allow non-factor statnames in utils::transformstats

[infotroph]

Description

PEcAn.utils::transformstats currently assumes the statname column if its input is a factor. This patch allows it to be character instead, if desired. Output keeps the same class it had on input.

Edit: while I was here, also cleaned up code formatting and fixed logic in the untransformed stats warning.

Motivation and Context

Noticed now because the development version of R no longer coerces character vectors in data.frame to factor, so our unit test started failing. Good catch, unit test!

Review Time Estimate

• Immediately
• Within one week
• When possible

Types of changes

• Bug fix (non-breaking change which fixes an issue)
• New feature (non-breaking change which adds functionality)
• Breaking change (fix or feature that would cause existing functionality to change)

Checklist:

• My change requires a change to the documentation.
• I have updated the CHANGELOG.md.
• I have updated the documentation accordingly.
• I have read the CONTRIBUTING document.
• I have added tests to cover my changes.
• All new and existing tests passed.

[infotroph]

@mdietze @dlebauer I think this is ready to merge -- previous Travis failures are hopefully now resolved.

[dlebauer]

I still don't understand travis well enough to know what to do with the travis failure - the log says "build exited with status 1" indicating success, but then travis says the build fails https://travis-ci.org/github/PecanProject/pecan/jobs/664465417?utm_medium=notification&utm_source=github_status

[infotroph]

@dlebauer the statuses are from bash, so anything not 0 is a failure.

Looks like this is yet another unrelated dependency failure, this time in the DB tests: apparently PEcAn.DB::build_insert_query is now returning doubled quotes because of something that changed between glue 1.3.1 and 1.3.2

[ashiklom]

As I mentioned on Slack, I think the easiest fix here is to replace the glue_sql statement in base/db/R/insert_table.R::build_insert_query with the following:

insert_string <- paste(insert_list, collapse = ",")
result <- glue::glue_sql(
  "INSERT INTO {`table`} ({`colnames(values)`*}) VALUES ",
  .con = .con
)
DBI::SQL(paste0(result, "(", insert_string, ")"))

It feel like a less-robust solution because it replaces supposedly robust backend-dependent SQL quoting with hard-coded quoting...but I think it should work with the SQLite backend in the test.

Also as I mentioned on Slack, the ideal solution here is to simplify the code in that script to just use prepared statements for everything, which should be faster, safer, and easier to implement.