Spanprocessor

[bogdandrutu]

Founded problems:

• Maps don't handle nil entries in the underlying slice. What should we do?
• Current data API do not allow to construct slices with nil entries for test - it should probably not allow doing this for normal conversions/pipeline.
• Current protocol does not support undefined type for Attributes, should we support that, or we consider an empty entry as being type string with empty string value?

[bogdandrutu]

@tigrannajaryan please take a look at the problems I found while converting Span processor.

[codecov-io]

Codecov Report

Merging #788 into master will increase coverage by 0.12%.
The diff coverage is 95.95%.

@@            Coverage Diff             @@
##           master     #788      +/-   ##
==========================================
+ Coverage   85.56%   85.68%   +0.12%     
==========================================
  Files         154      156       +2     
  Lines       11520    11673     +153     
==========================================
+ Hits         9857    10002     +145     
- Misses       1292     1297       +5     
- Partials      371      374       +3     

Impacted Files	Coverage Δ
processor/spanprocessor/span_processor.go	81.81% <86.95%> (ø)
internal/processor/filterspan/filterspan.go	98.29% <98.29%> (ø)
internal/data/common.go	95.14% <100.00%> (+0.30%)	⬆️
internal/processor/filterhelper/filter_helper.go	100.00% <100.00%> (ø)
processor/metrics.go	91.76% <100.00%> (+0.73%)	⬆️
processor/spanprocessor/factory.go	100.00% <100.00%> (ø)
translator/internaldata/resource_to_oc.go	68.11% <0.00%> (ø)

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update a8b83bc...8e52e10. Read the comment docs.

[bogdandrutu]

TODO: Add tests for:

• one-span-one-nil.

[tigrannajaryan]

Nice! I like how minimal changes are needed.