Process Help

SORTX - sort a file in ascending or descending order

 

Process Name

Menu Path

Link to Command Table

SORTX

Data ribbon | Data Tools | Tables | Sort

Data ribbon | Data Tools | Sort

Click here

 

Introduction

Sorts any file into ascending (or optionally descending) order of keyfields.

Ascending order is the default; descending order is specified by setting the optional parameter @ORDER=2.

At least one keyfield must be specified, and must appear as an explicit field. For alphanumeric fields, the collating sequence is the standard Datamine-specific sort sequence for text. The fields are sorted in the order in which they are specified. If a specified keyfield is not in the file, it is ignored. A warning message is given if @PRINT>=1.

From Studio MR22 Update 1 onwards, SORTX and MGSORT become identical, although separate process names were retained for legacy purposes. The new implementation provides significant performance benefits over earlier versions.

Where sufficient memory exists, the table is sorted entirely within memory. However, if the memory required exceeds that available, the table is sorted in separate 'chunks', each temporarily stored to disk and then merged together for the final write. The maximum number of records that can be sorted is therefore limited only by available disk space.

This process includes the KEYSFRST option which, by default, maintains legacy behavior by reordering key fields in the output file so that KEY1 becomes the first field, KEY2 the second field, and so on. If you specify that key fields are not reordered, then the order of fields in the OUT and IN file will be identical.

This process allows you to determine how the sort is performed with regards to duplicate key field values, using the ROWORDER parameter.

 

note.gif (1017 bytes)

Although the order of fields in a file does not affect subsequent processing, it makes it more difficult to review the file using Datamine Table Editor: for example, a borehole file which is sorted by BHID, FROM and TO is more difficult to manually validate if these fields are not adjacent.

 


Files, Fields and Parameters

Input Files

Name

Description

I/O Status

Required

Type

IN

File to be sorted.

Input

Yes

Table

Output Files

Name

I/O Status

Required

Type

Description

OUT

Output

Yes

Table

Sorted file.

Fields

Name

Description

Source

Required

Type

Default

KEY1

Keyfield 1 for sorting on.

IN

Yes

Any

Undefined

KEY2

Keyfield 2.

IN

No

Any

Undefined

KEY3

Keyfield 3.

IN

No

Any

Undefined

KEY4

Keyfield 4.

IN

No

Any

Undefined

KEY5

Keyfield 5.

IN

No

Any

Undefined

KEY6

Keyfield 6.

IN

No

Any

Undefined

KEY7

Keyfield 7.

IN

No

Any

Undefined

KEY8

Keyfield 8.

IN

No

Any

Undefined

KEY9

Keyfield 9.

IN

No

Any

Undefined

KEY10

Keyfield 10.

IN

No

Any

Undefined

Parameters

Name

Description

Required

Default

Range

Values

ORDER

Option

Description

1

for ascending order

2

for descending order (1)

No

1

1,2

1,2

KEYSFRST

Option

Description

0

output fields in the same order as the input table

1

output key fields first

No

1

0,1

0,1

ROWORDER

Option

Description

0

Rows which contain duplicate key field values could be in any order (faster)

1

Rows which contain duplicate key field values will be in the input file order (slower) (1)

No

1

0,1

0,1

KEYTOL

KEYTOL is the tolerance value used to test whether numeric key values are equal. It must be greater than or equal to zero. It replaces the previous heuristic comparison method.

If KEYTOL is set to a negative value then zero is used.

In a macro KEYTOL can be set to absent using -. "@KEYTOL=-" This will revert to legacy behaviour and use a heuristic comparison in relational commands and zero in sort.

No

0.00001

0,+

Undefined


Notes

The sort algorithm used by earlier versions of the SORTX process, through which data was allocated to different "histogram" bins before sorting, has been replaced. As a result, the BINS parameter is no longer used and is ignored when found in macros.

Both SORTX and MGSORT processes honor the standard Datamine-specific sort sequence for text, shown below.

note.gif (1017 bytes)

Note that "space" sorting is done before "A".

 

AaBbCcDdEeFfGgHhIiJjKkLlMmNnOoPpQqRrSsTtUuVvWwXxYyZz0123456789+-*/.,;:=?!@$%&()_' "<>[]

 

The above sequence has been extended by adding the following characters that were previously illegal for sorting purposes: #^`{|}~

For macros and script operation, any number of KEYn fields can be defined (the previous limit of ten has been removed). Key fields must be specified in consecutive order, and without gaps: for example, adding KEY1, KEY2 and KEY4 would not be allowed, and the reading of keyfields would stop after KEY2.

The previous limitation on keyfields being up to 10 words long has also been removed.


Example

!SORTX     &IN(ASSAYS),&OUT(SASSAYS),*KEY1(BHID),*KEY2(Z)


Error and Warning Messages

Message

Description

Solution

Warning: Keyfield:field_nameis implicit and will be ignored.

An implicit keyfield was provided. Since all values within it are identical, the keyfield is ignored.

 
Warning: Missing keyfield.

A requested keyfield could not be found in the file. The keyfield will be ignored.

 
Error: No valid keyfields have been found. Process aborted.

There were no valid keyfields to sort on. Keyfields may not have been specified, were not found in the file, or were implicit.

 
Error: Failed to open input file.

The required input file could not be found, or could not be opened for reading.

The file may not exist, be open in another application, or you may not have access rights to the file.

Error: Unable to create DmFile read buffer!

An unspecified error (possibly memory related) when reading the input file.

 
Error: Unable to open temporary chunk filefile_name.

An unspecified error when creating a temporary file - possibly due to a shortage of disk space.

 
Error: Error writing to temporary chunk filefile_name.

An unspecified error when writing to a temporary file - may be followed by a description of the problem, if available.

 
Error: Error whilst writing data tofile_name.

An unspecified error when writing to the output file - may be followed by a description of the problem, if available.

 
Error: Could not createfile_name.

An unspecified error when creating the output file - may be followed by a description of the problem, if available.

The output file may be open in another application, or you may not have access rights to overwrite or create it.

Error: Could not create an instance of DmTable.

An internal error when accessing the Datamine file.

 
Error: Could not create an instance of DmSchema.

An internal error when accessing the Datamine file.