[Distr-commits] r1380 - in branches/distr-2.9/pkg: . distr/R distr/inst
noreply at r-forge.r-project.org
noreply at r-forge.r-project.org
Sat Nov 12 19:41:34 CET 2022
Author: ruckdeschel
Date: 2022-11-12 19:41:34 +0100 (Sat, 12 Nov 2022)
New Revision: 1380
Modified:
branches/distr-2.9/pkg/20221112ReleaseText-distr-2.9.0.txt
branches/distr-2.9/pkg/distr/R/bAcDcLcDistribution.R
branches/distr-2.9/pkg/distr/inst/NEWS
Log:
[distr] devel branch: ported from trunk
During our first attempt to submit the package to CRAN a reverse dependency check against pkg distrEllipse revealed that our new code
checking for an appearance of "/" and "^" in the parsing tree could give paste()-results of length > 1... is fixed now...
Modified: branches/distr-2.9/pkg/20221112ReleaseText-distr-2.9.0.txt
===================================================================
--- branches/distr-2.9/pkg/20221112ReleaseText-distr-2.9.0.txt 2022-11-12 18:41:08 UTC (rev 1379)
+++ branches/distr-2.9/pkg/20221112ReleaseText-distr-2.9.0.txt 2022-11-12 18:41:34 UTC (rev 1380)
@@ -1,3 +1,11 @@
+This is a resubmit after having fixed a reverse dependency issue --
+<https://win-builder.r-project.org/incoming_pretest/distr_2.9.0_20221112_175501/reverseDependencies/summary.txt>
+
+During our first attempt to submit the package to CRAN a reverse dependency check against pkg distrEllipse revealed that our new code
+checking for an appearance of "/" and "^" in the parsing tree could give paste()-results of length > 1... is fixed now...
+
+Original comment:
+
This release is triggered by a mail from CRAN/Kurt Hornik 2022-11-07, drawing our attention to some flaws in the HTML versions of our manuals, detected by the most recent version of R CMD check. These flaws are fixed, together with some updated URLs; in addition, we fix some glitches detected by elio.campitelli at cima.fcen.uba.ar,
apro at unimelb.edu.au, Santhosh.V at se.com, and vmilchev at uni-koeln.de. For details as to these changes, see the NEWS file to our package.
Modified: branches/distr-2.9/pkg/distr/R/bAcDcLcDistribution.R
===================================================================
--- branches/distr-2.9/pkg/distr/R/bAcDcLcDistribution.R 2022-11-12 18:41:08 UTC (rev 1379)
+++ branches/distr-2.9/pkg/distr/R/bAcDcLcDistribution.R 2022-11-12 18:41:34 UTC (rev 1380)
@@ -110,7 +110,7 @@
i <- i + 1
trcall <- sys.call(sys.parent(i))
myc <- paste(as.list(trcall)[[1]])
- if(myc=="/"){
+ if(any(myc == "/")){
e2s <- as.character(deparse(match.call(call=trcall)$e2))
stopit <- TRUE
}
@@ -181,7 +181,7 @@
i <- i + 1
trcall <- sys.call(sys.parent(i))
myc <- paste(as.list(trcall)[[1]])
- if(myc=="/"){
+ if(any(myc == "/")){
e2s <- as.character(deparse(match.call(call=trcall)$e2))
stopit <- TRUE
}
@@ -300,7 +300,7 @@
i <- i + 1
trcall <- sys.call(sys.parent(i))
myc <- paste(as.list(trcall)[[1]])
- if(myc=="^"){
+ if(any(myc == "^")){
mc <- match.call(call=trcall)
e1s <- as.character(deparse(mc$e1))
e2s <- as.character(deparse(mc$e2))
@@ -398,7 +398,7 @@
i <- i + 1
trcall <- sys.call(sys.parent(i))
myc <- paste(as.list(trcall)[[1]])
- if(myc=="/"){
+ if(any(myc == "/")){
mc <- match.call(call=trcall)
e1s <- as.character(deparse(mc$e1))
e2s <- as.character(deparse(mc$e2))
@@ -491,7 +491,7 @@
i <- i + 1
trcall <- sys.call(sys.parent(i))
myc <- paste(as.list(trcall)[[1]])
- if(myc=="^"){
+ if(any(myc == "^")){
mc <- match.call(call=trcall)
e1s <- as.character(deparse(mc$e1))
e2s <- as.character(deparse(mc$e2))
Modified: branches/distr-2.9/pkg/distr/inst/NEWS
===================================================================
--- branches/distr-2.9/pkg/distr/inst/NEWS 2022-11-12 18:41:08 UTC (rev 1379)
+++ branches/distr-2.9/pkg/distr/inst/NEWS 2022-11-12 18:41:34 UTC (rev 1380)
@@ -23,6 +23,9 @@
in this deparsing within a try()-catch did not work any longer when called from within a function;
as way out, we now have code which no longer needs the try()-catch and instead climbs up the
parsing tree and searches for the first occurrence of "/" resp. "^" and deparses this
+ ... in the end this was more complicated than expected, as the parsing tree could give
+ paste()-results of length > 1... (only popped up in a reverse dependency check through
+ package distrEllipse)
under the hood:
+ triggered by an email by Santhosh V <Santhosh.V at se.com>, we added a patch to be more careful when producing
More information about the Distr-commits
mailing list