Printer Colon File Conventions

AIX Version 4.3 Guide to Printers and Printing

Printer Colon File Conventions

Printer and printer data stream attributes reside in colon files. Colon files reside in the /usr/lib/lpd/pio/predef and /var/spool/lpd/pio/@local/custom/* directories. The /usr/lib/lpd/pio/predef directory contains the predefined database and the /var/spool/lpd/pio/@local/custom/* directory contains the customized database.

The following sections describe the conventions for printer and attribute names and values in colon files.

Colon File Format

Colon files in both the predefined and customized databases have five fields (separated by colons) for each attribute. They are:

Message catalog ID

Identifies the message catalog where the attribute description is stored. The message catalog ID can take any of three forms:

Null string: The string value for the mD attribute is assumed to be the file name of the message catalog (for example, mydescriptors.cat ).
One character: An abbreviation for pioattrx.cat, where x is the one-character catalog ID. This form of the catalog ID is normally used only by the operating system.

Catalog file name: The file name of the message catalog (for example, mydescriptors.cat ).

Either the one-character form or the catalog file name form of the catalog overrides the catalog file name specified with the mD attribute.

Message number	Identifies the message index in the catalog that contains the description of this attribute. Leading zeros are ignored.
Attribute name	Specifies two characters, except for group header attributes, which are five characters.
Limits field	Specifies limits for the attributes.
Attribute value string	Specifies zero to 1000 characters.

Following is an example of one line in a colon file:

:023:_w::80

The attribute name is _w, the attribute value string is 80 , and the attribute description is stored in message number 23 in the message catalog specified by the mD attribute.

Note: All attribute descriptions are stored in message catalogs. If an attribute has the same description for multiple printers, the attribute in each printer's database can reference the same catalog and message number. If the same attribute name has a different description for different printers, separate message numbers are used.

Attribute Names

The following conventions have been established for virtual printer attribute names:

Each attribute name must be unique.
Attribute names can contain the characters a through z, A through Z, 0 through 9, and _ (underscore). The name cannot begin with a numeral.
All attribute names must be two characters long (except group header attribute names, which can be five characters long).
Attribute names for group headers begin with _ _ (two underscores) and must not be longer than five characters. A group header attribute (formerly called a comment attribute) marks the beginning of a group of related attributes. For example, the group header attribute _ _FLG marks the beginning of a group of attributes that define the default values for command line flags. The grouping of attributes is for readability purposes and does not affect how the attributes are processed.
An attribute name beginning with _ (an underscore), except for group headers, can be overridden by a command line flag of the same name as the second character of the attribute name. For example, -w 132 , specified with the qprt command, overrides with a value of 132 the value specified for the _w attribute in the colon file.

Automatic Attributes

Automatic attributes are names and values that are provided automatically and that cannot be in the database:

Always a null string. This attribute name can be used wherever an attribute name for a null string is needed.

A string containing the full path name of the file being printed. This attribute name is available only to attributes that define pipelines and attributes included by pipelines. The file being printed will be a temporary file if the -c flag is specified with the qprt command.

An integer containing the number of bytes to be passed through when %x is found in a command string by the piocmdout subroutine (obtained from the passthru parameter passed to the piocmdout subroutine).

An integer value indicating how the printer is attached:

0	Neither parallel nor serial
1	parallel
2	serial

The full path name of the pio directory whose subdirectories (burst, etc, fmtrs, fonts, predef, trans1, and trans2) contain STATIC data files and utility programs used to configure virtual printers and process print jobs. The directory must be a subdirectory of the directory containing the piobe command invoked by the qdaemon. The value for @4 is normally the /usr/lib/lpd/pio directory.

The full path name of the pio directory whose subdirectories (custom, ddi, dev, and flags) contain DYNAMIC data files used to configure virtual printers and process print jobs. The value for @5 is normally the /var/spool/lpd/pio directory.

The following attribute names are used for communicating from the piobe command (the print job manager) to the pioout command (the device driver interface program). The attribute values are referenced by flag arguments passed to the device driver interface program as specified in the pipelines.

@A	Number of bytes already printed.
@B	Total number of bytes to print.
@C	Number of times to send the cancel string (@D) to the printer at print job cancel.
@D	String to send to the printer if the print job is canceled.
@I	User to which to send `intervention required` messages.
@O	Name of file to be generated by the pioout command in which to store data instead of sending it to the printer.
@P	Name of file (usually the header page) to be sent to the printer before the first byte of the print file is sent.
@S	Name of file to be sent to the printer after the last byte of the print file has been sent.

Reserved Attribute Names

Reserved attribute names are names that are assumed by the print job manager:

First two characters are _ _	Group header attribute.
First character is @	Value is provided automatically.
First character is _	Default value for flag argument.
First character is i	Pipeline for input data stream.
First character is l	Flags prohibited for input data stream.
First character is f	Command string for the filter flag.

First character is z and second character is D, P, or S:

zD	Default state of the colon file when in the /var/spool/lpd/pio/custom/* directory (+ means expanded, ! means contracted).
zP	Name of the colon file's parent colon file. The parent colon file is assumed to be in the /usr/lib/lpd/pio/predef/* directory.
zS	Current state of the colon file (+ means expanded, ! means contracted).

First character is y

Values for terminal-attached printers.

Suggested Attribute Names

Suggested attribute names are names that are assumed by many formatter filters:

First character is s	System administrator value.
First character is d	Directory path.
First character is m	Miscellaneous value (constant).
First character is w	Work value (changes while formatting).
First character is c	Command aggregate.
First character is a	ASCII control code.
First character is e	Printer escape sequence.
First character is t and second character is 0-9
	Full path names of zero or more. Stage 2 translation tables used by formatter. Multiple values must be separated by commas.

Attribute Values

The following conventions have been established for attribute values:

Printer names are of the form 4201-3 , reflecting the printer name (4201 ) and model number (3 ).
File names in the Predefined Database are of the form PrinterType.DataStreamType; for instance, 4216-31.asc indicates a 4216 Model 31 printer and an ASCII data stream.
File names in the Customized Database are of the form QueueName:QueueDeviceName, such as proq:mypro .
Attribute values can contain a \ (backslash) followed by one to three octal digits to represent non-ASCII values. A \ (backslash) that does not begin an octal sequence should be represented by either \\ or \134 .
Characters can be represented by hexadecimal notation of the form \xXX , where XX represents a hexadecimal value.
Boolean values can be represented by a + (plus sign) for true, and an ! (exclamation point) for false.
Because attribute values reside in colon files, a colon character must not appear in the attribute value. Instead, a colon should be represented by \072 .
An attribute value that references an integer attribute requiring translation from a lookup table should always appear in a colon file after the referenced integer attribute: For example, from the string red to an equivalent integer value of 2. Integer values are created from a colon file in the same order they are defined in the colon file. Listing the attribute value first ensures that when the integer attribute is referenced, it has been converted before it is referenced by the %G escape sequence.
Run all the shell commands using ksh instead of bsh.

Limits Field

The limits field in the colon file contains two types of information. SMIT dialog information and validation information.

SMIT Dialog Information

Information used in building SMIT objects represent colon file attributes in the object data manager (ODM). These objects will be used in the Print a File, Printer Setup, and Default Job Characteristics dialogs.

The limits field gives you some control over the type of sm_cmd_opt ODM object that is built for every object. You can control whether or not an attribute is always displayed, never displayed, or displayed only if it is referenced in a pipeline. You can modify the following fields:

id_seq_num
entry_type
cmd_to_list_mode
required
op_type
multi_select
disp_values
aix_values
values_msg_file
values_msg_set
values_msg_id
help_msg_id
help_msg_loc

Validation Information

Validation information validates attribute values when the colon file is complete and a print job is submitted.