Skip to content

Compiling 1.2.1 to Windows binary

7 messages · Marc Feldesman, Peter Dalgaard, Marc R. Feldesman +1 more

#
I'm having a bit of trouble getting the 1.2.1 source to compile to a 
Windows binary.  The base compiles with no problem, but when the make file 
gets to the point of having to compile the HTML files (.chm), the make file 
quits, and neither the .chm files get built, nor does the make file 
continue on to build any of the packages beyond the base.  I can overcome 
this problem by changing MkRules to only do WinHelp.  The error message I 
get basically says that it can't find the hhc file.  Not only is the hhc 
correctly specified in the MkRules file, but it is also correctly located 
in the Path within the environment.  I can verify this by spawning a 
command prompt (cmd) and simply entering "hhc" at the command line.  It 
brings up hhc.exe without problem.

I've encountered this problem on both WinNT and Windows 2000, but never 
encountered it when I compiled the release prior to 1.2.0 (1.1.0?).

I seem to be missing something obvious here, but I can't figure out what is 
preventing the make file from finding hhc.exe.

I'd appreciate any advice you can offer.

Thanks.




=====================
Dr. Marc R. Feldesman
Professor and Chairman
Anthropology Department
Portland State University
1721 SW Broadway
Portland, Oregon 97201
email:  feldesmanm at pdx.edu
phone:  503-725-3081
fax:    503-725-3905
http://web.pdx.edu/~h1mf
PGP Key Available On Request
======================

"Anyway, no drug, not even alcohol, causes the fundamental ills of society.
If we're looking for the source of our troubles, we shouldn't test people
for drugs, we should test them for stupidity, ignorance, greed and love of
power."   P.J. O'Rourke

Powered by Optiplochoerus and Windows 2000 (scary isn't it?)

-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
r-help mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !)  To: r-help-request at stat.math.ethz.ch
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._
#
Marc Feldesman <feldesmanm at pdx.edu> writes:
Giving us the exact error message might be helpful (or maybe not, but...)

You could consider replacing the hcc command with its full path.
I'm not sure that paths in Makefile are guaranteed to be the same as
in a shell.
#
At 02:17 AM 1/20/01 +0100, Peter Dalgaard BSA wrote:
>Giving us the exact error message might be helpful (or maybe not, but...)

The exact error message is:

"Make[3]:  hhc:  command not found".

This is then followed by an abort of the attempt to compile the .chm files 
and a termination of the entire make.

Brian Ripley gave me a clue that helped.  I moved hhc.exe (but not all the 
30 odd additional files and 4 subdirectories associated with MS HTML Help 
Workshop) to my rtools directory.  This allowed the .chtml files to be 
built for base and all the packages.  I did get complaints from the OS that 
"hhc.exe is not properly registered", and a series of "Invalid HTML tag 
<JAVASCRIPT> at line xxxx" errors during the compilation.  I'm not sure 
what role some of the supporting files in HTML Help Workshop serve, but 
moving the hhc.exe stub to rtools does allow the make to complete.

As I explained to Brian, the path to HTML Help Workshop is pretty simple:

e:\htmlhelp in PATH and in MkRules as  e:/htmlhelp.

At the cmd prompt:

F:\rdevel\r-1.2.1\src\gnuwin32>hhc    (found immediately).

but

sh
$hhc
hhc not found
hhc not found
 >

Clearly hhc cannot be found from the information available in the PATH 
statement, or in the HHWDIR path within MkRules.  I don't know why.  Tcltk 
is found just fine with a much more complicated path (including spaces in 
the path name).




 >You could consider replacing the hcc command with its full path.
 >I'm not sure that paths in Makefile are guaranteed to be the same as
 >in a shell.
 >
 >--
 >   O__  ---- Peter Dalgaard             Blegdamsvej 3
 >  c/ /'_ --- Dept. of Biostatistics     2200 Cph. N
 > (*) \(*) -- University of Copenhagen   Denmark      Ph: (+45) 35327918
 >~~~~~~~~~~ - (p.dalgaard at biostat.ku.dk)             FAX: (+45) 35327907


Dr. Marc R. Feldesman
email:  feldesmanm at pdx.edu
email:  feldesman at attglobal.net
fax:    503-725-3905

"Don't know where I'm going.
Don't like where I've been.
There may be no exit.
But hell, I'm going in."  Jimmy Buffett

Powered by Superchoerus - the 700 MHz Coppermine Box

-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
r-help mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !)  To: r-help-request at stat.math.ethz.ch
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._
1 day later
#
Thanks to Brian Ripley and Peter Dalgaard for suggestions that led to *a* 
resolution of the problem compiling R 1.2.1 to a Windows binary.

The basic problem revolved around the inability of the makefile to locate 
the hhc.exe file from the Microsoft HTML Help Workshop.  Although the 
location was completely clear in the path statement at the NT level and 
from within the cygwin shell, it couldn't be located during the 
make.  Brian suggested moving hhc.exe to the Rtools directory, which 
partially solved the problem.  It permitted the .chtml files to be 
produced, but spawned a series of JScript errors along the way.

The crucial resolution lay in an offhand suggestion that Brian made (and 
which Peter hinted at) about other elements being on the same logical drive 
with HTML Help Workshop.  As it turned out, I placed HTML Help Workshop on 
my logical E: drive, while all the other elements needed to make R were 
located on the logical F: drive.  Again, even though the paths were quite 
clear in the shell, placing the different components on different logical 
drives is too confusing in the cygwin environment.  When I uninstalled HTML 
Help Workshop from the E: drive, rebooted and reinstalled it to the F: 
drive, the make proceeded without error.

I don't know whether this should be part of the instructions for building R 
from source or not, but it certainly does appear that everything needs to 
reside on the same logical drive.

Thanks again to Brian and Peter for their helpful suggestions.

-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
r-help mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !)  To: r-help-request at stat.math.ethz.ch
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._
#
On Sat, Jan 20, 2001 at 08:42:31PM -0800, Marc R. Feldesman wrote:
Not really sure about this. I have a almost full 'cygwin' in c:/cygwin, 
  'mingw' in c:/Programs/mingw32 but 'perl' and 'Ms Help Workshop'
  in d:/Program Files/. And, it works for me. Boh!!

  guido
-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
r-help mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !)  To: r-help-request at stat.math.ethz.ch
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._
#
Guido Masarotto <guido at sirio.stat.unipd.it> writes:
...
I haven't played with Cygwin for years [*], but isn't there some business
about mounting drives to make them visible to the shell?  

[*] and good riddance. Not that the Cygwin crowd haven't made an
incredible effort - it's just that the kludges necessary to emulate a
Unix environment under Windows get to you after a while. Much
admiration for Brian's and Guido's ability to stay on top of this for
so long.
#
On Sun, Jan 21, 2001 at 04:10:33PM +0100, Peter Dalgaard BSA wrote:
Yeah. That's can be the point. Indeed, my D: drive is mounted as
  /d (lot of fantasy, isn't it?) and hence it is accessed as a
  C: sub-dir (same for /floppy and /cdrom).

  guido 
-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
r-help mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !)  To: r-help-request at stat.math.ethz.ch
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._