Is there anything wrong with this?

Envrionment: Toolchain: stable-x86_64-pc-windows-msvc Rust/Cargo: 1.64.0

Steps to reproduce: cargo init app_name cd app_name cargo add plotly cargo add plotly --features kaleido cargo check

Output: ....

   Compiling plotly_kaleido v0.3.0
    Checking plotly v0.8.1
error[E0308]: mismatched types
   --> C:\Users\mwim\.cargo\registry\src\github.com-1ecc6299db9ec823\plotly-0.8.1\src\plot.rs:385:17
    |
383 |             .save(
    |              ---- arguments to this function are incorrect
384 |                 filename.as_ref(),
385 |                 &serde_json::to_value(self).unwrap(),
    |                 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ expected `str`, found enum `Value`
    |
    = note: expected reference `&str`
               found reference `&Value`
note: associated function defined here
   --> C:\Users\mwim\.cargo\registry\src\github.com-1ecc6299db9ec823\plotly_kaleido-0.3.0\src\lib.rs:124:12
    |
124 |     pub fn save(
    |            ^^^^

For more information about this error, try `rustc --explain E0308`.
error: could not compile `plotly` due to previous error

Hi,

I have this in Cargo.toml:

plotly = { version = "0.6", features = ["kaleido"] }

And, following the instructions, this in my code: plot.save(filename, ImageFormat::PNG, 1024, 768, 1.0);

The program crashes with this:

thread 'main' panicked at 'called `Result::unwrap()` on an `Err` value: Os { code: 2, kind: NotFound, message: "No such file or directory" }', /home/jgoerzen/.cargo/registry/src/github.com-1ecc6299db9ec823/plotly_kaleido-0.2.0/src/lib.rs:97:13

That backtrace wasn't much help, but strace was:

21466 lstat("/home", {st_mode=S_IFDIR|0755, st_size=10, ...}) = 0
21466 lstat("/home/jgoerzen", {st_mode=S_IFDIR|0755, st_size=722, ...}) = 0
21466 lstat("/home/jgoerzen/tree", {st_mode=S_IFDIR|0755, st_size=248, ...}) = 0
21466 lstat("/home/jgoerzen/tree/plotly_kaleido", 0x7ffd64e387b0) = -1 ENOENT (No such file or directory)

The current working directory was a different one under /home/jgoerzen/tree.

Why it expects ../plotly_kaleido to exist, I don't know. What's supposed to be there, I also don't know. The plotly documentation said there was some additional configuration necessary for plotly_kaleido but the plotly_kaleido documentation doesn't say what it is. I couldn't find any documentation for how to set up plotly_kaleido or why it would be looking for something in ...

Thanks for your work on this!

Hi, I have implemented some basic wrapping code for go.Mesh3D and go.Image as suggested in #49. I used @mrhrzg's scatter3d.rs as a starting point.

I am sending this pull request now to get some feedback. Should I do the following things to complete the added features? Anything else?

• add one example for each new trace type
• implement setters for all remaining parameters listed in the documentation
• fix all cargo warnings
• add entry in change log
• run test suite

Just before sending this I decided to also add some mapbox support, as suggested in #86. Specifically I started adding go.Scattermapbox. I am a little confused about layout.mapbox and layout.mapbox_style which seems to lack a documentation entry, and I have not been able to get layout.mapbox_style to work properly. I tried "stamen-watercolor" and "open-street-map" but they seem to have no effect. I guess it should be turned into an enum in the end, but this can't be the problem can it?

By the way I am using Evcxr.

Also by the way I think there is a fair bit of code duplication going on with parameters being repeated across different structs. Maybe this could be improved using some kind of rust macro? I guess the most difficult part of this might be to handle the documentation properly.

Hi, it's enjoyable to use this crate in plotting some scientific figures, and the experience was pleasant.

Just now, I found it crashed when adding color bar to the plot and I'm not sure if this was a matter of my code. Would it be possible to bother you have a look at it?

Here is the MWE:

use plotly;

fn main() {
    let x = (0 .. 100).collect::<Vec<_>>();
    let y = x.iter().map(|v| v * v).collect::<Vec<_>>();
    let c = x.iter().map(|v| (*v as f64).sqrt()).collect::<Vec<_>>();

    let tr = plotly::Scatter::new(x, y)
        .mode(plotly::common::Mode::Markers)
        .marker(plotly::common::Marker::new()
                .color_bar(plotly::common::ColorBar::new())
                .color_scale(plotly::common::ColorScale::Palette(plotly::common::ColorScalePalette::Jet))
                .color_array(c));

    let mut plot = plotly::Plot::new();
    plot.add_trace(tr);
    plot.use_local_plotly();

    plot.show();
}

And it crashed:

thread 'main' has overflowed its stack
fatal runtime error: stack overflow
[1]    39611 abort      cargo run --release

Here is my environment:

• macOS 10.15.7
• rustc 1.59 stable
• plotly 0.7

I'm using seed-rs to display plots with the inline html. seed-rs uses wasmpack which compile libraries with the wasm32-unknown-unknown target.

Repo to reproduce the error

error[E0599]: no function or associated item named `show_with_default_app` found for struct `plot::Plot` in the current scope
   --> C:\[...]\.cargo\git\checkouts\plotly-32f25712b3766afa\870dfc6\plotly\src\plot.rs:209:15
    |
96  | pub struct Plot {
    | --------------- function or associated item `show_with_default_app` not found for this
...
209 |         Plot::show_with_default_app(temp_path);
    |               ^^^^^^^^^^^^^^^^^^^^^ function or associated item not found in `plot::Plot`

Add a .to_inline_html() method to Plot which can then be sent elsewhere for rendering.

For example, when using in Evcxr this turns this construction:

let plotly_file = "/tmp/plot.html";
plot.to_html(plotly_file);
let plotly_contents = fs::read_to_string(plotly_file).unwrap();
println!("EVCXR_BEGIN_CONTENT text/html\n{}\nEVCXR_END_CONTENT", plotly_contents);
fs::remove_file(plotly_file)?;

into this one-liner:

println!("EVCXR_BEGIN_CONTENT text/html\n{}\nEVCXR_END_CONTENT", plot.to_inline_html());

I debated adding a helper method for that wrapping - I'm happy to do so but I feel it would be presumptive without an OK from @igiagkiozis since they might not wish to officially support such usage, and interested parties probably are using DARN anyway which is a better place for it.

Closes #12 as long as people are happy to do their own root div or Evcxr content wrapping.

The LegendGroupTitle seems only supported for Scatter3D and Sankey at this point. For example, the BoxPlot only offers to set a legend group. But there is no way to set the LegendGroupTitle? Is that on purpose or is it simply missing and should be added?

Yesterday I tried to write around 700 plots to an HTML file via calling to_inline_html() this many times into a vector of strings and then writing them all to an HTML template that only includes plotly.js once. My browser couldn't handle the resulting file very well. It is very slow. Are there any known tricks to make plotly behave more snappy in such situations? Or is this a complete misuse of the library on my part?

Tried to run this code

fn main() -> Result<(), Box<dyn Error>> {
    line_and_scatter_plot();
    Ok(())
}

fn line_and_scatter_plot() {
    let trace1 = Scatter::new(vec![1, 2, 3, 4], vec![10, 15, 13, 17])
        .name("trace1")
        .mode(Mode::Markers);
    let trace2 = Scatter::new(vec![2, 3, 4, 5], vec![16, 5, 11, 9])
        .name("trace2")
        .mode(Mode::Lines);
    let trace3 = Scatter::new(vec![1, 2, 3, 4], vec![12, 9, 15, 12]).name("trace3");

    let mut plot = Plot::new();
    plot.add_trace(trace1);
    plot.add_trace(trace2);
    plot.add_trace(trace3);
    plot.show();
}

It resulted in this error:

thread 'main' panicked at 'Could not find default application for HTML files.
Consider using the `to_html` method to save the plot instead. If using the `orca` feature the following
additional formats are available accessed by following methods:
- to_png
- to_jpeg
- to_webp
- to_svg
- to_pdf
- to_eps
: Os { code: 2, kind: NotFound, message: "No such file or directory" }', /home/nlam/.cargo/registry/src/github.com-1ecc6299db9ec823/plotly-0.5.1/src/plot.rs:454:14

Operating System: Windows 10 Home Version 2002 - WSL 2.0 Ubuntu Browser: Normal FireFox Version 78.02(64-bit)

Id like to see this crate support shapes in Plotly (https://plotly.com/javascript/shapes/).

Due to a failure to read the contributing file on my part I already made an implementation. My bad. I'll wait for discussion before making a PR.

This is a great addition to the broader scientific computing ecosystem for Rust. I've not used plotly.js extensively, and I suspect a bunch of folks would want to try to use this crate without browsing the plotly-documentation.

So I thought I'd attempt to explore this repo and make some documentation here and there. This issue is to gather ideas, strategy, or whatever needs to be taken into account in order to improve the documentation of this crate.

For instance: Is there a dataset prelude that can be used to make examples with?

By design, the API "plot.show()" would open a default web browser to display an interactive plot. However, under my environment(Windows 10, plotly="0.8.3"), it doesn't work. The default web browser was not opened either.

I found out that there was an issue message from cmd: The process cannot access the file because it is being used by another process

https://github.com/igiagkiozis/plotly/blob/49bd3a87fbaa3e507e9a2664f57feda89c35b95b/plotly/src/plot.rs#L465

//Line 465
    #[cfg(target_os = "windows")]
    fn show_with_default_app(temp_path: &str) {
        use std::process::Command;
        Command::new("cmd")
            .arg("/C")
            .arg(format!(r#"start {}"#, temp_path))
            .output()
            .expect(DEFAULT_HTML_APP_NOT_FOUND);
    }

After investigation, I suspect it's because, that the temp html file that should be opened was being used by another process, while this "another" process is the process itself which has not yet close the file handle in Rust program.

Could you fix this please? Thanks! :)

A few days ago I wrote and released a crate featuring a yew component wrapper that uses your library. Now I noticed you also just recently extended the wasm support and added examples for yew.

I wonder if it would make sense integrating this crate into your repo, or if we could move both repos to a joint github org?

This is a stub issue to track what is necessary to release a 1.0 version.

I'll be adding to this over time to show what has been implemented and what still needs doing before we can call it a 1.0 release.

Chart types

Basic Charts

• [x] Scatter
• [x] Line
• [x] Bar
• [ ] Pie
• [x] Bubble
• [x] Dot
• [ ] Filled area
• [ ] Sunburst
• [x] Sankey
• [ ] Point Cloud
• [ ] Treemaps
• [ ] Tables

Statistical Charts

• [x] Box
• [x] Histograms
• [ ] 2D density
• [ ] Violin
• [ ] Parallel Categories
• [ ] Splom

Scientific Charts

• [x] Contour
• [x] Heatmaps
• [ ] Ternary
• [ ] Parallel Coordinates
• [ ] Log
• [x] Scatter polar
• [ ] Bar polar
• [ ] Carpet
• [ ] Contour carpet

Financial Charts

• [ ] Waterfall
• [ ] Indicator
• [x] Candlestick
• [ ] Funnel
• [ ] Funnel area
• [x] OHLC

Maps

• [ ] Density heatmap
• [ ] Choropleth
• [ ] Scatter geo
• [x] Scatter mapbox

3D Charts

• [x] Scatter
• [x] Surface
• [x] Mesh
• [ ] Cone
• [ ] Streamtube
• [ ] Isosurface

API

• [ ] Make the library panic-safe by replacing unwrap() and expect with Result (#33)

Features

• [x] Wasm support (complete as of v0.8.0)
• [ ] Sliders

Meta

• [x] Add code coverage

This is a feature request for supporting a animated plot like seen in this python example. I would expect it to work by giving for x and y a slice of slices or a slice of Vecs instead of a single slice. Given this python example:

import plotly.express as px
df = px.data.gapminder()
px.scatter(df, x="gdpPercap", y="lifeExp", animation_frame="year", animation_group="country",
           size="pop", color="continent", hover_name="country",
           log_x=True, size_max=55, range_x=[100,100000], range_y=[25,90])

A usage example could look like the following code:

use plotly::{common::Mode,Plot,Scatter};
fn main() -> Result<(),BoxErr> {
    // Four different time points
    let time = vec![1.0,2.0,3.0,4.0]; 
    // For each timepoint generate data. This could be a position for example.
    let mut x = Vec::with_capacity(time.len());
    for i in 0..time.len() {
        x.push((0..1_000).map(|x|x as f64*0.1).collect::<Vec<f64>>());
    }
    // For each timepoint evaluate some function at each position
    let mut y = Vec::with_capacity(time.len());
    for (i,xi) in x.iter().enumerate() {
        y.push(
            xi.iter().map(|xk| xk*xk + time[i]).collect::<Vec<f64>>()
        );
    }
    // Generate traces from the data
    let traces = x.into_iter().zip(y).map(|(x,y)| Scatter::new(x,y).mode(Mode::Lines)).collect::<Vec<_>>();
    // Plot the data and animate the plot with respect to time
    let mut plot = Plot::new();
    plot.add_traces(traces).animate_by(time);
    plot.show();
    Ok(())
}