I've been playing with tplot for a few days now building a command-line tool to show some usage data, basically I've created a CLI connecting pandas.read_csv to tplot. This has turned out really well and the plots look great in the terminal.

Any thoughts on the best way to handle timeseries/datetime on the x-axis? One approach might be for the user to just cast the timestamps to Unix epoch times, but that's not very human readable on the axis labels. Do you have any sense of how much of a challenge it would be to construct a label formatter on the x-axis to show dates and times nicely (like months+days, or hours+minutes)?

Just sharing ideas, thanks.

Hello,

This PR aims to change termcolor package to termcolor-whl. Motivation: Installing via pip generates an error as below. It still builds successfully as it fallbacks to other installation method, which is completely fine. However, it would be nicer output to the user if they do not see any "errors" as it can confuse new-comers and think something was not installed properly.

 Building wheel for termcolor (setup.py) ... error
  ERROR: Command errored out with exit status 1: 
.                  ........................
 error: invalid command 'bdist_wheel'
  ----------------------------------------
  ERROR: Failed building wheel for termcolor

Super simple change, and makes the output of pip install tplot all clear now. Let me know if there is anything else I may have missed. Thank you for your time!

Switch back to Sphinx for documentation.

It's easier to work with than mkdocs once you get used to the reStructuredText syntax, it's more capable, it looks better on readthedocs, and it's the more popular choice for Python projects.

Hi,

Are there any low hanging fruits or TODOS that we could do? Perhaps a Contributing.MD to enable us to setup a develop environment, and something in the README to know which important issues could be tackled.

Thank you for your time!

Reference figure tests are failing, presumably because the ANSI escape characters for colors will be different on Windows.

I guess there will have to be a separate set of reference figures for Windows.

Write a new x axis tick label drawing method to:

• use space in y axis labels
• don't overwrite previously drawn labels with white space from next label
• shift labels so short labels make space for wider labels (within reason)
• get rid of annoying xticklabel_width argument and determine is automatically instead (if needed at all)

tplot makes heavy use of the functools.lru_cache decorator which has a memory leak where class instances don't get garbage collected if it's used on methods (see also this video).

Since we only cache one value in all cases (lru_cache(maxsize=1)), it might as well be a cached_property instead. There's a backport available for Python <3.8.

Thanks to @dmatos2012 for telling me about this!

Aside from a few unit tests for specific cases, testing currently consists of plotting a bunch of example datasets and manually looking at them. This should be automated.

I'm thinking about writing the output of each test figure to a file and checking the output is identical to the reference figures in the file at test time. This means tests will fail if the appearance of the figure is altered in any way. In the case that this is desired due to a fix or improvement, the reference figure can simply be regenerated and checked manually only once.

Currently, bar and hbar plots extend from the bottom of the graph or the left of the graph, respectively:

                                    Anscombe                                    
   15┤                                                                          
     │                                                                          
     │                                                          •               
   10┤                                                                         •
     │                                    •       •      •              •       
     │               •             •                                            
    5┤•      •              •                                                   
y    │                                                                          
     │                                                                          
l   0┤                                                                          
a    │                                                                          
b    │█                                                                         
e  -5┤█      █                                                                  
l    │█      █       █      █                                                   
     │█      █       █      █      █      █       █             █       █      █
  -10┤█      █       █      █      █      █       █      █      █┌────Legend───┐
     │█      █       █      █      █      █       █      █      █│• Anscombe I │
     │█      █       █      █      █      █       █      █      █│█ Anscombe II│
  -15┤█      █       █      █      █      █       █      █      █└─────────────┘
      ┬──────┬───────┬──────┬──────┬──────┬───────┬──────┬──────┬───────┬──────┬
      4      5       6      7      8      9       10     11     12      13    14
                                       x label                                  

                                    Anscombe                                    
   -3┤                                                                          
     │                                                                          
   -4┤                                                                          
     │                                                                          
     │███████                                                                   
   -5┤                                                                          
     │                                                                          
y  -6┤███████████████                                                           
     │                                                                          
l  -7┤                                                                          
a    │██████████████████████                                                    
b  -8┤█████████████████████████████████████████████████████████████████████████ 
e    │█████████████████████████████                                             
l  -9┤██████████████████████████████████████████████████████████████████        
     │██████████████████████████████████████████████████████████                
     │                                                                          
  -10┤                                                           ┌────Legend───┐
     │                                                           │█ Anscombe II│
  -11┤                                                           └─────────────┘
      ┬──────┬───────┬──────┬──────┬──────┬───────┬──────┬──────┬───────┬──────┬
      -4     -3      -2     -1     0      1       2      3      4       5      6
                                       x label                                  

They should instead extend from the zero line (y=0 for bar, x=0 for hbar).

anscombeA = [
    [10, 8, 13, 9, 11, 14, 6, 4, 12, 7, 5],
    [8.04, 6.95, 7.58, 8.81, 8.33, 9.96, 7.24, 4.26, 10.84, 4.82, 5.68]
]
anscombeB = [
    [10, 8, 13, 9, 11, 14, 6, 4, 12, 7, 5],
    [9.14, 8.14, 8.74, 8.77, 9.26, 8.10, 6.13, 3.10, 9.13, 7.26, 4.74]
]
# sort by X value
anscombeA = list(zip(*[(x, y) for x, y in sorted(zip(*anscombeA))]))
anscombeB = list(zip(*[(x, y) for x, y in sorted(zip(*anscombeB))]))

fig = Figure(xlabel="x label", ylabel="y label", title="Anscombe", legendloc="bottomright")
fig.scatter(x=anscombeA[0], y=anscombeA[1], label="Anscombe I")
fig.scatter(*anscombeB, label="Anscombe II", marker="+")
fig.show()

Output:

                            Anscombe                            
      |                                             o           
  10.0+                                                         
      |                                                        o
   9.0+                                       +                 
      |                            +     +          +    +      
   8.0+                                       o                 
y     |                      +           o                     +
   7.0+           o     +                                o      
l     |                      o                                  
a  6.0+                                                         
b     |           +                                             
e  5.0+      o                                                  
l     |                 o                                       
   4.0+      +                                   +----Legend---+
      |o                                         |o Anscombe I |
   3.0+                                          |+ Anscombe II|
       +                                         +-------------+
       +----------+----------+-----------+----------+----------+
       4.0       6.0        8.0         10.0       12.0     14.0
                               x label                          

Bumps certifi from 2021.10.8 to 2022.12.7.

Dependabot will resolve any conflicts with this PR as long as you don't alter it yourself. You can also trigger a rebase manually by commenting @dependabot rebase.

tplot currently assumes all characters are halfwidth. We could handle fullwidth characters by using wcwidth to get the displayed width of each character and then removing spaces from that row as needed to maintain alignment with the other rows.

I suspect this would have to be done in the __str__ method after converting the self._canvas array to a list of strings, since it will mean the canvas is no longer rectangular.

Currently, bar plots use the full block Unicode character (█) as markers by default. Resolution for the start and end of the bar could be improved by using partial bar characters, e.g. ▀ at the bottom and one of ▁▂▃▄▅▆▇ at the top. Similarly for horizontal bar plots we could use ▐ on the left and one of ▉▊▋▌▍▎▏ on the right.

Currently, the axis range is determined by the range of the input data, i.e. min(x), max(x) for the X axis. It would be nice to be able to override that, e.g. when plotting percentages you often want the range to be from 0 to 1.

Proposed syntax:

fig = tplot.Figure(xlim=(0, None), ylim=(0,1))

xlim and ylim each take a (min, max) tuple. None means the value should be taken from the input data. So in the above example, the X axis range would be from 0 to max(x) and the Y axis range would be from 0 to 1.

This may seem like a simple feature, but it affects the code in a bunch of different places. Care should also be taken to handle values in the input data that fall outside the specified range.