[wbuchanan]

Leonardo Guizzetti
isn’t this already possible if you drag one of the tabs outside of the window to create two windows for the files? Just trying to understand your ask a bit more clearly.

[Leonardo Guizzetti]

Two more suggested features for the do-file editor:

1. Split-view tabs/displays that allow the user to view and edit two separate files concurrently, without the need to switch tabs as is currently done.
2. Split-view tabs/displays that allow the user to view and edit the same file concurrently. This one in particular can be handy when programming -ado- or -mata- files.

[Sergiy Radyakin]

(After seeing v16.1 update of January 21 (Windows)):

Can we have the following improvements to the do-editor?

1. *.class filter to be added to the list of recognized file extensions in the do-editor? (I know this is quite exotic, but still less exotic than the supported business calendar files)
2. *.class recognized for Stata's ado-language syntax markup.
3. Syntax highlighter to allow for NONE, to quickly switch off syntax highlighting all together for the current file. There is a General style (alongside the python, ado, and markdown), but no way to apply it through the language menu (as far as I can see), instead one would need to switch off the option "Syntax Highlighting" - which affects all files that are open in the do-editor. (working with ini-files, text files, ascii-data files, etc).
4. Fold All/Unfold All callable via keyboard shortcuts.
5. Option to open files (with recognized extensions) with code-folding to fold all code.
6. This one I am not sure is possible, but still will try: I am mixing ado and python code in the same file. Naturally I want to have syntax highlighting and code folding behave appropriately for the corresponding portion of the code. But only one prevails, as the language setting is one for the whole file. I can survive with code folding not working, but the comments are not rendered correctly (e.g. text with // is rendered as a comment in python code, which is incorrect). The code folding recognizes both python and ado folds in python mode, but only recognizes ado folds in ado mode. Not sure I am getting it right, but there is definitely something confusing there.
7. a button (and a hot key) to switch from the do-editor to the Stata output window (and possibly back). Worked with a single monitor recently. This could have saved me some time clicking around.
8. I find it virtually impossible to program python without seeing the indentation and whitespaces. I'd vote for that to be the default for that language (or an option in the style definitions that the users could customize).
9. support for latex syntax highlighting perhaps? Could be a low hanging fruit, since the highlighter is already available for Scintilla/SciTE.
10. Export of source file with syntax highlighting (as RTF, HTML, LaTeX, or something else that is possible to manipulate otherwise). Currently have to do screenshots in PNG if want to show a fragment of the Stata code with syntax highlighting in a presentation.

And is there any chance we can see zooming with the Ctrl+Mousewheel in the output window, viewer and data editor windows? (this works in the do-editor and is very useful)

Thank you, Sergiy

[Fahad Mirza]

wbuchanan surely there is no standardized definition for modern looking graph but students/people alike have been thought to think that visualization looks better in all software other than Stata. So its like the first software that comes to their mind is R. That is mainly because it is probably very well marketed that way and also people are generally more aware of ggplot. Also, it goes without saying that brewscheme is awesome i found out about it when searching myself few months back. Also, for a new user, you want to excite them with how good the software visually depicts information out of the box to keep them hooked onto it. That is just my tuppence worth on the topic really, and these are my personal experience and opinion regarding it, i can be terribly wrong too.

[Mohammad Mansour]

Hello, I am trying to use collapse with MI data. I am imputing categorical variables (dependent variable is a count). I am trying to aggregate the data across all five imputations, but collapse wont work. Any idea on hoe I can perform collapse manually using mi xeq? I want to aggregate data by time and a stateID and by time alone.

[Leonardo Guizzetti]

Related to #427, a 1:m merge for frames would also be useful.

[wbuchanan]

One feature that I would imagine would be not terribly difficult to implement is adding an append/union type function to frames. If it is possible to implement the SQL equivalents of UNION (w/o duplicated records) and UNION ALL (w/duplicated records) that would be awesome.

[wbuchanan]

Fahad Mirza
I’m not sure there is a single standardized definition of what constitutes a modern looking graph, but I would recommend taking a look at https://wbuchanan.github.io/brewscheme which could be a solution for those looking to create schemes that suit their aesthetic needs.

[Fahad Mirza]

a really minor request. it will be nice to have modern style schemes for visualization as default. i come across many people who apparently dont like the default stata visual output because they believe thay this is the only way it is displayed.

[ryan melo]

More options for metanalysis.
One arm proportion for example

[Niamh Meagher]

Maybe it's already available and I just can't figure out how it's implemented in Stata, but would be great to see mixed effects zero-inflated count regression models (e.g. mixed-effect ZIP and mixed-effect ZINB).

[Leonardo Guizzetti]

Re #421, from someone who uses Python very little these days, what exactly could be improved with respect to Stata's integration of Python?

[Jonathan Afilalo]

Better python integration! (i.e. easier, documented, more flexible, more powerful)

[John Mullahy]

#419This is really useful. On a Mac: shift-command-uparrow

[Bjarte Aagnes]

#418 Ctrl+Shift+Home Extend selection to start of document. https://www.scintilla.org/SciTEDoc.html