TST: Check lossiness of floats with parse_dates

[gfyoung]

The examples don't fail anymore with read_csv, so let's add them as tests.

Closes #2697.

[codecov]

Codecov Report

Merging #18136 into master will decrease coverage by 0.03%.
The diff coverage is n/a.

@@            Coverage Diff             @@
##           master   #18136      +/-   ##
==========================================
- Coverage   91.28%   91.24%   -0.04%     
==========================================
  Files         163      163              
  Lines       50122    50122              
==========================================
- Hits        45752    45732      -20     
- Misses       4370     4390      +20

Flag	Coverage Δ
#multiple	89.05% <ø> (-0.03%)	⬇️
#single	40.32% <ø> (-0.06%)	⬇️

Impacted Files	Coverage Δ
pandas/io/gbq.py	25% <0%> (-58.34%)	⬇️
pandas/plotting/_converter.py	63.38% <0%> (-1.82%)	⬇️
pandas/core/frame.py	97.75% <0%> (-0.1%)	⬇️
pandas/io/parsers.py	95.57% <0%> (+0.06%)	⬆️

---

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 1181622...ff98f38. Read the comment docs.

[codecov]

Codecov Report

Merging #18136 into master will decrease coverage by 0.03%.
The diff coverage is n/a.

@@            Coverage Diff             @@
##           master   #18136      +/-   ##
==========================================
- Coverage   91.28%   91.24%   -0.04%     
==========================================
  Files         163      163              
  Lines       50122    50122              
==========================================
- Hits        45752    45732      -20     
- Misses       4370     4390      +20

Flag	Coverage Δ
#multiple	89.05% <ø> (-0.03%)	⬇️
#single	40.32% <ø> (-0.06%)	⬇️

Impacted Files	Coverage Δ
pandas/io/gbq.py	25% <0%> (-58.34%)	⬇️
pandas/plotting/_converter.py	63.38% <0%> (-1.82%)	⬇️
pandas/core/frame.py	97.75% <0%> (-0.1%)	⬇️
pandas/io/parsers.py	95.57% <0%> (+0.06%)	⬆️

---

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 1181622...ff98f38. Read the comment docs.

[jreback]

@gfyoung worth adding the parse_dates=True as additional xfails here? (or maybe just open a new issue with that case)? maybe even show a warning that things are losing precision?

[gfyoung]

@jreback : I'm not sure I follow you here. The tests are parametrized so that they pass regardless of whether parse_dates=True or False! 😄

[jreback]

right but the original issues shows a loss of precision when parse_dates=True
is that fixed (original issue gives a symptom that doesn’t have an easy fix)or is it just the way it is?

[gfyoung]

Yes, the issue is fixed on master (somehow). That's why my test checks the output for parse_dates=True (with no xfail).

[jreback]

thanks @gfyoung if you figure out where then add a reference here (not a big deal though)