[Bioc-devel] error: 2 arguments passed to .Internal(ls) which requires 3

On  4 November 2014 13:41, Martin Morgan wrote:

Hi Laurent --

On 11/04/2014 01:33 AM, Laurent Gatto wrote:

Something's fishy with my R devel/Bioc 3.1 installation and I fail to
see what is wrong.

[...]

bioclite("homo.sapiens")

## bioc_mirror: http://bioconductor.org
## using bioconductor version 3.1 (biocinstaller 1.17.1), r version 3.2.0.
## installing package(s) 'homo.sapiens'
## trying url 'http://bioconductor.org/packages/3.1/data/annotation/src/contrib/homo.sapiens_1.1.2.tar.gz'
## content type 'application/x-gzip' length 1511 bytes
## opened url
## ==================================================
## downloaded 1511 bytes
##
## * installing *source* package ?homo.sapiens? ...
## ** r
## ** data
## ** preparing package for lazy loading
## error : .onload failed in loadnamespace() for 'go.db', details:
##   call: ls(envir, all.names = true)
##   error: 2 arguments passed to .internal(ls) which requires 3
## error : package ?go.db? could not be loaded
## error: lazy loading failed for package ?homo.sapiens?
## * removing ?/home/lg390/r/x86_64-unknown-linux-gnu-library/3.2/homo.sapiens?
## * restoring previous ?/home/lg390/r/x86_64-unknown-linux-gnu-library/3.2/homo.sapiens?
##
## the downloaded source packages are in
## 	?/tmp/rtmprr15y4/downloaded_packages?
## warning message:
## in install.packages(pkgs = pkgs, lib = lib, repos = repos, ...) :
##   installation of package ?homo.sapiens? had non-zero exit status

I have the exact same error with ReportingTools/PFAM.db,
Mus.musculus/GO.db and probably others.

Session info with other relevant package versions below. I experienced
the same error with a slightly older R-devel (2014-10-29 r66897).

Any hints?

Somewhere I think there's a stale serialized R-3.1 reference to ls(), derived
from a package build with R-3.1 rather than R-devel.

A little detective work shows that packages are loaded in the following order

...
[43] "graph"                             "RBGL"
[45] "GO.db"                             "org.Hs.eg.db"
[47] "TxDb.Hsapiens.UCSC.hg19.knownGene" "Homo.sapiens"

so I wonder if org.Hs.eg.db or TxDb.... is actually the problem, and
biocLite(org.Hs.eg.db) or TxDb... (maybe with type="source") would help?

I tried to re-install all TxDb*, org* packages (using type="source"),
but without success.

Is type = "source" not the default (and only option) on linux? Does is
matter if I specify it or not?

It might help to (maybe before biocLite(), to find the source of the problem)

biocLite("Homo.sapiens", INSTALL_opts="--no-test-load")

Unfortunately that fails too.

Laurent

and then

    library(Homo.sapiens)

followed by traceback() -- this way you'll at least be able to explore the call
stack.

Martin

Best wishes,

Laurent

library("RSQLite")
library("AnnotationDbi")
sessionInfo()

## R Under development (unstable) (2014-11-01 r66923)
## Platform: x86_64-unknown-linux-gnu (64-bit)
##
## locale:
##  [1] LC_CTYPE=en_GB.UTF-8       LC_NUMERIC=C
##  [3] LC_TIME=en_GB.UTF-8        LC_COLLATE=en_GB.UTF-8
##  [5] LC_MONETARY=en_GB.UTF-8    LC_MESSAGES=en_GB.UTF-8
##  [7] LC_PAPER=en_GB.UTF-8       LC_NAME=C
##  [9] LC_ADDRESS=C               LC_TELEPHONE=C
## [11] LC_MEASUREMENT=en_GB.UTF-8 LC_IDENTIFICATION=C
##
## attached base packages:
## [1] parallel  stats4    stats     graphics  grDevices utils     datasets
## [8] methods   base
##
## other attached packages:
##  [1] GO.db_3.0.0          RSQLite_1.0.0        DBI_0.3.1
##  [4] AnnotationDbi_1.29.1 GenomeInfoDb_1.3.6   IRanges_2.1.4
##  [7] S4Vectors_0.5.2      Biobase_2.27.0       BiocGenerics_0.13.0
## [10] BiocInstaller_1.17.1
##
## loaded via a namespace (and not attached):
## [1] compiler_3.2.0 tools_3.2.0