CVS log for JSOC/proj/sharp/apps/update_sharp_keys.c
	Help

Request diff between arbitrary revisions

Figured out the memory allocation problem.
It turns out that the arrays are not constant per T_REC for any given HARPNUM, e.g.:

> show_info -d key=T_REC ds=hmi.sharp_cea_720s"[3311][]" seg={Bt}
T_REC	Bt	Bt_info
2013.10.24_00:48:00_TAI	Bt.fits	"Gauss, fits, 547x411"
2013.10.24_01:00:00_TAI	Bt.fits	"Gauss, fits, 547x411"
2013.10.24_01:12:00_TAI	Bt.fits	"Gauss, fits, 546x411"

So I moved all of the code that created one fixed array size per HARPNUM into the loop.
Now it creates a new array per T_REC.

fixed problem with cdelt1 (was calculating using testrec instead of per-record)
also fixed some formatting

fixed some issues with LOS keywords
added documentation

Added LOS keywords + cleanup

moved mallocs outside of the loop to save memory

freed two arrays

out of bounds problem still remained in Function 8; fixed it

There was a counting out-of-bounds problem with the functions that produce MEANGBZ, MEANGBT, and MEANGBH. I re-coded it such that the error terms are calculated separately; further, the error arrays are initalized to zero and the edge pixels are ignored (since they will be selected out by the bitmap and conf_disambig bitmaps anyway).

updated function calls for Jz to include the correct numerical derivative described in the sw_functions.c commit message for v.1.33.

added check on CEA series array dimensions
if nx by ny are not constant for any given CEA harpnum, the program will exit

added a debug flag

added lorentz force function to calculate keys TOTFX, TOTFY, TOTFZ, TOTBSQ, EPSX, EPSY, EPSZ

jesper's gaussian convolution function left a hwidth stripe of 0's along the edge of the image. i fixed this by padding with zeros prior to convolving.

There was a problem in the way that I was calling fsample(). fsample() takes several arguments, one of which are the x and y dimensions of the output array and the scale. If the following occurs:


nx_out > floor((ny_in-1)/scale + 1)
ny_out > floor((ny_in-1)/scale + 1),

where n*_out are the dimensions of the output array and n*_in are the dimensions of the input array, fsample() will usually result in a segfault (though not always, depending on how the segfault was accessed.)

Revert the versions of several sharps source-code files to make them intercompatible.

added ability to specify different input and output series
added functionality to populate R_VALUE

Revert accidental check-in.

Fix some bugs - orbit and hlzsaa were switched in one place; there was a space in a redirection clause, which resulted in a 1 being appended to a command line; pass in the orbit series name, not the fds series name, to ingestIrisOrbit.pl; was searching through the fds series for an obsdate instead of obsdate_index (use timeslot to get the index value); add the unit field to the ingested keyword.

 *      MODULE NAME: update_sharp_keys.c
 *
 *	DESCRIPTION: This module recalculates SHARP keywords. This is accomplished by
 *      cloning a record, recalculating keywords of choice (i.e., user input),
 *      and pointing to the same segments as the old record. Associated error keys
 *      are computed by default.
 *
 *      This module accounts for versioning by appending to the keyword CODEVER7 and HISTORY:
 *      CODEVER7 will contain multiple lines: the production build of sharp.c, the
 *      production build of include file sw_functions.c, and the production build
 *      of update_sharp_keys.c. HISTORY will include a human-readable sentence about which
 *      keywords were updated.
 *
 *      This module does not produce segments.
 *
 *      INPUTS     : -- DRMS SHARP series
 *                   -- DRMS SHARP CEA series
 *                   -- HARPNUM
 *                   -- comma separated list of keywords to recalculate
 *
 *	AUTHOR     : Monica Bobra
 *
 *	Version    :   v0.0	Jun 14 2013
 *
 *	EXAMPLE    :
 *      update_sharp_keys sharpseries=hmi.sharp_720s sharpceaseries=hmi.sharp_cea_720s //
 *      HARPNUM=1 keylist=USFLUX,TOTPOT
 *