Digital PDFs

AA-D346A-TC

July 1978

90 pages

Original

2.8MB

Document:	TRAX SORT Reference Manual
Order Number:	AA-D346A-TC
Revision:	0
Pages:	90
Original Filename:	http://bitsavers.org/pdf/dec/pdp11/trax/AA-D346A-TC_TRAX_V1.0_SORT_Reference_Manual_Jul78.pdf

OCR Text

July 1978

This document discusses the features and uses of the SORT
utility on TRAX systems. This document and the software
to which it pertains are for use with RMS-l1·formatted files
only.

TRAX
SORT Reference Manual
Order No. AA-D346A-TC

OPERATING SYSTEMS AND VERSIONS:

TRAX Version 1.0

SOFTWARE VERSION:

3.0

To order additional copies of this document, contact the Software Distribution
Center, Digital Equipment Corporation, Maynard, Massachusetts 01754

digital equipment corporation · maynard, massachusetts

First Printing, July 1978

The information in this document is subject to change without notice
and should not be construed as a commitment by Digital Equipment
Corporation. Digital Equipment Corporation assumes no responsibility
for any errors that may appear in this document.
The software described in this document is furnished under a license
and may only be used or copied in accordance with the terms of such
license.
No responsibility is assumed for the use or reliability of software on
equipment that is not supplied by DIGITAL or its affiliated companies.

The postage-prepaid READER'S COMMENTS form on the last page of this
document requests the user's critical evaluation to assist us in preparing future documentation.
The following are trademarks of Digital Equipment Corporation:

DIGITAL
DEC
PDP
DECUS
UNIBUS
COMPUTER LABS
COMTEX
DDT
DECCOMM
ASSIST-II

DECsystem-lO
DECtape
DIBOL
EDUSYSTEM
FLIP CHIP
FOCAL
INDAC
LAB-8
DECSYSTEM-20
RTS-8

MASSBUS
OMNIBUS
OS/8
PHA
RSTS
RSX
TYPESET-8
TYPESET-II
TMS-ll
ITPS-IO

CONTENTS

Page
vii

PREFACE
CHAPTER

CHAPTER

INTRODUCTION TO SORT

1-1

1.1
1.2
1.3
1.4
1.4.1
1.4.2

DATA FILES
COMMAND STRING AND SPECIFICATION FILE
OPERATION OF SORT
SORT PROCESSING OPTIONS
Record Sort (SORTR)
Tag Sort (SORTT)

1-2
1-4
1-7
1-8
1-8
1-9

USING THE SORT COMMAND

2-1

2.1
2.1.1
2.2
2.3

THE SORT COMMAND STRING
Multiple Line Input
INDIRECT COMMAND FILES
RMS-ll FILE SPECIFICATION INFORMATION

2-1
2-2
2-2
2-2

SORT EXECUTION OPTIONS

3-1

3.1
SWITCHES
ALLOCATION (/AL:n)
3.1.1
3.1.2
BLOCKSIZE (/BL:n)
BUCKETSIZE (/BU:n)
3.1.3
CONTIGUOUS (/CO)
3.1.4
3.1.5
DEVICE (IDE :x)
3.1.6
FILES (/FI: n)
3.1.7
FORMAT (/FO : x : n)
3.1.8
INDEXED SEQUENTIAL (/IN:X)
3.1.9
KEY (IKE: abm. n)
3.1.10
PROCESS (/PR: x)
3.1.11
RELATIVE (IRE)
3.1.12
SEQUENTIAL (/SE)
3.2
THE SPECIFICATION FILE
3.2.1
Header Specification
3.2.1.1
Notes and Comments on Header Specification
Entries
3.2.1.2
ALTSEQ Format and Notes
3.2.2
Record Type Specification
3.2.2.1
Notes and Comments on Record Type
Specifications
3.2.3
Field Specifications
3.2.3.1
Notes and Comments on Field Specification
Entries
3.2.4
Using a Specification File
3.3
SORT SPECIFICATIONS SUMMARY

3-15
3-16
3-18

ERROR CONDITIONS

4-1

4.1
4.2

ERRORS IN SORT: GENERAL COMMENTS
COMMAND DECODER ERRORS

4-1
4-1

iii

3-1
3-1
3-1
3-1
3-2
3-3
3-3
3-3
3-3
3-3
3-4
3-5
3-5
3-5
3-8
3-9
3-10
3-10
3-12
3-14

CONTENTS (Cont.)

Page
4.3
4.4
4.5
4.5.1
4.5.2
4.6
4.7
4.8

SPECIFICATION FILE ERRORS
I/O ERRORS
CONTROL PROGRAM ERRORS
Pre-Sort Errors
Merge Errors
OTHER ERRORS
RMS ERROR CODES
SORTS ERRORS

4-3
4-5
4-6
4-6
4-6
4-6
4-6
4-9

FILES

5-1

5.1
5.2
5.3
5.4
5.4.1
5.4.2

FILE NAME CONVENTIONS
FILE MAINTENANCE
FILE ALLOCATION
SCRATCH FILES
Scratch File Structure
Using Scratch Files

5-1
5-1
5-1
5-2
5-2
5-3

APPENDIX A

CHARACTER SETS USED BY SORT

A-I

APPENDIX B

PRINTABLE CHARACTERS

B-1

APPENDIX C

SORT PROGRAM EXAMPLES

C-1

APPENDIX D

RMS I/O STATUS CODES

D-1

CHAPTER

GLOSSARY

G1ossary-1

INDEX

Index-1

FIGURES
FIGURE

1-1
3-1
3-2
3-3
C-1
C-2
C-3

Sample Record Types
Specification File Format
SORT Specification Form
Header Specification Entries
Sales List
Overtime Analysis
Employee List

1-3
3-5
3-7
3-17
C-2
C-3
C-4

TABLES
TABLE

1-1
1-2
1-3
2-1
2-2
3-1
3-2
3-3
3-4

Selecting the Sorting Process and Devices
that Best Suit the Processing Environment
Sorted Output File
Sorting Process Options
Switch Summary
Header Specifications
Record Type Specifications
Field Specifications

1-3
1-4
1-9
2-4
2-4
3-2
3-18
3-19
3-21

CONTENTS ( Con t • )

Page
TABLES (Cont.)
TABLE

4-1
4-2
4-3
A-l
A-2

B-1
B-2
0-1

RMS Status Codes for TRAX
RMS File Attribute Mismatch status Codes
Common Non-RMS Status Codes
The ASCII Character Set with Corresponding
EBCDIC Codes
The Subset of the EBCDIC Character Set Used by
SORT when EBCDIC Sorting is Requested
Printable Characters Grouped by Equal Digits
Printable Characters Grouped by Equal Zones
RMS Error Status Codes

4-7
4-7
4-8
A-2
A-3
B-2
B-3

0-1

PREFACE

This manual describes the features and operation of the SORT utility
program under TRAX on the PDP-II.
Chapter I describes the SORT
program, its operation and environment. Chapter 2 is an operator's
guide for running SORT. Chapter 3 is a reference for those who write
the instructions for the SORT program.
Chapter 4 describes error
conditions, and Chapter 5 describes the use of files in SORT.
Appendix A lists the ASCII character set and a subset of the EBCDIC
code.
Table A-I lists the ASCII code values in ascending octal
sequence with the corresponding EBCDIC values.
This is useful for
comparing values when requesting a modified ASCII sorting sequence or
when defining forced key fields.
Table A-2 lists the EBCDIC subset
used when you request an EBCDIC sorting sequence. The EBCDIC values
are in ascending order with the corresponding ASCII values.
Appendix B contains two tables to help you determine the digit and
zone values of printable characters. For example, the characters B,
K, and S have an equal absolute digit value of 2:
however,
the
"eleven" punch denotes a negative value.
The digit values of J
through R are negative (-) I through 9.
Appendix C gives examples of SORT applications.
Appendix 0 lists RMS status codes.
A glossary is included to define terminology used in this manual.

vii

CHAPTER 1
INTRODUCTION TO SORT

The SORT utility program allows you to read any input file,
sort the
contents, and write out the sorted data onto an output file.
The
sorting sequence is determined by control fields, or key fields,
within the data itself. If you do not wish to sort your data base,
you can still use SORT to extract key information, sort that
information, and store the sorted information on a permanent file.
You can later use that file to access your data base in the order of
the key information on the sorted file. The contents of the sorted
file may be entire records, key fields, or record indices relative to
the position of each record within the file (the first record on the
data base is record 1, the second, 2, etc.).
SORT provides two sorting techniques:
1.

Record Sort (SORTR)
Record Sort produces a reordered file by using the entire
contents of each record as the record key. A record sort can
be used on any acceptable input device and can process any
valid RMS (Record Management Services) format.

Tag Sort (SORTT)
Tag Sort produces a reordered file by sorting only the record
keys. SORT then randomly reaccesses the input file to create
a resequenced output file according to those record keys.
The tag sort method conserves temporary storage, but can only
accept input files residing on disk.

The SORT utility program may be controlled by a command string and an
optional specification file. There is a simple format for each.
If
your SORT application does not require that records be restructured or
that only a subset of the input file be sorted, then you need only a
command string to control SORT.
SORT can handle any RMS-valid file organization.
Different file
organizations are distinguished by the ordering of the records they
contain and the way they handle the retrieval process.
The order of the records in a sequential file is determined by the
order in which the records are read from the file. The first record
in the file is the first record read out, regardless of whether the
records are written to the file in some sorted order or not.
A relative file consists of record areas that are identified by
relative record numbers. The first record area in the file is record
number 1, the second is 2, etc., much the same as an apartment house
where the first apartment is 1, and so forth.
But, as in an apartment
house, if you want the record that is in the twelfth record area, for

1-1

INTRODUCTION TO SORT
instance, you must ask for record number 12, even though there may not
be records in areas 1 through 11. Relative files can reside only on
disk.
An indexed file contains prologue information, one or more indices,
and the data records themselves.
To retrieve information, you ask for
the proper record by primary or alternate key. The system looks up
the key in the appropriate index and retrieves the record using the
record pointer associated with the key.
Indexed files can reside only
on disk.
Table 1-1 shows the devices that you may use to supply data
Data may be stored in binary, ASCII, or EBCDIC form.

1.1

SORT.

DATA FILES

SORT may accept a file from anyone
available in the system configuration:
1.

Disk unit

Magtape unit

Support Environment terminal

the

peripheral

devices

A record is usually divided into several logical areas called data
fields.
The data in each field mayor may not be relevant to SORT.
Each field may be interpreted as a record identifier,
key data, or
general data related to the logical content of the record and not
relevant to the sorting process.
SORT uses record identifiers to
distinguish the various types of records in a file.
SORT uses the key
fields in each record to reorder an input file. Any other data field
in a record may be retained in the output file or ignored by SORT.
Figure 1-1 shows three different types of input records, each with a
different format.
The record identifiers are the letters in position
1: S means Sales record, 0 means Order record, and R means Restock
record.
In this case, the keys chosen for sorting the Sales record
types are the "item number code" in positions 2 to 7, and the "number
of items sold" in positions 8 to 13. The "total amount of sale" is an
example of a data field not relevant to the sorting process.

1-2

INTRODUCTION TO SORT
Table 1-1
Selecting the Sorting Process and Devices
that Best Suit the Processing Environment
WORK
FILE

OUTPUT
FILE

INPUT
FILE

SORTING
TECHNIQUE

Disk (3-8 files)

SORTR

Disk

(Record Sort)

Magtape

Magtape*

Terminal

SORTT

Disk

(Tag Sort)

Magtape

Magtape*
Printer
Terminal

* Provided records are at least 18 bytes long.
ANSI format.

Magtape

must

CUSTOMER
NO.
CODE

DATE
OF
SALE

RECORD
IDENTIFIER

SALES
RECORD

ITEM
NUMBER
CODE

NUMBER
OF ITEMS
SOLD

UNIT
CODE

PRICE
PER
UNIT

TOTAL
AMOUNT
OF SALE

ORDER
RECORD

DE SCRIPTION
OF
ITEM

ITEM
NO.
CODE

PRICE
OF
ITEM

NUM8ER
OF ITEMS
ORDERED

PURCHASE
ORDER
NO.

RESTOCK
RECORD

ITEM
NO.
CODE

UN IT
CODE

COST
PER
UNIT

MINIMUM
STOCK
QUANTITY

PRESENT
AMT.
OF STOCK

REORDER
NUMBER
11-'520

Figure 1-1

Sample Record Types

If you request a sort in ascending order on the sales records
illustrated above, the sort is based on the item number code first and
then on the number of each item sold within that item number.
In
order of decreasing significance, the keys are:
1.

Item number

Number of items sold

The output file contains all sales records in the order shown in Table
1-2.
Data
fields other than those you use for keys, e.g.,
salesperson's code, may be in any order.

1-3

INTRODUCTION TO SORT
Table 1-2
Sorted Output File
First Key:Item Number

Second Key:Quantity

lowest item no.

lowest quantity

lowest item no.

next higher quantity

·
·
·

·
·
·
lowest item no.

highest quantity

next higher item no.

lowest quantity

next higher item no.

next higher quantity

·
·

·
·
·

next higher item no.

highest quantity

highest item no.

lowest quantity

highest item no.

next higher quantity

··

·
highest item no.

1.2

highest quantity

COMMAND STRING AND SPECIFICATION FILE

You invoke the SORT program by entering a command string.
string has three functions:
system

To reference devices in the
current sort.

file

the

To specify switches that define file parameters used
sorting process.

the

To reference a specification
switches to control the sort.

file

for

each

The command

specify

other

Several command string switches define the sorting process parameters.
One switch describes record formats and the maximum record size.
Another delimits the internal work files.
Others provide detailed
file information to RMS.
Normally, you direct the sort with a specification file.
Or, you may
use two additional switches instead of a specification file to control
a sort.
One switch specifies the sorting process option;
the other
identifies the key fields.
The use of these switches is limited to
sorting an input file of uniform format:
1-4

INTRODUCTION TO SORT
1.

The key fields must reside in
record of the input file.

The file must contain only the records to be included in the
sort. The figure below illustrates a general sort that would
require only a command string and switches.

I
I

the

same

location

D
B

every

I
B

SORT USING COMMAND STRING AND SWITCHES

The specification file is the supplement to the command string, which
provides the basis for controlling and directing the sorting process.
The specification file provides the following controlling features:
1.

Record Selection

I
I

D
C

RECORD SELECTION

You can include or omit any records from the sorting process.
The output file will contain only the records that you
specify.

1-5

INTRODUCTION TO SORT
2.

Alternate Collating Sequence

0
C

B
C

_____ 8-0,0-8---+

I-

ALTERN~.TE

COLLATING SEQUENCE

If necessary, you can specify an
alternate
collating
sequence. The normal sequence is that implied in ASCII code.
One alternate choice is EBCDIC values.
The other
is an
individual alternate collating sequence (ALTSEQ). An ALTSEQ
can be used to change the ASCII values of the normal
sequence.
It applies to all the alphanumeric key data in the
records, but only during the actual sorting process.
The
output record remains unchanged.

Forced Keys

333
102

351

242
---+3nn-/nn---'

242
102

351
333

FORCED KEY

An ALTSEQ applies to all positions of the key.
Forced
keys
allow you to specify an alternate sequence for particular
positions within the key.
You can specify an alternate
sequence by substituting a lower-valued character, such as
the slash (I) in the example above.
Since the slash comes
before 0,
the 300-series records in the example are brought
to the front of the file.
Notice that the records so treated
are in sequence and in front of the rest of the sorted file.
The net effect is that of two sorted files,
one behind
the
other.

1-6

INTRODUCTION TO SORT
4.

Input Format variation

Abbby

I .....

N 141
A bbby

A 0001

N 207

N 207
N 141

INPUT FORMAT VARIATION

If the input file contains records with several different
formats,
you can identify those records by type so that they
may be properly handled.
In the example above, A and N are record identifiers.

Output Format variation

I
I

B mn 19

0 UW 23

C Sf 17

uw 0 23
sf C 17

mn B 19
'y A 11

A 'y 11

OUTPUT FORMAT VARIATION

You can change the format of the data file during the sort,
but you cannot change the contents of any given data item.

1.3

OPERATION OF SORT

The SORT program consists of two basic parts:
a control program and a
subroutine package called SORTS.
The control program directs the
overall processing. SORTS serves as a collection of subroutines that
the control program uses during its processing.
There are three phases of operation in the SORT control program.'* In
the first phase, SORT reads the command string, decodes, and stores
the switch values and the specification file, if present. Any errors
in the command string or specification file are reported at this
point.

* Each phase of operation corresponds
program structure.
1-7

overlay

the

SORT

INTRODUCTION TO SORT
The second phase begins the pre-sort operation. The control program
is called to open and read the input file and establish the keys.
Then the SORTS subroutine begins the initial sorting process. At this
point,
the amount of available internal storage space becomes
important to the efficiency of the sort.
If that space is not
sufficient to hold all the records, SORT builds strings of sorted
records and transfers them to scratch files on bulk storage devices.
In order to merge these files and complete the sort, space for at
least three scratch files must be available.
The SORT program
normally provides for a maximum of eight scratch files.
Either a
switch in the command string or the amount of available internal work
space can reduce the number of scratch files used.
The merge phase rebuilds the intermediate scratch files into a merged
file.
Another subroutine reads the records in the proper sequence.
The records are then written into the output file.
If there are no
scratch files to merge because main memory was sufficient to hold all
the records, the sorted records are written directly into the output
file. After the last record is written, the control program cleans up
the scratch files and returns to the first phase; SORT is then ready
to accept another job.

1.4

SORT PROCESSING OPTIONS

There are two SORT processes: Record Sort and Tag Sort.
You can
specify the SORT process in the specification file or with a switch in
the command string. The default process is the Record Sort.
Each
process has its particular input requirements, processing methods, and
resultant output files.

1.4.1

Record Sort

(SORTR)

The Record Sort (SORTR) outputs all specified record data in a sorted
sequence.
Each record is kept intact throughout the entire sorting
process. Since it moves the whole record, SORTR is relatively slow
and may require considerable main memory or external storage work
space for large files.
NOTE
Records can be FIXED, STREAM ASCII, or
VARIABLE length in any valid RMS (Record
Management Services) format.
The size
of the records on a fixed length format
file is determined when the file is
created.
Variable length records carry
a count along with them that specifies
their length. This length data is used
by the RMS routines and is transparent
to SORT.

1-8

INTRODUCTION TO SORT
Table 1-3
Sorting Process Options
Type of
SORT

Type of
File

Record Size
and Format

SORTR
(RECORD
SORT)

Input
and
Output

any

Any appropriate device
disk,
including:
magtape*

Slow

SORTT
(TAG
SORT)

Input

any

disk

Output

any

any appropriate device
(including magtape*)

Slow for
large
files,
large keys

Device

Speed

Magtape labels are ANSI standard.

1.4.2

Tag Sort (SORTT)

The Tag Sort (SORTT) produces the same kind of output file as SORTR,
but it only handles record pointers and key fields.
Since SORTT moves
a smaller amount of data than SORTR, SORTT usually performs a
faster
sort than SORTR.
The input file must be randomly re-accessed to
create the entire output file, which may be a lengthy process for
large files.

1-9

CHAPTER 2
USING THE SORT COMMAND

The SORT utility is installed as part of the TRAX Support Environment.
It can be invoked from any TRAX support terminal by using the SORT
command string. This chapter describes the available forms of the
SORT command. See Chapter 3 of this manual for details about specific
properties of the SORT utility.

2.1

THE SORT COMMAND STRING

Two forms of the SORT command are possible.
the syntax:
)SORT[/Qualifier(s)]

The

standard

form

uses

Input-Filespec[/file-qualifier(s)]

Qualifiers and File-qualifiers are discussed in Chapter 3.
They are
used to specify the type of sort and the attributes of the file being
sorted, as well as passing control information to the SORT utility.
A second form of the SORT command allows for interactive prompting
the input file specifications:

)SORT[/Qualifier(s)]
FILE?Input Filespec [/file-qualifier (s)]
To illustrate the SORT command string, consider a payroll program
example.
The payroll program has generated a disk-based file of
SO-character fixed-length records called WAGES.DAT which contains the
Social Security Account Number along with wage and tax data for each
employee of a firm.
Prior to reporting wages to the Social Security
Administration, you must verify that the file is in ascending SSAN
sequence. The following command string will sort the file WAGES.DAT
on a nine-character SSAN key beginning in character position one of
the wage record.
)SORT/KEYS: (1.9)/PROCESS:RECORD WAGES.DAT/FORMAT:FIXED:80
The qualifiers /KEYS and /PROCESS, as well
/FORMAT are explained in detail in Chapter 3.

the

file-qualifier

After you enter the SORT command string,
the SORT utility begins
processing.
Any
errors
detected
in
the command string or
specification file are printed immediately on your support terminal.
At the end of a successful run, SORT returns the following messages:
SRT
SRT

M:ELAPSED REAL TIME:
hh:mm:ss
M:TOTAL RECORDS SORTED:
nn

2-1

USING THE SORT COMMAND
This message reports the wall-clock time necessary to perform the sort
and the number of records in the output file.
You are then returned
to the TRAX operating system where you may enter additional SORT
command strings, or go on to some other operation.

2.1.1

Multiple Line Input

In the event that your SORT command string becomes too long to fit
in
a single line, division of the command string is possible. Simply use
a hyphen (-) as the last printing character of a line, enter a
carriage return, and continue your command string on the next input
line.
You should avoid splitting qualifiers and their arguments over
more than one line.
Referring again to the example of the
divide that command string as follows:

wages

file

sort,

you

could

>SORT/KEYS(1.9)DCL>/PROCESS:RECORD DCL>/WAGES.DAT/FORMAT:FIXED:80
DCL> is the prompt of the DEC Command Language interpreter, which
decodes your command string and passes this information to the
operating system. Note that the second line PROCESS:RECORD has a
space after
the word RECORD.
This space is required to distinguish
the command qualifiers from the input file specification and should be
present at all times.
You may also subdivide indirect command file
input strings in the manner described above.

2.2

INDIRECT COMMAND FILES

The TRAX Support Environment allows you to create indirect command
files which reside on your system device and are made up of command
string input. With the SORT command,
this facility allows you to
create a command file which can then be executed at regular intervals
for certain recurring types of processing.
In the case of the wages
example cited earlier
in this chapter, you might create a file
WAGSRT.CMD which contains the following information:
SORT/KEYS: (1.9)/PROCESS:RECORD

WAGES.DAT/FORMAT:FIXED:80

You execute this indirect command file by typing the following string:
>@WAGSRT
The file WAGSRT.CMD is then retrieved by the indirect command file
processor and decoded to invoke the SORT command with the qualifiers
you have specified.

2.3

RMS-ll FILE SPECIFICATION INFORMATION

A TRAX file specification conforms to standard RMS-ll conventions.
has the following form:
device: [ufd]filename.filetypeiversion

2-2

USING THE SORT COMMAND
where:
device

is the name of the physical device on which the volume
containing the desired file is mounted.
The name
consists of two ASCII characters followed by a 1- or
2-digit octal unit number and a colon(:); for example,
LPO: or OBI:. A logical device name may also be used.
Dev:

Device

SY:

system disk, default device

TI:

user's terminal

TTn:

Support Environment Terminal

LPn:*

line printer

DBn:

RP04/05/06 disk

DMn:

RK06/07 disk

DRn:

RM02/03

MMn:

magnetic tape (TU45/TE16)

* n specifies the device unit number;
specified, n is assumed to be o.
[ufd]

not

is the user file directory specification consisting of
two octal numbers in the range of 1 through 377
(octal). These numbers must be enclosed in brackets
and separated by a comma and must be in the following
format:
[group,member]
For example, member 225 of
following entry:

group

300

would

use

the

[300,225]
filename

is the name of the desired file.
The file name can be
from one to nine alphanumeric characters; for example,
BILLRTN. You must always specify the file name. There
is
no
default specification for this component.
Failure to specify the file name causes an error to be
generated.

filetype

is the 3-character file type identification.
Separate
the file name and filetype with a period (.). Files
with the same name but different
functions
are
distinguished from one another by the filetype;
for
example, BILL.TSK and BILL.OBJ might be the task file
and the object file,
respectively, for the program
BILL. You may omit the file type, but you should avoid
this practice when dealing with system programs which
generally assume a default filetype
for
various
operations. The wild card (*) specifier may be used in
place of a file type. The action specified will be
applied to all file types associated with a given file
name or wild card file name.
(e.g., *.* specifies the
current
version
of all files in the user file
directory.)

2-3

USING THE SORT COMMAND
version

is the octal version number of the file in the range 1
through 77777
(octal).
Separate the extension and
version by a semicolon (i).
Various versions of the
same file are distinguished
from each other by the
version numberi
for example, BLL.OBJil and BILL.OBJi2
are successive versions of the same file.
The version
number may be omitted,
in which case the current
(highest-numbered) version is assumed.
To act upon all
versions of a file, you can use the wild card
(*)
specification.

The device, user file directory
specification are all optional.

code,

filetype,

and

the

version

Table 2-1 lists the default assumptions applied to missing components
of a
file
specification.
Table 2-2 lists the filetypes assumed by
TRAX system software.
Table 2-1
Default

Item
device

user's current default device

[ufd]

user's current default [ufd]

version

for an input file, the default version number is
the existing version with the highest (octal)
number for that file.
for an output file, the default is calculated as
one greater than the highest existing version
number for that file.
Table 2-2
File Contents
Description

Default
Filetype

task image file
memory allocation file
symbol definition file
object module
object module library file
overlay description file
indirect command file
COBOL source text file
BASIC-PLUS-2 source file
MACRO-II source file

.TSK
.MAP
.STB
.OBJ
.OLB
.ODL
.CMD
.CBL
.B2S
.MAC

2-4

CHAPTER 3
SORT EXECUTION OPTIONS

3.1

SWITCHES

SORT switches describe the input and output files and allow you to
change SORT defaults, such as the number of scratch files used. All
numeric switch values are decimal, not octal. All switches are valid
for both input and output files unless otherwise noted. Table 3-1
summarizes the switches and other pertinent information for your
convenience.

3.1.1

ALLOCATION (/AL:n)

The ALLOCATION switch is valid for the output file only. You can use
this switch to specify the initial space allocation for the output
file. Legal values range from 0 to 65,535. If you do not use this
switch, the output file allocation defaults to the input file size for
the record and tag sort processes.

3.1.2

BLOCKSIZE (/BL:n)

The BLOCKSIZE switch is valid for magtape files only and is used to
specify a nonstandard magtape blocksize. If this switch is not used,
the blocksize defaults to the standard 512-byte block.

3.1.3

BUCKETSIZE (/BU:n)

The BUCKETSIZE switch specifies the RMS bucket size
(the number of
512~byte
blocks per bucket). Please note that the size of the block
is the standard 512 bytes even if the BLOCKSIZE switch is used.
The
default value of this switch depends on the organization 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.

3-1

SORT EXECUTION OPTIONS
Table 3-1
Switch Summary
Switch Name

Code

Default Value

Validity
Output

ALLOCATION

/AL:n

BLOCKSIZE

/BL:n

See Section 3.1.1
512 bytes

BUCKETSIZE
CONTIGUOUS

/BU:n
/CO

Input file or 1
Non-contiguous

Output
Output

DEVICE

/DE :X

SY:

Input

FILES
FORMAT

/FI:n
/FO:x:n

Both
Both

INDEXED SEQUENTIAL

/IN:x

5
(See Section 3.1.7)
(See footnote)

KEY

/KE:abm.n

C, N, first position
of field, length of
field

Input

OUTPUT

/OU:fi1espec

Output

PROCESS

/PR:x

Input file
R (record sort)

RELATIVE

/RE

(See footnote)

Output

SEQUENTIAL

/SE

(See footnote)

Output

SIZE

/SI:n

(See Section 3.1.14)

Output

* The default for input files is the type of
time.

file

Both
(magtape)

Input

present

OPEN

The default for output files is SEQUENTIAL. The output record format
will default to the characteristics of the input file unless you use
STREAM ASCII input record format.
STREAM ASCII will produce a
VARIABLE output record format.
Default record sizes are the size of the input record.

3.1.4

CONTIGUOUS (/CO)

The CONTIGUOUS switch is valid for output files only.
It specifies
that the output file will be contiguously allocated. This means that
each successive block assigned to a file will be physically located
between its logical predecessor and its logical successor with no
filler or extraneous material separating the blocks. The default is
non-contiguous.

3.1.5

DEVICE (/DE:x)

The DEVICE switch lets you specify which device you want to be
associated with the scratch files. This switch overrides any device
specification from task build options. The parameter x is any valid
1- to 4-character device name with the colon that normally follows
that name omitted (e.g. DB3).

3-2

SORT EXECUTION OPTIONS
3.1.6

FILES (/FI:n)

The FILES switch specifies the maximum number of intermediate
files.
The default is 5.

3.1.7

scratch

FORMAT (/FO:x:n)

The FORMAT switch specifies the record format
(x)
and the maximum
record size (n) in a file. The record format x may take the following
values:
FIXED,
STREAM,
VARIABLE, or
UNKNOWN
Record format values may be truncated to the first letter.
The
maximum record size 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. The record size is required only on input.
This switch must be present in input files even if the format is
UNKNOWN.
The default of the record format (x) is VARIABLE. The
record format on output can be used to override the input record
format.
For the default of n, see the footnote of Table 3-1.

3.1.8

INDEXED SEQUENTIAL (/IN:X)

The INDEXED SEQUENTIAL switch specifies the INDEXED SEQUENTIAL file
organization.
For default values, see the footnote of Table 3-1. X
specifies the number of keys (no default). This switch is valid on
input only.

3.1.9

KEY (/KE:abm.n)

The KEY switch is the most important switch in the SORT parameters.
It tells SORT which fields are to 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. Here is an
example of two key fields:
/KE:BNl.6:C8.2
The field description sequence in the
which breaks down as follows:

basic

format

above

abm.n

specifies the way in which the data is to be handled.
-If it is omitted, the default value is C as described
below. But first, if you have not used SORT before,
take a look at Appendix B and become acquainted with
the concepts of zones and digits.

Here are the legal values for the
parameter and their interpretations:
B

two's complement binary

alphanumeric

3-3

data

handling

SORT EXECUTION OPTIONS
D

1.
if the characters are alphabetic, numeric with the
sign superimposed over the units digit, or contain
slashes (I), use the value of the digits group
(see
Table B-1).
Here are two examples and their values:
A2CD5

A/47J

(+)12345

(-)11471

If the characters represent a standard number, such as
12, -35,
42.98, or -O.76E+3, convert the number to
binary for storage or evaluation
F

2- or 4-word floating point binary

same as D, but with the sign leading and separate,
that the first byte of the field is a + or -

same as I but with the sign trailing and separate

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.)

packed decimal format

ASCII zone (see Table B-2)
defines the general
(ascending order) .

sort

order.

ascending order

opposite, or descending, order

The

default

is a decimal number giving the first byte of the key
field.
Number from the first byte of the record which
is byte 1. This item must be present.

is a decimal number giving the length of the key
in bytes. This item must be present.

3.1.10

field

OUTPUT (IOU: filespec)

The OUTPUT switch specifies the file that will receive the sorted
records.
Default is the input file.
3.1.11

PROCESS (/PR:x)

The PROCESS switch specifies the
Legal values are:

type

sorting

process

use.

switch,
for output only,
specifies
RELATIVE
For defaults, see the footnote of Table 3-1.

file

R - RECORD (the default)

T - TAG
3.1.12

RELATIVE (IRE)

The RELATIVE
organization.

3-4

SORT EXECUTION OPTIONS
3.1.13

SEQUENTIAL (/SE)

The SEQUENTIAL switch, for output only, specifies SEQUENTIAL
organization. For defaults, see the footnote of Table 3-1.

3.1.14

file

SIZE (/SI:n)

The SIZE switch specifies the size of the retrieval window. The range
and default for the switch are determined by your system manager.

3.2

THE SPECIFICATION FILE

The specification file offers a variety of controls for the sorting
process.
These controls are entered in a format consisting of up to
three types of records or lines in the specification file;
a fourth
type may be used if nonstandard collation is required. Figure 3-1
shows two possible arrangements of the specification file whose
elements are explained below.

DATA FIELD
DATA FIELD

RECORD TYPE

FIELD
SPECIFICATIONS

r - - - RECORD TYPE
SPECIFICATION

DATA FIELD
DATA FIELD

RECORD TYPE

------------..,
ALTSEQ
HEADER

r - ALT-;E; -

FIELD
SPECIFICATIONS

I
I
I

-l

~<L"'NATE

HEADER
RECORD TYPE
SPECIFICATION

L - ALTERNATE COLLATING

--1

SEQUENCE RECORDS
HEADER RECORD

FIELD
SPECIFICATIONS

COCCA'''G

SEQUENCE RECORDS

HEADER RECORD

~------------------~

With Record Type Specifications

Figure 3-1

Without Record Type Specifications

Specification File Format

3-5

SORT EXECUTION OPTIONS
1.

The Header
The first record in a specification file must be the
The header tells the SORT program:

•

the kind of sorting process to be used,

•

the key field and output record size,

•

whether an alternate collating sequence is
and

•

whether the key fields are to
output.
There
is only one
execution.

Header.

used,

be stripped from the
Header for each SORT

ALTSEQ Records
If you use an alternate collating sequence, ALTSEQ records
follow the header.
ALTSEQ records allow you to change the
order in which a character or characters is sorted.
You can
use these records to:

Move certain records to the front or back of a file

Group them in one place within the file

Change the order in which records appear in a
sequence within the file

given

Record Type Specifications
If you do not use an alternate collating sequence,
the next
type of record or line in the specification file is record
type specifications.
They control record selection and allow
SORT to work with variable record formats.
If all records
have an identical format (or the format is not important) and
all are to be included in the output file, the record type
specification may be omitted.

Field Specifications
If record type specifications appear, each is followed by any
applicable
field
specifications.
If
not,
the field
specifications follow the header (behind any ALTSEQs).
The
two
kinds
of
field
specifications
are
key
field
specifications and data field
specifications.
Key field
specifications define and locate each key.
They are listed
in
order
of
decreasing
significance.
Data
field
specifications define and locate all the data fields to be
written to the output \file.
The data will appear in the same
order in the output file as the data field specifications.
Key field specifications appear first
in the specification
file and are followed by any data field specifications.
The format of each type of specification record is fixed.
The SORT specification form,
is shown in Figure 3-2.
The
following entries are common to all
three
types
of
specification file lines.

3-6

SORT EXECUTION OPTIONS
Column

Entry

Notes

1-2

Page number

Required only when
different
types
of records are to _be
described.
A separate
page,
numbered in ascending sequence,
should be used for each record
type and its corresponding Field
Specifications. Only the first
page has a header specification.

3-5

Line number

Depicting line sequence.
If
column 5 is blank, 0 is assumed.
Thus a digit entry in
this
column can be used to identify
later line insertions.

Specification Type

H, I, 0, or F as shown below

40, etc. Comments

Ignored by SORT processing

SORT SPECIFICATIONS
Page

HEADER SPECIFICATION

75 76 77 78 79 80

CD ~:~;7.:"ion I I I I I I I

Programmer

SORTR
SORTT
SORTA
SORTI

~
3.

01 0/

Total

iii 0

i
£"

Length of
Key Fields

SORTR,T

Mode of Processing

Line

£~~ Record
8~~
_ z ?

NOT USED

Output

Comments

U m 91h

IProgram Identification!

89101112 i3i41!.1611 18192021 22?3?4?!s 2671 7879]011 17 33301 350]631]8 H

S/OIRITI I

/11

II/

4041474] 44 450 46.' 48 4Q!SO 501502 5ol!t. 50550657508596061626364656661 6a 69 101172 731.

[Ill

1/1

I /I

RECORD TYPE SPECIFICATION
o
Line

Factor 1

~ ~

~ ~ ~

~ 8~ §

Field Location
From
To

Factor 2

I-,.---------------i

LT
LE ~

~..:...-_--=-~onstant - - - -_ _ _ _ _ _ - ,

i
I

Field Location

From

910 1112 1114 15161718 ·9102122 2) 14:?5o 26

Comments

Record Name

nln ,") 1031 J? 11)4]50 36 11 JB l<.l 40 41 4243 U 45046 411.18.950501 50? 5354 5556 ,)1 &859606162636465666168697071 1'! 13 74

o 1

f---r+-r-~t1~L+-~~~~~~~~~~~~~~+,~~~~~~~~

o 2

- ~f---~+-

o 3
o 4

-I--f--

'- -+-j'-++lf--t--+-+-+-++-++-H++-t++-H-+~~H+-+--1H-~H~+-+---+-l

o 5
o 6
FIELD SPECIFICATION
Forced
Lme

0 u...

u...~~

Field Location

-----,

! ~ §l----.--~a:
~ U

From

Field Ndme

10 )112131415 1611

lO l i n n24 250

16 1118 14.1031 11 11 J.t

)S 3631181940414143 U

Comments

I
I

45 464748 .Ill SO 51 ')2515455565758509606161 63U 65 66 6168 69 10 11127314

o 7
f- -

o 8

1---

o 9

-f--I-- -f- -- -

1 0

f--1 2 I1 3

- - - f--

~~-+f----l
1 4

-t
-

-f---

- i-

Ii i j 1 t

f--

-i-

Figure 3-2

r- -

j
l-

I--

f--f-- -

- f-- r-f--

-- H--

SORT Specification Form

3-7

- I-f-

SORT EXECUTION OPTIONS
NOTE
Whole lines of comments may also be
included in the file;
these are denoted
by an asterisk (*)
in column 7.
The
comment lines may appear anywhere in the
specification file and serve to document
the file in its printed listings.
They
do not affect SORT operations.
In the following material,
criteria apply:

unless

otherwise

stated,

the

following

•

Numeric data is decimal, not octal.

•

Either leading zeroes or leading
right-justified entries.

•

All field position definition records begin at column 1.

3.2.1

blanks

are

acceptable

Header Specification

See Section 3.2.1.1 for notes and comments:
Columns

Explanation and Legal Entries

1 - 5

Page and line numbers: described in
Section 3.2

Header record identification
Legal value: H

7 - 12

Type of SORT (must be left-justified)
Legal values:
SORTR or blanks - Record SORT
SORTT - Tag SORT

13-17

Total of all key field sizes
Legal values: 1 - 16383
N.B.: must be equal to the total size in
bytes of the largest record key on the file
and right-justified

Normal sort order sequence
Legal values:
A or blank - ascending
D - descending

19 - 25

(not used)

3-8

SORT EXECUTION OPTIONS
Alternate collating sequence

Legal values:
blank - standard ASCII sequence
E - EBCDIC sequence
X - user-modified ASCII sequence

(not used)

Output option
Legal values:

x - key fields created by SORT will
be stripped from the output record
blank - no action
29-32

Output record length
Legal values:
decimal number equal to the
number of bytes in the largest
output record

33-39

3.2.1.1

(not used by SORT - may be used for
comments)

Notes and Comments on Header Specification Entries

Column of
Entry Affected

Notes and Comments

This

field

may

qualified by N or
the field
specification (q.v.)

This field identifies the sort order for
fields identified as alphanumeric (C
in
column
8
of
the
field
specification) •
If X is entered
here, the header must immediately be
followed by ALTSEQ
cards.
See
Section
3.2.1.2 for ALTSEQ card
format.
Standard ASCII
is assumed
for
forced key field specifications
and this field cannot be used to
amend that assumption.
SORT does
not perform file
transliteration.
Files
that
are not already in
standard ASCII characters must be
transliterated before the soit.

o entered in column 7 of

The EBCDIC
characters
sequence.

3-9

option specifies ASCII
sorted
into
EBCDIC

SORT EXECUTION OPTIONS
28

If a blank appears here,
the key fields
may be treated as data fields and
moved to the output record as any
other data field would be, whether
relocated within the output record
with respect to the input record or
not.

29 - 32

To determine this number, add the sizes
of all the data fields in the field
specifications
(plus the sizes of
the key fields if column 28 is
blank) for the largest record in the
file.
If neither SORTT nor SORTR is
to be run at this time, then an
entry in this field is not needed.

3.2.1.2

ALTSEQ Format and Notes
Notes

Column

1 - 6

ALTSEQ - mandatory entry

7 -

(not used)

9 - 80

3.2.2

Replacement character groups (12), of 6 positions each;
each group contains the octal representations of
two ASCII characters: the character being sorted
out of
sequence,
and
the
character
that
corresponds to its new position in the sorting
sequence.
For example, the correct entry to
replace an ASCII space with an ASCII zero (0)
would be 040060;
040 for the space and 060 for
the zero.
The result of this would be to sort
spaces into the same relative position in the
sequence as Os.
Notice that preexisting Os are
not affected by this entry, but they may be
affected by another entry on the same or another
ALTSEQ card under the same header.

Record Type Specification

See Section 3.2.2.1 for notes and comments.
Columns

Explanation and Legal Entries

1 - 5

Page and line numbers. See Section 3.2
for description

Inclusion or Omission character
Legal values:
I - Include the records described
in this line in the sort

o - omit the records described
in this line from the sort

3-10

SORT EXECUTION OPTIONS
7

Continuation character (allows for cases
in which several conditions define a
single record type)
Legal values:
A - logical AND
o - logical INCLUSIVE OR

Field mode (defines the processing to be
applied to the data field defined)
Legal values:
B C D F I J -

K P Z 9 - 12

binary
alphanumeric
numeric
binary floating point in
2 to 4 words
numeric with leading and
separate sign
numeric with trailing and
separate sign
numeric with sign leading
and overpunched in first byte
packed decimal
zone, see Table B-2, Appendix B

Factor 1 (position of the base comparison
factor is in the first byte; the first
byte in the record is byte 1)
Legal values:
a decimal number - location of the
first byte
blanks - specifies a single byte
field

13 - 16

Factor 1 endpoint
Legal values:
a decimal number - location of
last or only byte in the
field

17 - 18

Relationship of Factor 1 to Factor 2
Legal values:
EQ NE LT GT LE GE -

equal
not equal
less than (i.e. Fl LT F2)
greater than
less than or equal to
greater than or equal to

3-11

SORT EXECUTION OPTIONS
Comparison factor format (indicates whether
Factor 2 is a field or a constant)

Legal values:
F - identifies a field
C - identifies a constant
20 - 39

Factor 2 (This takes two different formats
depending on whether Factor 2 is a field
or a constant.)
If Factor 2 is a field:
20 - 23

location of first byte - a
decimal number or blanks
as for Factor 1.

24 -

location of the last or only
byte - a decimal number as
for Factor 1.

(not used)

28 - 39

If Factor 2 is a constant:
the value of the constant

20 - 39
40 -

3.2.2.1

(not used - available for comments)

Notes and Comments on Record Type Specifications

Columns

Explanation

(general)

If all the records in the input file are to be
included in the sorted output file and they all
have the same format or you are using the entire
record
as
the
key,
then
no record type
specification is necessary.
If all the records
have the same format and some are to be omitted,
then only one record type specification is needed.
If several different formats of records are to be
sorted, then there must be
a
record
type
specification
for
each.
Each
record type
specification used should be followed by its
appropriate field specifications.

If this column contains an I
and there are no
further entries in this line, all records not
previously described are to be included in the
sort input.
Records that are not described in
Include lines will be ignored.
Because SORT
processes Include and Omit lines sequentially, you
should be very careful to get your lines in the
right order, particularly if any records are
described on more than one line. The last Record
Specification line before a Field Specification
must be an Include line.
If you describe the
records to be processed by omitting the unwanted
records from the file, then you can use a line
with an I
in column 6 and no further entries in
the line as the final record specification.
3-12

SORT EXECUTION OPTIONS
7

Multiple AND and OR lines are permissible, but the
series will be processed strictly in sequence,
each line ANDed or ORed to the accumulated Boolean
expression in turn:
If A is False and Band C are True,
the expression A OR (B AND C) is
True logically, but in order to
get
the correct results from
SORT, arrange it this way:
BAND C OR A
SORT will react:
(B AND C) OR A,
which gives
value.

the

required

True

On the other hand,
A OR BAND C
gives a True value in both the
SORT and non-computerized logic
as it stands.

C - alphanumeric information limited to 256 bytes
D - numeric value, maximum size 8 bytes after
conversion to binary.
If the number is
presented in standard ASCII numeric format,
as for a decimal number with or without
decimal point, floating point exponent, or
sign at the beginning, the SORT will convert
the number to binary. But the maximum size
is still 8 bytes.
F - maximum size 8 bytes - two or four
word
floating
binary
representation
before
conversion
Z - ASCII zone, maximum size 1 byte
B-2 in Appendix B

see

Table

9 - 16

Factor 1 - Sort identifies records by comparing
the value of one or more fields in a record
with constants or other fields in
the
record. These other fields or constants are
referred to as Factor 2.

17 - 18

defines the Factor 1 - Factor 2 relationship

20 - 39

Factor 2 - See notes under Factor 1 above.
If
Factor 2 is a field, it must be ihe same
length as Factor 1. If a constant, after
conversion to the corresponding internal
format, be the same length as Factor 1 and
be
in the same mode
(see Field Mode,
position 8).

3-13

SORT EXECUTION OPTIONS
3.2.3

Field Specifications

There are normally a set of field specifications following each record
type specification in the specification file set.
(For one exception,
see paragraph 3.2.2.1, general notes.) If there are no record type
specifications,
any appropriate field specifications follow the
header. Field specifications have two main jobs:
•

to describe the output file format

•

to determine, for each record type, the
sort is to be based.

keys

List the key field specifications first,
in order
significance.
Then add data field specifications
portions of the original input record.

which

the

of decreasing
to put out any

List the Key and Data field specification lines in the specification
file in the same order as the required output record format.
SORT
creates the output record according to the order, size, and contents
of the key and data specification lines.
Place an X in column 28 of
the Header to strip unwanted key fields from the output record.
If
the key field is specified twice, as a key field and as a data field,
the key field can remain in the output record where the data field
specification puts it, even if the key fields are stripped. See
Section 3.2.3.1 for further commentary.
Columns

Explanations and Legal Entries

1 - 5

Page and line numbers - see Section 3.2

F - specifies a field specification

Field type - specifies keys and their sort
sequences and data fields
Legal values:
N - key field, normal sort
sequence (see position 18 of
Header)

o - key field opposite sequence
F - key field, special treatment
required (see positions 17 - 19)
D - data field (if used, then
column 8 should be C)
8

Field type - should be C if column 7 is
D - these options are the same as those
for column 8 of the record type specifications

9 - 12

Field location - location of first byte of
multi-byte field
Legal values:
a decimal number - location of first
byte of field
blanks - specifies one-byte field

3-14

SORT EXECUTION OPTIONS
13 - 16

Field location - last or only byte
Legal values:
a decimal number - location of last
or only byte in the field

17 - 19

Forced key fields
17

Trigger character - identifies the
character needed to trigger the
force:
Legal values:
blank - the force always occurs
any other character - the force
occurs only if this character is
found

Replacement character
Legal values:
any legal character

Continuation character
Legal values:
blank - this is a new forced
key
any other character - this
key is continued from
the previous line

20 -

3.2.3.1

(not used - available for comments)

Notes and' Comments on Field Specification Entries

Columns

Explanation

Enter key fields first in order of
decreasing significance.

17 - 19

Use this field to change the content of
a
I-character
key
field,
thereby
changing the sorting sequence.
The
original data will still appear on the
output record, since only the key will
be changed.
To specify a forced key,
put F in column 7
of
the
field
specification and the nature of the
force in columns 17
19.
Since the
force is based on characters, there must
be a C in column 8. A force does not
apply to any other keys or fields than
the one stated.

3-15

SORT EXECUTION OPTIONS
To help you understand how th~se forced key fields affect the
here is a description of how the SORT handles these fields:

SORT,

When column 7 contains an F, SORT sets up a one-byte field in its key
area and sets its value to either 377(octal) for ascending or 0 for
descending sequence. Note that the alternative collating sequence
does not apply to forced key characters.
If column 17 is not blank, SORT compares its contents to the content
of the field defined in columns 13 - 16.
If an equal condition
occurs, SORT moves the character to the reserved key byte previously
described.
By using more than one Field Specification, you can force more than
one character out of normal sequence, such as sorting the entire
alphabet out ahead of numerics. But beware:
1.

Unless the displaced character is itself displaced to another
position in the sorting sequencE, it will appear with the
character displacing it in the order in which the two
characters were found
in the input file.
Thus,
if A
displaces 1 and a series appears in the input file as (A, B,
1,
A, 1, 1, A, C, D), the sorted output will be (A, 1, A, 1,
1, A, B, c, D), so ...

When in doubt,
force the displaced character elsewhere,
particularly when it is known to appear in the input file and
is not omitted elsewhere in the SORT specification file.

If, on the other hand, column 17 contains a blank, no comparison is
performed and the value of column 18 always replaces the initial value
in the reserved key byte. This has the following results:
1.

When associated with conditional keys in preceding lines,
records not satisfying the prescribed conditions can be
grouped into any position in the output file, not necessarily
the end of the file.

When this specification stands alone,
it
becomes
an
unconditional force of all records in the group.
This form
can be useful when the group is already identified by a
record type specification.
It has no effect on the entire
file if applied to the entire file.

If column 19 of the field specification is blank or the preceding line
does not specify a force key, SORT only reserves a new key byte. The
same byte is modified in all other cases. This makes it possible to
combine several conditions within the same key, such as arranging an
output file in the order 4, 2, 3, 1, rather than 1, 2, 3, 4 or 4,
3,
2, 1. However, SORT still expects the field location to be defined in
columns 13 - 16, even if each continuation line covers the same field.

3.2.4

Using a Specification File

The sample input file contains records in the format shown in Figure
1-1.
The file may contain null records such that there may be a
record identifier and no other data in the record.
SORTR is entered in columns 7 through 11 of the Header.
The header
specification also indicates that the key is to be sorted in ascending
sequence and that an alternate collating sequence is to be used.
The
ALTSEQ line following the header tells SORT to interpret all nulls as
spaces.
3-16

SORT EXECUTION OPTIONS
The null records are omitted from the sorting process with an Omit
line in the record type specification. The Include lines direct SORT
to process only the RESTOCK records where the present amount of stock
is less than the stock quantity allowed.
ORDER Records, SALES
Records, and RESTOCK records where the present amount of stock is
greater than the stock quantity allowed are ignored.
The field specification shows that the selected records will be sorted
on the reorder number. Since this is a SORTR, the Item Number Code,
Unit Code, and Cost per unit data fields, in positions 2-14, transfer
to the output records.
The output records will have the reorder
number in positions 1-6, and the Item Number Code, Unit Code, and Cost
per Unit in positions 7-19. Note that the total length of the key
fields and the output record length in the header specification are
calculated from the field sizes indicated in the field specification.

SORT SPECIFICATIONS
Page

HEADER SPECIFICATIDIII

Progr.".mer
~ode of Proceu;n~
Line

SORTR
SORTT
SORTA
SORTI

'"!

~
3.

5 6

Total
Length of
Key Fields

SORTR,T

~Dx

~
~

£~! OutPUt
Recore

NOT USED

o~~
~z~

Com~nts
(Program Identification)

Length

9101112 1314151611 1.192021 221)242!t 26?1 21 29)0 II 32 Jl 34 ]$ 36 31 31 39 40 4. 42 U .... 45 '6 41 n 4, SO $1 52 5) S. 55 56 $7 SIS' 60 61 62 til '465 "

01 01 H slolRITI I

1111 I I

1111

III

I I I I I I [ I I I I I I I I

67 ' I 19 70 11

n 73 ,.

II I I I I I I I I I I I I I I I I I I I I I I I I

RECORD TYPE SPECIFICATION
o

<!!:

L,ne

~ (J

•

Field Location
From
To
9 10 II 12

Factor 2

h--------------I

NE
GT
Constant -----~
GE
~- - - - - ,
- - - --...,
LT
Field Location
j
I
LE ~
From
To
I
Record Name
I

o g !!!

~;
~ .§ ~
1

Factor 1

Comments

l~ 14 IS .6 11 II ., 20 2122 211425 26 111,1 19 10 31 32 13 34 3536 3111 ]IJ 40 41 4243 Ie 45 t6 41 1ea 49 SO $1 52 5354 55 56 57 51 59606151 U 6465" 6' 61 69 70 71 1Z 13 1.

o 2
o 3
o 4
o 5

o 6
FIELD SPECIFICATION
Forced

-5g
~

NOT USED

13 8
1 4 5 6 7 ' 910 I1J2 )l141516

o 7

FrC~

o 8
o 9

F tole
F"ill
F

1 0

Field Name

Comments

I
I

II 19 2021 222124 25 26 21 281fJ ]0 11 12 II 14 1!1 16 1J 38 19 4~ 41 '141 U 45 46 41 .. I 49 SO 51 '2 5) !Ie 55 56 51 51596.0 6J 62 U 64 656651 61 "

1'i!l19l
lie!

I-+-+-t-t-t--+--H-t--+-t-t-I-++-+-t-f-+++ -+- ,-. -1---.,1--+-+-++-+
12

1 3

1 4

- I-- - -

1-+-+-t-t-I--+--H-t--+-t---l--I-+-t-+-+4 1--

-~-

- -_.-

- - -. -

Figure 3-3

- r--

--- -

f.--

Header Specification Entries

3-17

10 1) 12 711'

SORT EXECUTION OPTIONS

3.3

SORT SPECIFICATIONS SUMMARY

Table 3-2
Header Specifications
Entry

Explanation

Header Specification

7-12

SORTR
SORTT

Record Sort
Tag Sort

13-17

1 - 16383

Decimal number specifies total
length
of
all key fields
listed in Field Specifications
(must
be
the maximum for
SORTR)

A or Blank

Processing sequence: Ascending
or Descending

Column
6

ASCII collating sequence
EBCDIC collating sequence
ASCII with modification listed
in following lines

Blank
E

29-32

Blank

The key fields will reside at
the beginning of each output
record

The
key
fields
will
be
stripped from the beginning of
the output records

Decimal

Number

3-18

This
entry
specifies
the
number of bytes of all key and
data fields listed in
the
Field Specifications (must be
maximum for SORTR)

SORT EXECUTION OPTIONS

Table 3-3
Record Type Specifications
Column

Entry

The records described in this record
specification will be included in the
sorting process.
If there are no
entries in the following columns, all
records not yet described will be
included

The records described in this record
specification will be omitted from
the sorting process

The condition identified by
this
record specification is a logical AND
with the condition from the previous
line

The condition identified by
this
record specification is a logical OR
with the condition from the previous
line

Binary - the
field
contains
complement binary data

Character - the field in columns 9-16
(Factor 1) is alphameric

Digit - use the
digit
value
convert the number to binary

Floating - treat the field as a 2- or
4-word floating binary representation

Same as D but with sign leading and
separate, i.e., the first byte of the
field is a + or a -

Same as D but with sign trailing and
separate, i.e., the last byte of the
field is + or -

Sa~e
as D but with sign leading
overpunch, i.e., sign is superimposed
upon first byte of the field

Packed - the
decimal data

Zone - use the
entry

Explanation

field
zone

contains
value

2's

packed
for

the

(continued on next page)

3-19

SORT EXECUTION OPTIONS
Table 3-3 (Cont.)
Record Type Specifications
Column

Entry

Explanation

9-12

Decimal Number

Location of the first
field (Factor 1)

13-16
17-18

the

Blanks

Single byte field

Decimal Number

Location of last byte of field
location of single byte field

EQ
NE
LT

GT
LE
GE

20-23

byte

Factor 1 must equal Factor 2
Factor 1 must not equal Factor 2
Factor 1 must be less than Factor 2
Factor 1 must be greater than Factor
2
Factor 1 must be less than or equal
to Factor 2
Factor 1 must be greater than or
equal to Factor 2

The following entry describes a field
to
be
compared
with the field
specified in columns 9 through 16

The following entry
describes
a
constant to be compared with the
field specified in columns 9 through
16

Decimal Number

Location of the first byte of the
field (Factor 2) compared with Factor
1

Blanks

Factor 2 is a single byte field

24-27

Decimal Number

position of last byte of Factor 2
field
or location of single-byte
field

20-39

Character String

The constant compared with Factor 1
entry must agree with mode and length
of Factor 1

3-20

SORT EXECUTION OPTIONS
Table 3-4
Field Specifications
Explanation

Entry

Column
6

Field Specification

Normal - key field sequenced as indicated
in column 18 of header

Opposite - key field sequenced opposite
to column 18 of header

Forced - single byte key field with
special sequence

Data field

Binary - the key field is in 2's
complement binary notation

Character - alphanumeric key or data
field

Digit - use digit value or convert to
binary for FORTRAN IV numbers

Floating - the key field is in 2- or
4-word floating binary representation

Same as 0 but with sign leading and
separate, i.e., the first byte of the
field is a + or a -

Same as 0 but with sign trailing and
separate, i.e., the last byte of the
field is + or a -

Same as 0 but. with sign leading overpunch, i.e., sign is superimposed upon
first byte of the field

Packed - the key field is in packed
decimal

Zone - zone value

9-12

Decimal
Number

Location of first byte of field

Replacement
Character

The character which will be substituted
when the force condition (if any) is met

Continuation
Character (any
character other
than blank)

This specification line covers the same
forced key field as the preceding line(s)

Blank

This line designates a new forced key

3-21

CHAPTER 4
ERROR CONDITIONS

4.1

ERRORS IW SORT: GENERAL COMMENTS

When SORT exits, it always prints the status of the sort operation.
The status message is either SUCCESS or an error message for one of
the following errors:
•

Command decoder errors

•

Specification file errors

•

I/O errors

SORT errors are reported at the terminal only and are not recoverable.
The format of a SORT error is as follows:
[-RMS-status-code]

SRT -- control-phase:?message
where:
control-phase

is the SORT phase in control at the time
error occurred. Legal values are:

the

C - command decoder
M - merge
P - presort

4.2

message

is a one-line
happened.

brief

explanation

RMS-status-code

is a decimal status code returned by RMS for
additional information on file errors only.
If RMS is not impacted by the SORT error,
this status code does not appear. Status
codes likely to be seen are listed with their
meanings in Section 4.7.

what

COMMAND DECODER ERRORS

If your error message has a control-phase of C, you have made a
command decoder error. The command decoder error messages and their
probable causes are listed below:
1.

SORT COMMAND ERROR
a.

Too many input files
(more
than
two,
including
specification file) or output files (more than one)

4-1

ERROR CONDITIONS

General syntax error

Too many switches

Erroneous switches on the specification file

An undefined switch

IMPROPER SWITCH:

IFI

Less than three or greater than eight scratch files.

Invalid terminator
NOTE
Valid terminators are period, comma, slash, equal
sign and <RET>. "INVALID TERMINATOR" means that
some other character was used as a terminator or
that SORT expected to find a terminator where
none existed.

IMPROPER SWITCH:

IKE

Invalid letter or value

Start location or size is 0

No period

Illegal size for data mode

Invalid terminator (See NOTE above)

(

. ) between start location and size

TOO MANY KEYS
Buffer space overflowed
NOTE
SORT reserves a buffer area for storage of a table
based on the input specifications in order to control
the processing of each record. This space should be
ample for all situations to make th~s error unlikely.

NO KEYS SPECIFIED
There are no key switches in the command
specification file has been declared.

string

and

KEY AFTER LAST BYTE OF RECORD
The end of an input record key field goes
record size (switch or specification).
4-2

past

the

stated

ERROR CONDITIONS
7.

NO /FO SWITCH
You omitted the /FORMAT switch on the input file.

IMPROPER SWITCH:

/FO

You have not specified a valid format type.

IMPROPER SWITCH:

/PR

You specified an invalid sort process.

4.3

SPECIFICATION FILE ERRORS

In general, the detection of an error in the
results in an appropriate message followed by:

Specification

•

a printout of the entire record, or

•

a printout of the record up to the point where the
detected.

INVALID CHARACTER

error

Column 6 is not H,I,O,F and record is not ALTSEQ.

Process is not SORTR, SORTT, or blank.

Collating sequence is not blank, E, or

Data type is not B, C, 0, F, I, J, K, P,

Key type is not D, F, N, O.

Logical entry is not A, 0, blank, or

File

was

x.
z.

ILLEGAL FIELD
a.

A numeric field in specification
decimal digits or blanks.

No key size is given in Header specification.

No output size is given in Header Specification.

ALTSEQ is misspelled.

ALTSEQ entries do not represent 7-bit octal values.

Last location is less than first location in record field
identification.

Size is invalid for data mode.

Sizes of Factors 1 and 2 in Record Specification
match.

4-3

contains

other

than

not

ERROR CONDITIONS

Compare relation is undefined.

Forced field is other
position.

than

one

Constant given in Factor 2 is greater than 20 characters.

Mode of constant does not agree with mode of Factor.

Invalid characters appear in constant
if the constant is numeric) .

Sign is omitted from binary or packed constant.

(e.g.,

non-digits

NO HEADER
First record of
specification.

specification

file

not

INCORRECT SEQUENCE
a.

Numeric record sequence is lower than sequence previously
encountered.

No valid data specification appears when keys are
stripped from output.

Record specification after
should be last).

Key specifications appear after data specifications.

"include-all"

("include-all"

NO ALTSEQ
a.

type

ILLEGAL CONSTANT

than

Specification for alternate collation entered
column 26 but no ALTSEQ Specifications follow.

Header

Number of specifications for a particular type of
have overflowed the buffer space.

record

TOO MANY SPECIFICATIONS
a.

NOTE
SORT reserves a buffer space for storage
of
a
table
based
on
the input
specifications and used to control the
processing of each record type. This
space should be ample for all situations
to make the error unlikely other than in
very exceptional circumstances.

4-4

ERROR CONDITIONS
8.

LAST RECORD SPEC NOT "I"
a.

Last record specification in file was OMIT.

ILLEGAL KEY
a.

A forced field specification was
SORT (SORTI).

included

Index

10. NO KEYS SPECIFIED
a.

4.4

No key specification appeared before data
in a specification file set.

specifications

I/O ERRORS

Once the sort is under way the only normal errors that can
due to I/O failure, as indicated below:
1.

are

xxxx

OPEN (IN/OUT) FAILURE ON filename.ext xxx x
The file could not be opened by RMS.
the appropriate RMS error code.

occur

The parameter

INPUT/OUTPUT ERROR ON filename.ext xxxx
RMS generated an error while reading or writing this
The parameter xxxx is the appropriate RMS error code.

BAD RECORD SIZE ON filename.ext
RMS tried to read a record on this file that is
the size specified in the /FO switch.

than

The storage capacity of the device
while writing the output file.

has

been

exhausted

INAPPROPRIATE FILE ORGANIZATION
a.

larger

NO ROOM IN filename.ext
a.

file.

/IN switch was not specified.

TEMP FILE ERROR xxxx
a.

An I/O error occurred on a scratch file.
error code.

4-5

xxxx is the RMS

ERROR CONDITIONS
4.5

CONTROL PROGRAM ERRORS

4.5.1

Pre-Sort Errors
1.

NO DATA IN INPUT FILE filename.ext
a.

There are no records in the input file.

The file is not an RMS file.

INVALID KEY FIELD DATA filename.ext
a.

4.5.2

An inappropriate byte has been found in a field described
as D, I, J, K or Z mode.

Merge Errors
1.

# OF OUTPUT RECORDS DOES NOT MATCH # INPUT
a.

The number of records released to the sort does not equal
the number returned.
NOTE
This is a warning message only.
The
output file contains as many records as
specified in the end-of-sort message.

4.6

OTHER ERRORS

Occasionally, errors may occur due to the rejection of data passed
the SORTS subroutine package. They are reported as:

SORT ERROR -- CODE nn
where nn is the code number returned by a SORTS subroutine
in paragraph 4.8).

4.7

(described

RMS .ERROR CODES

A complete listing of RMS status codes can be found in Appendix D.
The status codes that appear in this section represent errors that you
will be most likely to encounter.
The status codes in Table 4-1 are due to file primitive failure.
Check to see if your file is readable and in the correct format.
If
these status codes recur, please send an SPR to DIGITAL.

4-6

ERROR CONDITIONS
Table 4-1
RMS Status Codes for TRAX
Code

Brief Explanation

-160

Read error on file header attributes

-176

Write error on file header attributes

-400

Deaccess error during $CLOSE

-520

Device positioning error

-560

Enter function failure

-624

File extension failure

-1088

File could not be marked for deletion

-1456

Remove function failure

-1520

Error while reading prologue

-1776

File write error

-1792

Error while writing prologue

The status codes listed in Table 4-2 appear because the
attributes you gave did not match those of the existing file.

file

Table 4-2
RMS File Attribute Mismatch Status Codes
Code

Brief Explanation

-192

Buc·ketsize exceeds maximum

-480

Dynamic memory exhausted

-976

Key size equals zero or is too large (indexed
not 4 (relative file)

-1168

Not enough room to open an indexed file

file)

Table 4-3 gives commonly occurring status codes other than RMS.

4-7

ERROR CONDITIONS
Table 4-3
Common Non-RMS Status Codes
Code

Brief Explanation

~112

Variable length records on
ANSI D format

-432

RMS tried to access a deleted record

-448

Syntax error, no such device, or
operation

-464

Syntax error in directory name

-496

Directory not found

-512

Device not ready

-672

RMS attempted to create an already existing file

-704

File locked by another user - access denied

-736

File not found

-752

Syntax errQr in file name

-768

Invalid file options (also occqrs if you
contiguous file)

-784

Device full:

-880

Illegal operation - see Appendix D for examples

-896

Illegal record in sequential file or invalid count field

-1008

Magtape is not ANSI-labelled

-1024

Logical channel busy

-1040

Invalid logical

-1120

Maximum record size equals zero during $CREATE

-1152

Not at end of file

-1408

Invalid record address

-1424

Invalid or illegal record format

-1440

Target bucket locked by another task or stream

-1536

Invalid record in indexed file - file may be damaged

-1568

Record size error during $PUT or $UPDATE

-1744

Syntax error in file version number

-1784

Device write-locked

ANSI-labelled

device

magtape

are

not

inappropriate

for

try

extend

cannot create or extend file

chann~l

number

See Appendix D for a more complete explanation of each status code.

4-8

ERROR CONDITIONS
4.8

SORTS ERRORS

Whenever SORTS detects an error, it returns an octal non-zero code in
the location specified by <Location of Error Code> in the most recent
call. In addition, for I/O errors, Rl contains the address of file
control block for the file in error. The status supplied by RMS is
set into the error status word of that file control block.
(See
Appendix D for error code values.)
The error codes (octal) and their meanings are listed below:
Meaning

Error Code
00

No errors.

Device input error.

Device output error.

OPEN(IN) failure.

OPEN (OUT) failure.

Size of current record is greater than maximum size.

Not enough work area.

RETRN was called after it had exited with a negative
error code (end of sort).

SORT routine called out of order (The order 'of the
calls should be RSORT, RELES, MERGE, RETRN, ENDS).

Sort already in progress (To do a second sort,
must be called to clean up the first sort).

ENDS

Key size is not positive, SORTS detected a zero
negative key size in its calling parameter.

Record size not positive.

Key address is not even (the keys must start
even address because SORT uses word moves) .

Record address is not even.

Scratch records will be too large (the size of the
keys plus the size of the largest record must be
less than 37776 octal).

Too few scratch files are given
scratch files must be specified).

minimum

Too many scratch files are given (a
scratch files may be specified).

maximum

End-of-string record was
expected.

where

none

was

Unexpected end-of-file.

SORT found a record larger than expected.

Record length is not standard for SORTT.

4-9

detected

CHAPTER 5
FILES

5.1

FILE NAME CONVENTIONS

Input, output, and specification files may have any name acceptable to
TRAX.
SORT generates its scratch files by using the temporary file creation
mechanisms of TRAX. Scratch files are automatically deleted from your
file directory by the operating system when SORT is no longer running.

5.2

FILE MAINTENANCE

At the end of a successful run, SORT stores the new output file on the
device you specified.
The new output file may then be used as
immediate input to some other program. The input and specification
files are unchanged.

5.3

FILE ALLOCATION

To optimize you~ sort for speed, SORT tries to allocate in advance the
output file and each scratch file at maximum size. SORT does this by
making an educated guess about the space needed. Under the tag and
record sorting processes, SORT allocates as much space to the output
file as to the input file.
Each scratch file is arbitrarily given 50
percent more space than the input file, no matter what process is
used.
Usually, this pre-allocation process causes no problems.
But you
should
be
aware
of the instances in which problems due to
over-allocation do arise and what you can do about them.
Your pre-allocation size may be too
following conditions are true:

large

one

You are sorting from an indexed file.

You are using OMIT lines in the specification file.

You are using the reformatting option
file.

You are using a smaller record size in the output
in the input file.

the

specification
file

than

Some wasting of space in the output file can usually be tolerated, but
on a crowded disk you may be notified of insufficient space when there

5-1

FILES
really was enough to hold your file.
Try cutting down on the amount
of space by making a conservative estimate of what you need and
including that estimate in the /AL switch on the output file
side of
the command string.
You may perhaps be more concerned about the situation in which you get
the following error message when you try to sort a large file:
P:?TEMP FILE ERROR -784
This means that there is no more room for the allocation of scratch
files.
Here,
too, there may be space wasted due to over-allocation.
Try cutting down the size of the scratch file to one third that of the
input file.
If the error persists, trim the size again.
If this does
not work, then the disk may be too full to handle the sort.

5.4

SCRATCH FILES

5.4.1

Scratch File Structure

The scratch files consist of one or more sequenced strings of records,
followed by an end-of-file flag word.
Each string consists of zero or
more records, in order by key, followed by an end-of-string flag word.
The number of strings on each file varies from time to time as the
sort progresses, until, at the end of the merge phase,
there is one
string on each file.
If there is not enough data in the input file to
produce at least one string on each scratch file,
the unused files
will contain only an end-of-file flag word.
The format of the records stored in these files is as follows:
Word 1

Size of the data portion in bytes.
The
maximum acceptable to SORT is 16K (up to
37777 octal).

Words 2 through n

After possible conversion to a form more
suitable for logical comparison, the key
fields are stored in reverse.

Words n+l* through m

The data portion for record sort
(SORTR)
is the record as read
from the input
file;
for SORTT, it is a 2-word relative
record pointer.

Two unique records also appear in the scratch files as markers.
is one word only containing all 0 bits except as below:
•

Bit 15

indicates end-of-string flag

•

Bit 14

indicates end-of-file flag

* When controlled by specification file, word n+l is used to hold
internal control pointer, and the data starts at n+2.

5-2

Each

FILES
5.4.2

Using Scratch Files

The scratch files switch, FILES (see Section 3.1.6), is only effective
when operating under special conditions.
For each scratch file
potentially to be used, a certain amount of core is required to
establish internal control blocks. This necessarily makes the sort
take longer by:
1.

Restricting the space available for record storage during the
sort, and

Forcing the usage of scratch files even for small input files

By specifying a smaller number of scratch files than the default, you
may be able to release sufficient space to do the whole sort in core
and thereby speed up the sort of a small file.

5-3

APPENDIX A
CHARACTER SETS USED BY SORT

Table A-I:

The ASCII Character Set with Corresponding
Codes

Table A-2:

The Subset of the EBCDIC Character Set
SORT When EBCDIC Sorting is Requested

A-I

EBCDIC

Used

caARACTER SETS USED BY SORT
Table A-I
The ASCII Character Set with Corresponding EBCDIC Codes
ASCI!
Code

EBCDIC
Code

000
001
002
003
004
005
006
007

000
001
002
003
067
055
056
057

010
011
012
013
014
015
016
017

026
005
045
013
014
015
016
017

ETB

020
021
022
023
024
025
026
027

020
021
022
023
074
075
062
046

CAN
EM
SUB
ESC
FS
GS
RS
US

030
031
032
033
034
035
036
037

030
031
077
047
034
035
036
037

SPC

040
041
042
043
044
045
046
047

100
117
177
173
133
154
120
175
115
135
134
116
153
140
ll3

050
051
052
053
054
055
056
057

141

0
1
2
3
4
5
6
7

060
061
062
063
064
065
066
067

360
361
362
363
364
365
366
367

P
q

070
071
072
073
074
075
076
077

370
371
172
136
114
176
156
157

Character
NUL
SOH
STX
ETX
EOT
E~Q

ACK
BEL
BS
HT
LF
VT

FF
C~

SO
51
DLE
DCl
DC2
DC3
DC4
NAK

"
#
$
%

(
)

9
:
;

<
=
>
?

Character
@

A
B
C
D

E
F
G
H
I
J

K
L
M
N
0

P
Q
R

S
T
U

V
W

X
y

Z
[

\
]

\
a
b

c
d
e
f
g

h
i
j
k

1
rn

r
s
t
u
v
w

x
Y

DEL

A-2

ASCII
Code

EBCDIC
Code

100
101
102
103
104
105
106
107

174
301
302
303
304
305
306
307

110
III
ll2
113
114
115
116
117

310
311
321
322
323
324
325
326

120
121
122
123
124
125
126
127

327
330
331
342
343
344
345
346

130
131
132
133
134
135
136
137

347
350
351
112
340
132
137
155

140
141
142
143
144
145
146
147

171
201
202
203
204
205
206
207

150
151
152
153
154
155
156
157

210
211
221
222
223
224
225
226

160
161
162
163
164
165
166
167

227
230
231
242
243
244
245
246

170
171
172
173
174
175
176
177

247
250
251
300
152
320
241
007

CHARACTER SETS USED BY SORT

Table A-2
The Subset of the EBCDIC Character Set Used by SORT
when EBCDIC Sorting is Requested 1
EBCDIC
Code

ASCII
Code
000
001
002
003
011
177
013
014

c
d
e
f
g
h
i

000
001
002
003
005
007
013
014

CR
SO
SI
DLE
DC1
DC2
DC3
BS

015
016
017
020
021
022
023
026

015
016
017
020
021
022
023
010

CAN
EM
FS
GS
RS
US
LF
ETB

030
031
034
035
036
037
045
046

030
031
034
035
036
037
012
027

ESC
ENQ
ACK
BEL
SYN
EOT
DC4
NAK

047
055
056
057
062
067
074
075

033
005
006
007
026
004
024
025

SUB
SPC

077
100
112
113
114
115
116
117

032
040
133
056
074
050
053
041

120
132
133
134
135
136
137
140
141
152

046
135
044
052
051
073
136
055
057
174

153
154
155
156
157
171

054
045
137
076
077
140

172
173
174
175
176
177
201
202

072
043
100
047
075
042
141
142

2
3
4
5
6
7

Character
NUL
SOH
STX
ETX
HT
DEL
VT

[

<
(

+
!
&

1
$

*)
;

t or

,
%

or
>
?
:

#
@

,
=
"

a
b

Character

1
m

n
0

P
q
r

s
t
u
v

w
x
Y

A
B
C
D
E

F
G
H
I
J

K
L
M
N
0
P

Q
R
\
S
T
U
V

X
y
Z

0
1

8
9

EBCDIC
Code

ASCII
Code

203
204
205
206
207
210
211
221

143
144
145
146
147
150
151
152

222
223
224
225
226
227
230
231

153
154
155
156
157
160
161
162

241
242
243
244
245
246
247
250

176
163
164
165
166
167
170
171

251
300
301
302
303
304
305
306

172
173
101
102
103
104
105
106

307
310
311
320
321
322
323
324

107
110
111
175
112
113
114
115

325
326
327
330
331
340
342
343
344
345

116
117
120
121
122
134
123
124
125
126

346
347
350
351
360
361

127
130
131
132
060
061

362
363
364
365
366
367
370
371

062
063
064
065
066
067
070
071

1 This table shows only those EBCDIC characters for which there are
ASCII equivalents. It therefore illustrates the collating sequence
SORT applies when the Header Specification contains E in column 26.

A-3

APPENDIX 8

PRINTABLE CHARACTERS

Table 8-1:

Printable Characters Grouped by Equal Digits

Table 8-2:

Printable Characters Grouped by Equal Zones

B-1

PRINTABLE CHARACTERS
Table B-1
Printable Characters Grouped by Equal Digits

Group

Character

DEC 029
Card Code

Group

Character

blank

No punches

12-5

11-5

12
11

0-5

12-1
11 .. 1

12-6

11-6

DEC 029
Card Code

0-1

0-6

12-2

12-7

11-2

11-7

0-2
2

0-7
7

12-3

12-8

11-3

11-8

0-3

0-8

12-4

12-9

11-4

11-9

0-4

0-9

B-2

PRINTABLE CHARACTERS

Table B-2
Printable Characters Grouped by Equal Zones
DEC 029
Card Code

Group

Character

DEC 029
Card Code

Group

Character

(

12-5-8

0-4-8

12-6-8

•

0-3-8

·<

12-3-8

0-1

12-4-8

0-6-8

12-7-8

0-7-8

12-1

0-2

12-2

0-3

12-3

0-4

12-4

0-5

12-5

0-6

12-6

0-7

12-7

0-8

12-8

0-9

12-9

Blank

No punches

·$•

11-2-8

7-8

11-3-8

"
#

)

11-5-8

•

5-8

3-8

11-4-8

2-8

;

11-6-8

6-8

11-1

4-8

11-2

11-3

11-4

11-5

11-6

11-7

11-8

11-9

B-3

APPENDIX C
SORT PROGRAM EXAMPLES

EXAMPLE 1:

SALES LIST

Input Record positions
1-5
6-39
52-59
60-65
66-70
71-77
78-80

Customer Number (CUSTNO)
Customer Name (CUSTNM)
Date
Item Number (ITEMNO)
Item Quantity (ITEMQY)
Price
Blanks

This Record Sort is primarily intended to produce reformatted records.
The structure of the input file is indicated above.
The input file contains only one record type, therefore no Record Type
Specifications are needed. The keys for the sort are CUSTNO, ITEMNO,
and ITEMQY, with the last key sorted in descending sequence.
The
output file will have records ordered from lowest to highest customer
number, with the lowest to highest item number records within each
customer number category, and the highest to lowest item quantity in
each item number category.
The order of the data field specifications reformat the output record.
Note the use of the blank in columns 78 and 79; the data will be set
off in field columns.

C-l

SORT PROGRAM EXAMPLES

SORT SPECI FICATIONS

page~

Dare _ _ __

HEADER SPECIFICATION

75 76 77 78 79 80

~;~;~f:atoon 1s1~1&..IElSl I

Programmer

SORTR,T

Mode of ProceSSIng

0
Line

!
3

•

a I0 J

..
,

SORTR
SORTT
SORTA
SaRTI

Total

NOT USED

Length of

Recor(l

l en llttl

/Ill! .?'llU

?71!.?4?'l

Comments
(Progrdm Identification)

313:'

j I

38 3'1 40 41 47 4.! 44

46 ., I

., 'I ')0 !:II S? "13 S4 5!'.o 56 ') 7 58 59 60 6 I 62 63 64 65 66 67 68 69 70 71

72 73 7.

~ 11715 1~~TLJ~lsl~ I I I I I I I I I I I I I I I I I I I I I I I I I I I I J I I

i J Jllb A I II I II

SJoJRITIRJ

IS 1611 181920?i

9 101112 IJl4

OUHhJt

Key Fields

FIELD SPECIFICATION
Forced

t:lu.

Lifle

! ]§

., . , .

o 7

Field Locdtion

~a~

~ 6~
~o ; ~
:::l.

From
~

10 11 12

Ft(j}

a 8

F~ J)

F~ ~- -~~

2_~

I a

FtJ

I I

FJ>~o

FJ>~

F~~

,--I

1
I

F)j)

--I
f.--

~--

t t

III I' I

t 'l~~

''is

---

-~ ::r-~ ~~ t.~
-- CO 5 , "'ttl

IIIi! I

j I

1
I

co 51 .. NO

-----

ae ~

~. £.1'11

]

56}

a 9

I aFt))
f-+-

F~C

F~<::

1-----

FIeld Name

Page ~

Comments

n ;?J 2425 ?6 ?J 78 1'J JO 31 3(' 3! 34 J':> 36 J' 383'14041 4} 43 44 liS 46 4148 49 !:IO ~l 52 53 ~4 555651 585960616261 64 6566616869 10 11 1213 74

___ c+-

f- -

HJ ---~I~~j;~fy

+-+-+--!-f-H-++-+-+-+-+-+++-+--!-+-+-+++-+--!-+-l--+-4

~.~ - - , ; fI ::'E IfI' ~~, .
j'

:::

II iii II I filii I ~ , II j- ~~
Figure C-l

EXAMPLE 2:

757677787980

~;~;~f:.tiOn 1~IAI~Ie:;f51 I

10 II 12 13 14 IS 1611 18 192021

z~ ~:;

-- -

r<q

-----,

~o~i
~~-~u: : ~--r--~... ~
..---From
To
~ § g

:; r0:-2

Field Locdllun

u..

FIELD SPECIFICATION

Lme

7l7t

c.'\.J 5T rio

:!~ En! 1'(0

_ 7 1C}

n 23242516 }1?8}'l )03137 ll14 353631]839 404147434445 46414849 SO "II "125] 5455565158 "196061626364656661681>97071 n

f---

3'i c_

Comments

Flel.l N..Jrne

IJ14\516 1118 192021

a 9
f----

-----,

NOT USED

~ j 8

+-+--+-1-++++-+-+----1--1-+-++-+-+--1

--+--+-++"--+-+-+-+--+-+-+-+-+-+-+-+-+-+-+-+-H

Sales List

OVERTIME ANALYSIS

Input Record positions
1-3
4-8
9-39
40-43
50-53
80

Department Number (DEPTNO)
Employee Number (EMPNUM)
Employee Name (EMPNAM)
Standard Number of Working Hours (STDHRS)
Total Number of Hours Worked (HOURS)
Plant Identification Code

This Record Sort produces a record of employees by department number
in the main plant
(M)
who incurred overtime.
The Record Type
Specification shows the use of both an Include and an Omit line; with
a field and a constant Factor 2 comparison.

C-2

SORT PROGRAM EXAMPLES

"'.-au-oEOU
WIIW iii

I PM E NT

=~~:~~~A~~I.~~

SORT SPECIFICATIONS

page~

Oate _ _ _ __

HEADER SPECIFICATION

7576n787980

~~~;~f~ation lo&"§I~I~Il..1

Programmer

SORTR,T
0
Line

SORTR
SORTT
SORTA
SORTI

Output
Record
Leng1h

NOT USED

Comments
(Program IdentificatIon)

RECORO TYPE SPECIFICATION
Factor 1

Factor 2

~~~--------~~~~----------------------------~
GT
- Constant
-- - -1

Line

Q g!!!

i ;~
~ u~o~ §

GE
Field Location

From
5

-:- -

F leltJ Location

LE ~

From

..,
_j

....,

Comments

Record Name

a I) 10 11 12 Il 14 IS 16 17 18 "20 1122 21 24 ,,5, 26 211:>8 29 30 31 J? II J4 )') l6 )13839' 40 41 4243 U 4~ 46 47 148 49')051 52 5354 5S 56 57 !II 5960 61 62 63 U 65 66 6761 69 70 11 1~ 11 14

o 1

ISaJ

o 2

--f-

1---

o 3

o 4

-- -- -

-- --

- --I---

FIELD SPECIFICATION
Forced
Lme

u.~~
~ Z 0
~ ~ ~

Field Location

From

9101112 13141S 16\1

F ~tp

o B

FOi:P

sl¢

5.3

o 9

F"~
F

»r_

11+

1 0

'B
3<l

)3)

'"
a:

,:.
u

'"0

1 2

I--1
I---

f---

-f-

---

------,

o
0

j
191021 72232425

26 21 ?8 ?q ]0 JI J' lJ 34

)6 )7 38 394041 4143 u

c--+- -

r--

-- - --

-1--- f---

I---

r-- f-

lJ>e

I---

c---- "-

Figure C-2

4S 46 4148 49 SO SI S2 S3 S4 ~S S6 SJ S8 S960 6162636465666161 69 70 11 12 1314

, nO

Hlo 11.l~il'>

1.11'1)

E~ ",c AIO

-r---

.-~ _.L f!!~-~

-c---

EXAMPLE 3:

Comments

Field Name

r---

"- -

f--I--- -1----

Overtime Analysis

EMPLOYEE LIST

This Tag Sort illustrates the use of forced key field specifications.
The sort is based on the plant identification code in column 80 of the
input records.
In order to sort them in the order M, W, P, the M in
column 80 is forced to a 1, the W is forced to a 2, and the P is
forced to a 3. The output record entry is unchanged.

C-3

SORT PROGRAM EXAMPLES

"'DIIDD"D
~ l!J
~ =~~~::.~.~~I.<;',:!
E

a U I P MEN T

SORT SPECI FICA TlONS

75 76 17 78 79 80

~:~;~f:at,on Ie: IMI.. I. . ls ITI

page~

Date _ _ _ _

HEADER SPECIFICATION
rogrammer

SORTR,T

Mode of Processing

0
Line

SORTR
SORTT
SORTA
SORTI

.
101

Length of

NOT USED

I ~ 16 I' 18 19 2021

9101112 13 14

I I

Cl x
~ s
::l

Key Fields

SIOIRITtrl

Total

~~A

n 1; 24 l~

Record

~ ~

U'nqHl

7'1303131

Comments
{Program Identification)

33 34

18 3'1 40 41 41 43 44 4:' 46 4 I 48 49

j j

:'3 ~4 505 56 51 58 59 60 61 62 63 64 65 66 61 68 69 70 71 72 73 ,.

x 1181zs EI~I~I~ol~EI~1 IJ~~~ I I I I I I I I

11111

Output

I- 0

1I 1

RECORD TYPE SPECIFICATION

~i~

., .

EQ
NE
GT
GE

Field Location

~ ~ U

F,-jctor 2

Factor 1

Line

From

LE L)

-~-

-~ COrlSlant

.-~-

f------,
Field LocatIOn

Flom

Comments
r-----,
I
R>?corrl Ndmt'
I
34 3536 31 3B 3') 40 41 4? 43 44 4:' 46 4,1 48 4'1 SO ') 1 52 ~3 54 55 ')6 51 58 59 60 61 62 6] 64 65 66 61 &8 69 70 11 12 7] 14

c-Jt J lr
mHlH
m
+
lllJvf
~l ~:It
v~
-rl i

9101112 13141516 1718

a 2
a 3
0

f--~-

r--

t--- -

31 32

' i I

f-f--

1--1--

~--

-+
+~

T11114'
~~t++

1--+

o 6

26 71178

~H-~ ill

1- ~,.-' ~V

a 5

20t12? }] <'4

~r--

I--

I
I

--~

FIELD SPECIFICATION

FOf(erj

Line

Field LOCdt Ion

~6~

a 7
a 8
a 9
r, a
'--"

1?U
~~

a: u

~l¢ Pli

B!¢ wi.; X

F ~
F

,31

F
F

~¢ Pi3 )(

''1

~~- L

.3'=1_ r-I-~~-

3¢

t,
11

- .-

n 2425

.-" tJtt
?6

281'1 30 31]7 J; 34

J5 36 31 3839 40 4! 47 43 44

L-O~f t!:_'!

t--- -

Comments

Field NJme

910 11!2 1314}') 16 1118 197021?2

FF C __

-----,

NOT USED

'2,

r-~
,4

From

~ Bg
~~ ~~ §~

46 41 48 4'1 <;.0 S I ~2 53 54 55 56 S I 58 59 60 61 62 63 6e 65 66 67 68 69 70 71 12 13 74

Sol~ it

- f4 W P
I'

~~f-

t 1··-llt
Irn1IIIil
df Flt.

~f-r-

-t.

·i
I
'
-t
j
-t t IH
t t!

r ; t I r ; til
Iii j j 1 tr 1 j

t; +t f

Figure C-3

Employee List

,1\5 :

-- 1-,-r- I- -rr- f-- -. - J:IIIP
A~ +:-:+ A ~ &>4\ "e.~, :ele All-. [h"i
~ c~ '-+ WHP'- -~_~E:: ~b R"P ou T~ oT

, I

C-4

til !

+ tII

~,.~

r--~T

t 1-[

APPENDIX D
RMS I/O STATUS CODES

This appendix describes I/O status codes that can be returned by RMS.
For some error conditions, RMS uses the RMS-status-code field of the
SORT error message to communicate additional information. Table D-l
lists the RMS status codes and what they mean.
Table D-l
RMS Error Status Codes
Description

Decimal
Value
-16

Operation aborted: out-of-stack save area or
structures corrupted.

-32

ACP could not access the file.

-48

File activity precludes action (e.g., attempting to close
file with outstanding asynchronous record operation).

-64

Bad area identification number
(i.e., out of sequence).

-80

Illegal value in alignment
allocation XAB.

data

-96

Value in allocation quantity
(ALQ)
field
in FAB
(or
allocation XAB)
exceeds maximum or, during an explicit
$EXTEND operation, equals o.

-112

Records in a file on ANSI-labeled magnetic tape are variable
length but not in ANSI D format.

-128

Illegal value in
allocation XAB.

-144

Invalid operation at AST level:
attempting to issue a
synchronous operation from an asynchronous record operation
completion routine.

-160

Read error on file header attributes.

-176

write error on file header attributes.

(AID) field in allocation XAB

boundary

allocation

in-core

type

options

(ALN)

(AOP)

field

(Continued on next page)

D-l

RMS I/O STATUS CODES

Table D-l (Cont.)
RMS Error Status Codes
Description

Decimal
Value

-192

Bucket size (BKS)
maximum.

field

FAB

contains

-208

Bucket size (BKZ) field in
exceeding maximum.

-224

Block length (BLN)

-232

Beginning of file detected on $SPACE operation
tape file.

-240

Private buffer pool address not a double word boundary.

-256

Private buffer pool size not a mUltiple of 4.

-272

Internal error detected in RMS-ll;
contact a Software Specialist.

-288

Cannot connect RAB (i.e.,
only one
permitted for sequential files).

record

-304

$UPDATE attempting to change a key field that does not
the change attribute.

have

-320

Index file bucket check-byte mismatch.
The bucket has
corrupted. No recovery possible for the bucket.

been

-352

Invalid COD field in XAB or this XAB type is illegal for the
organization or operation.

-36H

ACP could not create file.

-384

No current record:
operation not immediately preceded by
successful $GET or $FIND.

-400

ACP deaccess error during $CLOSE.

-416

Invalid area number in DAN field of key definition XAB.

-448

1.
2.
3.

allocation

value

XAB

exceeding

contains

value

field in a FAB or RAB is incorrect.
to

recovery

magnetic

possible;

access

stream

Syntax error in device name.
No such device.
Inappropriate device for operation (e.g., attempting
create an indexed file on magnetic tape).

-464

Syntax error in directory name.

-480

Dynamic memory exhausted:
insufficient
space pool or private buffer pool.

space

central

(Continued on next page)

D-2

RMS I/O STATUS CODES
Table D-l (Cont.)
RMS Error Status Codes
Description

Decimal
Value
-496

Directory not found.

-512

Device not ready.

-520

Device positioning error.

-544

Duplicate key detected, duplicates allowed attribute not set
for one or more key fields.

-560

ACP enter function failed.

-576

Environment error;

-592

End of file.

-608

Expanded string area in NAM block too short.

-616

File expiration date not reached.

-624

File extend failure.

-640

Not a valid FAB:

-656

1.
2.
3.

operation not selected in ORG$ macro.

BID field does not contain FB$BID.

Record operation attempted was not declared in FAC field
of FAB at open time.
Invalid contents in FAC field.
FB$PUT not present in FAC for $CREATE operation.

-672

File already exists (attempted $CREATE operation).

-680

Invalid file ID.

-688

Invalid combination of values in FLG field of key definition
XAB (e.g., no duplicates and keys can change).

-704

File locked by another user -- your program cannot access
the file because its sharing specification cannot be met.

-720

ACP Find function failed.

-736

File not found.

-752

Syntax error in file name.

-768

Invalid file options.

-784

Device full:

-800

Invalid area number in IAN field of key definition XAB.

-816

Index not initialized (this code can only occur in
field when STS contains ER$RNF).

cannot create or extend file.

the

STV

(Continued on next page)

0-3

RMS I/O STATUS CODES
Table D-l (Cont.)
RMS Error Status Codes
Description

Decimal
Value

-832

Invalid IFI field in FAB.

-848

Maximum number (254) of key definition or allocation XABs
exceeded or multiple summary, protection,
or date XABs
present during operation.

-864

$INIT or $INITIF macro call never issued.

-880

Illegal operation:
1.
2.

3.
4.

examples include:

Attempting a $TRUNCATE operation to a non-sequential
file.
Attempting an $ERASE or $EXTEND operation to a magnetic
tape file.
Issuing a block mode operation (e.g., $READ or $WRITE)
to a stream not connected for block operations.
Issuing a record operation
(e.g.,
$GET,
$PUT)
to a
stream connected for block mode operations.

... 896

Illegal record
count field.

-912

Invalid internal stream identifier (lSI) field in RAB (field
may have been altered by user) or $CONNECT never issued for
stream.

-928

Key buffer address (KBF)

-944

Record identifier (i.e., the 4-byte location addressed by
KBF) for random operation to relative file is 0 or negative.

-960

Invalid key of reference (KRF)

1.
2.

encountered

sequential

file:

invalid

field equals O.

in RAB:

As input to random $GET or $FIND operation, or
As input to $CONNECT or $REWIND (in this case, ER$KRF is
returned for
the first record operation following the
$CONNECT or $REWIND).

-976

Key size equals a or too large (indexed file)
to 4 (relative file).

-992

Invalid area number in LAN field of key definition XAB.

-1008

Magnetic tape is not ANSI labeled.

-1024

Logical channel busy.

-1040

Invalid value in logical channel number

-1048

Attempt to extend an area containing an unused extent.

-1056

Invalid value in LOC field of allocation XAB.

(LCH)

not

equal

field of FAB.

(Continued on next page)

D-4

RMS I/O STATUS CODES
Table 0-1 (Cont.)
RMS Error Status Codes
Description

Decimal
Value
-1072

In-core data structures (e.g., I/O buffers) corrupted (this
code can only occur in the STV field when STS contains
ER$ABO).

-1088

ACP could not mark file for deletion.

-1104

-1120

Maximum record number field contains a negative value
during $CREATE of relative file.
Record identifier (pointed to by KBF)
for
random
operation to relative file exceeds maximum record number
specified when file created.

Maximum record size (MRS) field contains
operation and:
1.
2.

during

$CREATE

Record Format is fixed, or
File organization is relative.

-1136

Odd address in Name Block address
$OPEN, $CREATE, or $ERASE.

-1152

Not at end-of-file:
attempting a $PUT operation
sequential file when stream is not positioned to EOF.

-1168

Cannot allocate internal index descriptor:
insufficient
room in space pool while attempting to open an indexed file.

-1184

No primary key definition
indexed file.

-1216

XABs in chain not in correct order:
1.
2.

XAB

(NAM)

present

field

during

Allocation or key definition XABs not in
densely ascending) order.
XAB of another type intervenes in key
allocation XAB sub-chain.

FAB

$CREATE

ascending

(or

definition

-1232

Invalid value in file organization (ORG) field of FAB.

-1248

Error in file prologue:
reconstructed.

-1264

Key position (POS) field in key definition
value exceeding maximum record size.

-1280

File header contains bad date
and
time
information
(retrieved by RMS-ll because a date and time XAB is present
during a $OPEN or $DISPLAY operation);
file may
be
corrupted.

file

corrupted
XAB

and

must

contains

(Continued on next page)

D-5

RMS I/O STATUS CODES
Table D-l (Cont.)
RMS Error Status Codes
Description

Decimal
Value
-1296

Privilege violation:
operating system.

-1312

Not a valid RAB:
BID field does not contain RB$BID.
to Section A.3 of this Appendix.

-1328

1.
2.

Illegal values in record access mode (RAC) field of RAB.
Illogical value in RAC field
(e.g., RB$KEY with a
sequential file).

-1344

Illegal values in record attributes (RAT) field of
during $CREATE.
Illogical combination of attributes
(e.g.,
FB$CR
FB$FTN) in RAC field during $CREATE.

access

the

file

denied

Refer

FAB
and

Record address (RBF) field in RAB contains
(block mode access only).

-1376

File read error.

-1392

Record already exists: during a $PUT operation in random
mode to a relative file, an existing record found in the
target record position.

-1408

Invalid RFA in RFA field of RAB during RFA access.

-1424

1.
2.

RFM

field

odd

the

-1360

Invalid record format in
$CREATE.
Specified
record
format
organization.

address

FAB

during

illegal

for

file

-1440

Target bucket locked by another task or
the same program.

another

-1456

ACP Remove function failed.

-1472

Record identified by KBF/KSZ fields of RAB for
random $GET
or $FIND operation does not exist in relative or indexed
file (for indexed files only, STV may contain ER$IDX).
Record may never have been written or may have been deleted.

-1488

$FREE operation issued but no bucket was locked by stream.

-1504

Record options
(ROP)
field contains
illogical combination of values.

-1520

Error while reading prologue.

-1536

Invalid PRV record encountered in indexed file;
corrupted.

-1552

Record stream active,
i.e.,
in asynchronous environment,
attempting to issue a record operation to a stream that has
a request outstanding.

illegal

stream

values

file may be

(Continued on next page)

D-6

RMS I/O STATUS CODES
Table D-l (Cont.)
RMS Error Status Codes
Description

Decimal
Value
-1568

Record size specified in RSZ of RAB during $PUT
is invalid:
1.
2.
3.
4.

$UPDATE

RSZ equals O.
RSZ exceeds maximum record size
(MRS)
specified when
file created.
RSZ not equal to size of Current Record for $UPDATE
operation to a sequential file on disk.
RSZ does not equal MRS (for fixed format records).

-1584

Record too big for your buffer:
RMS-ll could not move
entire record retrieved by $GET operation to work area
(UBF/USZ). Note that this error does not destroy the
current context of the stream. Rather, the stream's context
is updated as if the operation had
been
completely
successful.

-1600

During $PUT operation, key of record to be written is not
equal to or greater than key of previous record (and RAC
field contains RB$SEQ).

-1616

Illogical value in SHR field of FAB (e.g., FB$WRI
for sequential file).

specified

-1632

Invalid SIZ field in key definition XAB during
(e.g., specified size exceeds maximum record size).

$CREATE

-1648

During asynchronous record operation, RMS-ll has found that
the stack is too big to be saved (this code can only occur
in the STV field when STS contains ER$ABO).

-1664

System directive error.

-1680

Index tree error:

-1696

Syntax error in file type (e.g., more than three
specified) •

-1712

Invalid address in UBF field of RAB:
1.
2.

indexed file is corrupted.
characters

UBF contains 0, or
UBF not word aligned (for block mode access only).

-1728

Invalid USZ field in RAB (i.e., USZ contains 0).

-1744

Syntax error in file version number.

-1760

Invalid VOL field in allocation
contain 0).

-1776

File write error.

-1784

Device is write locked.

-1792

Error while writing prologue.

-1808

XAB field in FAB (or NXT
address.

field

D-7

XAB

(i.e.,

XAB)

VOL

does

not

contains

odd

GLOSSARY

Alphanumeric Characters - the numbers and
ASCII character set.

letters

contained

the

ASCII Character Set - the set of 128 seven-bit American Standard
for Information Interchange characters (see Appendix A) •

Code

Byte - the smallest addressable unit of information; eight bits.
Each ASCII character usually resides in a single byte.
Collating Sequence - the order into which characters are sorted
upon numeric values assigned to each.

based

Device - any of the peripherals referenced by the operating system.
Digit - in EBCDIC representation, the numeric value of
four bits of a byte.

the

low-order

EBCDIC - an alternate character set used widely in the computing
industry (see Table A-2 for comparison with ASCII characters).
Field - a logically distinguishable area within a record in which data
is located.
File - a collection of related records treated as a unit.
Floating-Point Binary - a method for storing numeric data,
the
exponent occupies the first byte and the mantissa resides in the next
three or seven bytes.
Format - the arrangement of any record or file;
fields reside in a record.

the

order

which

Key, Key Field - the sorting key is the data element chosen from a
record to control the sort. For example, if a list of names is sorted
alphabetically, the keys would be each successive letter in the name
field.
Key, Major - the most important field in the total key.
If you were
sorting a list by name, department and salary, name would be the major
key.
Key, Minor - the least significant field in the
example above, salary is the minor key.

total

key.

the

Line - In this document, a line generally refers to a line in the SORT
specifications form or a record on the specification file.
Packed Decimal - method for compact storage of numeric data;
two
digits are stored in each 8-bit byte and the sign resides in the last
byte of the low-order digit.

Glossary-l

RMS - Record Management Services
Record - the unit of information in a file;
treated as a unit

a group of related fields

System Device - the device on which the Executive resides.
Word -

two bytes or sixteen bits of data

Work File - a collection of sorted records created
processing cycle and released after the sort is finished

during

the

Zone - the top three punches of a character as represented on a
card or the numeric value of the upper four bits of a byte

punch

Glossary-2

INDEX

IAL,
IBL,
IBU,
ICO,
IDE,

3-1
3-1
3-1
3-2
3-3
IFI, 3-3
IFO, 3-3

lIN, 3-3
IKE, 3-3
IOU, 3-4
IPR, 3-4
IRE, 3-5
ISE, 3-5
lSI, 3-5

Allocation,
file, 5-1
Allocation switch, 3-1
Alternate collating sequence,
1-6, 3-6
Alternate key, 1-2
ALTSEQ, 1-6, 3-6
ALTSEQ format and notes, 3-10
ALTSEQ records, 3-6
ANSI format, 1-3
Ascending order, 1-4
ASCII character set, table, A-2
ASCII data, 1-2
ASCII zone, 3-4
Available internal storage
space, 1-8
Binary data, 1-2
Block,
5l2-byte, 3-1
Blocks per bucket, 3-1
Blocksize switch, 3-1
Boolean expression, 3-12
Bucket,
blocks per, 3-1
Bucket size, 3-1
Bucketsize switch, 3-1
Changing SORT defaults, 3-1
Changing sort order within
a sequence, 3-6
Character set, tabl~,
ASCII, A-2
EBCDIC, A-3
Characters, equal digits,
printable, table, B-2
Characters, equal zones,
printable, table, B-3

Code,
device, 2-3
user identification, 2-3
Codes,
RMS error, 4-6
RMS error status, 0-1
Codes, table,
device specification, 2-4
Collating sequence,
alternate, 1-6, 3-6
normal, 1-5
Collation,
nonstandard, 3-5
Command decoder errors, 4-1
Command file,
indirect, 2-2
Command string, 1-1, 1-4, 2-2
errors, 1-7
format, 2-1, 2-2
functions, 1-4
SORT, 2-1
switches, 2-2
Comments, 3-7
Conditional keys, 3-16
Conditions,
error, 4-1
Configuration,
system, 1-2
Contiguous switch, 3-2
Control fields, 1-1
Control program, 1-7
errors, 4-6
Control-phase, 4-1
Controlling SORT, 1-1
Controlling SORT options, 3-1
Controls,
sorting process, 3-5
Conventions,
file naming, 5-1

Data,
ASCII, 1-2
binary, 1-2
EBCDIC, 1-2
field, 1-2
defining, 3-6
locating, 3-6
specifications, 3-6
handling, 3-4
irrelevant to SORT, 1-2
key, 1-2
output, 1-9
relevant, 1-2
restricted-format key, 3-6
Index-l

INDEX (Cont.)

Data base, 1-1
Decoder errors,
command, 4-1
Defaults,
changing SORT, 3-1
file, footnote, 3-2
file types, 2-3
sort process, 1-8
switch, table, 3-2
Defining a data field, 3-6
Defining a key, 3-7
Description sequence,
field, 3-3
Device,
code, 2-3
file, input and output, 2-2
peripheral, 1-2
specification codes, table,
2-4
switch, 3-2
types, table, 2-4
Different record formats, 1-7
Digits,
printable characters,
equal, B-2
Direct access from a random
file, 1-1
Directing SORT, 1-5
Discussion,
specification file example,
3-16
Disk, 1-2
scratch files on, 1-8
wasted space on, 5-2

EBCDIC,
character set, table, A-3
data, 1-2
Entry,
common, of specification
lines, 3-6
field specification, 3-14
header specification, 3-8
notes, 3-9
record type specification,
3-16
SORT specification, figure,
3-17
Error,
codes,
RMS, 4-6
RMS status, 0-1
command decoder, 4-1
command string, 1-7
control program, 4-6
I/O, 4-1, 4-5
merge, 4-6

Error (Cont.),
messages, 4-1
other, 4-6
presort, 4-6
SORT, 4~1
specification file, 1-8,
4-1, 4-3
Example,
SORT, C-l
SORTR, 3-16
specification file discussion,
3-16
Expression,
Boolean, 3-12

Field,
control, 1-1
data, 1-2
defining a data, 3-6
description sequence, 3-3
first byte, key, 3-3
interpretation, 1-2
key, 1-1, 1-5, 3-3
length, key, 3-3
locating a data, 3-7
location, key, 1-5
specification, 3-6, 3-14
data, 3-6
entries, 3-15
key, 3-6
notes, 3-15
stripping key, 3-14
table, 3-21
File,
allocation, 5-1
defaults, footnote, 3-2
devices,
input and output, 2-2
errors,
specification, 1-8, 4-1,
4-3
example, discussion,
specification, 3-16
index, 1-1
indexed, 1-2
indirect command, 2-2
information,
RMS, 1-5
input, 1-1, 1-9, 2-2, 3-16,
5-1
input and output, 3-1
internal work, 1-4
merging scratch, 1-8
moving records within a,
3-6
name, 2-3

Index-2

INDEX (Cont.)

File (Cont.),
naming conventions, 5-1
needed, number of scratch,
1-8
on disk, scratch, 1-8
opening the input, 1-8
optional specification, 1-1
organization,
input, 3-1
RMS.,..valid, 1-2
output, 1-8, 2-1, 2-2, 3-6,
5-1
parameters, 1-4
random, direct access, 1-1
relative, 1-1
scratch, 5-2
sequential, 1-1
sequential access from a
random, 1-1
sorted, 2-1
specification, 1-4, 1-8,
2-2, 2-3, 3-5, 3-14,
3-16, 4-3, 5-1
structure, scratch, 5-2
switch, 3-3
type, 2-3
types, default, 2-3
uniform format input, 1-4
usage, scratch, 5-3
First byte,
key field, 3-4
FIXED format, 3-3
FIXED records, 1-8
Forced keys, 1-6
Form,
SORT specification, 3-7
Format,
ALTSEQ, and notes, 3-10
ANSI, 1-3
command string, 2-1, 2-2
FIXED, 3-3
output, 1-7
packed decimal, 3-4
record, 1-5
different, 1-7
STREAM, 3-3
switch, 3-3
uniform input file, 1-4
UNKNOWN, 3-3
VARIABLE, 3-3
variable record, 3-6
variation, input, 1-7
Functions,
command string, 1-4

General data, irrelevant to
SORT, 1-2

Handling,
data, 3-4
Header specification, 3-6,
3-17
entries, 3-9
notes, 3-9
table, 3-18
High-speed sorting, 1-9

I/O errors, 4-1, 4-5
Identification by type,
record, 1-7
Identification code,
user, 2-3
Identifier,
record, 1-2
Index,
file, 1-1
record, 1-1
sort, 1-1, 1-8
Indexed file, 1-2
Indexed sequential switch,
3-3
Indirect command file, 2-2
Information,
key, 1-1
prologue, 1-2
RMS file, 1-4
storage of sorted, 1-1
Initial sorting process, 1-8
Input and output files, 3-1
devices, 2-3
Input file, 1-1, 1·9, 2-2,
3-16, 5-1
organization, 3-1
uniform format, 1-4
Input format variation, 1-7
Interactive mode, 2-1
Internal operations,
SORT, 5-1
Internal work files, 1-4
Interpretation,
field, 1-2
Invoking SORT, 2-1
Irrelevant to SORT,
general data, 1-2

Key,
alternate, 1-2
conditional, 3-16
data, 1-2
restricted-format, 3-6
defining, 3-7

Index-3

INDEX (Cont.)

Key (Cont.),
fields, 1-1, 1-5, 3-3
first byte, 3-3
length, 3-3
location, 1-5
specifications, 3-6
stripping, 3-14
forced, 1-6
specification entries, 3-16
information, 1-1
locating a, 3-6
primary, 1-2
significance,
sort, 1-4
switch, 3-3

Null records, 3-16
Number,
line, 3-7
page, 3-7
Number of scratch files
needed, 1-8

Opening the input file, 1-8
Operations,
SORT, 3-8
SORT internal, 5-1
Optimization,
sort, 5-1
Optional specification file,
1 .... 1

Length,
key field, 3-3
Line number, 3-7
Lines,
common entries of
specification, 3-7
Locating a data field, 3-6
Locating a key, 3-6
Location,
key field, 1-5

Magtape, 1-2, 3-1
Management Services,
Record, 1-1, 1-8
Maximum record size switch,

Options,
controlling SORT, 3-1
sort processing, 1-8
sorting process, 1-5
sorting process, table,
1-9

Order, 1-3
ascending, 1-3
ascending sort, 3-4
descending sort, 3-4
general sort, 3-4
Organization,
input file, 3-1
Organization, RMS-valid,
file, 1-2
Other errors, 4-6, 4-8
Output data, 1-9
Output file, 1-8, 2-1, 2-2,
3-7

1-5

Merge errors, 4-6
Merge phase, 1-8
Merging scratch files, 1-8
Messages
SORT error, 4-1
Method,
sorting, 1-1
Mode,
interactive, 2-1
Moving records within a file,
3-6

Naming conventions,
file, 5-1
Nonstandard collation, 3-5
Normal collating sequence, 1-6
Notes,
ALTSEQ format and, 3-10
field specification, 3~15
record type specifications,
3-12

devices, 2-2
parameters, 2-2, 3-1, 5-1
Output format variation, 1-7
Output record, 1-11, 3-14
Output switch, 3-4
Packed decimal format, 3-4
Page number, 3-7
Parameters,
file, 1-4
input and output file, 2-1
SORT, 3-3
Parts of SORT, 1-7
Peripheral device, 1-2
Phase,
merge, 1-8
SORT, 1-7
Pointer,
record, 1-2
Preallocation problems, 5-1
Presort errors, 4-6
Presort operation, 1-8

Index-4

INDEX (Cont.)

Primary key, 1-2
Printable characters, equal
digits, table, B-2
Printable characters, equal
zones, table, B-3
Printer, 1-3
Problems,
prea11ocation, 5-1
Process,
controls, sorting, 3-5
default sort, 1-8
initial sorting, 1-8
options, 1-5
table, 1-9
SORT, 1-8
sorting, 1-4, 1-9
switch, 3-4
Processing,
options, sort, 1-8
Program, control, 1-7
errors, 4-6
Prologue information, 1-2
Prompt,
SORT, 2-1

Relative file, 1-2
Relative switch, 3-5
Relevant data, 1-2
Restricted-format key data,
3-6
RMS, 1-1, 1-8, 3-1
error codes, 4-6
error status codes, 0-1
file information, 1-5
RMS-va1id,
file organization, 1-1

Scratch files, 5-2
merging, 1-8
needed, number of, 1-8
on disk, 1-8
structure, 5-2
usage, 5-3
Selection,
record, 1-5, 3-6
Sequence,
alternate collating, 1-6,
3-6
changing sort order within
a, 3-6
field description, 3-3
normal collating, 1~5
Sequential,
access from a random file,
1-1
file, 1-2
switch, 3-5
Significance,
sort key, 1-4
Size,
bucket, 3-1
Size switch,
maximum record, 1-5, 3-5
SORT, 1-1, 1-4, 1-8, 2-1,
3-16, 5-1
address routing, 1-1, 3-4
command string, 2-1
controlling, 1-1
defaults, changing, 3-1
directing, 1-5
error messages, 4-1
errors, 4-1
examples, C-1
general data, irrelevant to,
1-2
Index, 1-1
index, 1-8, 1-10, 1-11, 3-4
internal op~rations, 5-1
inVOking, 2-1
key significance, 1-4
operations, 3-8
optimization, 5-1

Random file,
direct access from a, 1-1
sequential access from a,
1-1
Record Management Services,
1-1, 1-8
Records,
ALTSEQ, 3-6
FIXED, 1-8
format,
different, 1-7
switch, 1-5
variable, 3-6
identification by type, 1-7
identifier, 1-2
index, 1-1
moving, within a file, 3-6
null, 3-17
output, 3-14
pointer, 1-2
selection, 1-5, 3-6
size switch, maximum, 1-5
sort, 1-1, 1-8
sorted, 2-1
STREAM ASCII, 1-8
strings of sorted, 1-8
type specifications, 3-6
entries, 3-10
notes, 3-12
table, 3-19
VARIABLE length, 1-8
within a file, moving, 3-6

Index-5

INDEX (Cont.)

SORT (Cont.),
options, controlling, 3-1
order,
ascending, 3-4
descending, 3-4
general, 3-4
within a sequence,
changing, 3-6
parameters, 3-3
parts of, 1-7
phases, 1-7
process, 1-8
default, 1-8
processes, 1-8
processing,
options, 1-8
Record, 1-1
record, 1-8
running, 2-1
specification,
entries, figure, 3-18
form, 3-6
figure, 3-7
switches, 3-1
Tag, 1-1, 1-8, 1-9, 3-4
Sorted file, 2-1
Sorted information,
storage of, 1-1
Sorted records, 2-1
strings of, 1-8
Sorting,
high-speed, 1-9
method, 1-1
process, 1-4, 1-9
controls, 3-5
initial, 1-8
options, 1-5
table, 1-9
SORTR, 1-1, 1-8, 3-14
example, 3-16
SORTT, 1-1, 1-9, 3-14
Space,
available internal storage,
1-8
Space on disk,
wasted, 5-1
Specification,
device codes, table, 2-3
entries,
field, 3-14
notes, 3-15
header, 3-8
notes, header, 3-9
record type, 3-10
SORT, figure, 3-17
field, 3-6, 3-14
file, 1-5, 1-8, 2-2, 2-3, 3-5,
3-14, 3-17, 4-3, 5-1
errors, 1-8, 4-1, 4-3

Specification (Cont.),
file (Cont.),
example, discussion, 3-16
optional, 1-1
swi tch., 3- 3
form, SORT, 3-7
header, 3-6, 3-17
key field, 3-6
lines, common entries of, 3-7
record type, 3-6
notes, 3-12
table,
field, 3-22
header, 3-18
record type, 3-19
type, 3-7
SRT, 2-1
Status codes,
RMS error, D-l
Storage of sorted information,
1-1
Storage space,
available internal, 1-8
STREAM ASCII records, 1-8
STREAM format, 3-3
String,
command, 1-1, 1-4, 2-1
of sorted records, 1-8
SORT command, 2-1
Stripping key fields, 3-14
Structure,
scratch file, 5-2
Subroutine package,
SORTS, 1-7
Switch,
allocation, 3-1
b1ocksize, 3-1
bucketsize, 3-1
command string, 2-4
contiguous, 3-2
defaults, table, 3-2
device, 3-2
files, 3-3
format, 3-3
indexed sequential, 3-3
key, 3-4
maximum record size, 1-4
process, 3-4
record format, 1-5
relative, 3-5
sequential, 3-5
SORT, 3-1
specification, 1-4
System configuration, 1-2
Table, see also Table of
Contents
EBCDIC character set, A-3

Index-6

INDEX (Cont.)

Tag sort, 1-1, 1-8, 1-9, 3-4
Terminal, 1-2
Terminators,
valid, 4-2
Time,
elapsed, 2-1
wall clock, 2-2
Type,
file, 2-3
record identification by, 1-7
specification, 3-6
Type specifications,
record, 3-10
entries, 3-10
notes, 3-12
table, 3-19
Types,
default file, 2-4
Types, table,
device, 2-4
Uniform format input file, 1-4
UNKNOWN file format, 3-3
Usage,
scratch file, 5-3

Valid terminators, 4-2
Values,
switch, 1-7, 3-1
VARIABLE format, 3-3, 3-6
VARIABLE length records,
1-8
Variation,
input format, 1-7
output format, 1-7

Wall clock time, 2-2
Wasted space on disk, 5-1
Work files,
internal, 1-4

Zone,
ASCII, 3-4
Zones,
printable characters,
equal, table, B-3

Index-7

TRAX SORT
Reference Manual
AA-D346A-TC

READER'S COMMENTS
NOTE:

This form is for document comments only. DIGITAL will
use comments submitted on this form at the company's
discretion. Problems with software should be reported
on a Software Performance Report (SPR) form. If you
require a written reply and are eligible to receive
one under SPR service, submit your comments on an SPR
form.

Did you find errors in this manual?

If so, specify by page.

Did you find this manual understandable, usable, and well-organized?
Please make suggestions for improvement.

I:
I.~

1'£

Ig>
1...2

1°

IB
ICD

III

I~
10::

Is there sufficient documentation on associated system programs
required for use of the software described in this manual? If not,
what material is missing and where should it be placed?

Please indicate the type of user/reader that you most nearly represent.

o Assembly language programmer
o Higher-level language programmer
o Occasional programmer (experienced)
o User with little programming experience
o Student programmer
o Non-programmer interested in computer concepts and capabilities
Name

Date _________________________

Organization _______________________________________________________________
Street ________________________________________________________________----__
City __________________________ State _____________ Zip Code ______________
or

-------------------------------------------------------------Fold lIere------------------------------------------------------------

------------------------------------------------ Do Not Tear - Fold lIere and Staple -----------------------------------------------

FIRST CLASS
PERMIT NO. 33
MA YNARD, MASS.
BUSINESS REPLY MAIL
NO POSTAGE STAMP NECESSARY IF MAILED IN THE UNITED STATES

Postage will be paid by:

Software Documentation
146 Main Street ML5-5/E39
Maynard, Massachusetts 01754