Process 1.2.0 revdep issues

[DavisVaughan]

• archiveRetriever

  • False alarm. Spurious HTTP test failure unrelated to us.

• causact

  • replace_na() issue, will be fixed by DiagrammeR
  • Only call replace_na(replace = "") on character columns rich-iannone/DiagrammeR#448
  • Unfortunately seems to have another bug related to replace_na(). This is related to ifelse() returning a logical result if the input is purely logical, then they try and replace missings in a logical vector with a character string https://github.com/flyaflya/causact/blob/df48f06a1f2d2f4c8a96f5289af043612dbd4638/R/dag_diagrammer.R#L217
  • Avoid needing replace_na() entirely flyaflya/causact#40

• cogmapr

  • replace_na() issue
  • Sent an email with a patch, but it was applied incorrectly so the problem still persists.
  • Followed up with another email on 2021-12-21
  • Fix on CRAN

• covid19br

  • See pivot_longer() ptype arg updates and S3 extensions #1291 (through sf, which it uses in a vignette and calls its pivot_longer() method)
  • Possibly also issues with sf's generic-method signature consistency for pivot_longer()

• covidregionaldata

  • Expected an exact column ordering from the result of complete(), but now grouping columns are pulled to the front
  • Don't rely on exact column ordering when grouped data is involved epiforecasts/covidregionaldata#445
  • Additional issue about them calling complete() on a grouping column Don't rely on exact column ordering when grouped data is involved epiforecasts/covidregionaldata#445 (comment)

• crosstable

  • Calls replace_na(NULL, "NA") and vec_equal_na(NULL) is an error Should vec_equal_na(NULL) return logical(0)? r-lib/vctrs#1494
  • Return logical() from vec_equal_na(NULL) r-lib/vctrs#1497
  • We should probably just patch this in replace_na.default() temporarily
  • Patch replace_na(NULL) until vctrs fix is released #1293
  • Found an additional issue after this that was an actual bug in their package
  • Replace missings by iterating over list of level vectors DanChaltiel/crosstable#6

• dbplyr

  • See pivot_longer() ptype arg updates and S3 extensions #1291
  • Possibly also issues with generic-method signature consistency for pivot_longer() and pivot_wider()

• DiagrammeR

  • replace_na() issue
  • Only call replace_na(replace = "") on character columns rich-iannone/DiagrammeR#448

• DSSAT

  • replace_na() issue
  • Convert $width to character before calling replace_na() palderman/DSSAT#35
  • Never made it on CRAN? Convert $width to character before calling replace_na() palderman/DSSAT#35 (comment)

• emayili

  • Related to hoist() when the elements are scalar objects. We now call vec_size() rather than length() on the elements, so this fails on the scalar elements. Sent in a PR that recommends either unclassing or explicitly inheriting from "list".
  • Strip scalar "header" class before hoist()ing datawookie/emayili#111

• eudract

  • Was completing on grouping columns unnecessarily
  • Don't complete on existing grouping variables DavisVaughan/eudract-1#1
  • Broken package detected in tidyr 1.2.0 revdeps shug0131/eudraCT#18

• ftExtra

  • Improper use of names_ptypes in ftExtra:::fill_header() where they set names_ptypes = integer() which should never have worked before now
  • Remove broken names_ptypes usage atusy/ftExtra#66

• heuristicsmineR

  • replace_na() issue, will be fixed by DiagrammeR
  • Only call replace_na(replace = "") on character columns rich-iannone/DiagrammeR#448

• iotables

  • There seem to be quite a few general issues with the package. I can't reproduce the exact problem, and I'm going to say that this is not tidyr's fault. It seems to be a failure in the vignette environmental_impact.Rmd which is also failing on CRAN in some cases. That vignette downloads a huge amount of data so its possible that is the issue.

• lcsm

  • Turned out to be a broom issue related to calling pivot_wider(id_cols = terms, names_from = terms) (i.e. duplication in the id_cols and names_from). This used to silently remove the id_cols, but now is an error because this is almost always unintentional.
  • Remove duplicate id_cols in lavaan glance() method tidymodels/broom#1067

• PheVis

  • Basically just this issue Should complete() and expand() disallow accessing the group columns? #1299 (the groups are wrong after completing because of this issue). But we should recheck after fixing this.

• processanimateR

  • replace_na() issue, will be fixed by DiagrammeR
  • Only call replace_na(replace = "") on character columns rich-iannone/DiagrammeR#448

• r2dii.analysis

  • Basically Should complete() and expand() disallow accessing the group columns? #1299. We should review again after fixing that.

• RavenR

  • replace_na() issue, will be fixed by DiagrammeR
  • Only call replace_na(replace = "") on character columns rich-iannone/DiagrammeR#448

• simpr

  • Seems like the same issue as sparklyr with the pivot-wider method spec changes
  • Prepare for the updated pivot_wider() signature statisfactions/simpr#68

• skater

  • replace_na() issue
  • https://github.com/DavisVaughan/skater/pull/1
  • Sent an email on 2021-12-21 because the package isn't on GitHub

• sparklyr

  • See pivot_longer() ptype arg updates and S3 extensions #1291
  • Possibly also issues with generic-method signature consistency for pivot_longer() and pivot_wider()
  • Prepare for the updated pivot_wider() signature sparklyr/sparklyr#3215

• ssimparser

  • Interesting usage of group-by-all passed to expand, where they then used the group columns
  • Use summarise() rather than expand() cran/ssimparser#1
  • Had to send an email, not on GitHub as far as I can tell

• swissdd

  • This looks like a flaky test unrelated to us

• SWMPrExtension

  • Was calling complete on a grouped data frame when they really didn't need to
  • Drop groups before completing NOAA-OCM/SWMPrExtension#54

• tabshiftr

  • Various issues related to pivot_wider() and pivot_longer() that were resolved by matching arguments by name rather than by position
  • Fix argument specification in pivot_wider() and pivot_longer() calls luckinet/tabshiftr#1

• tidygate

  • Calls replace_na(<chr>, 0)
  • Use a character "0" as the replacement value stemangiola/tidygate#15
  • Sent a follow up message encouraging a CRAN submission on 2021-12-21
  • Responded on 2021-12-21 saying they'd release today

• tidyseurat

  • See pivot_longer() ptype arg updates and S3 extensions #1291
  • Possibly also issues with generic-method signature consistency for pivot_longer() and pivot_wider()

• tidytidbits

  • Calls replace_na(<dbl>, "missing")
  • Convert var_name column to character before replacing NAs DavisVaughan/tidytidbits#1
  • Had to send an email linking to that PR, as they are not on GitHub (2021-11-18)
  • Send a follow up email on 2021-12-21
  • Got a response on 2021-12-21 saying they will send in a patch after Christmas

• Tplyr

  • Was calling complete on a grouped data frame with all the grouping columns
  • Ungroup before completing atorus-research/Tplyr#15

• xml2relational

  • Calls replace_na(<chr>, 0)
  • Replace with "0" not 0, since the columns are always character jsugarelli/xml2relational#2
  • Sent a follow up message on 2021-12-21

• wpa

• Needed to ungroup before completing, looks like they just forgot
  *. Ungroup before completing microsoft/wpa#199

[DavisVaughan]

Round 1

## New problems (20)

|package                                        |version |error  |warning |note |
|:----------------------------------------------|:-------|:------|:-------|:----|
|[causact](problems.md#causact)                 |0.4.0   |__+1__ |        |     |
|[cogmapr](problems.md#cogmapr)                 |0.9.2   |__+2__ |        |     |
|[covid19br](problems.md#covid19br)             |0.1.3   |       |__+1__  |2    |
|[crosstable](problems.md#crosstable)           |0.3.2   |__+2__ |__+1__  |     |
|[dbplyr](problems.md#dbplyr)                   |2.1.1   |__+2__ |        |     |
|[DiagrammeR](problems.md#diagrammer)           |1.0.6.1 |       |__+1__  |2    |
|[DSSAT](problems.md#dssat)                     |0.0.4   |__+1__ |        |     |
|[emayili](problems.md#emayili)                 |0.7.0   |__+2__ |        |     |
|[ftExtra](problems.md#ftextra)                 |0.2.0   |__+2__ |__+1__  |     |
|[heuristicsmineR](problems.md#heuristicsminer) |0.2.5   |__+1__ |        |1    |
|[lcsm](problems.md#lcsm)                       |0.1.2   |       |__+1__  |1    |
|[processanimateR](problems.md#processanimater) |1.0.4   |__+1__ |__+1__  |1    |
|[RavenR](problems.md#ravenr)                   |2.1.4   |__+1__ |        |1    |
|[skater](problems.md#skater)                   |0.1.0   |       |__+1__  |     |
|[sparklyr](problems.md#sparklyr)               |1.7.3   |       |__+1__  |2    |
|[tabshiftr](problems.md#tabshiftr)             |0.3.0   |__+2__ |__+1__  |     |
|[tidygate](problems.md#tidygate)               |0.4.7   |__+1__ |        |     |
|[tidyseurat](problems.md#tidyseurat)           |0.3.0   |__+2__ |__+1__  |     |
|[tidytidbits](problems.md#tidytidbits)         |0.2.3   |__+1__ |        |1    |
|[xml2relational](problems.md#xml2relational)   |0.1.0   |__+1__ |        |     |

Round 2:

## New problems (19)

|package                                            |version |error  |warning |note |
|:--------------------------------------------------|:-------|:------|:-------|:----|
|[archiveRetriever](problems.md#archiveretriever)   |0.1.0   |__+1__ |        |     |
|[causact](problems.md#causact)                     |0.4.0   |__+1__ |        |     |
|[covidregionaldata](problems.md#covidregionaldata) |0.9.2   |__+1__ |        |1    |
|[crosstable](problems.md#crosstable)               |0.3.2   |__+1__ |        |     |
|[DiagrammeR](problems.md#diagrammer)               |1.0.6.1 |       |__+1__  |2    |
|[DSSAT](problems.md#dssat)                         |0.0.5   |__+1__ |        |     |
|[emayili](problems.md#emayili)                     |0.7.0   |__+2__ |        |     |
|[heuristicsmineR](problems.md#heuristicsminer)     |0.2.5   |__+1__ |        |1    |
|[iotables](problems.md#iotables)                   |0.4.7   |       |__+1__  |     |
|[PheVis](problems.md#phevis)                       |1.0.2   |__+1__ |        |     |
|[processanimateR](problems.md#processanimater)     |1.0.4   |__+1__ |__+1__  |1    |
|[r2dii.analysis](problems.md#r2diianalysis)        |0.1.12  |__+1__ |        |     |
|[RavenR](problems.md#ravenr)                       |2.1.4   |__+1__ |        |1    |
|[skater](problems.md#skater)                       |0.1.0   |       |__+1__  |     |
|[sparklyr](problems.md#sparklyr)                   |1.7.4   |       |__+1__  |2    |
|[tabshiftr](problems.md#tabshiftr)                 |0.3.0   |__+2__ |__+1__  |     |
|[tidygate](problems.md#tidygate)                   |0.4.8   |__+1__ |        |     |
|[tidytidbits](problems.md#tidytidbits)             |0.2.3   |__+1__ |        |1    |
|[xml2relational](problems.md#xml2relational)       |0.1.0   |__+1__ |        |     |

Round 3:

(Dropped a few from flaky revdeps and fixing #1299, but that restriction on grouping columns also broke a few more)

## New problems (21)

|package                                            |version |error  |warning |note |
|:--------------------------------------------------|:-------|:------|:-------|:----|
|[causact](problems.md#causact)                     |0.4.0   |__+1__ |        |     |
|[covidregionaldata](problems.md#covidregionaldata) |0.9.2   |__+1__ |__+1__  |1    |
|[crosstable](problems.md#crosstable)               |0.3.2   |__+1__ |        |     |
|[DiagrammeR](problems.md#diagrammer)               |1.0.6.1 |       |__+1__  |2    |
|[DSSAT](problems.md#dssat)                         |0.0.5   |__+1__ |        |     |
|[emayili](problems.md#emayili)                     |0.7.0   |__+2__ |        |     |
|[eudract](problems.md#eudract)                     |0.9.3   |__+2__ |__+1__  |     |
|[heuristicsmineR](problems.md#heuristicsminer)     |0.2.5   |__+1__ |        |1    |
|[processanimateR](problems.md#processanimater)     |1.0.4   |__+1__ |__+1__  |1    |
|[RavenR](problems.md#ravenr)                       |2.1.4   |__+1__ |        |1    |
|[skater](problems.md#skater)                       |0.1.0   |       |__+1__  |     |
|[sparklyr](problems.md#sparklyr)                   |1.7.4   |       |__+1__  |2    |
|[ssimparser](problems.md#ssimparser)               |0.1.0   |__+1__ |        |     |
|[swissdd](problems.md#swissdd)                     |1.1.4   |__+1__ |        |     |
|[SWMPrExtension](problems.md#swmprextension)       |2.1.0   |__+1__ |        |     |
|[tabshiftr](problems.md#tabshiftr)                 |0.3.0   |__+2__ |__+1__  |     |
|[tidygate](problems.md#tidygate)                   |0.4.8   |__+1__ |        |     |
|[tidytidbits](problems.md#tidytidbits)             |0.2.3   |__+1__ |        |1    |
|[Tplyr](problems.md#tplyr)                         |0.4.3   |__+2__ |__+1__  |     |
|[wpa](problems.md#wpa)                             |1.6.3   |__+1__ |        |     |
|[xml2relational](problems.md#xml2relational)       |0.1.0   |__+1__ |        |     |

2022-01-11:

• 5 related to DiagrammeR (Merged PR)
• 6 related to complete() changes (6 open PRs)
• 6 related to replace_na() changes (4 merged PRs, 1 unresponsive email, 1 unresponsive PR)
• 1 related to hoist() changes (1 merged PR)
• 1 related to pivot_longer() and pivot_wider() (1 merged PR)
• 1 sparklyr issue

Round 4

## New problems (14)

|package                                            |version |error  |warning |note |
|:--------------------------------------------------|:-------|:------|:-------|:----|
|[causact](problems.md#causact)                     |0.4.0   |__+1__ |        |     |
|[covidregionaldata](problems.md#covidregionaldata) |0.9.2   |__+1__ |__+1__  |1    |
|[crosstable](problems.md#crosstable)               |0.3.2   |__+1__ |        |     |
|[DSSAT](problems.md#dssat)                         |0.0.5   |__+1__ |        |     |
|[emayili](problems.md#emayili)                     |0.7.0   |__+2__ |        |     |
|[iotables](problems.md#iotables)                   |0.4.7   |       |__+1__  |     |
|[simpr](problems.md#simpr)                         |0.2.1   |       |__+1__  |     |
|[skater](problems.md#skater)                       |0.1.0   |       |__+1__  |     |
|[sparklyr](problems.md#sparklyr)                   |1.7.4   |       |__+1__  |2    |
|[SWMPrExtension](problems.md#swmprextension)       |2.1.0   |__+1__ |        |     |
|[tidygate](problems.md#tidygate)                   |0.4.8   |__+1__ |        |     |
|[Tplyr](problems.md#tplyr)                         |0.4.3   |__+2__ |__+1__  |     |
|[wpa](problems.md#wpa)                             |1.6.3   |__+1__ |        |     |
|[xml2relational](problems.md#xml2relational)       |0.1.0   |__+1__ |        |     |

• 1 false alarm (iotables)
• 2 related to pivot_wider() signature changes (sparklyr and simpr)
• 2 related to patches to CRAN that somehow didn't contain my PR (DSSAT, tidygate)
• 9 with PRs that are either open or are merged but not on CRAN yet

Round 5

## New problems (11)

|package                                            |version |error  |warning |note |
|:--------------------------------------------------|:-------|:------|:-------|:----|
|[covidregionaldata](problems.md#covidregionaldata) |0.9.2   |__+1__ |__+1__  |1    |
|[crosstable](problems.md#crosstable)               |0.3.2   |__+1__ |        |     |
|[emayili](problems.md#emayili)                     |0.7.0   |__+2__ |        |     |
|[iotables](problems.md#iotables)                   |0.4.7   |       |__+1__  |     |
|[simpr](problems.md#simpr)                         |0.2.1   |       |__+1__  |     |
|[skater](problems.md#skater)                       |0.1.0   |       |__+1__  |     |
|[sparklyr](problems.md#sparklyr)                   |1.7.4   |       |__+1__  |2    |
|[SWMPrExtension](problems.md#swmprextension)       |2.1.0   |__+1__ |        |     |
|[tidygate](problems.md#tidygate)                   |0.4.8   |__+1__ |        |     |
|[Tplyr](problems.md#tplyr)                         |0.4.3   |__+2__ |__+1__  |     |
|[xml2relational](problems.md#xml2relational)       |0.1.0   |__+1__ |        |     |

2022-01-20

• 1 false alarm (iotables)
• 2 related to pivot_wider() signature changes (sparklyr and simpr)
• 1 related to patches to CRAN that somehow didn't contain my PR (tidygate, which just went to CRAN this morning)
• 7 with PRs that are either open or are merged but not on CRAN yet