JMP Wish List

☑ cool new feature ☑ could help many users! ☐ removes a „bug“ ☐ nice to have ☐ nobody needs it   #myTop10_2022   If a user changes the name of a column, references to the column in table scrips are (sometimes) automatically corrected, references in user scrips unfortunately not: Column() vs As Column() vs datable:column vs dt:As name("column")    So, column name changes are very dangerous: there is always the risk that a jsl code doesn't work anymore after the change - like illustrated in the example below. JMP detects that a noc-existing column is requested and throws an error:   My wish: Instead of throwing an error, JMP could open a window and provide a list of column names. Then the user can select a replacement column. There should also be a toggle box, allowing the user to decide if the JSL code should be automatically adjusted wit the new column name. The new behaviour could be enabled/disabled in the preferences menu such that users which got used to error message still have the choice to get it.     NamesDefaultToHere(1); dt = Open("$SAMPLE_DATA/Big Class.JMP"); Data Table( "Big Class" ):age << Set Name( "age_" ); s = dt << runScript("Distribution"); // column names automatcally adjusted wait(2); s = dt << runScript("Set Age Value Labels"); // this script throws an error   more wishes submitted by   ... View more

What inspired this wish list request? Challenges with establishing the ability to pull data from SharePoint lists and/or files stored on sharepoint.  SharePoint is a tool that is used for managing tasks and incoming work.  There are several instances where we would like to be able to pull from these lists to use the data for metrics and dashboards.  Connecting to the lists and having the ability to pull in files (.jmp, .xlxs, .csv) that are stored on the SharePoint is very important to use.   What is the improvement you would like to see?  Something similar to what is available to Google Docs to allow easy access to import SharePoint lists and documents stored on a SharePoin   Why is this idea important?  To ensure metrics are accurate and up to date.   ... View more

What inspired this wish list request?  Losing information in column formulas when I'm troubleshooting or explaining a complex jsl is lost when things are commented out   What is the improvement you would like to see?  It's important because I like to troubleshoot column formulas with show/print commands and i don't want to have to redo all that work every time if i save time/overhead by commenting out lines when they are not immediately needed   Why is this idea important?  It's important because column formulas can become complex and it's important to have good tracking of what a formula does as well as being able to easily troubleshoot when time time arises.   ... View more

JMP17 has new function called New Formula Column. It would be nice if it were to return the references to the newly created columns in a list, in similar manner as << New Column does for single column. This would make using this function easier   Names Default To Here(1); dt = Open("$SAMPLE_DATA/Big Class.jmp"); new_col = dt << New Column("Log2[height]", Numeric, Continuous, Formula( Log(:height, 2) )); new_cols = dt << New Formula Column(Operation("Log 2"), Columns(:height, :weight)); Show(new_col, new_cols); ... View more

What inspired this wish list request?    It seems like it would be simple to use currentdatatable() at the start of a script embedded in a table and it would refer to that table. But if the table is private, currentdatatable() will not return it. I have cases where I want to keep the table private but be able to use <<runscript() to run its script and make some behind-the-scenes updates.   An exchange I had with JMP support on that led to no good ideas on how I could reference the private owning table and  the suggestion that it could be a wish list item.   What is the improvement you would like to see?    I would like a table script to have some simple shortcut to refer to the table it is embedded in.  This could be something similar to the "self" you can reference in a script embedded in a button.     Why is this idea important?    I just made a similar request about <<setmenuscript where it's just a convenience to have a simple reference to the containing outlinebox object. But in this case I think there is no good way to have a private table's script refer to itself even though you put the script there for a reason - it likely wants to specifically operate on that table. ... View more

What inspired this wish list request?  I had a situation similar to that described in this post by @AlterEgo  and @jthi :  Solved: Self-reference in the hot-spot script ("Set Menu Script"). Is it possible? - JMP User Community and was surprised there is no simple way to refer to the "self" outline box the scripts are attached to.  Yes, it's possible to use a robust reference to that outline in the embedded script (as shown in that thread) but it seems overly complicated. And that's not reflected well in the scripting index where the example just uses the same variable inside the embedded script in a way that seems a recipe for disaster.   What is the improvement you would like to see?    Something like the "self" that's built-in when adding a script to a button just like @AlterEgo  was hoping for.   Why is this idea important?    Ease of use.  And, as I noted, I think the scripting index example is misleading because it doesn't want to suggest you need to eval(evalexpr()) to use the function.   ... View more

I'm currently working on some larger JMP-scripting projects where version control is a must. With JMP14 I started to work with projects.    Committing & pushing my code to github manually is a bit of a pain. It would be great to have some very basic functionality like commit/push/pull integrated in JMP projects. ... View more

From time to time I have to look matches from long strings and usually regex is the method I will (would) use for. Below is an example where I try to find all words starting with f Names Default To Here(1); str2 = "which foot or hand fell fastest"; show(Regex(str2, "f[a-z]+")); // wanted result {"foot", "fell", "fastest"} This will return me only the first match ("foot"). I would request a flag to be added, which could be used to find all possible non-overlapping occurrences, in similar manner as Python's re.findall works.   ... View more

As you know, many of data scientist use XGBoost, LightGBM, and CatBoost (gradient boosting decision tree) to solve their problems by using Python. Could you add these modelings in the decision tree menu (JMP) as well? If it can be provided by JMP Script Language (JSL), it will be great as well. Thank you. : )   Reference: https://papers.nips.cc/paper/6907-lightgbm-a-highly-efficient-gradient-boosting-decision-tree.pdf         ... View more

☐ cool new feature ☑ could help many users! ☑ removes a „bug“ ☐ nice to have ☐ nobody needs it   Workflow Builder just records part of the steps which a user performs in Jmp. Especially commands from Add-ins are not recorded.   My wish: Enable Workflow Builder to record as many actions as possible. Options: Provide an easy way for Add-In Programmers to make the add-ins "suitable" to be recorded via Workflow Builder. add a functionality to workflow builder to record the actions of existing add-ins (without adjustment within the Add-In) add the steps as Main Menu("Add-Ins: command") - i.e. not fully automatic, but with some user interaction add a dummy step to the workflow idicating that Workflow Builder was not able to grab an action (then the user can decide if this action was important and just in case try to re-do it using an official function)   https://community.jmp.com/t5/Discussions/Workflow-Builder-Add-Ins/m-p/573991/highlight/true#M78301       more wishes submitted by   ... View more

Currently depending on the size of data / data structure which I want to get the uniques from I use following: Data table column values Associative Array (if I'm lazy, but gets slow with lots of values) Summarize (a bit better, also slows down) Summary table and get values (I can get counts this way if I want to) Distribution platform and get values from Report layer (nice thing with this, is that I can get other statistics at the same time and it is fast) Lists Associative Array Insert values to table and use methods mentioned above My suggestion is to add specific function to get unique values from from data tables and lists in a list/matrix. It could be also data structure like Python's Set (and if were to be data structure  A set is an unordered collection with no duplicate elements. Basic uses include membership testing and eliminating duplicate entries. Set objects also support mathematical operations like union, intersection, difference, and symmetric difference. ... View more

☐ cool new feature ☐ could help many users! ☐ removes a „bug“ ☑ nice to have ☐ nobody needs it     In Mathematica, one can use "%" as a placeholder for the last returned value. Could you please implement something similar in JSL? Could help during debugging: Run single lines of the original code, add some test code important: don't mess up the original code lines, but instead: get access to the return values via the placeholder . Aditional benefit: work gets focused on "forward direction": Ah, there is some returned value, how to continue and get most out of it ... instead of  How do I have to change the previous code to get access to the returned value.     Names Default to Here(1); dt = Open( "$SAMPLE_DATA/Big Class.jmp" ); dt << Graph Builder( Size( 437, 413 ), Graph Spacing( 4 ), Variables( X( :height ), Y( :weight ), Overlay( :sex ) ), Elements( Points( X, Y, Legend( 1 ) ), Smoother( X, Y, Legend( 2 ) ) ) ); // during debugging: gb = %;(gb << XPath( "//AxisBox" )) << select;// or even better: gb << get XML(); New Window("xml", %) https://community.jmp.com/t5/Discussions/placeholder-for-the-last-returned-value/m-p/568699         more wishes submitted by   ... View more

When I write code, I try to make it easy to read, understand, and maintain.  I write comments.  I indent code blocks.  I use line breaks to spread the code out over multiple lines.  I try to write code with a consistent style.  However, I frequently find that JMP reformats my code without asking.  It simply decides to obliterate all the effort that I put into good coding practices.   For example, I write a script and attach it to a data table.  Then I right-click on the script and press "Copy" and then right-click in the script window and press "Paste".  The copy of the script does not contain any of the comments or code block indentions that were in the original script.   Another example:  I create a formula column on a data table.  Inside of the formula editor block, I write some JSL code to perform an iterative calculation (this is really cool. I just figured out you could do this!).  Of course I want to use comments and whitespace in this code, like I would in literally any code that I write anywhere.  Well, after I hit enter, the comments are invisible until I double-click on the formula editor block.  But they do exist.  Until I click OK and JMP evaluates the formula.  Next time I open the formula, comments and whitespace are gone.   It gets worse.  I have multiple scripts attached to my table that all reference a certain column.  Then I rename that column.  All of these scripts are updated with the new column name.  But in doing so, they are all "reformatted" by the JSL parser and all of my comments and whitespace are destroyed.  ಠ_ಠ   What is the point of even having comments in the JSL language if JMP is going to immediately wipe them out at the first opportunity?  Am I supposed to edit my JSL scripts in a less-hostile text editor (i.e. Notepad++) and store the readable versions as separate files?  I just don't understand why it has to be this way.  Please leave my comments alone!   Note: @Cwalla reported the same issue in the forums here  ... View more

What inspired this wish list request?  JMP has multiple different ways to indicate if values are empty, missing or null and functions/operations might return different types when they have no values when compared to when they do have (and in Python the "pythonic" way would be just to check value in If-statement as the objects will be false Built-in Types/Truth Value Testing).   What is the improvement you would like to see? More universal JSL function to check if value is empty/missing/null/has no value. Below is one example for the function (hopefully others have additional ideas): Names Default To Here(1); Is None = function({obj, keep_zero = 1}, {Default Local}, /*"""Attempt to check if object is None (Empty(), ., 0, "", {}, [], [=>], [](0,1)... Args: obj: object to check Returns: int: 0 if non-zero and 1 if zero Examples: >> show(Is None([1 => 2])); Is None([1 => 2]) = 0; >> show(Is None(Associative Array())); Is None(Associative Array()) = 1; """*/ cur_type = Type(obj); return( If(Contains({"List", "Matrix", "Associative Array"}, cur_type), // check items in list, matrix and associative array !N Items(obj); , cur_type == "String", !Length(obj); // check string length, Is Missing() should also work , Contains({"Integer", "Number"}, cur_type), If(IsMissing(obj), 1 , keep_zero & obj == 0, 1 , 0 ); , cur_type == "Empty", 1; , 0; ); ); ); Print("Zeros:"); show(Is None([1 => 2])); show(Is None({"a"})); show(Is None([1])); show(Is None(1)); show(Is None("a")); Write("\!N"); Print("Ones:"); show(Is None(Associative Array())); show(Is None({})); show(Is None([])); show(Is None(0)); show(Is None("")); show(Is None(.)); show(Is None(Empty()));   Why is this idea important? Makes it easier to build comparisons without guessing which check should be implemented this time (IsMissing(), Is Empty(), N Items() > 0, Zero Or Missing()...) ... View more

What inspired this wish list request?  Variables are used in scripts. Currently the scripter has always write each variable name in to the script.   What is the improvement you would like to see?  It would be better if there is an autofill function, which suggests the scripter at the first letters of the variable name a list of variable names already existing in the script.     Why is this idea important?  It's important to reduce the time for scripting and have more time for solving problems :-).   @martindemel  JMP 16.2.0 ... View more

Could you please provide a function which takes as input: A jsl script The jsl version in which the script was written An provides a list of changes to all used functions as output  ... View more

What inspired this wish list request?  A Community Discussion item came in that had the need to separate into columns, each individual character in found in one of the columns in a data table.      What is the improvement you would like to see?  I would like to enhance the Text to Columns function to be able to specify an empty delimiter, which would indicate to the function to use each individual character as a new column   Why is this idea important?  It adds nice functionality to the function and it also makes the function more consistent with other JLS functions, such as the Words function.   ... View more

Some built in column formulas like Col Mean() are very fast because they only calculate expensive values once and return or modify them for each row.  With custom formulas this is harder to achieve because each row is evaluated separately.  Sometimes using a separate column to store intermediate values can speed things up (for example storing the row indexes that meet a certain criteria, instead of re-calculating them every row), but that gets messy and hard to follow. If there were some concept of a column context such that the evaluation of one row could access intermediate values calculated in the previous row, then a script could calculate results for all rows at once and then quickly return them and pass the intermediate results to the next row. The formula would also need access to the whole column instead of just the current row, so the column reference should be passed unevaluated. ... View more

☐ cool new feature ☑ could help many users! ☑ removes a „bug“ ☐ nice to have ☐ nobody needs it   The scripting index is a great source of knowledge and the examples are very useful to understand what the functions do and how they can be used (with some code context around the actual function).   On the other hand, for a user who wants to find a function for a specific application, the functionality of Scripting Index feels a bit like: If you don't know the correct function name, I cannot give you the correct info. https://community.jmp.com/t5/Discussions/Collection-of-quot-funny-quot-JSL-newbie-questions/m-p/559302/highlight/true#M77281    With some more "links" between the entries, a user can be guided through the Scripting Index and finally get to the function that he wants.   -> Please add fill "see also" with more alternatives along the idea: What possible confusions could have led users to search this function? e.g. there should be a link between Length/NItems  and  NRow/Dim   examples: https://www.mathworks.com/help/matlab/ref/length.html https://reference.wolfram.com/language/ref/Length.html compare: https://www.jmp.com/support/help/en/16.2/index.shtml#search/length    more wishes submitted by   ... View more