The R Project SVN R

Rev

Show changed files | Details | Compare with Previous | Blame | RSS feed

Filtering Options

Rev Age Author Path Log message Diff
52864 5808 d 16 h urbaneks /branches/R-exp-R5/ Experinental branch implementing "R5" ideas - extending S language to
create a more intuitive object system
 
52406 5875 d 10 h murrell /trunk/ added support for polygons with holes  
51121 6011 d 9 h murrell /trunk/ vectorized grid.raster(); added raster() to graphics package (with a fairly clunky interface to be consistent with other low-level graphics package functions, but something more flexible would require a radical departure from other low-level graphics package functions)  
45051 6690 d 23 h ripley /trunk/src/ more cleaning up of headers, including moving createAtVector  
44986 6695 d 13 h ripley /trunk/ new interface for marked encodings  
44369 6746 d 23 h ripley /trunk/src/ tidy up  
44312 6750 d 17 h ripley /trunk/src/ more tidying up of code for colors.  
44309 6750 d 22 h ripley /trunk/src/ separate out base graphics initialization.  
44307 6751 d 1 h ripley /trunk/src/include/ clean up  
44288 6753 d 11 h ripley /trunk/src/ don't use opaque pointer for GEDevDesc internally in R  
44285 6753 d 12 h ripley /trunk/ use pGEDevDesc as pointer type.
remove unused do_replay.
don't set dd->open on devices (unused) nor newDevStruct.
tidy up duplication in header files.
 
44264 6753 d 22 h ripley /trunk/src/ move RGBpar and other device-level functions to GraphicsDevice.h  
44262 6753 d 23 h ripley /trunk/src/ start phasing out use of Rdevice.h (old devices interface)  
44259 6754 d 1 h ripley /trunk/ clean up some of the graphics internals:
- DevDesc is no longer a real structure, and DevDsec* is an opaque pointer
to GEDevDesc.
- add functions like GEaddDevice to avoid the need for (pointless) casts.
- add conveneience funtion desc2GEDesc
 
44258 6754 d 11 h ripley /trunk/src/include/ revert - just committed to get on record  
44257 6754 d 11 h ripley /trunk/src/ add desc2GEDesc and make use of it  
44154 6759 d 16 h ripley /trunk/src/ more graphics tidying up  
44142 6760 d 1 h ripley /trunk/src/ A little re-organization to start moving towards making plot/graphics/base
solely the base graphics system. In paticular, calls from grid are now in
other files.
Colour handling has been moved from graphics.c to colors.c, and
device-handing split out into devices.c.
The header files reflect these changes.
 
43926 6774 d 22 h ripley /trunk/src/ pass encoding down to graphics engine (and eventually to device)  
43924 6775 d 10 h ripley /trunk/src/ clean up Hershey font handling  
42308 6943 d 22 h ripley /trunk/src/include/ change reference to GPL  
41877 6991 d 18 h ripley /trunk/src/ add comments
remove unused declaration
 
41875 6991 d 19 h ripley /trunk/src/ tidy up the treatment of vfonts (including removing duplicate declarations)  
41852 6992 d 14 h ripley /trunk/src/ tidy up graphics events, don't use prompt as format string  
41784 6996 d 23 h ripley /trunk/src/ more 'const char *'  
37012 7507 d 15 h ripley /trunk/src/ tidy up the graphics interface a bit  
36820 7525 d 15 h ripley /trunk/ update FSF address  
35499 7631 d 8 h murrell /trunk/src/ export col2name

grid now uses device startcol and startfill for top-level viewport
 
32444 7876 d 21 h ripley /trunk/src/ shut up warnings from gcc4 and Solaris cc  
32205 7896 d 11 h murdoch /trunk/src/ getGraphicsEvent changes: less in NewDevDesc  
32163 7901 d 14 h murdoch /trunk/ Keyboard support in getGraphicsEvent  
32151 7903 d 4 h murdoch /trunk/ getGraphicsEvent() preliminary implementation  
30998 8000 d 10 h murrell /trunk/ Patch by Uwe Ligges

Adds "padj" arg to mtext() and axis()
and to GMText and GMMathText

Fixes PR#1659 and PR#7188
 
30711 8016 d 5 h murrell /trunk/src/ Reversed internal alpha semantics; 0 now means transparent and 255 means opaque
Also implemented full alpha channel in devQuartz
 
30129 8062 d 6 h murrell /trunk/src/ Added support for graphics engine font families in PDF device
Added support for alpha transparency in PDF device
Changed mapping of R-level NA colour to internal C-level colour:
was NA goes to NA_INTEGER
now NA goes to R_RGBA(255, 255, 255, 255) - transparent white
Plus some small grid patches
 
22931 8579 d 17 h ripley /trunk/ tidy up headers, make R_NO_REMAP useful, add STRICT_R_HEADERS  
20431 8784 d 0 h maechler /trunk/src/include/ 2 lines comments "fix"  
19912 8824 d 5 h duncan /trunk/src/ Some removal of global variables in src/main  
15641 9107 d 21 h ripley /trunk/ transparency support phase I  
14945 9150 d 18 h duncan /trunk/src/ The modules now explicitly register their function pointers rather than have the centralized loader resolve the symbols dynamically by looking in the shared library. This was motivated by the failure on the Dec T Tru64 reported by Brad McNeney. However, it is just a good thing to do as it reduces the dependency on the operating systems linker and dynamic loader and is more in keeping with the registration of native routines introduced recently. Tested on RH, YDL, Solaris. Needs checking on the Mac and Windows. Also needs to be verified on a machine without X11