Scientific notation write issue in v0.5

[chris-german]

I've noticed when writing to CSV files in 0.5, it changes numbers and no longer puts them in scientific notation.

In the code below, the issue is reproduced.

1.9263083035570026e-11 is displayed as 1.9263083035570026e-11, but when looking at the written file, it is written as 19263083035570026e-27. This makes written files a bit more difficult to scan through.

using DataFrames, CSV

y = collect(1.0:4.0)
x = [1.0, 0.0, 1.9263083035570026e-11, 0.5]
df = DataFrame(y = y, x = x)

CSV.write("test.csv", df)

[quinnj]

Thanks for reporting @chris-german; this changed because we're trying to use a lower-level float printing algorithm to customize the decimal point we print (',' for european locales, for example).

Luckily, I recently worked on porting/writing an entirely new (and faster!) float printing algorithm to Julia, and I'm planning on completely re-writing the CSV.write machinery under the hood to be much much faster. So sorry for the delay in responding here, and I'm sorry I don't have an immediate solution, but it's coming soon.

[quinnj]

This was fixed in the most recent release