[Chicago-talk] Help with Net::Amazon::MechanicalTurk

[Alan Mead]

I need to install Net::Amazon::MechanicalTurk on my CentOS 6 server, but
it's failing. If you're interested, the results are here:
http://gooseburger.homeunix.org/alan/files/net-amazon-mturk.buildfails.text

Looking at the open bugs on CPAN, this code looks unmaintained. But I
did find this: 
http://daveviner.blogspot.com/2009/12/amazon-mechanical-turk-perl-library.html

But the patch seems to be gone away and brian d foy's unofficial version
seems to have gone away:
http://search.cpan.org/~bdfoy/Net-Amazon-MechanicalTurk-1.01_01/

But I did find a suggestion to delete the function 'defined' on line 35
of ModuleUtil. The only comment in that routine was "Symbol table black
magic" and I have no idea what it's doing. I've reproduced the relevant
routine ("packageExists") with the change, although I think I'd need
more context (like a call) to understand what it's doing... Can anyone
explain this change?

sub packageExists {
    my ($class, $package) = @_;
    # Symbol table black magic
    no strict 'refs';
    return %{*{"${package}::"}};          # <-- here's the change
    #return defined %{*{"${package}::"}}; # <-- here's the old line
    # based on Pavel Karoukin's suggestion
    #
http://stackoverflow.com/questions/701093/how-can-i-fix-netamazonmechanicalturks-failing-tests
}

When I run the accountbalance sample script, it returns "MechanicalTurk
account balance: 10000.000" while the website says I have about
$102.00.  So, I'm assuming the code is not working...

A number of the package's tests were deferred until I configured the
package with AWS credentials,  and one test failed when I re-ran them:

t/93-RetrieveResults.t .................... 1/4
#   Failed test 'Actual reviewable hits'
#   at t/93-RetrieveResults.t line 15.
# Looks like you failed 1 test of 4.
t/93-RetrieveResults.t .................... Dubious, test returned 1
(wstat 256, 0x100)
Failed 1/4 subtests
t/boilerplate.t ........................... ok
t/pod-coverage.t .......................... skipped: Test::Pod::Coverage
1.04 required for testing POD coverage
t/pod.t ................................... ok

Test Summary Report
-------------------
t/93-RetrieveResults.t                  (Wstat: 256 Tests: 4 Failed: 1)
  Failed test:  3
  Non-zero exit status: 1
Files=33, Tests=112, 28 wallclock secs ( 0.09 usr  0.05 sys +  3.01
cusr  0.42 csys =  3.57 CPU)
Result: FAIL
Failed 1/33 test programs. 1/112 subtests failed.
make: *** [test_dynamic] Error 255

Any suggestions about how I go about investigating this?

Finally, I also found this: 
https://aws.amazon.com/items/443?externalID=443&categoryID=24

Which also doesn't work (no output) but I feel like I can probably debug
that script myself.

Any thoughts, suggestions, or experiences with this would be much
appreciated.  I'd really like to do this using Perl, if I can.

-Alan

-- 

Alan D. Mead, Ph.D.
President, Talent Algorithms Inc.

science + technology = better workers

+815.588.3846 (Office)
+267.334.4143 (Mobile)

http://www.alanmead.org

I've... seen things you people wouldn't believe...
functions on fire in a copy of Orion.
I watched C-Sharp glitter in the dark near a programmable gate.
All those moments will be lost in time, like Ruby... on... Rails... Time for Pi.

          --"The Register" user Alister, applying the famous 
            "Blade Runner" speech to software development