[Feature Request]: Specify plot sizes with values

[matyasbukva]

Description

It would be nice (actually essential) if the "Edit Image" window would allow you to specify the height and width of a plot by value.

Purpose

No response

Use-case

No response

Is your feature request related to a problem?

No response

Is your feature request related to a JASP module?

No response

Describe the solution you would like

No response

Describe alternatives that you have considered

No response

Additional context

Currently, we use JASP for performing statistical analyses. Often, we need to create multiple plots of the same type, such as residual plots, scatterplots, etc., for different variables. It is incredibly frustrating to have to manually adjust the plots using the cursor. The most straightforward solution would be to include a scaling option under the "Edit Image" section where we can adjust dimensions using numerical values (without cluttering the GUI), or to allow resizing a plot incrementally, for example, by holding down Shift while resizing. This would also address the issue where, in certain cases (e.g., when creating scatterplots with side densities and splitting by a categorical variable), the legend sometimes gets cut off. Enhancing the ability to resize plots would significantly improve the user experience and make the software much more versatile for comprehensive use.

[matyasbukva]

Alternatively, holding down Shift could at least maintain the aspect ratios during resizing. This is a fairly common complaint in our biostatistics classes, unfortunately. :S

[tomtomme]

old duplicates:
#1444
#1674
#2800

and one workaround for the time being by vandenman:

There is no straightforward way to get the R command used to generate the plot (it's possible but it's not easy). However, you can access the plot objects saved in the state the following hacky way:

• unzip the .jasp file, let's assume it's now called "unzipped_jasp"
• in R, do load("unzipped_jasp/resources/0/state"). Here 0 is an analysis number, if there are multiple analyses then these all have their own state. So you may need to adjust 0 to something else. There are png files in each folder which should help you find the plot of interest.
• after load there is now an object called state. You can access the plot objects in there via state$figures$resources/0/_0_t1646864985384.png$obj
  Here resources/0/_0_t1646864985384.png is actually the name of the png file that you want to save differently.
• Once you've found the plot you're interested in, you can resave it in the usual way:

 pdf(file, width, height)
 print(plot) # assuming the plot object is a ggplot object
 dev.off()

I hope that helps!

[matyasbukva]

i understand, thank you very much. is there any prospect of you working on this in the future?

…

On Wed, 11 Dec 2024 at 10:12, Thomas Langkamp ***@***.***> wrote: old duplicates: #1444 <#1444> #1674 <#1674> #2800 <#2800> and one workaround for the time being by vandenman There is no straightforward way to get the R command used to generate the plot (it's possible but it's not easy). However, you can access the plot objects saved in the state the following hacky way: unzip the .jasp file, let's assume it's now called "unzipped_jasp" in R, do load("unzipped_jasp/resources/0/state"). Here 0 is an analysis number, if there are multiple analyses then these all have their own state. So you may need to adjust 0 to something else. There are png files in each folder which should help you find the plot of interest. after load there is now an object called state. You can access the plot objects in there via state$figures$`resources/0/_0_t1646864985384.png`$obj Here resources/0/_0_t1646864985384.png is actually the name of the png file that you want to save differently. Once you've found the plot you're interested in, you can resave it in the usual way: pdf(file, width, height) print(plot) # assuming the plot object is a ggplot object dev.off() I hope that helps! — Reply to this email directly, view it on GitHub <#3078 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AVW2D7N6NBZA4RWDC2E2LD32E76ZFAVCNFSM6AAAAABTM4YOY6VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDKMZVGI3DCOJYGE> . You are receiving this because you authored the thread.Message ID: ***@***.***>

[tomtomme]

Sure. It is a usefull feature. But I can give no timeline, sorry. So many things to do...
Things like long-wide-conversion and broader missing value handling also were requested ages ago and may have a higher priority atm.

Here is our ToDo list of requests only related to data wrangling and plot editing:
https://github.com/jasp-stats/jasp-issues/issues?page=1&q=is%3Aopen+is%3Aissue+label%3A%22Component%3A+Data+Editing%22%2C%22Component%3A+Plot+Editing%22+sort%3Acreated-asc

[matyasbukva]

What you said is a great solution, but if you have fill or color in the plot, it gets a bit confusing... the variable names are not assigned to the axes correctly. Do you have any tips for this? I attach the plots. load("próba/resources/18/state") state$figures$`resources/18/_5_t-1346600528.png`  state$figures$`resources/18/_17_t-1346596799.png` 

…

On Wed, 11 Dec 2024 at 10:54, Thomas Langkamp ***@***.***> wrote: Sure. It is a usefull feature. But I can give no timeline, sorry. So many things to do... Things like long-wide-conversion and broader missing value handling also were requested ages ago and may have a higher priority atm. Here is our ToDo list of requests only related to data wrangling and plot editing: https://github.com/jasp-stats/jasp-issues/issues?page=1&q=is%3Aopen+is%3Aissue+label%3A%22Component%3A+Data+Editing%22%2C%22Component%3A+Plot+Editing%22+sort%3Acreated-asc — Reply to this email directly, view it on GitHub <#3078 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AVW2D7JTA5EMHSYTB2CGRJD2FADVRAVCNFSM6AAAAABTM4YOY6VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDKMZVGM2TIOBQGI> . You are receiving this because you authored the thread.Message ID: ***@***.***>

[tomtomme]

@vandenman is the expert here. Any hints?

[tomtomme]

progress on bmatyasb module to implement tidyplot see here:
#3083

@bmatyasb
As this is essentially now a duplicate, we could close this one?