistical sports information and puts it into varying models in the
hopes
of analyzing and researching the data in my own way. Though, to test a
lot
of the functionality, I require that the database have many records
across
many database tables (including one that has about 15 columns of
information).

Can anyone point to an easy repeatable way to post data when I run my
tests, using RSpec, so that feature testing can easily be run.

Thanks in advance

2

On Apr 17, 2016 2:35 PM, “John L.” [email protected] wrote:

I’m currently trying to build through TDD an app that scrapes daily
statistical sports information and puts it into varying models in the
hopes
of analyzing and researching the data in my own way. Though, to test a
lot
of the functionality, I require that the database have many records
across
many database tables (including one that has about 15 columns of
information).

Can anyone point to an easy repeatable way to post data when I run my
tests, using RSpec, so that feature testing can easily be run.

in advance
Thanks

You received this message because you are subscribed to the Google G.
“Ruby on Rails: Talk” group.
To unsubscribe from this group and stop receiving emails from it, send an
email to [email protected].
To post to this group, send email to [email protected].
To view this discussion on the web visit
https://groups.google.com/d/msgid/rubyonrails-talk/a68b0c87-172a-4e0d-8abe-e8636e4e5f3d%40googlegroups.com
.
For more options, visit https://groups.google.com/d/optout

I’m responding on the phone so I will be brief.

A large data set is concerned with the number of records in your tables,
not the number of columns. If you can test with just a few or a single
row,
then you shouldn’t have to do anything special. If your tests will
actually
require large number of rows, you maybe want to do something different.
However, your tests and close could probably benefit from more unit
testing
with tiny data sets, or even separating things out into POROS instead of
having everything in the data models. In addition to that, if your code
will be working on statistics on large number of records, consider
writing
direct sql to take full advantage of the dbms.