Fixing LocalTime rounding (losing precision)

[franetw]

When converting the Java LocalTime to PostgreSQL time value we have to consider the following:

• Java LocalTime has nano and time has micro precision
• PostgreSQL time max value is 24:00:00 while LocalTime is 23:59:59.999999999

Conversion logic:

• truncate and round up when nanos >= 500
• rounding up all >= 23:59:59.999999500 to 24:00:00

The current code intends to do this but it has a bug where it rounds too much. This PR fixes it.

I've added test for toLocalTime() which demostrates that these two methods are asymmetrical, you can lose/change information if you store and retrieve the time.
Example:

timestampUtils.toLocalTime(timestampUtils.toString(LocalTime.parse("23:59:59.999999500")))) -> LocalTime.parse("23:59:59.999999999")

I'm open to suggestions on how can this be consolidated and if there's anything hidden to consider with this kind of change.

All Submissions:

• Have you followed the guidelines in our Contributing document?
• Have you checked to ensure there aren't other open Pull Requests for the same update/change?

New Feature Submissions:

1. Does your submission pass tests?
2. Does mvn checkstyle:check pass ?
3. Have you added your new test classes to an existing test suite?

Changes to Existing Features:

• Does this break existing behaviour? If so please explain.
• Have you added an explanation of what your changes do and why you'd like us to include them?
• Have you written new tests for your core changes, as applicable?
• Have you successfully run tests with your changes locally?

[AppVeyorBot]

❌ Build pgjdbc 1.0.449 failed (commit fb32c1c4fc by @franetw)

[AppVeyorBot]

✅ Build pgjdbc 1.0.450 completed (commit c108576bc8 by @franetw)

[AppVeyorBot]

❌ Build pgjdbc 1.0.451 failed (commit aded3817c2 by @franetw)

[codecov-io]

Codecov Report

Merging #1570 into master will decrease coverage by <.01%.
The diff coverage is 100%.

@@             Coverage Diff              @@
##             master    #1570      +/-   ##
============================================
- Coverage      68.9%   68.89%   -0.01%     
+ Complexity     3972     3969       -3     
============================================
  Files           179      179              
  Lines         16565    16557       -8     
  Branches       2695     2694       -1     
============================================
- Hits          11414    11407       -7     
+ Misses         3899     3897       -2     
- Partials       1252     1253       +1

[franetw]

I assume the failing build is due to these tests running on JRE7 when LocalTime did not exist.
I'd appreciate some help on how to include/exclude these tests to get the builds passing.

[AppVeyorBot]

❌ Build pgjdbc 1.0.452 failed (commit e665359ca8 by @franetw)

[davecramer]

search for
//#if mvn.project.property.postgresql.jdbc.spec >= "JDBC4.2"

it is used in before TimestampUtilsTest.toLocalTimeBin()

[AppVeyorBot]

❌ Build pgjdbc 1.0.457 failed (commit dd1960fbe7 by @franetw)

[AppVeyorBot]

❌ Build pgjdbc 1.0.458 failed (commit f3bf18cb50 by @franetw)