Gridlines and inverted mode

[z-m-k]

Two switches in the plot() function which allow to:

• add gridlines with gridlines=true
• invert the plot with invert=true
• both

It comes in handy if Braille blank, 0x2800, has different width than
the remaining characters in the range.

Example without gridlines:

Example with gridlines:

[waldyrious]

I wonder if there is another way to workaround the inconsistent width of the Braille blank without using the gridlines. Maybe one of these consistently matches the proper width?

[sunetos]

In a monospaced font, when would the width ever be inconsistent? What font are you using? I think it's safe to say that TextPlots is pretty dependent on a monospaced font.

[z-m-k]

Source Code Pro/Consolas. I think it is a Windows thing, though.

[sunetos]

Even the regular space character is sized incorrectly in your screenshots: the axis label (max x value) in the bottom-right is normally right-aligned with the edge of the graph, but it extends further right in your screenshot. I'm not sure how that's possible with a monospaced font.

[sunetos]

Either way, thanks for the pull request! I can't see any reason not to support these parameters. I hope you don't mind that I rewrote it to be quite a bit shorter, and fixed 2 bugs I found:

1. Combining invert=true, gridlines=true wasn't an exact inversion of invert=false, gridlines = true (the gridline was in a different place).
2. Since you were setting the char each time in showdot() instead of bitwise ORing, it failed when plotting multiple Y values with the same X value.

I will update the version in METADATA.jl in a day or so with this addition after making sure no bugs pop up. You can test it with Pkg.clone("https://github.com/sunetos/TextPlots.jl/").

[z-m-k]

Nice. I will admit it took me some time to actually figure out what is going on there, and since I assumed this might be localized to my machine only, it felt stupid just to complain something is not working without proposing a way to fix it. It does feel silly how little you changed compared to the original code to get the same/better effect than after what I did.