Go to the previous, next section.

The Stack -- Why, When, How

Often, it is desired to cycle around many scans, performing the same operation, such as ACCUMing, displaying and scaling, on each of them. If the scan numbers are consecutive, or are spaced by a constant increment, then a FOR-loop, in either "execute mode" or a procedure, can be used to achieve one's ends. However, if the scan numbers are more randomly distributed, things would not be so simple, were it not for the presence of THE STACK.

The stack is actually an internal array of 5120 elements. The numbers of the scans you wish to cycle over are entered into the stack using either of the verbs ADDSTACK or ASTACK. A scalar adverb ACOUNT keeps count of the total number of scans entered on the stack. A FOR-loop can now be used to access the scans through the internal-array verb ASTACK. For example, to list the header parameters of a series of scans, type,

	>SCALAR IZ
	>FOR IZ = 1 TO ACOUNT; GET ASTACK(IZ); HEADER; END

The mechanisms for manipulating and using the stack will now be described in detail.

NOTE : The stack can be used as a temporary array to contain any set of numbers. They need not be only scan numbers. For example, if you are going to be manipulating data stored in the "save" file, you could use the stack to store save-bin numbers. For example:

	>SCALAR IZ
	>FOR IZ = 1 TO ACOUNT; NSAVE=ASTACK(IZ); RECALL; HEADER; END

Putting Scan Numbers on the Stack

To place a series of scan numbers into the stack, the verb ADDSTACK is used. ADDSTACK takes three arguments, namely the numbers of the first and last scans in a series to be added to the stack, and the increment between each of the scan numbers. The scans numbers are placed on the top of the stack, and the adverb ACOUNT, containing the total number of scans on the stack, is incremented appropriately. Suppose it is desired to add scans 600 to 620, 700, 710, 720, and 903 to the existing contents of the stack, type,

	>ADDSTACK(600, 620, 1)
	>ADDSTACK(700, 720, 10)
	>ADDSTACK(903, 903, 1)

The value of ACOUNT will be incremented by 25.

If you want to change a scan number in an already-allocated location in the stack, then use the internal-array verb ASTACK, which takes a location in the stack as its argument, in a direct assignment statement. (See the next section to find how to list the contents of the stack.) For example, if you wish to change the 34-th location in the stack to the value 4210, type,

	>ASTACK(34) = 4210

Of course, ASTACK can also be used to place a number on the top of the stack as follows,

	>ACOUNT = ACOUNT + 1
	>ASTACK(ACOUNT) =  7000

NOTE : Neither ADDSTACK nor ASTACK check whether data actually exist for any scan number they enter on the stack !

Listing the Contents of the Stack

To find out which scan numbers are currently entered on the stack, the verb TELL STACK should be used to list the ACOUNT scan numbers on the stack, i.e.,

	>TELL STACK

Emptying the Stack

If you wish to clear all scan numbers from the stack, and set ACOUNT to zero, use the verb EMPTY by typing,

	>EMPTY

Deleting a Scan Number from the Stack

If you want to remove a scan from the stack, then use the verb DELETE for this. DELETE takes as argument the number of the scan to be removed. Suppose you choose to remove the entry for scan 609 from the stack, then type,

	>DELETE(609)

The adverb ACOUNT will be decremented by one, if an entry with the value 609 is found.

The Power of the Stack

The contents of the stack are accessed via the internal-array verb ASTACK, which takes a location in the stack as argument, and in this case returns the scan number held in that location. As an example of how to put these building blocks together, and use the full power of the stack, suppose that you wish to check out scans 600 - 620, 700, 710, 720 and 903 using SHOW. From this, you decide that scans 610 - 614 are of inferior quality and should be discarded, while the rest of the scans should be averaged, and this average displayed. This could be achieved as follows,

	>SCALAR I_I
	>EMPTY
	>ADDSTACK(600,620,1)
	>ADDSTACK(700,720,10)
	>ADDSTACK(903,903,1)
	>FOR I_I = 1 TO ACOUNT; GET ASTACK(I_I); PAGE SHOW\
	 PAUSE(30); END
	>FOR I_I = 610 TO 614; DELETE I_I; END
	>SCLEAR
	>FOR I_I = 1 TO ACOUNT; GET ASTACK(I_I) ACCUM; END
	>AVE
	>PAGE SHOW

Selecting Sets of Scans

Much manual effort can often be saved if it is possible to select a set of scans from your total database, all of which meet a particular set of criteria. The UniPOPS verb SELECT enables you to do just this, leaving the scan numbers of the selected data on the stack. The new stack entries will begin at location ACOUNT + 1 in the stack, and ACOUNT will be set to the new top-of-the-stack position on completion. SELECT always takes an attribute, which currently can be INFO, ONDATA, OFFDATA, KSCANS, SSCANS, or RSCANS (for Green Bank individual records data).

Select INFO

SELECT INFO displays the current selection criteria, which on entry to UniPOPS should look something like the following for LINE,


SELECT INFO : Status of adverbs used by SELECT verb.

  Object : All Objects (S_OBJECT)
  Mode   : All modes (S_MODE)
  Program: LINE

       Field             From            To                Adverbs

  Scan Number           DEFAULT        DEFAULT      SCAN_MIN, SCAN_MAX
  Subscan Number        DEFAULT        DEFAULT      FEED_MIN, FEED_MAX
  Source X Position     DEFAULT        DEFAULT      X_MIN, X_MAX
  Source Y Position     DEFAULT        DEFAULT      Y_MIN, Y_MAX
  (decimal degrees
  LST (hours)           DEFAULT        DEFAULT      LST_MIN, LST_MAX
  Univ. Time Date       DEFAULT        DEFAULT      UT_MIN, UT_MAX
  (years)
  Rest Freq. (MHz)      DEFAULT        DEFAULT      F_MIN, F_MAX
  Bandwidth (MHz)       DEFAULT        DEFAULT      BW_MIN, BW_MAX

  Additions to ASTACK will begin at location    1 (ACOUNT + 1)
  Additions will not be sorted (S_SORT)

Or the following for CONDAR,


  SELECT INFO : Status of adverbs used by SELECT verb.

    Object : All Objects (S_OBJECT)
    Mode   : All modes (S_MODE)
    Program: CONT

          Field             From            To                Adverbs

    Scan Number            DEFAULT        DEFAULT      SCAN_MIN, SCAN_MAX
    Subscan Number         DEFAULT        DEFAULT      FEED_MIN, FEED_MAX
    Source X Position      DEFAULT        DEFAULT      X_MIN, X_MAX
    Source Y Position      DEFAULT        DEFAULT      Y_MIN, Y_MAX
    (decimal degrees)
    LST (hours)            DEFAULT        DEFAULT      LST_MIN, LST_MAX
    Univ. Time Date        DEFAULT        DEFAULT      UT_MIN, UT_MAX
    (years)
    Rate ("/s)             DEFAULT        DEFAULT      RATE_MIN, RATE_MAX
    Int. Time (s)          DEFAULT        DEFAULT      IT_MIN, IT_MAX

    Additions to ASTACK will begin at location    1 (ACOUNT + 1)
    Additions will not be sorted (S_SORT)

The single-valued object and mode (see below) criteria are set via the 16- and 4-character adverbs, S_OBJECT and S_MODE. The initial defaults for these are blank strings, which signify "all objects" and "all modes", respectively. They can be set to various values as follows,

	>S_OBJECT = '3C454.3'
	>S_OBJECT = ' '
	>S_MODE = 'TLPW'
	>S_MODE = ' '

The available modes at present include:

	 "    " = All Modes	
	 "PS  " = Position Switched
	 "APS " = Absolute Position Switched
	 "FS  "	= Frequency Switched
	 "BSP "	= Beam-Switch + Position-Switch
	 "TPON"	= Total Power On
	 "TPOF"	= Total Power Off
	 "ATP "	= Absolute Total Power
	 "PSM " = Position Switched Map
	 "APM "	= Absolute Position Switched Map
	 "FSM "	= Frequency Switched Map
	 "TPMO" = Total Power Map On
	 "TPMF"	= Total Power Map Off
	 "DRF " = Drift Map
	 "PCAL"	= Position Calibrate
	 "BCAL"	= Beam Calibrate
	 "BLNK"	= Blanking
	 "SEQ "	= Sequence
	 "FIVE"	= Five Point
	 "MAP "	= Continuum Map
	 "FOC "	= Focalize
	 "NSFC" = North-South Focalize
	 "TTIP"	= Total Power Tip
	 "STIP" = Switched Power Tip
	 "DON "	= Continuum On
	 "CAL "	= Calibration
	 "FSPS"	= Freq-Switch + Position-Switch
	 "BSPS"	= Beam-Switch + Position-Switch",
	 "ZERO"	= Zero Check
	 "TLPW"	= Total Power
	 "FQSW"	= Frequency Switched
	 "NOCL"	= No Calibration
	 "PLCL"	= Pulse Cal
	 "ONOF"	= Continuum On-Off Scan
	 "BMSW" = Nutation
	 "PSSW" = Position-Switched, Tucson, old
	 "DRFT" = Continuum Drift Scans, Tucson
	 "OTF"  = On-the-fly, Tucson

"PSPZ" = Position switched, polarizer, Tucson

New modes are constantly being added.

The other criteria are defined by pairs of adverbs (see the middle section of the above table) which define lower and upper limits for each selection item. For example, if you wanted to select all scan numbers from 2000 through 2343, then type,

	>SCAN_MIN = 2000; SCAN_MAX = 2343

The default values for each range are those to give "all-inclusive" selection, i.e. for LST it would be all possible hours, while for scan numbers, it would all available data.

NOTE : Selection by rest frequency and bandwidth is only possible in LINE, while selection by rate and integration time is only possible in CONDAR.

SELECT ONDATA, SELECT OFFDATA, SELECT KSCANS, SELECT SSCANS, and SELECT RSCANS

To select the appropriate set of scans, you invoke SELECT with the attribute ONDATA for the scans in the "on-line data" file, OFFDATA for the "off-line" file, KSCANS for the "keep" file, SSCANS for the "save" file or RSCANS for the Green Bank "individual records" data (both on-line and off-line forms). On completion, the scan numbers in the set will be added to those already on the stack. If the adverb S_SORT is set to 0, the added scan numbers will not be sorted before being placed on the stack, while if S_SORT is set to -1 or +1, the scan numbers will be placed on the stack sorted in descending or ascending order, respectively.

As an example, suppose that for a spectral line database you wanted the stack to contain just the numbers of the scans in the "on-line data" file meeting the following criteria,

	i) Source = W49A.
       ii) Total-power mode of observing.
      iii) No restriction on scan numbers.
       iv) Observation positions in the range,
	    19h 7m 40s < RA < 19h 08m 00s ; 9deg 00" < Dec < 9deg 03".
	v) All LST's.
       vi) Observations made in 1989 and 90.
      vii) Frequencies between 1400 and 1440 MHz.
     viii) All bandwidths.
       ix) All the numbers in the stack to be sorted in ascending
	   order.

To obtain this, assuming nothing about the existing settings of the select adverbs, type,

	>EMPTY
	># First clear the stack.
	>
	>S_OBJECT = 'W49A'
	>S_MODE = 'TLPW'
	>SCAN_MIN = DEFAULT; SCAN_MAX = DEFAULT
	>FEED_MIN = DEFAULT; FEED_MAX = DEFAULT
	>X_MIN =286.916666667 ; X_MAX = 287.0
	># R.A.'s turned into degrees.
	>
	>Y_MIN = +9.00000; Y_MIN = +9.05000
	>LST_MIN = DEFAULT; LST_MAX = DEFAULT
	>UT_MIN = 1989.00; UT_MAX = 1990.99999
	>F_MIN = 1400.00; F_MAX = 1440.00
	>BW_MIN = DEFAULT; BW_MAX = DEFAULT
	># Last of the criteria.
	>
	>S_SORT = +1
	># Sort the selected scan numbers in ascending order.
	>
	>SELECT INFO
	># Check the criteria got entered correctly.
	>
	>SELECT ONDATA
	># Enter the selected scans into the stack.	
	>
	>TELL STACK
	># See what scan numbers have been selected.

The criteria could be filled more conveniently using a procedure (Chapter 13) to prompt you for the values. The SELECT INFO command should show the following,

SELECT INFO : Status of adverbs used by SELECT verb.

  Object : "W49A            " (S_OBJECT)
  Mode   : "TLPW" (S_MODE) Total Power
  Program: LINE

        Field             From            To                Adverbs

  Scan Number            DEFAULT        DEFAULT      SCAN_MIN, SCAN_MAX
  Subscan Number         DEFAULT        DEFAULT      FEED_MIN, FEED_MAX
  Source X Position    286.9167         287.0000     X_MIN, X_MAX
  Source Y Position    9.000000         9.050000     Y_MIN, Y_MAX
  (decimal degrees)
  LST (hours)            DEFAULT        DEFAULT      LST_MIN, LST_MAX
  Univ. Time Date      1989.000         1991.000     UT_MIN, UT_MAX
  (years)
  Rest Freq. (MHz)     1400.000         1440.000     F_MIN, F_MAX
  Bandwidth (MHz)        DEFAULT        DEFAULT      BW_MIN, BW_MAX

  Additions to ASTACK will begin at location    1 (ACOUNT + 1)
  Additions will be sorted in ascending order (S_SORT)

Go to the previous, next section.



Webmaster: Ronald J. Maddalena
Send questions or comments to


Cookbook table of contents               NRAO information