Commit summary from repository starlink
------------------------------------

  (Ed Chapin) smurf: quality now exported with any 3d model component
     5.9% applications/smurf/examples/
    94.0% applications/smurf/libsmf/
  
  (Ed Chapin) smurf: fix an accounting error in smf_qualstats_report
   100.0% applications/smurf/libsmf/
  
  (Ed Chapin) smurf: whitespace / formatting
   100.0% applications/smurf/libsmf/
  
  (Ed Chapin) smurf: slightly weaker DC step-finding
   100.0% applications/smurf/examples/
  
  (Ed Chapin) smurf: display quality report both before and after pre-processing
   100.0% applications/smurf/libsmf/
  
  (Ed Chapin) smurf: whitespace cleanup
   100.0% applications/smurf/libsmf/
 

Commits from repository starlink
-----------------------------

  commit 821f6ab8d4e382afdb96a9e7e2de65837342ffb2
  Author: Ed Chapin <[log in to unmask]>
  Date:   Tue Mar 30 17:07:37 2010 -0700
  
      smurf: quality now exported with any 3d model component
      
      Re-jigged the logic for attaching quality to exported model
      components. Now any model component with the same dimensions
      as the residual (i.e. 3d data cube) will have quality attached...
      as long as 'qua' is specified in exportndf in the dimmconfig file.
  
   applications/smurf/examples/dimmconfig.lis       |    9 +-
   applications/smurf/libsmf/Makefile.am            |    1 +
   applications/smurf/libsmf/smf.h.source           |    3 +
   applications/smurf/libsmf/smf_iteratemap.c       |   22 ++++-
   applications/smurf/libsmf/smf_samedims_smfData.c |  100 ++++++++++++++++++++
   5 files changed, 127 insertions(+), 8 deletions(-)
  
  commit 4da0ae3e3d1c600a40dbebe5071b62944171f931
  Author: Ed Chapin <[log in to unmask]>
  Date:   Tue Mar 30 15:36:02 2010 -0700
  
      smurf: fix an accounting error in smf_qualstats_report
      
      The reason BADDA and BADBOL were reporting different numbers of
      total flagged bolometers is because BADBOL flags *every* sample,
      whereas BADDA skips the padding at the start and end of a time
      series. Therefore the total number of samples flagged BADDA
      must be divided by the total time slices minus the padding to
      correctly count bolometers.
  
   applications/smurf/libsmf/smf_qualstats_report.c |   23 +++++++++++++++----
   1 files changed, 18 insertions(+), 5 deletions(-)
  
  commit 26304b20b88d6ecb5d1d11d1ab2e2ff184a0e220
  Author: Ed Chapin <[log in to unmask]>
  Date:   Tue Mar 30 14:43:53 2010 -0700
  
      smurf: whitespace / formatting
  
   applications/smurf/libsmf/smf_update_quality.c |    3 ++-
   1 files changed, 2 insertions(+), 1 deletions(-)
  
  commit 578a20a7def5a45d362acc87bb599e239750a79b
  Author: Ed Chapin <[log in to unmask]>
  Date:   Tue Mar 30 14:25:31 2010 -0700
  
      smurf: slightly weaker DC step-finding
      
      Looking at blank-field data, erroneous DC steps were being found
      in many locations during the pre-processing. Backing off the
      threshold to 5-sigma from 2-sigma seems to be much better. For
      the faint dimmconfig, it can safely iteratively identify
      down to 3-sigma steps in the much flatter residual.
  
   applications/smurf/examples/dimmconfig.lis       |    2 +-
   applications/smurf/examples/dimmconfig_faint.lis |    2 +-
   2 files changed, 2 insertions(+), 2 deletions(-)
  
  commit 93945bcbd0094bbe4dfae1fc5a1d6fcad0cd60b3
  Author: Ed Chapin <[log in to unmask]>
  Date:   Tue Mar 30 14:24:10 2010 -0700
  
      smurf: display quality report both before and after pre-processing
  
   applications/smurf/libsmf/smf_iteratemap.c |    8 ++++++--
   1 files changed, 6 insertions(+), 2 deletions(-)
  
  commit 1369705cd22a5e461763a03d2b764549abcbb087
  Author: Ed Chapin <[log in to unmask]>
  Date:   Tue Mar 30 10:45:44 2010 -0700
  
      smurf: whitespace cleanup
  
   applications/smurf/libsmf/smf_filter_execute.c |   86 +++++++++++-----------
   1 files changed, 43 insertions(+), 43 deletions(-)