SORT-11 is an utility program that allows you to read an input file, to sort the contents, and to write out the sorted data into an output file. The sorting sequence is determined by control fields, also known as key fields, within the data itself. The general form of a SORT-11 command line is: outfilespec=infilespec<,specfilespec> where: outfilespec = output file specification infilespec = input file specification specfilespec = specification file specification /sw = one or more of SORT's switches For a list of valid switches type HELP SORT SWITCHES For help on a particular switch type HELP SORT 2 SWITCHES Switch Position in Line Function ====== ================ ======== /AL:n Output Initial Allocation in Blocks /BL:n Both Magtape Blocksize /BU:n Output Bucketsize for output file /CO Output Make output file Contiguous /DE:x Input Put Scratch files on named device /FI:n Both Use n Scratch files /FO:x:n Both Format and recordsize of file /IN:x Input Indexed file with x keys /KE:abm.n Input Key specification /PR:x Input Type of sort process required /RE Output Relative File /SE Output Sequential File /SI:n Output Retrieval Window Size 2 AL /AL:n Output Initial Allocation in Blocks /AL:n is used to specify the initial space allocation for the output file. N must be between 0 and 65,535. The default is taken from the input file for record and tag sort processes. For index and address routing sort processes, the default size is based on the number of records sorted. 2 BL /BL:n Both Magtape Blocksize /BL:n valid for magtape files only and specified the nonstandard magtape block size in bytes. The default is /BL:512 2 BU /BU:n Output Bucketsize for output file /BU:n specified the RMS bucket size (the number of 512-byte blocks per bucket). Note that the size of the block is the standard 512 bytes even if the /BL switch is used. The default value of this switch depends on the organi- zation of the input file. If the input and output files are of the same organization, the default for the output file is the input file value. If the input files differ in organization or the default is required for the input file, the default value is 1. 2 CO /CO Output Make output file Contiguous /CO specifies that the output file will be contiguously allo- cated. The default is /-CO (noncontiguous). 2 DE /DE:x Input Put Scratch files on named device /DE:x specifies the device on which you want the scratch files. This switch overrides any device specification from the task build options. X is any 1- to 4-character device name with the colon that normally follows that name omitted (for example, DB3). 2 FI /FI:n Both Use n Scratch files /FI:n specifies the maximum number of intermediate scratch files. Legal values are from 3 to 8. The default is /FI:5 2 FO /FO:x:n Both Format and recordsize of file /FO:x:n specifies the record format (x) and the maximum record size (n) isn a file. X may take the following values: FIXED STREAM VARIABLE UNKNOWN Record format values may be truncated to the first letter. N is the exact record size in bytes for FIXED length records. For the other formats, n is the size in bytes of the largest record present. N is required only on input. This switch must be present in input files even if the format is UNKNOWN. The default for x is VARIABLE. 2 IN /IN:x Input Indexed file with x keys /IN:x specifies the INDEXED SEQUENTIAL file organization. X specifies the number of keys (no default). 2 KE /KE:abm.n Input Key specification /KE:abm.n specifies which fields control the sequence of the output file. You can specify more than one key field (up to 10) if you separate each description from the next with a colon. For example: /KE:BN1.6:C8.2 a specifies the way in which the data is to be handled. If it is omitted, the default value is C. The possible values are: B, C, D, F, I, J, K, P, Z For information about any of them, type HELP SORT KE letter b defines the general sort order. The possible values are: N Ascending order O Opposite, or descending, order m is a decimal number giving the first byte of the key field starting from 1. This item must be present. n is a decimal number giving the length of the key field in bytes. This item must be present. 3 B /KE:B twos complement binary 3 C /KE:C Alphanumeric 3 D #K 3 F /KE:F 2- or 4-word floating point binary 3 I #K 3 J #K 3 K /KE:D If the characters are alphabetic, numeric with the sign superimposed over the units digit, or contain slashes (/), use the value of the digits group. For example: A2CD5 = (+) 12345 and A/47J = (-) 11471 If the characters represent a standard FORTRAN IV number, such as 12, -35, 42.98, or -0.76E+3, convert the number to binary for storage or evaluation. /KE:I Same as D, but with the sign leading and separate, so that the first byte of the field is a + or - /KE:J Same as I but with the sign trailing and separate /KE:K Same as D but with the sign leading and overpunched (54321, for instance if positive, would come out as 5432A. the negative of 54321 would be 5432J). 3 P /KE:P Packed decimal format 3 Z /KE:Z ASCII zone 2 PR /PR:x Input Type of sort process required /PR:x specifies the type of sorting process to use. The legal values are: R - RECORD (the default) T - TAG A - ADDRESS ROUTING I - INDEX 2 RE /RE Output Relative File /RE specifies RELATIVE file organization 2 SE /SE Output Sequential File /SE specifies SEQUENTIAL file organization 2 SI /SI:n Output Retrieval Window Size /SI:n specifies the size of the retrieval window. The ranges and defaults for this switch are determined by the operating system. Check with your system manager.