Feat: options to customise outputs in ppt power point (#972, compatible with #1033 changes) and output/zip file validation in modal

[bachapman]

Issue

Closes #972, iterates towards #1033 solution with additional analysis slide inclusion but lack of design features as there is a lot of spillover.

Description

• Slide content customisation: a new "Customise Slide Contents" modal lets users
  pick which slide sections (individual plots, PK parameters, summary statistics,
  additional analysis) and which PK parameters appear in exported PPTX/QMD decks
  before downloading.
• Additional analysis outputs: additional_analysis data frames are threaded
  through create_pptx_dose_slides() and create_qmd_dose_slides() and surfaced as
  selectable slide sections. Formatting is currently minimal (plain tables) — full
  styling is tracked in Add Additional Analysis slides to exported PPT/QMD results decks #1033.
• ZIP export validation: the export modal now blocks submission when no results are
  selected or a result type is checked but its format list is empty; errors render as a
  centred red pill badge above the footer with the "Export ZIP with Results" button
  disabled until resolved.
• UX polish: two-step modal flow (export options → slide customisation → download),
  visible download button in the confirmation step, easyClose = FALSE throughout,
  cancel and back navigation.

Definition of Done

The requirements for the feature to be complete:

• When downloading the zip file, have a pop-up window where you can select which plots/sections of the PowerPoint are needed (with select of all of them by default)
• Make sure that only what is selected appears in the Powerpoint

How to test

Test Plan

• Open ZIP export modal — all results pre-selected, no validation error, button enabled
• Deselect all tree nodes → red badge "Select at least one result" appears, button
  disabled; re-select any item → badge clears, button re-enables
• Clear all graphics formats with a plot node selected → badge shows graphics error;
  restore a format → clears
• Clear all table formats with a table node selected → badge shows table error
• Clear all slide formats with results_slides selected → badge shows slide error
• Trigger two errors simultaneously → both messages appear as a bulleted list inside
  the badge
• With results_slides selected, click "Export ZIP with Results" → "Customise Slide
  Contents" modal opens; deselect sections and parameters, click Export → ZIP contains
  only selected content
• Without results_slides (or no slide formats), click Export → "Ready to Export"
  confirmation modal, Back returns to export modal with state preserved
• Verify additional_analysis sections appear in the slide-sections tree when
  present in session data

Contributor checklist

• Code passes lintr checks
• Code passes all unit tests
• New logic covered by unit tests
• New logic is documented
• App or package changes are reflected in NEWS
• Package version is incremented
• R script works with the new implementation (if applicable)
• Settings upload works with the new implementation (if applicable)

Notes to reviewer

Let me know thoughts on the current iteration of this and I'll bump the version.

[bachapman]

Quick update: just double checked the PowerPoint and while users can select which parameters to include/remove from tables, slide type inclusion/exclusion (e.g., including only summary plots but not individual plots) is currently not working.

[bachapman]

@Gero1999 , let me know if this is in line with the utility you envisioned. Code might need to be refactored to ensure future utility and expansion is easy (and to pass tests)

[Gero1999]

This functionality is exactly what I was thinking, amazing work! The point would be now to refactor a bit as you said and that is it 💪🏽

[Gero1999]

Hey thanks for the good job!

Functionally speaking seems great, I tested again, using this time DrugA, Metab-DrugA + Dose1, Dose2, Dose3 + SERUM, URINE. I then tried different combinations of PPT parameters/slide-types:

• PPTX is fine for me. Displays are correct and links keep working.
• QMD needs small fixes to be tested. The problem however is from main, we unnoticed in some recent implementation the removal of the input file needed for the QMD to run in the ZIP. I fixed that locally during my testing (see this comment). The content seems fine and well structured, although the display is not perfect in all slide types, see the next images:

---

Regarding the code, there is a lot of refactoring. But I reviewed carefully and I think all seems in good shape, also thanks a lot for making the tests!

However, because there are many changes that in-practice can cause unnoticed breaks I would like that for this case we add another reviewer as well to test that the functionality is indeed fine using different inputs once you make your changes, perhaps @js3110 can add herself if she has time (no need to check the code though).

[bachapman]

@Gero1999 implemented your feedback on ensuring that the rda is created when qmd is generated.

[js3110]

thanks for the changes, looks great now!

I can't get the QMD file to work, when i try to render I get the following error:

Error running Lua:
YAML parse exception at line 4, column 0,
while scanning for the next token:
found character that cannot start any token
stack traceback:
	...RESOUR~1\app\bin\quarto\share\pandoc\datadir\readqmd.lua:157: in function 'readqmd.readqmd'
	...dio\RESOUR~1\app\bin\quarto\share\filters\qmd-reader.lua:13: in function 'Reader'

Not sure if thats just on my end or a global issue, so I think I will ask @Gero1999 if he is able to render it, and if yes then i am happy to approve, as I don't use/am not super familiar with qmd files anyway

[bachapman]

thanks for the changes, looks great now!

I can't get the QMD file to work, when i try to render I get the following error:

Error running Lua:
YAML parse exception at line 4, column 0,
while scanning for the next token:
found character that cannot start any token
stack traceback:
	...RESOUR~1\app\bin\quarto\share\pandoc\datadir\readqmd.lua:157: in function 'readqmd.readqmd'
	...dio\RESOUR~1\app\bin\quarto\share\filters\qmd-reader.lua:13: in function 'Reader'

Not sure if thats just on my end or a global issue, so I think I will ask @Gero1999 if he is able to render it, and if yes then i am happy to approve, as I don't use/am not super familiar with qmd files anyway

I was able to render on my end. For reference, I am using quarto version 1.7.32. I think the issue is with how the title is parsed in the zip-utils function.

aNCA/inst/shiny/functions/zip-utils.R

Line 419 in ced3be3

slide_title <- if (pn == "") "NCA Results" else paste0("NCA Results\n", pn)

My hypothesis is for some people and their environment/settings, the quarto document can render the new line but for others it cannot. @js3110 , can you test replacing that new line with a dash? If that still provides an error, I'll keep it with a new line and we'll need to debug further. If that fixes the error, I can package that fix with this update.

[js3110]

I was able to render on my end. For reference, I am using quarto version 1.7.32. I think the issue is with how the title is parsed in the zip-utils function.

aNCA/inst/shiny/functions/zip-utils.R

Line 419 in ced3be3

slide_title <- if (pn == "") "NCA Results" else paste0("NCA Results\n", pn)

My hypothesis is for some people and their environment/settings, the quarto document can render the new line but for others it cannot. @js3110 , can you test replacing that new line with a dash? If that still provides an error, I'll keep it with a new line and we'll need to debug further. If that fixes the error, I can package that fix with this update.

@bachapman I tried changing it to a dash but unfortunately I still get the same error. I've tried debugging for a bit but have had no success :(

[Gero1999]

Hey @bachapman so for me is also breaking the QMD now with the same error as @js3110 . I suspect it might be related with the column structuring getting broken when custom selections are chosen... But I cannot say for sure what it is

[bachapman]

I'll be out of office until Tuesday so I can take more time to debug, fix, and/or revert qmd functionality once I get back, but for reference are you coding on Windows or Linux (I'm using the latter) and what version of quarto are you using? @js3110 @Gero1999

[Gero1999]

for reference are you coding on Windows or Linux (I'm using the latter) and what version of quarto are you using?

Windows with quarto_1.5.1, I will just take note of Linux though later to see if it works

I'll be out of office until Tuesday so I can take more time to debug, fix, and/or revert qmd functionality once I get back

Enjoy the time off, see you soon!

[js3110]

for reference are you coding on Windows or Linux (I'm using the latter) and what version of quarto are you using?

Windows with quarto_1.5.1, I will just take note of Linux though later to see if it works

I'll be out of office until Tuesday so I can take more time to debug, fix, and/or revert qmd functionality once I get back

Enjoy the time off, see you soon!

+1

[Shaakon35]

Looks good :)

[Shaakon35]

@bachapman Any news for this PR?

Changes have been made since most recent review. I have resolved the requested changes that I can see. Re-request with most recent version as necessary.

[bachapman]

@bachapman Any news for this PR?

Apologies, took a little bit longer to get back to this PR than anticipated.

I have updated the code to be able to merge into main. In addition, I added handling that attempts to fix the quarto rendering issue experienced by @js3110 and @Gero1999 with how titles/lines are parsed. Let me know if the issue still persists.

Otherwise, I have gone through the test plan myself and everything works as intended in my environment, to be replicated in others. Happy to make changes as things are brought up in review

[Gero1999]

my current problem is this with the qmd:

processing file: results_slides.qmd
                                                                                                             
output file: results_slides.knit.md

pandoc 
  to: revealjs
  output-file: results_slides.html
  standalone: true
  wrap: none
  default-image-extension: png
  html-math-method:
    method: mathjax
    url: >-
      https://cdn.jsdelivr.net/npm/mathjax@2.7.9/MathJax.js?config=TeX-AMS_HTML-full
  slide-level: 2
  toc: true
  toc-depth: 1
  variables: {}
  
metadata
  link-citations: true
  width: 1050
  height: 700
  margin: 0.1
  center: false
  navigationMode: linear
  controlsLayout: edges
  controlsTutorial: false
  hash: true
  history: true
  hashOneBasedIndex: false
  fragmentInURL: false
  transition: none
  backgroundTransition: none
  pdfSeparateFragments: false
  lang: en
  auto-stretch: true
  title: NCA Results S1
  '::': null
  theme: default
  scrollable: true
  smaller: true
  
Error running Lua:
YAML parse exception at line 8, column 0,
while scanning for the next token:
found character that cannot start any token
stack traceback:
	...a\Local\Programs\Quarto\share\pandoc\datadir\readqmd.lua:170: in function 'readqmd.readqmd'
	...pData\Local\Programs\Quarto\share\filters\qmd-reader.lua:13: in function 'Reader'
WARN: Should not have arrived here:
ERROR: [non-error-thrown] undefined

can you also check this @js3110?

[Gero1999]

hey @bachapman actually the core team reunited and we concluded that as an exception it will benefit merging this one even if the QMD is not rendering...

So opened a new issue to address this problem apart (#1167) and approve this one. The rest seems to work perfectly, great job!

[bachapman]

hey @bachapman actually the core team reunited and we concluded that as an exception it will benefit merging this one even if the QMD is not rendering...

So opened a new issue to address this problem apart (#1167) and approve this one. The rest seems to work perfectly, great job!

Thanks! I'll merge this PR today. Apologies for the qmd bug. Note, I tested on both linux and mac and for both, the qmd was able to render without issues. For #1167, I can serve as an extra reviewer to make sure that the expected behavior continues for both OS's.