Fixed issue with model fitting measurement leftovers with masked sources (thank
s to E. Rykoff).
Pushed version number to 2.19.3.

Fixed model sub-sampling issues with large objects, including check-image resizing.
Fixed initialization of DETMODEL size guess (thanks to Eli Rykoff).
Fixed image extension pb in double weighting/single image mode.
Improved debug info for memory allocations in FFT and FITS libraries.
Fixed crashes with DETMODEL magnitudes on null measurement weight maps.
Added Eli Rykoff's fix to discard sources with NaNs in the fitted positions.
Improved MKL autoconfiguration.
Improved vectorization in FITS data transformations and FFT-based convolutions.
Pushed SExtractor version number to 3.18.8.

Added control of the number of threads in MKL calls.
Added support for negative NTHREADS (= max number of threads).
Pushed version number to 2.18.1.

Updated LevMar library to V2.6.
Pushed version number to 2.18.0.

Preliminary implementation of a RESCALE_WEIGHTS configuration switch.

Added more complete headers to all files.
Clarified licensing issues.
Updated FITS library.
Updated configuration files.
Added configuration files to the list of data that get installed in ${prefix}/share/sextractor.
Added background offset and point-source model components: FLUX_BACKOFFSET,FLUXERR_BACKOFFSET,FLUX_POINTSOURCE,FLUXERR_POINTSOURCE,MAG_POINTSOURCE,MAGERR_POINTSOURCE.
Tweaked model-fitting convergence parameters.
Fixed catalogue source ordering/number issue in dual-image mode.
Pushed version number to 2.13.0.

Changed the management of tolerances in Hessian inversion SVD.
Pushed Lanczos interpolator to 4th order in PSF and model-fitting.
Normalized the interpolating function so that the DC component gain is 1.
Pushed version number 2.12.2.

Fixed model-fitting issue with empty images in dual-image mode (thanks to V. de Lapparent for reporting).
Updated LevMar to version 2.5.
Updated URLs in various places.
Pushed version number to 2.12.1.

Fixed TFORM error message in fitshead.c (thanks to S.Guieu).
Fixed model-fitting error variance/covariance issue (bug in matrix inversion).
Added ELLIP1ERRMODEL_IMAGE, ELLIP2ERRMODEL_IMAGE, ELLIPCORRMODEL_IMAGE,
POLAR1ERRMODEL_IMAGE, POLAR2ERRMODEL_IMAGE, POLARCORRMODEL_IMAGE (more testing is needed and WORLD versions are not yet functional).
Narrowed the tolerance constraint for stopping windowed centering iterations.
Fixed incorrect reading of the DATE-OBS FITS keyword when EPOCH and MJD-OBS are missing.
Pushed version number to 2.11.11.

Fixed various compilation warnings.

Fixed compilation warnings in pattern.c and lm.h.
Pushed version number to 2.9.9.

Added support for 64 bit binary table elements in FITS library (thanks to J.P. McFarland).

Updated the LevMar library to V2.4.
Pushed version number to 2.9.7.

Removed singular matrix warning.
Added MATRIX_MODELERR measurement parameters.
Added DURATION_ANALYSIS measurement parameter.

Changed configure behaviour towards CFLAGS and LDFLAGS: automatic settings are now applied only if the --enable-automatic-flags configure option is set (thanks to S.Pascual).
Version number pushed to 2.9.0.

Replaced configuration error with warning if FFTW is not multithreaded and multithreading has been activated (unused).
Added autoconfig support for FFTW multithreaded versions combined in libfftw3 (unused).
Added runtime warning if ATLAS is not multithreaded and multithreading is on (unused).
Version number pushed to 2.8.8.

Cleaned up Makefile.am RPM options.
Fixed rpm build issue on Fedora Core 10.
Fixed display of max thread number in configure help.
Updated icc compilation flags to 11.0 syntax.
Set the license string to CeCILL in the .spec file.
Pushed version number to 2.8.5.