Navigation Bar
Technical Support

Document Structuring Conventions

PostScript Language Document Structuring Conventions Specification
As discussed in Chapter 3 of the PostScript Language Reference Manual, Second Edition, the PostScript (TM) language standard does not specify the overall structure of a PostScript language program. Any sequence of tokens conforming to the syntax and semantics of the PostScript language is a valid program that may be presented to a PostScript interpreter for execution.
For a PostScript language program that is a page description (in other words, a description of a printable document), it is often advantageous to impose an overall program structure.
A page description can be organized as a prolog and a script, as discussed in section 2.4.2, "Program Structure" of the PostScript Language Reference Manual, Second Edition. The prolog contains application-dependent definitions. The script describes the particular desired results in terms of those definitions. The prolog is written by a programmer, stored in a place accessi-ble to an application program, and incorporated as a standard preface to each page description created by the application. The script is usually generated automatically by an application program.
Beyond this simple convention, this appendix defines a standard set of document structuring conventions (DSC). Use of the document structuring conventions not only helps assure that a document is device independent, it allows PostScript language programs to communicate their document structure and printing requirements to document managers in a way that does not affect the PostScript language page description.
A document manager can be thought of as an application that manipulates the PostScript language document based on the document structuring con-ventions found in it. In essence, a document manager accepts one or more PostScript language programs as input, transforms them in some way, and produces a PostScript language program as output. Examples of document managers include print spoolers, font and other resource servers, post-processors, utility programs, and toolkits.
If a PostScript language document properly communicates its structure and requirements to a document manager, it can receive certain printing services. A document manager can offer different types of services to a document. If the document in question does not conform to the DSC, some or all of these services may be denied to it.
Specially formatted PostScript language comments communicate the docu-ment structure to the document manager. Within any PostScript language document, any occurrence of the character % not inside a PostScript language string introduces a comment. The comment consists of all characters between the % and the next newline, including regular, special, space, and tab charac-ters. The scanner ignores comments, treating each one as if it were a single white-space character. DSC comments, which are legal PostScript language comments, do not affect the destination interpreter in any manner.
DSC comments are specified by two percent characters (%%) as the first characters on a line (no leading white space). These characters are immedi-ately followed by a unique keyword describing that particular comment- again, no white space. The keyword always starts with a capital letter and is almost always mixed-case. For example:

%%BoundingBox: 0 0 612 792
%%Pages: 45

Note that some keywords end with a colon (considered to be part of the keyword), which signifies that the keyword is further qualified by options or arguments. There should be one space character between the ending colon of a keyword and its subsequent arguments.
The PostScript language was designed to be inherently device independent. However, there are specific physical features that an output device may have that certain PostScript operators activate (in Level 1 implementations many of these operators are found in statusdict). Examples of device-dependent operators are legal, letter, and setsoftwareiomode. Use of these operators can render a document device dependent; that is, the document images properly on one type of device and not on others.
Use of DSC comments such as %%BeginFeature:, %%EndFeature (note that the colon is part of the first comment and that this comment pair is often referred to as %%Begin(End)Feature) and %%IncludeFeature: can help reduce device dependency if a document manager is available to recognize these comments and act upon them.
The DSC are designed to work with PostScript printer description (PPD) files, which provide the PostScript language extensions for specific printer features in a regular parsable format. PPD files include information about printer-specific features, and include information about the fonts built into the ROM of each printer. The DSC work in tandem with PPD files to provide a way to specify and invoke these printer features in a device-independent manner. For more information about PPD files, see the PostScript Printer Description Files Specification available from the Adobe Systems Developers' Association.
Note Even though the DSC comments are a layer of communication beyond the PostScript language and do not affect the final output, their use is considered to be good PostScript language programming style.
Troubleshooting index Postscript Index DSC Index

1 Using the Document Structuring Conventions
Ideally, a document composition system should be able to compose a document regardless of available resources-for example, font availability and paper sizes. It should be able to rely on the document management system at printing time to determine the availability of the resources and give the user reasonable alternatives if those resources are not available.
Realistically, an operating environment may or may not provide a document management system. Consequently, the DSC contain some redundancy. There are two philosophically distinct ways a resource or printer-specific feature might be specified:
. The document composition system trusts its environment to handle the resource and feature requirements appropriately, and merely specifies what its particular requirements are.
. The document composer may not know what the network environment holds or even that one exists, and includes the necessary resources and printer-specific PostScript language instructions within the document. In creating such a document, the document composer delimits these included resources or instructions in such a way that a document manager can recognize and manipulate them.
It is up to the software developer to determine which of these methods is appropriate for a given environment. In some cases, both may be used.
These two methods are mirrored in the DSC comments:
. Many DSC comments provide %%Begin and %%End constructs for identifying resources and printer-specific elements of a document. The document then prints regardless of whether a document manager is present or not.
. Many of the requirement conventions provide a mechanism to specify a need for some resource or printer-specific feature through the use of %%Include comments, and leave the inclusion of the resource or invocation of the feature to the document manager. This is an example of complete network cooperation, where a document can forestall some printing decisions and pass them to the next layer of document management. In general, this latter approach is the preferred one.
Troubleshooting index Postscript Index DSC Index

2 Document Manager Services
A document manager can provide a wide variety of services. The types of services are grouped into five management categories: spool, resource, error, print, and page management. The DSC help facilitate these services. A document that conforms to this specification can expect to receive any of these services, if available; one that does not conform may not receive any service. Listed below are some of the services that belong to each of these categories.

2.1 Spool Management
Spooling management services are the most basic services that a document manager can perform. A category of DSC comments known as general conventions-specifically the header comments-provide information concerning the document's creator, title, pages, and routing information.

The basic function of spool management is to deliver the document to the specified printer or display. The document manager should establish queues for each device to handle print job traffic in an effective manner, giving many users access to one device. In addition, the document manager should notify the user of device status (busy/idle, jammed, out of paper, waiting) and queue status (held, waiting, printing). More advanced document managers can offer job priorities and delayed-time printing.

Banner and Trailer Pages
As a part of spool management, a document manager can add a banner or trailer page to the beginning or end, respectively, of each print job to separate the output in the printer bin. The document manager can parse information from the DSC comments to produce a proper banner that includes the title, creator, creation date, the number of pages, and routing information of the document.

Print Logging
If a document manager tracks the number of pages, the type of media used, and the job requirements for each document, the document manager can produce a comprehensive report on a regular basis detailing paper and printer usage. This can help a systems administrator plan paper purchases and estimate printing costs. Individual reports for users can serve as a way to bill internally for printing.

Troubleshooting index Postscript Index DSC Index

2.2 Resource Management
Resource management services deal with the inclusion, caching, and manipu-lation of resources, such as fonts, forms, files, patterns, and documents. A category of DSC comments, known as requirement conventions, enables a document manager to properly identify instances in the document when resources are either needed or supplied.

Resource Inclusion
Frequently used resources, such as company logos, copyright notices, special fonts, and standard forms, can take up vast amounts of storage space if they are duplicated in a large number of documents. The DSC support special %%Include comments so a document manager can include a resource at print time, saving disk space.

Supplied resources can be cached in a resource library for later use. For example, a document manager that identifies a frequently used logo while processing a page description subsequently stores the logo in a resource library. The document manager then prints the document normally. When future %%IncludeResource: comments are found in succeeding documents, the document manager retrieves the PostScript language program for the logo from the resource library. The program is inserted into the document at the position indicated by the DSC comment before the document is sent to the printer.

Resource Downloading
Another valuable service that a document manager can provide is automatically downloading frequently used resources to specific printers so those resources are available instantly. Transmission and print time of documents can be greatly reduced by using this service.

For example, the document manager judges that the Stone-Serif font program is a frequently used resource. It downloads the font program from the resource library to the printer. Later, the document manager receives a document that requests the Stone-Serif font program. The document manager knows this resource is already available in the printer and sends the document to the printer without modification. Note that the resource can be downloaded persistently into VM or onto a hard disk if the printer has one. For Level 2 interpreters, resources are found automatically by the findresource operator.

Resource Optimization
An intelligent document manager can alter the position of included resources within a document to optimize memory and/or resource usage. For example, if an encapsulated PostScript (EPS) file is included several times in a document, the document manager can move duplicate procedure set defini-tions (procsets) to the top of the document to reduce transmission time. If a document manager performs dynamic resource positioning, it must main-tain the relative order of the resources to preserve any interdependencies among them.

Troubleshooting index Postscript Index DSC Index

2.3 Error Management
A document manager can provide advanced error reporting and recovery services. By downloading a special error handler to the printer, the document manager can detect failed print jobs and isolate error-producing lines of PostScript language instructions. It can send this information, a descriptive error message, and suggestions for solution back to the user.
There may be other instances where a document manager can recover from certain types of errors. Resource substitution services can be offered to the user. For example, if your document requests the Stone-Serif font program and this font program is not available on the printer or in the resource library, a document manager could select a similar font for substitution.

Troubleshooting index Postscript Index DSC Index

2.4 Print Management
Good print management ensures that the requested printer can fulfill the requirements of a particular document. This is a superset of the spool management spooling function, which is concerned with delivering the print job to the printer regardless of the consequences. By understanding the capabilities of a device and the requirements of a document, a document manager can provide a wide variety of print management services.

Printer Rerouting
A document manager can reroute documents based on printer availability. Heavily loaded printers can have their print jobs off-loaded to different printers in the network. The document manager can also inform a user if a printer is busy and suggest an idle printer for use as a backup.

If a specified printer cannot meet the requirements of a document (if for example, the document requests duplex printing and the printer does not support this feature), the document manager can suggest alternate printers.

For example, a user realizes that a document to be printed on a monochrome printer contains a color page. The user informs the document manager that the document should be rerouted to the color printer. Any printer-specific portions are detected by the document manager via the %%Begin(End)Feature: comments. The document manager consults the appropriate PostScript printer description (PPD) file, the printer-specific portion is replaced in the document, and the document is rerouted to the appropriate queue.

Feature Inclusion

This service is similar in concept to resource inclusion. Instead of using PostScript language instructions that activate certain features of a target printer, an application can use the %%IncludeFeature: comment to specify that a fragment of feature instructions should be included in the document at a specific point. A document manager can recognize such a request, consult the PPD file for the target printer, look for the specified feature, and insert the code into the document before sending it to the printer.

Parallel Printing

Parallel printing, another possible feature of a document manager, is especially useful for large documents or rush orders. Basically, the document manager splits the document based on the %%Page: comment, sending different pieces of the document to different printers simultaneously. The document is printed in parallel.

For example, a user requests that the first 100 pages of a document be printed in parallel on five separate printers. The document manager splits the document into five sections of 20 pages each, replicating the original prolog and document setup for each section. Also, a banner page is specified for each section to identify the pages being printed.

Page Breakout

Color and high-resolution printing are often expensive propositions. It does not make sense to send an entire document to a color printer if the document contains only one color illustration. When the appropriate comments are used, document managers can detect color illustrations and detailed drawings that need to be printed on high resolution printers, and split them from the original document. The document manager sends these pages separately to a high-resolution or color printer, while sending the rest of the document to lower-cost monochrome printers.

Troubleshooting index Postscript Index DSC Index

2.5 Page Management

Page management deals with organizing and reorganizing individual pages in the document. A category of comments known as page comments facilitate these services. See section 4.5, "Convention Categories," for a thorough description of page-level comments.

Page Reversal

Some printers place output in the tray face-up, some face-down. This small distinction can be a nuisance to users who have to reshuffle output into the correct order. Documents that come out of the printer into a face-up tray should be printed last page first; conversely, documents that end up face-down should be printed first page first. A document manager can reorder pages within the document based on the %%Page: comment to produce either of these effects.

n-Up Printing

n-up, thumbnail, and signature printing all fall under this category. This enables the user to produce a document that has multiple virtual pages on fewer physical pages. This is especially useful when proofing documents, and requires less paper.

For example, suppose a user wants a proof of the first four pages of a docu-ment (two copies, because the user's manager is also interested). Two-up printing is specified, where two virtual pages are mapped onto one physical sheet. The document manager adds PostScript language instructions (usually to the document setup section) that will implement this service.

Range Printing

Range printing is useful when documents need not be printed in their entirety. A document manager can isolate the desired pages from the document (using the %%Page: comment and preserving the prolog and document setup) before sending the new document to the printer. In the previous example, the user may want only the first four pages of the document. The document manager determines where the first four pages of the document reside and discards the rest.

Collated Printing

When using the #copies or setpagedevice features to specify multiple copies, on some printers the pages of the document emerge uncollated (1-1-1-2-2-2-3-3-3). Using the same mechanics as those for range printing, a document manager can print a group of pages multiple times and obtain collated output (1-2-3-1-2-3-1-2-3), saving the user the frustration of hand collating the document.


Underlays are text and graphic elements, such as draft and confidential notices, headers, and images, that a document manager can add to a document so they appear on every page. By adding PostScript language instructions to the document setup, each page of the document renders the underlay before drawing the page itself.

Troubleshooting index Postscript Index DSC Index

3 DSC Conformance

The PostScript interpreter does not distinguish between PostScript language page descriptions that do or do not conform to the DSC. However, the struc-tural information communicated through DSC comments is of considerable importance to document managers that operate on PostScript page descrip-tions as data. Because document managers cannot usually interpret the PostScript language directly, they must rely on the DSC comments to properly manipulate the document. It is necessary to distinguish between those documents that conform to the DSC and those that do not.

Note In previous versions of the DSC, there were references to partially conforming documents. This term has caused some confusion and its use has been discontinued. A document either conforms to the conventions or it does not.

3.1 Conforming Documents

A conforming document can expect to receive the maximum amount of services from any document manager. A conforming document is recognized by the header comment %!PS-Adobe-3.0 and is optionally followed by keywords indicating the type of document. Please see the description of this comment in section 5, "General Conventions," for more details about optional keywords.

A fully conforming document is one that adheres to the following rules regarding syntax and semantics, document structure, and the compliance constraints. It is also strongly suggested that documents support certain printing services.

Syntax and Semantics

If a comment is to be used within a document, it must follow the syntactical and semantic rules laid out in this specification for that comment.

Consider the following incorrect example:

%%BoundingBox 43.22 50.45 100.60 143.49

This comment is incorrect on two counts. First, there is a colon missing from the %%BoundingBox: comment. Abbreviations for comments are not accept-able. Second, floating point arguments are used instead of the integer argu-ments this comment requires.

Document Structure

The document structure rules described in section 4, "Document Structure Rules," must be followed. The following comments delineate the structure of the document. If there is a section of a document that corresponds to a particular comment, that comment must be used to identify that section of the document.














For example, if there are distinct independent pages in a document, the %%Page: comment must be used at the beginning of each page to identify those pages.

Structure of a conforming PostScript language document


...DSC comments only...



%%BeginResource: procset name 1

...PostScript code and DSC comments...


. .

%%BeginResource: procset name n

...PostScript code and DSC comments...




...PostScript code and DSC comments...


%%Page: label 1 ordinal 1

...DSC comments only...


...PostScript code and DSC comments...


...PostScript code and DSC comments...


...PostScript code and DSC comments... . . .

%%Page: label n ordinal n

...DSC comments only...


...PostScript code and DSC comments...


...PostScript code and DSC comments...


...PostScript code and DSC comments...


...PostScript code and DSC comments...

Compliance Constraints

The compliance constraints described in section 4.3, "Constraints," including the proper use of restricted operators, must be adhered to..

Where sections of the structure are not applicable, those sections and their associated comments need not appear in the document. For example, if a document setup is not performed inside a particular document, the %%BeginSetup and %%EndSetup comments are unnecessary. Figure 1 illustrates the structure of a conforming PostScript language document.

Printing Services

There are document manager printing services (such as those described in section 2, "Document Manager Services") that can be easily supported and add value to an application. Although it is not a requirement of a conforming document, it is strongly suggested that applications support these services by using the comments listed below. Note that 20 comments will ensure support of all services.

Spool Management (Spooling, Banner and Trailer Pages, and Print Logging)

%%Creator: %%PageMedia:

%%CreationDate: %%PageRequirements:

%%DocumentMedia: %%Requirements:

%%DocumentPrinterRequired: %%Routing:

%%For: %%Title:

Resource Management (Resource Inclusion, Downloading, and Optimization)

%%DocumentNeededResources: %%IncludeResource:

%%DocumentSuppliedResources: %%Begin(End)Resource:


Error Management (Error Reporting and Recovery)

%%Extensions: %%ProofMode: %%LanguageLevel:

Printer Management (Printer Rerouting, Feature Inclusion, Parallel Printing, Color Breakout)

%%Begin(End)Feature: %%IncludeFeature:

%%Begin(End)Resource: %%IncludeResource:

%%DocumentMedia: %%LanguageLevel:

%%DocumentNeededResources: %%PageMedia:

%%DocumentPrinterRequired: %%PageRequirements:

%%DocumentSuppliedResources: %%PageResources:

%%Extensions: %%Requirements:

Page Management (Page Reversal, N-up Printing, Range Printing, Collation, Underlays)

%%Pages: %%Page:

%%EndComments %%Begin(End)PageSetup

%%Begin(End)Setup %%PageTrailer

%%Begin(End)Prolog %%Trailer

Troubleshooting index Postscript Index DSC Index

3.2 Non-Conforming Documents

A non-conforming document most likely will not receive any services from a document manager, may not be able to be included into another document, and may not be portable. In some cases, this may be appropriate; a PostScript language program may require an organization that is incompatible with the DSC. This is especially true of very sophisticated page descriptions composed directly by a programmer.

However, for page descriptions that applications generate automatically, adherence to the structuring conventions is strongly recommended, simple to achieve, and essential in achieving a transparent corporate printing network.

A non-conforming document is recognized by the %! header comment. Under no circumstances should a non-conforming document use the %!PS-Adobe-3.0 header comment.

Troubleshooting index Postscript Index DSC Index

4 Document Structure Rules

One of the most important levels of document structuring in the PostScript language is the distinction between the document prolog and the document script. The prolog is typically a set of procedure definitions appropriate for the set of operations a document composition system needs, and the script is the software-generated program that represents a particular document.

A conforming PostScript language document description must have a clearly defined prolog and script separated by the %%EndProlog comment.

4.1 Prolog

The prolog consists of a header section, an optional defaults subsection, and the prolog proper, sometimes known as the procedures section.

The header section consists of DSC comments only and describes the environment that is necessary for the document to be output properly. The end of the header section is denoted by the %%EndComments comment (see the note on header comments in section 4.5, "Convention Categories").

The defaults section is an optional section that is used to save space in the document and as an aid to the document manager. The beginning of this section is denoted by the %%BeginDefaults comment. Only DSC page comments should appear in the defaults section. Information on the page-level comments that are applicable and examples of their use can be found in section 5.2, "General Body Comments" under the definition of %%Begin(End)Defaults. The end of the defaults section is indicated by the %%EndDefaults comment.

The beginning of the procedures section is indicated by the %%BeginProlog comment. This section is a series of procedure set (procset) definitions; each procset is enclosed between a %%BeginResource: procset and %%EndResource pair. Procsets are groups of definitions and routines appropriate for different imaging requirements.

The prolog has the following restrictions:

. Executing the prolog should define procsets only. For example, these procsets can consist of abbreviations, generic routines for drawing graphics objects, and routines for managing text and images.

. A document-producing application should almost always use the same prolog for all of its documents, or at least the prolog should be drawn from a pool of common procedure sets. The prolog should always be constructed in a way that it can be removed from the document and downloaded only once into the printer. All subsequent documents that are downloaded with this prolog stripped out should still execute correctly.

. No output can be produced while executing the prolog, no changes can be made to the graphics state, and no marks should be made on the page.

Troubleshooting index Postscript Index DSC Index

4.2 Script

The document script consists of three sections: a document setup section, page sections, and a document trailer.

. The document setup section is denoted by the %%Begin(End)Setup comments. The document setup should consist of procedure calls for invoking media selections (for example, setting page size), running initialization routines for procsets, downloading a font or other resource, or setting some aspect of the graphics state. This section should appear after the %%EndProlog comment, but before the first %%Page: comment.

. The pages section of the script consists of 1 to n pages, each of which should be functionally independent of the other pages. This means that each page should be able to execute in any order and may be physically rearranged, resulting in an identical document as long as the information within it is the same, but with the physical pages ordered differently. A typical example of this page reordering occurs during a page-reversal operation performed by a document manager.

The start of each page is denoted by the %%Page: comment and can also contain a %%Begin(End)PageSetup section (analogous to the document setup section on a page level), and an optional %%PageTrailer section (similar to the document trailer). In any event, each page will contain between the setup and the trailer sections the PostScript language program necessary to mark that page.

. The document trailer section is indicated by the %%Trailer comment. PostScript language instructions in the trailer consists of calls to termination routines of procedures and post-processing or cleanup instructions. In addition, any header comments that were deferred using the (atend) notation will be found here. See section 4.6, "Comment Syntax Reference," for a detailed description of (atend).

There are generally few restrictions on the script. It can have definitions like the prolog and it can also modify the graphics environment, draw marks on the page, issue showpage, and so on. There are some PostScript language operators that should be avoided or at least used with extreme caution. A thorough discussion of these operators can be found in Appendix I of the PostScript Language Reference Manual, Second Edition.

The end of a document should be signified by the %%EOF comment.

Troubleshooting index Postscript Index DSC Index

4.3 Constraints

There are several constraints on the use of PostScript language operators in a conforming document. These constraints are detailed below and are not only applicable to documents that conform to the DSC. Even a non-conforming document is much more portable across different PostScript interpreters if it observes these constraints.

Page Independence

Pages should not have any inter-dependencies. Each page may rely on certain PostScript language operations defined in the document prolog or in the document setup section, but it is not acceptable to have any graphics state set in one page of a document on which another page in the same document relies on. It is also risky to reimpose or rely on a state defined in the docu-ment setup section; the graphics state should only be added to or modified, not reimposed. See Appendix I of the PostScript Language Reference Manual, Second Edition for more details on proper preservation of the graphics state with operators like settransfer.

Page independence enables a document manager to rearrange the document's pages physically without affecting the execution of the document description. Other benefits of page independence include the ability to print different pages in parallel on more than one printer and to print ranges of pages. Also, PostScript language previewers need page independence to enable viewing the pages of a document in arbitrary order.

For the most part, page independence can be achieved by placing a save-restore pair around each page, as shown below:


...Header comments, prolog definitions, document setup...

%%Page: cover 1

%%BeginPageSetup /pgsave save def

...PostScript language instructions to perform page setup...


...PostScript language instructions to mark page 1...

pgsave restore showpage

...Rest of the document...


The save-restore pair will also reclaim any non-global VM used during the page marking (for example, text strings).

Note If pages must have interdependencies, the %%PageOrder: Special comment should be used. This ensures that a document manager will not attempt to reorder the pages.

Line Length

To provide compatibility with a large body of existing application and document manager software, a conforming PostScript language document description does not have lines exceeding 255 characters, excluding line-termination characters. The intent is to be able to read lines into a 255-character buffer without overflow (Pascal strings are a common example of this sort of buffer).

The PostScript interpreter imposes no constraints as to where line breaks occur, even in string bodies and hexadecimal bitmap representations. This level of conformance should not pose a problem for software development. Any document structuring comment that needs to be continued on another line to avoid violating this guideline should use the %%+ notation to indicate that a comment line is being continued (see %%+ in section 5.2, "General Body Comments").

Line Endings

Lines must be terminated with one of the following combinations of characters: CR, LF, or CR LF. CR is the carriage-return character and LF is the line-feed character (decimal ASCII 13 and 10, respectively).

Use of showpage

To reduce the amount of VM used at any point, it is common practice to delimit PostScript language instructions used for a particular page with a save-restore pair. See the page-independence constraint for an example of save-restore use.

If the showpage operator is used in combination with save and restore, the showpage should occur after the page-level restore operation. The motivation for this is to redefine the showpage operator so it has side effects in the printer VM, such as maintaining page counts for printing n-up copies on one sheet of paper. If showpage is executed within the confines of a page-level save-restore, attempts to redefine showpage to perform extra operations will not work as intended. This also applies to the BeginPage and EndPage parameters of the setpagedevice dictionary. The above discussion also applies to gsave-grestore pairs.

Document Copies

In a conforming document, the number of copies must be modified in the doc-ument setup section of the document (see %%BeginSetup and %%EndSetup). Changing the number of copies within a single page automatically breaks the page independence constraint. Also, using the copypage operator is not recommended because doing so inhibits page independence. If multiple copies of a document are desired, use the #copies key or the setpagedevice operator.

In Level 1 implementations, the #copies key can be modified to produce multiple copies of a document as follows:


%%Pages: 23

%%Requirements: numcopies(3) collate


...Prolog with procset definitions...


%%BeginSetup /#copies 3 def


...Rest of the Document (23 virtual pages)...


In Level 2 implementations, the number of copies of a document can be set using the setpagedevice operator as follows:

<< /NumCopies 3 >> setpagedevice

The %%Pages: comment should not be modified if the number of copies is set, as it represents the number of unique virtual pages in the document. However, the %%Requirements: comment should have its numcopies option modified, and the collate option set, if applicable.

Restricted Operators

There are several PostScript language operators intended for system-level jobs that are not appropriate in the context of a page description program. Also, there are operators that impose conditions on the graphics state directly instead of modifying or concatenating to the existing graphics state. How-ever, improper use of these operators may cause a document manager to process a document incorrectly. The risks of using these operators involve either rendering a document device dependent or unnecessarily inhibiting constructive post-processing of document files for different printing needs- for example, embedding one PostScript language document within another.

In addition to all operators in statusdict and the operators in userdict for establishing an imageable area, the following operators should be used carefully, or not at all, in a PostScript language page description:

banddevice framedevice quit

setpagedevice clear grestoreall

renderbands setscreen cleardictstack

initclip setglobal setshared copypage

initgraphics setgstate settransfer

erasepage initmatrix sethalftone

startjob exitserver nulldevice

setmatrix undefinefont

For more specific information on the proper use of these operators in various situations, see Appendix I of the PostScript Language Reference Manual, Second Edition.

There are certain operators specific to the Display PostScript system that are not part of the Level 1 and Level 2 implementations. These operators are for display systems only and must not be used in a document. This is a much more stringent restriction than the above list of restricted operators, which may be used with extreme care. For a complete list see section A.1.2, "Display PostScript Operators, of the PostScript Language Reference Manual, Second Edition."

Troubleshooting index Postscript Index DSC Index

4.4 Parsing Rules

Here are a few explicit rules that can help a document manager parse the DSC comments: Troubleshooting index Postscript Index DSC Index

4.5 Convention Categories

The DSC comments are roughly divided into the followingsix categories of conventions: Typically, some subsets of the general, requirement, and color separation conventions are used consistently in a particular printing environment. The DSC have been designed with maximum flexibility in mind and with a mini-mum amount of interdependency between conventions. For example, one may use only general conventions in an environment where the presence of a document manager may not be guaranteed, or may use the requirement con-ventions on a highly spooled network.

General conventions delimit the various structural components of a PostScript language page description, including its prolog, script, and trailer, and where the page breaks fall, if there are any. The general convention comments include document and page setup information, and they provide a markup convention for noting the beginning and end of particular pieces of the page description that might need to be identified for further use.

Requirement conventions are comments that suggest document manager action. These comments can be used to specify resources the document sup-plies or needs. Document managers may make decisions based on resource frequency (those that are frequently used) and load resources permanently into the printer, download them before the job, or store them on a printer's hard disk, thus reducing transmission time.

Other requirement comments invoke or delimit printer-specific features and requirements, such as paper colors and weights, collating order, and stapling. The document manager can replace printer-specific PostScript language fragments based on these comments when rerouting a print job to another printer, by using information in the PostScript printer description (PPD) file for that printer.

Color separation conventions are used to complement the color extensions to the PostScript language. Comments typically identify PostScript language color separation segments in a page, note custom color ratios (RGB or CMYK), and list document and page level color use.

Query conventions delimit parts of a PostScript language program that query the current state or characteristics of a printer, including the availability of resources (for example, fonts, files, procsets), VM, and any printer-specific features and enhancements. The type of program that uses this set of conven-tions is usually interactive-that is, one that expects a response from the printer. This implies that document managers should be able to send query jobs to a printer, and route an answer back to the application that issued the query. Query conventions should only be used in %!PS-Adobe-3.0 Query jobs.

Open structuring conventions are user-defined conventions. Section 9, "Open Structuring Conventions," provides guidelines for creating these vendor-specific comments.

Special conventions include those comments that do not fall into the above categories.

The general, requirement, and color separation conventions can be further broken down into three classes: header comments, body comments, and page comments.

Header Comments

Header comments appear first in a document file, before any of the executable PostScript language instructions and before the procedure definitions. They may be thought of as a table of contents. In order to simplify a docu-ment manager's job in parsing these header comments, there are two rules that apply:

. If there is more than one instance of a header comment in a document file, the first one encountered takes precedence. This simplifies nesting documents within one another without having to remove the header comments.

. Header comments must be contiguous. That is, if a document manager comes across a line that does not begin with %, the document manager may quit parsing for header comments. The comments may also be ended explicitly with the %%EndComments convention.

All instances of lines beginning with %! after the first instance are ignored by document managers, although to avoid confusion, this notation should not appear twice within the block of header comments (see %%BeginDocument: and %%EndDocument for examples of embedded documents).

Body Comments

Body comments may appear anywhere in a document, except the header sec-tion. They are designed to provide structural information about the organiza-tion of the document file and should match any related information provided in the header comments section. They generally consist of %%Begin and %%End constructs to delimit specific components of the document file, such as procsets, fonts, or emulation code, and %%Include comments that request the document manager to take action when encountering the comment, such as including a document, resource, or printer-specific fragment of code.

Page Comments

Page comments are page-level structure comments. They should not span across page boundaries (see the exception below). That is, a page comment applies only to the page in which it appears. The beginning of a page should be noted by the %%Page: comment. The other page comments are similar to their corresponding header comments (for example, %%BoundingBox: vs. %%PageBoundingBox:), except for %%Begin or %%End comments that are more similar to body comments in use (e.g., %%Begin(End)Setup vs. %%Begin(End)PageSetup).

Note Some page comments that are similar to header comments can be used in the defaults section of the file to denote default requirements or media for all pages. See the %%Begin(End)Defaults comments for a more detailed explanation.

Technical Support | Developer Services | Production Tools | Corporate Information | What's New | Site Map | Home