[Rd] Most recent R manuals [Was: Likely disruption to R-devel builds on Windows]

2010-04-30 Thread Simon Urbanek

On Apr 30, 2010, at 4:20 AM, Prof Brian Ripley wrote:

 We are planning to phase in some major changes to the R build process on 
 Windows shortly, so expect problems and temporary unavailability of binary 
 builds of R and of packages, and if you are building from sources, check out 
 the latest version of the R-admin manual (in the sources) for the current 
 state.
 

To make sure no one has an excuse for not reading the current manuals - they 
are available (built nightly) at

http://r.research.att.com/man/

Cheers,
Simon


 The planned changes are
 
 - to move the 32-bit builds to MinGW's recent release of gcc 4.5.0 (but with 
 static linking of the Fortran and C++ runtimes: dynamic linking of C++ in 
 packages crashes R).
 
 - to merge the 32-bit and 64-bit builds into a single installer and (for at 
 least 98% of CRAN) a single binary for each package.
 
 NB: R-devel is not due for release until ca October and changes such as this 
 are done early in the development cycle to allow plenty of time for them to 
 be bedded down.  Please do take seriously the description as
 
 R version 2.12.0 Under development (unstable) (2010-04-29 r51864)
 
 
 -- 
 Brian D. Ripley,  rip...@stats.ox.ac.uk
 Professor of Applied Statistics,  http://www.stats.ox.ac.uk/~ripley/
 University of Oxford, Tel:  +44 1865 272861 (self)
 1 South Parks Road, +44 1865 272866 (PA)
 Oxford OX1 3TG, UKFax:  +44 1865 272595
 
 __
 R-devel@r-project.org mailing list
 https://stat.ethz.ch/mailman/listinfo/r-devel
 
 

__
R-devel@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-devel


Re: [Rd] Most recent R manuals [Was: Likely disruption to R-devel builds on Windows]

2010-04-30 Thread Gabor Grothendieck
Some easy way to find out what has changed would be desirable here.

On Fri, Apr 30, 2010 at 9:03 AM, Simon Urbanek
simon.urba...@r-project.org wrote:

 On Apr 30, 2010, at 4:20 AM, Prof Brian Ripley wrote:

 We are planning to phase in some major changes to the R build process on 
 Windows shortly, so expect problems and temporary unavailability of binary 
 builds of R and of packages, and if you are building from sources, check out 
 the latest version of the R-admin manual (in the sources) for the current 
 state.


 To make sure no one has an excuse for not reading the current manuals - they 
 are available (built nightly) at

 http://r.research.att.com/man/

 Cheers,
 Simon


 The planned changes are

 - to move the 32-bit builds to MinGW's recent release of gcc 4.5.0 (but with 
 static linking of the Fortran and C++ runtimes: dynamic linking of C++ in 
 packages crashes R).

 - to merge the 32-bit and 64-bit builds into a single installer and (for at 
 least 98% of CRAN) a single binary for each package.

 NB: R-devel is not due for release until ca October and changes such as this 
 are done early in the development cycle to allow plenty of time for them to 
 be bedded down.  Please do take seriously the description as

 R version 2.12.0 Under development (unstable) (2010-04-29 r51864)
                 

 --
 Brian D. Ripley,                  rip...@stats.ox.ac.uk
 Professor of Applied Statistics,  http://www.stats.ox.ac.uk/~ripley/
 University of Oxford,             Tel:  +44 1865 272861 (self)
 1 South Parks Road,                     +44 1865 272866 (PA)
 Oxford OX1 3TG, UK                Fax:  +44 1865 272595

 __
 R-devel@r-project.org mailing list
 https://stat.ethz.ch/mailman/listinfo/r-devel



 __
 R-devel@r-project.org mailing list
 https://stat.ethz.ch/mailman/listinfo/r-devel


__
R-devel@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-devel


Re: [Rd] Most recent R manuals [Was: Likely disruption to R-devel builds on Windows]

2010-04-30 Thread hadley wickham
Maybe Duncan could apply the same script that's being used for RNEWS
to the manuals?  An RSS feed of changes to the manuals would be really
useful.

Hadley

On Fri, Apr 30, 2010 at 8:16 AM, Gabor Grothendieck
ggrothendi...@gmail.com wrote:
 Some easy way to find out what has changed would be desirable here.

 On Fri, Apr 30, 2010 at 9:03 AM, Simon Urbanek
 simon.urba...@r-project.org wrote:

 On Apr 30, 2010, at 4:20 AM, Prof Brian Ripley wrote:

 We are planning to phase in some major changes to the R build process on 
 Windows shortly, so expect problems and temporary unavailability of binary 
 builds of R and of packages, and if you are building from sources, check 
 out the latest version of the R-admin manual (in the sources) for the 
 current state.


 To make sure no one has an excuse for not reading the current manuals - they 
 are available (built nightly) at

 http://r.research.att.com/man/

 Cheers,
 Simon


 The planned changes are

 - to move the 32-bit builds to MinGW's recent release of gcc 4.5.0 (but 
 with static linking of the Fortran and C++ runtimes: dynamic linking of C++ 
 in packages crashes R).

 - to merge the 32-bit and 64-bit builds into a single installer and (for at 
 least 98% of CRAN) a single binary for each package.

 NB: R-devel is not due for release until ca October and changes such as 
 this are done early in the development cycle to allow plenty of time for 
 them to be bedded down.  Please do take seriously the description as

 R version 2.12.0 Under development (unstable) (2010-04-29 r51864)
                 

 --
 Brian D. Ripley,                  rip...@stats.ox.ac.uk
 Professor of Applied Statistics,  http://www.stats.ox.ac.uk/~ripley/
 University of Oxford,             Tel:  +44 1865 272861 (self)
 1 South Parks Road,                     +44 1865 272866 (PA)
 Oxford OX1 3TG, UK                Fax:  +44 1865 272595

 __
 R-devel@r-project.org mailing list
 https://stat.ethz.ch/mailman/listinfo/r-devel



 __
 R-devel@r-project.org mailing list
 https://stat.ethz.ch/mailman/listinfo/r-devel


 __
 R-devel@r-project.org mailing list
 https://stat.ethz.ch/mailman/listinfo/r-devel




-- 
Assistant Professor / Dobelman Family Junior Chair
Department of Statistics / Rice University
http://had.co.nz/

__
R-devel@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-devel


Re: [Rd] Most recent R manuals [Was: Likely disruption to R-devel builds on Windows]

2010-04-30 Thread Duncan Murdoch

On 30/04/2010 10:22 AM, hadley wickham wrote:

Maybe Duncan could apply the same script that's being used for RNEWS
to the manuals?  An RSS feed of changes to the manuals would be really
useful.
  


My script is very NEWS specific.  The manuals would be harder.  You 
could get the diffs from svn pretty easily, but displaying them would be 
trickier:  do you display the diffs in the .texi source, or convert to 
HTML and display the diffs there, or what?  How much context around 
them?  Not impossible, but not trivial.


Duncan



Hadley

On Fri, Apr 30, 2010 at 8:16 AM, Gabor Grothendieck
ggrothendi...@gmail.com wrote:
 Some easy way to find out what has changed would be desirable here.

 On Fri, Apr 30, 2010 at 9:03 AM, Simon Urbanek
 simon.urba...@r-project.org wrote:

 On Apr 30, 2010, at 4:20 AM, Prof Brian Ripley wrote:

 We are planning to phase in some major changes to the R build process on 
Windows shortly, so expect problems and temporary unavailability of binary builds of R 
and of packages, and if you are building from sources, check out the latest version of 
the R-admin manual (in the sources) for the current state.


 To make sure no one has an excuse for not reading the current manuals - they 
are available (built nightly) at

 http://r.research.att.com/man/

 Cheers,
 Simon


 The planned changes are

 - to move the 32-bit builds to MinGW's recent release of gcc 4.5.0 (but 
with static linking of the Fortran and C++ runtimes: dynamic linking of C++ in packages 
crashes R).

 - to merge the 32-bit and 64-bit builds into a single installer and (for at 
least 98% of CRAN) a single binary for each package.

 NB: R-devel is not due for release until ca October and changes such as 
this are done early in the development cycle to allow plenty of time for them to be 
bedded down.  Please do take seriously the description as

 R version 2.12.0 Under development (unstable) (2010-04-29 r51864)
 

 --
 Brian D. Ripley,  rip...@stats.ox.ac.uk
 Professor of Applied Statistics,  http://www.stats.ox.ac.uk/~ripley/
 University of Oxford, Tel:  +44 1865 272861 (self)
 1 South Parks Road, +44 1865 272866 (PA)
 Oxford OX1 3TG, UKFax:  +44 1865 272595

 __
 R-devel@r-project.org mailing list
 https://stat.ethz.ch/mailman/listinfo/r-devel



 __
 R-devel@r-project.org mailing list
 https://stat.ethz.ch/mailman/listinfo/r-devel


 __
 R-devel@r-project.org mailing list
 https://stat.ethz.ch/mailman/listinfo/r-devel







__
R-devel@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-devel


Re: [Rd] Most recent R manuals [Was: Likely disruption to R-devel builds on Windows]

2010-04-30 Thread Gabor Grothendieck
On Fri, Apr 30, 2010 at 10:39 AM, Duncan Murdoch murd...@stats.uwo.ca wrote:
 On 30/04/2010 10:22 AM, hadley wickham wrote:

 Maybe Duncan could apply the same script that's being used for RNEWS
 to the manuals?  An RSS feed of changes to the manuals would be really
 useful.


 My script is very NEWS specific.  The manuals would be harder.  You could
 get the diffs from svn pretty easily, but displaying them would be trickier:
  do you display the diffs in the .texi source, or convert to HTML and
 display the diffs there, or what?  How much context around them?  Not
 impossible, but not trivial.


Just making the source diffs easily accessible would be sufficient IMHO.

For example, go to this page:
   http://rwiki.sciviews.org/doku.php?do=recent
and click on any of the eyeglass symbols.

__
R-devel@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-devel


Re: [Rd] Most recent R manuals [Was: Likely disruption to R-devel builds on Windows]

2010-04-30 Thread hadley wickham
 My script is very NEWS specific.  The manuals would be harder.  You could
 get the diffs from svn pretty easily, but displaying them would be trickier:
  do you display the diffs in the .texi source, or convert to HTML and
 display the diffs there, or what?  How much context around them?  Not
 impossible, but not trivial.

I suspect the best combination of ease and quality of output would be
to output the texi to plain text, and then diff that.   I really like
the existing RNEWS format, but I guess you split the news into
individual entries and only display those with any changes.  It would
be a bit harder to do with general output, although maybe splitting
into paragraphs would be sufficient.

Hadley

-- 
Assistant Professor / Dobelman Family Junior Chair
Department of Statistics / Rice University
http://had.co.nz/

__
R-devel@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-devel


Re: [Rd] Most recent R manuals [Was: Likely disruption to R-devel builds on Windows]

2010-04-30 Thread Duncan Murdoch

On 30/04/2010 10:48 AM, Gabor Grothendieck wrote:

On Fri, Apr 30, 2010 at 10:39 AM, Duncan Murdoch murd...@stats.uwo.ca wrote:
 On 30/04/2010 10:22 AM, hadley wickham wrote:

 Maybe Duncan could apply the same script that's being used for RNEWS
 to the manuals?  An RSS feed of changes to the manuals would be really
 useful.


 My script is very NEWS specific.  The manuals would be harder.  You could
 get the diffs from svn pretty easily, but displaying them would be trickier:
  do you display the diffs in the .texi source, or convert to HTML and
 display the diffs there, or what?  How much context around them?  Not
 impossible, but not trivial.


Just making the source diffs easily accessible would be sufficient IMHO.


That's already available, but not too pretty.  E.g.

svn log -v https://svn.r-project.org/R/trunk/doc/manual | less

will show something like


r51865 | ripley | 2010-04-30 00:56:00 -0400 (Fri, 30 Apr 2010) | 1 line
Changed paths:
  M /trunk/doc/manual/R-admin.texi
  M /trunk/src/gnuwin32/CHANGES
  M /trunk/src/library/tools/R/install.R

tests starssrc/library/tools/R/install.R

r51858 | ripley | 2010-04-29 00:51:01 -0400 (Thu, 29 Apr 2010) | 1 line
Changed paths:
  M /trunk/doc/manual/R-exts.texi

tweak wording

r51854 | ripley | 2010-04-28 10:41:26 -0400 (Wed, 28 Apr 2010) | 1 line
Changed paths:
  M /trunk/doc/manual/R-exts.texi

more on portability
(etc.)

and for particular details, you can do something like

svn diff -c 51858 https://svn.r-project.org/R/trunk/doc/manual

which will give

Index: R-exts.texi
===
--- R-exts.texi (revision 51857)
+++ R-exts.texi (revision 51858)
@@ -10813,10 +10813,10 @@

The @R{} for Windows installers have for a long time allowed the value
of @code{R_HOME} to be recorded in the Windows Registry: this is
-optional but the default. Where is it is recorded has changed over the
-years to allow for multiple versions of @R{} to be installed at once,
-and as from @R{} 2.11.0, to allow 32- and 64-bit versions of @R{} to be
-installed on the same machine.
+optional but selected by default.  @emph{Where} is it is recorded has
+changed over the years to allow for multiple versions of @R{} to be
+installed at once, and as from @R{} 2.11.0, to allow 32- and 64-bit
+versions of @R{} to be installed on the same machine.

The basic Registry location is @code{Software\R-core\R}.  For an
administrative install this is under @code{HKEY_LOCAL_MACHINE} and on a

The non-trivial part is converting it into a nice display like what you 
pointed out in the Wiki.


Duncan Murdoch

__
R-devel@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-devel


Re: [Rd] Most recent R manuals [Was: Likely disruption to R-devel builds on Windows]

2010-04-30 Thread Gabor Grothendieck
On Fri, Apr 30, 2010 at 11:01 AM, Duncan Murdoch murd...@stats.uwo.ca wrote:
 On 30/04/2010 10:48 AM, Gabor Grothendieck wrote:

 On Fri, Apr 30, 2010 at 10:39 AM, Duncan Murdoch murd...@stats.uwo.ca
 wrote:
  On 30/04/2010 10:22 AM, hadley wickham wrote:
 
  Maybe Duncan could apply the same script that's being used for RNEWS
  to the manuals?  An RSS feed of changes to the manuals would be really
  useful.
 
 
  My script is very NEWS specific.  The manuals would be harder.  You
  could
  get the diffs from svn pretty easily, but displaying them would be
  trickier:
   do you display the diffs in the .texi source, or convert to HTML and
  display the diffs there, or what?  How much context around them?  Not
  impossible, but not trivial.

If you could mirror the manuals to googlecode you would automatically
get an HTML interface to the diffs. e.g.

http://code.google.com/p/sqldf/source/list

__
R-devel@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-devel


Re: [Rd] Most recent R manuals [Was: Likely disruption to R-devel builds on Windows]

2010-04-30 Thread Duncan Murdoch

On 30/04/2010 2:11 PM, Gabor Grothendieck wrote:

On Fri, Apr 30, 2010 at 11:01 AM, Duncan Murdoch murd...@stats.uwo.ca wrote:
 On 30/04/2010 10:48 AM, Gabor Grothendieck wrote:

 On Fri, Apr 30, 2010 at 10:39 AM, Duncan Murdoch murd...@stats.uwo.ca
 wrote:
  On 30/04/2010 10:22 AM, hadley wickham wrote:
 
  Maybe Duncan could apply the same script that's being used for RNEWS
  to the manuals?  An RSS feed of changes to the manuals would be really
  useful.
 
 
  My script is very NEWS specific.  The manuals would be harder.  You
  could
  get the diffs from svn pretty easily, but displaying them would be
  trickier:
   do you display the diffs in the .texi source, or convert to HTML and
  display the diffs there, or what?  How much context around them?  Not
  impossible, but not trivial.

If you could mirror the manuals to googlecode you would automatically
get an HTML interface to the diffs. e.g.

http://code.google.com/p/sqldf/source/list

Go for it.  You've got more experience with googlecode than I do.

Duncan Murdoch

__
R-devel@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-devel