Not showing the last-modified date

[cgoo4]

Bug description

Publishing a qmd file with format hugo, the last-modified date is showing the published date ("March 19, 2022") rather than the date the qmd was last updated (October 6, 2022).

---

date: 2022-03-19
date-format: long
format: hugo
lastmod: last-modified

The whole qmd file is available here.

The whole output is published here, with both dates printed in the details twistie at the foot of the page per the above snippet.

lastmod: 'r Sys.Date()' in my other Rmarkdown files works okay.

Update 25/11 -- I've rebuilt the site with the html format to make fuller use of all the Quarto features including code-linking and date-modified etc.

RStudio IDE Version running on Mac OS:

RStudio 2022.07.1+554 "Spotted Wakerobin" Release (7872775ebddc40635780ca1ed238934c3345c5de, 2022-07-22) for macOS
Mozilla/5.0 (Macintosh; Intel Mac OS X 12_6_0) AppleWebKit/537.36 (KHTML, like Gecko) QtWebEngine/5.12.10 Chrome/69.0.3497.128 Safari/537.36

Checklist

• Please include a minimal, fully reproducible example in a single .qmd file? Please provide the whole file rather than the snippet you believe is causing the issue.
• Please format your issue so it is easier for us to read the bug report.
• Please document the RStudio IDE version you're running (if applicable), by providing the value displayed in the "About RStudio" main menu dialog?
• Please document the operating system you're running. If on Linux, please provide the specific distribution.

[dragonstyle]

Thanks for reporting this. The issue here is that we preserve the document metadata 'as is' in the case of formats like Hugo. We do this so that we don't create side effects or mutations in the final metadata - you can trust that what you wrote in the qmd is what will make its way to the rendered markdown. But this has the unfortunate effect of meaning that features that 'resolve' that markdown (for example our date handling or author and affiliation resolution) don't function in these cases.

We looked a little at changing this approach, but I think there will be significant risk of issues where we inadvertently mutate metadata and cause issues for downstream systems. I'm leaving this open for us to consider again.