
2008.01.31

  Adding extended sources:

  * create a libautocode extended source table description (DONE)
  * load the extended source measurements:
    this could be done within LoadStars or as a separate pass over the file, loading the extended source tables
    * we can call this function only optionally
    * need to make the association between the psf and ext measurements via det ID
  * in find_matches, we could/should record the reference from Stars to the matched average entry
    that would let us use the PSF position to get the average object match:

  * find_matches_extended
    * loop over extended sources
    * use ID to find the corresponding Stars entry
    * use the averef to find the corresponding Average entry
    * update the extended source table

  * Write out ext sources with the rest of the data

  where do we put the table? require split?

  (160 byte per detection; PS1 will have 5e8 => 400 GB)

2007.01.04

  Use of options.photcode:
    - M_REFLIST: used to assign a photcode value to the loaded data
    - M_REFCAT: used to select data from the reference of the given photcode
    - M_FAKEIMAGE: used to assign a photcode value to the generated data
    - M_RESORT: unused
    - M_IMAGE: is used to override or supply the image header photcode
	       * only valid for single chip runs

  Re: MOSAIC_PHU
      - the mosaic phu needs to have the time to enable the image/mosaic match

      - NX,NY are used for plotting the mosaic outline.  the center of
        the mosaic is 0,0, and the mosaic outline is drawn from
        -NX/2,-NY/2 to NX/2,NY/2.  A problem: NX,NY in Image is using
        a short int.  If the units of the FPA/TPA coords are microns, this overflows.
	* change NX,NY to int?
	* require the use of mm not microns?
	* for GPC, NX,NY ~ 384000 um excluding gaps

      - only the astrometry, time, and NX,NY are loaded for MOSAIC_PHU headers.

2007.01.03

  I am working on adding the ability to read MEF files containing
  multiple tables and multiple image headers.  Here are the details:

  SIMPLE CMP:
    PHU:
      NAXIS = 2 or
      SIMPLE = FALSE or
      TEXTMODE = TRUE (use option; global variable)

  SIMPLE CMF:
    PHU: 
      NAXIS = 0
      NAXIS1 exists, > 0
      NAXIS2 exists, > 0
    EXT:
      EXTNAME = SMPDATA, PS1DATA, etc (defines layout)

  SIMPLE MEF:
    PHU: 
      NAXIS = 0
      NAXIS1 does not exist
      NAXIS2 does not exist
      &CTYPE[4] is not "-DIS"
      EXTEND is true
    EXT HEAD:
      EXTTYPE = IMAGE
      EXTNAME = image.name
      EXTDATA = table.name
    EXT DATA:
      EXTTYPE = SMPDATA, PS1DATA, etc (defines layout)
      EXTNAME = table.name
      EXTHEAD = image.name

  MOSAIC MEF:
    PHU: 
      NAXIS = 0
      NAXIS1 does not exist
      NAXIS2 does not exist
      &CTYPE[4] is "-DIS"
      EXTEND is true
    EXT HEAD:
      EXTTYPE = IMAGE
      EXTNAME = image.name
      EXTDATA = table.name
    EXT DATA:
      EXTTYPE = SMPDATA, PS1DATA, etc (defines layout)
      EXTNAME = table.name
      EXTHEAD = image.name

  MOSAIC PHU:
    PHU: 
      NAXIS = 0
      NAXIS1 does not exist
      NAXIS2 does not exist
      &CTYPE[4] is "-DIS"
      NEXTEND = 0 or  
      EXTEND is false

    

2006.04.02

  I am considering an option to insert fake camera images into the
  database.  The camera would be described with a file specifying the
  dimensions of each chip (Nx,Ny,pixel scale), the location of each
  chip in the focal plane, and the optical parameters of the camera
  (plate scale at center, x,y distortion terms).  Each call to addstar
  would specify the boresite center:

  addstar -fakeimage ra dec theta

  the camera parameters should be loaded via the config information.

2005.10.24

  I have changed addstar to use the new, more flexible, SkyRegion
  concept to define the sky layout rather than just the GSCRegion
  files.  This includes the ability to subdivide the existing
  GSCRegion concept, and to have a more flexible sky table
  arrangement.  

  Names of catalog files are a bit tricky.  The SkyRegion functions
  return base names without fullpath or extensions (ie, n0000/0012).
  There are a few places in addstar which need the fullname:

  SkyListExitingSubset () requires the path and adds .cpt

  load_pt_catalog () takes the path and adds .cpt.  it also allocates
  a filename entry for catalog, which is freed later in the loop in
  addstars

  mkcatalog uses the name in the catalog structure and replaces .cpt
  with .cpm, .cpn, .cps as needed.

  *** I need to fix the name method: common function to set it once 
  in the dvo functions.  perhaps a pointer in the SkyRegion structure.
  This needs some 32/64 bit care, though.  use a long long (or other 
  type guaranteed to be 64 bit), and case to the appropriate type (size_t) 
  when it is used?  also needs to be freed.

todo:

  * sky db tools 
  * parallel servers
  * threaded addstard
  * dvo_load daemon
  * getstar (program and c-client)
  * parallal / proper-motion data
  * other additional Average / Measure columns
  * deal with Myyyy as sky value
  * better way to interpret flips header info for sky?
  * fix unlink empty file problem
  * programs like relphot and dvo need to be provided 
    with sorted measure tables; add this as a feature 
    of the load_catalog API?

done:

  o move addstar/gstars to AddstarClientOptions
  o add thiscode to AddstarClientOptions
  o address MOSAIC somehow
  o replace skyprobe mode
  o complete server db interactions
  o address the cat & ref modes in client/server
  o add airmass quality to options

2005.10.19

	I have added the client / server mode, and tested it out to a
	fair degree.  It handles all of the available addstar modes,
	including -ref and -cat.  A handful of things still need to be
	done, including pushing some of the logic to enforce sorting
	into the dvo load_catalog functions.  Also, I need to correct
	the idiosyncratic problem with sky vs Myyyy for the skyprobe
	data.  There are a handful of other improvements that are
	needed before addstar / DVO is ready for IPP release.
	However, at this point, it is ready for internal use by the
	grad students, but not yet ready for CFHT use with Elixir.

2005.10.14

  the current db interactions are somewhat in efficient wrt the image
  table.  the basic outline is as follows:

  - load image overlaps (needs entire image table)
  - find detection matches (uses overlap information for missed)
  - save new image (needs calculated calibration value from detection
    matches)
  
  currently, I access the image table twice, and both times I load the
  entire thing.  I really only need to append in the second case.  To
  get to that point, I need to:

  - split image overlaps from image db load
  - define image partial load (analogous with catalog LOAD_MEAS_META)
  
  for the moment, I'm being wasteful so I can defer this issue.

2005.10.13

  I am working to minimize usage of globals and cleanup addstar
  functions to make them more appropriate for the client / server
  model.  

  I have dropped save_pt_catalog, and modified load_pt_catalog, as
  well as the load_catalog functions.  Now, the choice of elements to
  load is made by setting flags in the catalog structure itself
  (catalog.catflags).  Thus this information is carried around by the
  catalog variable.  the load_pt_catalog function now only returns
  true.  it is now necessary to check the state of the disk file by
  examining Nave_disk to decide if the file is empty or not.  This is
  perhaps too limiting: it prevents us from using only one of the
  catalog tables (say measure without average).  The concern I have is
  to manage the process of cleaning up the catalogs if no data is
  added to them:  I don't want addstar to leave behind empty files
  (plus, I think this breaks the lock/load state logic).  Thus,
  whenever addstar unlocks / closes a file, if the Nave_disk +
  Naverage is 0, the file should be unlinked.

2005.10.12

  By friday (2005.10.07), I managed to get the update process
  working well, and tested it with some example data to demonstrate
  that the partial update improved speed hugely.  Saturday, I started
  to work on the client/server separation of addstar.  The first step
  was to test out socket communication issues.  I set up a simple
  client/server pair, in which one program listens on a socket while
  the client connects to the socket and sends some command.  I set up
  two autocode functions to send and receive complete structures
  (Send_*, Recv_*) and added these to libautocode.  I have been able
  to demonstrate sending Stars, Image, and Options to the server.

  The next difficulty is in pulling out the command-line and
  configuration options.  My original, somewhat lazy, design over-used
  global variables to pass information between program sections.  This
  needs to be cleaned up, especially if the server needs to have
  multiple threads interacting with the db tables.  

  Of the global variables that get set, some simply carry options on
  how gstars should behave when it loads the data (including the
  header interpretation information).  This can probably stay in the
  globals for now.  Another group are configuration information used
  only by the server-related functions, such as defining the location
  of the database files and their format, etc.  These can also stay as
  global, at least for now.  Another set of variables basically pass
  information from the client to the server on how to behave *this
  time*.  I have defined the AddstarClientOptions structure to carry
  these, and will pass them from the client to the server along with
  the stars and the image data.

  Some outstanding issues:

  sidereal time / latitude

  I am carrying these as globals simply because the image structure
  did not have them.  This is silly: I can define them as part of the
  internal image structure without breaking any of the I/O issues.  

  thiscode

  I have been using this as a static photcode structure.  The
  Send_*/Recv_* mechanisms are better suited to simple structures.  I
  probably need to change this to just the code, and use the lookup
  functions to the full structures where needed.  It can be part of
  the options, and I should re-write functions which use it to accept
  the options as an argument rather than rely on it as a global.

  mosaic

  this stores the coords structure for a needed mosaic.  this should
  be kept, filled as needed by the client, and sent to the server with
  the image.  perhaps it can be moved out of global.

2005.10.06

  I am getting seg faults from find_matches_closest, apparently
  related to the partial Measure load.  They occur when trying to
  realloc secfilt.  I tried to debug using the ohana_allocate stuff,
  but it causes problems because fitsio does not fall under the ohana
  memory system.  This is fairly bad; I should probably deal with that
  issue before pursuing the other stuff.  Do I have to make libfits
  depend on libohana, in which case I should probably split off libdvo
  from libohana.  That is probably not a bad plan in any case...

2005.10.07

	I have finished the basic implementation of the update mode.
	I have been able to demonstrate substantial improvements in
	speed when the number of existing measurements dominates the
	total number of measurements and the number of averages is
	typically small compared to the number of measurements (ie,
	most objects are real, detected in most images, and each new
	image supplies many new measurements of objects which exist
	and not many of objects which don't exist already).  The speed
	gain is significant in this context because the average table
	is small compared to the measure table; since both update and
	full-load methods require the complete average table, there is
	no difference in the load time for the average table.  

	I was having some memory collision problems, and attempting to
	use the ohana_allocate functions reminded me that the libFITS
	functions were not supported under ohana_allocate.  This was
	unhelpful.  I bit the bullet and split libohana into libohana
	(base functions only, including ohana_allocate) and libdvo
	(functions based on the libautocode structures).  Doing this
	allowed me to make libFITS depend on libohana (including
	ohana_allocate).  BUT, this forced me to change all LDFLAGS
	entries in ohana to swap -lohana -lFITS for -lFITS -lohana,
	and to add include <fitsio.h> in some cases.

2005.10.06
	split / nosort / update
	I have added a few new concepts to addstar recently: split
	catalog files, nosort for the measurement table, and
	update-only.

	split mode

	The split mode is quite straightforeward.  In this mode, each
	catalog is represented by a set of four files: *.cpt, *.cpm,
	*.cpn, *.cps.  Each file contains only one FITS table of the
	data, along with basic header and empty matrix.  Having
	individual tables for each component of the database lets me
	add entries without re-writing the entire table.  This should
	save on I/O operations in the long run.  

	The first file contains the table of averages, and is the file
	normally identified in the table lookup functions.  The header
	of this file contains the names of the other table files
	(paths relative to the directory containing the cpt file).
	The names and extensions are specified in 'mkcatalog.c'; all
	other functions use the defined filename references, rather
	than expecting a naming convention.

	The additional files contain the measures (cpm), missings
	(cpn), and secfilt (cps) elements of the catalog tables.  

	To facilitate the handling of the additional filenames, file
	pointers, and headers, Catalog was extended to include
	pointers to the measure, missing, and secfilt files as
	additional catalogs.  When the data are loaded into memory,
	these catalogs are locked (as usual), and file information is
	stored in the individual Catalog entries; the data segements
	are all loaded into the main catalog pointers (eg, measures
	are loaded into catalog[0].measure, rather than
	catalog[0].measure_catalog[0].measure).
	
	The function 'load_catalog' auto-recognizes the SPLIT format
	by looking for the header keyword MEASURE, identifying the
	file containing the measures.  The identification of the RAW
	format and the SPLIT format are not cross-checked: if the
	NAXIS keyword is set to 2, the file is assumed to be RAW, even
	if the MEASURE keyword is present.  Careful with this (though
	there is no reason the main matrix should be used in a basic
	database table).

	nosort 

	the nosort option by itself provides a minor processing
	speed-up by deferring the re-sorting of the measurement table
	until after multiple addstar processes are run.  addstar
	should not require the measurements to be sorted, so this step
	can be safetly deferred if only addstars are being performed.
	the other DVO operations require the sorted table, so the sort
	must be performed before they are run (either as part of the
	catalog load, not implemented yet, or with a call to addstar
	without the -nosort option set.  the real goal of the nosort
	option is to enable the -update concept in addstar, in which
	only the new rows are written out; this will only work if
	addstar can handle unsorted measures. 

	the nosort option required the addition of a 'sorted' element
	in the Catalog structure to track if the data are sorted or
	not.  On load, this flag is set based on the value of the
	header keyword SORTED; if the data is sorted during addstar,
	the flag is appropriately set, otherwise it is set FALSE be
	default.  

	The nosort option requires a function which can generate the
	'next_meas' link sequence based on the measure table.  there
	is now a function called 'build_meas_link' which generates a
	correct link list; there is also the pair of functions
	'init_meas_links' and'init_miss_links' to generate the links
	in the event that the table is sorted (should be must
	quicker).

	The 'missing' table is problematic: the LONEOS and ELIXIR
	formats do not carry an averef entry, thus they do not have
	enough information to define the links based only on the
	missing table.  This means we are forced to write out a sorted
	missing table; the nosort option is invalid for the missing
	table.  One future upgrade path is to add the averef entry to
	the PANSTARRS format and then only require the missing table
	to be sorted if the format is old and does not support
	-nosort.  (Note also that, for the moment, the missing table
	has only a single valid format).

	In the process of defining the nosort option, I also cleaned
	up a bit the find_matches functions to use clearer functions
	for the links.  
	
	update

	The 'update' process in principle allows addstar to
	substantially reduce the amount of I/O it needs to perform by
	only requiring addstar to write out new measures and new
	average/secfilt entries.

	The 'missing' table is problematic: since the format does not
	support the 'nosort' option, it is not possible to use update
	with the missing table.  This means we are forced to write out
	a complete, sorted missing table.  This is currently
	implemented in update_catalog_split by simply writing out the
	complete missing table.  In fact, this choice is still flawed
	because the average table, since it is not written out in full
	each time, is inconsistent with the missing table: the Nn
	entries for each average, which identifies the number of
	missing entries, are not updated.  In practice, this means
	that the -update option forces the use of the -missed option,
	though at the moment, this is not forced or checked in any
	way.

	Note that the 'missed' table contains duplicate information
	and can, in principle, be completely regenerated at any time.
	This should be an addstar option: to re-construct the missing
	table, potentially with constraints on the images which are
	searched for matches.  

2005.10.04
	- moved measure/missing list manipulation to separate functions
	- added concept of sorted / unsorted measure catalog
	- defined build_meas_links and reorder_measure,missing
	- some cleanup of both find_matches.c and find_matches_closest.c

2005.10.03:
	- dropping GSCRegion *region entry from find_matches (unused!)
	- adding function find_matches_closest (alternate matches)

2005.08.19:
	changed load_photcode to handle CATMODE and CATFORMAT variations
	- addstar.h: added CATMODE and CATFORMAT globals
	- ConfigInit: read CATMODE and CATFORMAT from config
	- gcatalog: set catalog.catmode from CATMODE
	- mkcatalog: set CATFORMAT and CATMODE for new catalog
	- wcatalog: set CATFORMAT for new catalog
	
	using full photometry conversions in find_matches
	added SetZeroPoint to gstars to enable phot conversions
	
2005.08.15:
	cleanup of the minor Wall,Werror messages

2005.03.07 : notes related to new version of addstar

- USNO seems to work fine now:

  addstar -region 9.0 12.0 9.0 12.0 -cat usno -p USNO_RED
  addstar -region 9.0 12.0 9.0 12.0 -cat usno -p USNO_BLUE

- 2MASS seems to work as well:

  addstar -region 3.0 12.0 12.0 14.0 -cat 2mass -p 2MASS_J
  addstar -region 3.0 12.0 12.0 14.0 -cat 2mass -p 2MASS_H
  addstar -region 3.0 12.0 12.0 14.0 -cat 2mass -p 2MASS_K

- both of these are somewhat inefficient since they need to search
  through the (larger) USNO/2MASS data files for each output ptolemy
  region file.

- GSC seems to work fine now:

  addstar -region 3.0 12.0 12.0 14.0 -cat gsc 
  (photcode is not needed here since there is only one)

- images seems to work fine now

- general concerns:

  - RA 0,360 boundary is likely to have problems, especially with
    region selection
  - mosaic astrometry concepts still need to be added.

  - average magnitudes are being set incorrectly in addstar (not
    taking account of the zero-points).  I don't really want to apply
    the colors here, but I need to do something, perhaps?

    one option is to only apply the correction for the first
    measurement of an object, unlike the way astrometry works
    currently.  both of these should be updated based on
    relphot/uniphot and eqiuvalent relastro/uniastro independently
    from the addstar step.

