I suggest to replace the RCurl routines with httr which may be less sensitive to miss-configurations:

library(httr)

Attaching package: ‘httr’

The following object is masked from ‘package:curl’:
handle_reset

x <- GET("https://cactus.nci.nih.gov:443/chemical/structure/7529-22-8/stdinchikey")

content(x)

[1] "InChIKey=LFTLOKWAGJYHHR-UHFFFAOYSA-N"

Maybe related? I have just had a report of a failure downstream with our test data ... with our standard workflow (default settings) Babeldir is not set and CACTUS should be the back-up, it appears this now fails too...

@tsufz thanks for the work, I have provisionally added the workaround (slightly adapted) in 0c82f3b. It is building successfully on Windows. I am waiting for a successful Bioc build in order to close the issue.

RCurl should probably be removed completely at this point; up to now it's only a minimal workaround.

Building on Bioconductor.

It might be building on BioC but I am only getting NAs for all our test data and everything else?

`> getCactus("C1=CC=CC=C1", "cas")
[1] NA

> getCactus("C1=CC=CC=C1", "stdinchikey")
[1] NA

https://cactus.nci.nih.gov/chemical/structure/aspirin/stdinchikey
InChIKey=BSYNRYMUTXBXSQ-UHFFFAOYSA-N

> getCactus("aspirin","stdinchikey")
[1] NA

> sessionInfo()
R version 3.4.2 (2017-09-28)
Platform: x86_64-w64-mingw32/x64 (64-bit)
Running under: Windows >= 8 x64 (build 9200)

Matrix products: default

locale:
[1] LC_COLLATE=English_Australia.1252  LC_CTYPE=English_Australia.1252    LC_MONETARY=English_Australia.1252
[4] LC_NUMERIC=C                       LC_TIME=English_Australia.1252    

attached base packages:
[1] grid      stats     graphics  grDevices utils     datasets  methods   base     

other attached packages:
 [1] ReSOLUTION_0.1.7   nontarget_1.9      mgcv_1.8-20        nlme_3.1-131       nontargetData_1.1  mzR_2.10.0        
 [7] readxl_1.1.0       OrgMassSpecR_0.5-3 RChemMass_0.1.12   enviPat_2.2        rsvg_1.2           curl_3.2          
[13] rcdk_3.4.3         rcdklibs_2.0       rJava_0.9-10       RMassBank_2.9.1    Rcpp_0.12.19       devtools_1.13.3   

loaded via a namespace (and not attached):
 [1] lattice_0.20-35       colorspace_1.3-2      stats4_3.4.2          fingerprint_3.5.7     yaml_2.2.0           
 [6] vsn_3.44.0            XML_3.98-1.16         rlang_0.1.2           withr_2.0.0           MSnbase_2.2.0        
[11] BiocParallel_1.10.1   affy_1.54.0           BiocGenerics_0.22.1   affyio_1.46.0         foreach_1.4.3        
[16] plyr_1.8.4            mzID_1.14.0           zlibbioc_1.22.0       ProtGenerics_1.8.0    cellranger_1.1.0     
[21] munsell_0.4.3         pcaMethods_1.68.0     gtable_0.2.0          codetools_0.2-15      memoise_1.1.0        
[26] Biobase_2.36.2        IRanges_2.10.5        doParallel_1.0.11     BiocInstaller_1.26.1  parallel_3.4.2       
[31] itertools_0.1-3       preprocessCore_1.38.1 scales_0.5.0          limma_3.32.10         S4Vectors_0.14.7     
[36] impute_1.50.1         rjson_0.2.20          ggplot2_2.2.1         png_0.1-7             digest_0.6.18        
[41] tools_3.4.2           bitops_1.0-6          lazyeval_0.2.0        RCurl_1.95-4.11       tibble_1.3.4         
[46] Matrix_1.2-11         httr_1.3.1            iterators_1.0.10      R6_2.2.2              MALDIquant_1.16.4    
[51] git2r_0.19.0          compiler_3.4.2       

`

OK our bad ... fresh install of BioC for Loulou didn't pull the latest RMB for some reason ... httr cactus is working...

Hi, remember that BioC/RMB package versions are tied to the R version.
So "old" (as in not-brand-new) R will fetch old RMBs. Installing cutting-edge RMB
in "normal" (as in not-brand-new) R requires manual install_guthub() action. Yours, Steffen