gregleleu

Also tried the following approach in my case – nested morph to subgraph filtering on edges and then on nodes – by making my own morpher. In theory the morpher...

Get it to work by manually protecting the index columns. Not very pretty. ``` to_my_morpher % mutate(.tidygraph_node_index_protect = .tidygraph_node_index) %>% activate(edges) %>% mutate(.tidygraph_edge_index_protect = .tidygraph_edge_index) %>% ## doing the morph...

Sure. A more general solution could be added to the package, by merging the two approaches: * When morphing on a morphed graph, passing the morpher to a purrr::modify_depth, with...

Also posted here: https://github.com/rstudio/shiny/issues/1728 I added a repro and tested a few ideas -> I think the link to the leaflet map is broken when the DOM element is recreated...

After a bit more digging: The "shiny-bound-output" class is present when the htmlwidget factory is called the first time around, but the second time the class only gets added _after_...

Related to https://github.com/sparklyr/sparklyr/issues/2009 dbplyr translation is not perfect, one way to force it to do the right thing is to use "collapse" (in the dbplyr sense https://dplyr.tidyverse.org/reference/compute.html) ``` sdf_test %>%...

The transaltion for `sdf %>% summarise(distinct_count = n_distinct(num[id == "1-A"]))` creates nulls, so the distinct function sees 2 distinct values ``` sdf %>% summarise(distinct_count = n_distinct(num[id == "1-A"])) %>% show_query()...

You're instantiating a new object, so you'll want to use `invoke_new`not `invoke_static`. As per the java documentation the constructor needs two parameters (not sure why). Hence you're looking at something...

OK, a bit of a tricky one: the translation of the quotes is not perfect, so a workaround is to wrap the pattern in `sql(...)`so that there is no translation....

Using `scale_y_continuous(labels = scales::percent, breaks = scales::pretty_breaks(n = 10))` does the trick for me