Linting, Part 2 #81
Linting, Part 2 #81
Conversation
The Generate tests were failing after I changed the lines with multiple assignment statements (eg. `x = y = 5`) to be two lines. I had previously tried to preserve the syntax, by assigning one object as null, and then assigning the second object as the first, but this failed because the first object didn't exist (NULL), so the second assignment operation failed. Change to explicitly assign second object as NULL.
Codecov Report
@@ Coverage Diff @@
## master #81 +/- ##
==========================================
- Coverage 100% 98.52% -1.48%
==========================================
Files 35 35
Lines 1394 1631 +237
==========================================
+ Hits 1394 1607 +213
- Misses 0 24 +24
Continue to review full report at Codecov.
|
|
I think I've buttoned up the My plan for tests is to do a literal linting of what is currently there, and then go back through and supplement until coverage is back up to 100%. |
@wlandau-lilly: please check this one extra carefully. I don't think I changed anything structural, but as near as I can tell, there aren't an actual unit test on this one.
|
Looks that way to me. Definitely much improved. style warning error
inst/examples/basic/basic.R 54 0 0
tests/testthat/test-command-changes.R 13 0 0
tests/testthat/test-envir.R 26 0 0
tests/testthat/test-full-build.R 16 0 0
tests/testthat/test-generate.R 76 0 0
tests/testthat/test-import-object.R 18 0 0
tests/testthat/test-intermediate-file.R 7 0 0
tests/testthat/test-Makefile.R 35 0 0
tests/testthat/test-namespaced.R 17 0 0If you rerun |
|
I ran I'm a bit nervous about touching them (which is why I'm doing it last), so if you could give them a second set of eyes, just to make sure that I don't drop any arguments or parens or such, I would appreciate it greatly. |
|
All my checks came out totally clear:
I will merge your work so far to allow others to develop |
More progress on #40. Unless, I've misunderstood, this should wrap up what can be done without merge conflicts before 4.1.0?
I've tried to be consistent with the changes that I made in #72, including keeping each function's lints as its own commits, to make
git-blamea bit easier.All checks pass on my local machine, but I expect coverage to drop, like last time.