Installation via "R CMD INSTALL --no-multiarch --with-keep.source data.table" fails on dev

[MichaelChirico]

Platform info

R version 3.4.1 (2017-06-30)
Platform: x86_64-apple-darwin15.6.0 (64-bit)
Running under: OS X El Capitan 10.11.6

R CMD INSTALL --no-multiarch --with-keep.source data.table

fails for me on the current dev. (this is also the default command run by Install and Restart in RStudio). The error is:

Error: package or namespace load failed for ‘data.table’ in dyn.load(file, DLLpath = DLLpath, ...):
 unable to load shared object '/Library/Frameworks/R.framework/Versions/3.4/Resources/library/data.table/libs/datatable.so':
  dlopen(/Library/Frameworks/R.framework/Versions/3.4/Resources/library/data.table/libs/datatable.so, 6): Symbol not found: _ENC2UTF8
  Referenced from: /Library/Frameworks/R.framework/Versions/3.4/Resources/library/data.table/libs/datatable.so
  Expected in: flat namespace
 in /Library/Frameworks/R.framework/Versions/3.4/Resources/library/data.table/libs/datatable.so

I assume this is related to the work on #2566/#2462.

If instead we just run

R CMD build data.table
R CMD INSTALL data.table_1.10.5.tar.gz --no-multiarch --with-keep.source data.table

it runs fine...

[HughParsonage]

Permissions issue?

[MichaelChirico]

sudo R CMD INSTALL --no-multiarch --with-keep.source data.table also fails

[mattdowle]

Is this one still a problem in dev as of now? Back in January, ENC2UTF8 was a function and that might have been the problem (it is mentioned in the error you pasted). Now it's a macro, I'm hoping it's gone away? Perhaps related the the fact that R itself has a function of the same name, and perhaps linking the datatable.so into R was a conflict somehow on Mac.

[MichaelChirico]

@mattdowle actually no. I do however get these warnings:

fread.c:2206:144: warning: format specifies type 'void *' but the argument has
      type 'const char *' [-Wformat-pedantic]
  ...jump=%d, headPos=%p, thisJumpStart=%p, sof=%p", jump, headPos, thisJumpS...
                      ~~                                   ^~~~~~~
                      %s
/usr/include/secure/_stdio.h:57:62: note: expanded from macro 'snprintf'
  __builtin___snprintf_chk (str, len, 0, __darwin_obsz(str), __VA_ARGS__)
                                                             ^~~~~~~~~~~
fread.c:2206:153: warning: format specifies type 'void *' but the argument has
      type 'const char *' [-Wformat-pedantic]
  ...headPos=%p, thisJumpStart=%p, sof=%p", jump, headPos, thisJumpStart, sof);
                               ~~                          ^~~~~~~~~~~~~
                               %s
/usr/include/secure/_stdio.h:57:62: note: expanded from macro 'snprintf'
  __builtin___snprintf_chk (str, len, 0, __darwin_obsz(str), __VA_ARGS__)
                                                             ^~~~~~~~~~~
fread.c:2206:168: warning: format specifies type 'void *' but the argument has
      type 'const char *' [-Wformat-pedantic]
  ...thisJumpStart=%p, sof=%p", jump, headPos, thisJumpStart, sof);
                           ~~                                 ^~~
                           %s
/usr/include/secure/_stdio.h:57:62: note: expanded from macro 'snprintf'
  __builtin___snprintf_chk (str, len, 0, __darwin_obsz(str), __VA_ARGS__)
                                                             ^~~~~~~~~~~
3 warnings generated.

[MichaelChirico]

Not sure what changed, but fixed.