Fix filepath wildcard for avro/parquet

[steveroy0226]

Below changes have been tested and confirmed to work. Also proposing alternative approach since I'm not a big fan of guessing whether we are dealing with Avro vs Parquet:

Approach (1)

• This PR, minimal change. Basically checking to see if the matching files has the relevant extensions and routing it to either AvroSampler or ParquetSample

Approach (2)

• Not in this PR, but I think we should create a new arg called mode (which can be either parquet, avro or bigquery) and have that be a required field. It will be a breaking change, but then it will help avoid guessing which Sampler to use.

[codecov]

Codecov Report

Merging #564 (670b28c) into master (1d816bd) will increase coverage by 0.32%.
The diff coverage is 100.00%.

@@            Coverage Diff             @@
##           master     #564      +/-   ##
==========================================
+ Coverage   70.34%   70.66%   +0.32%     
==========================================
  Files          40       40              
  Lines        1703     1708       +5     
  Branches      150      146       -4     
==========================================
+ Hits         1198     1207       +9     
+ Misses        505      501       -4     

Flag	Coverage Δ
ratatoolCli	3.07% <0.00%> (-0.02%)	⬇️
ratatoolCommon	?
ratatoolDiffy	29.89% <0.00%> (-0.11%)	⬇️
ratatoolExamples	15.96% <0.00%> (-0.06%)	⬇️
ratatoolSampling	62.57% <100.00%> (+0.57%)	⬆️
ratatoolScalacheck	81.53% <ø> (ø)
ratatoolShapeless	4.41% <0.00%> (-0.02%)	⬇️

Flags with carried forward coverage won't be shown. Click here to find out more.

Impacted Files	Coverage Δ
...ala/com/spotify/ratatool/samplers/BigSampler.scala	80.41% <100.00%> (+5.41%)	⬆️

[benkonz]

I'm a fan of adding an optional parameter called --input-mode, similar to how we do it in BigDiffy https://github.com/spotify/ratatool/tree/master/ratatool-diffy#usage, just to keep the names consistent. In general, I'm not a fan of "guessing" what the input type is, but I also don't want to break existing users' workflows, so adding an optional parameter that we use, and inferring the input type if that parameter isn't there would work for me!

[steveroy0226]

I'm a fan of adding an optional parameter called --input-mode, similar to how we do it in BigDiffy https://github.com/spotify/ratatool/tree/master/ratatool-diffy#usage, just to keep the names consistent. In general, I'm not a fan of "guessing" what the input type is, but I also don't want to break existing users' workflows, so adding an optional parameter that we use, and inferring the input type if that parameter isn't there would work for me!

@benkonz, I like the approach. I will create a ticket to add --input-mode. I agree that we shouldn't break existing users' workflows.

[idreeskhan]

just the one typo

[idreeskhan]

can we log which one we resolve it as? might help us later on