Current File : //usr/share/ghostscript/Resource/Init/pdf_base.ps
% Copyright (C) 2001-2018 Artifex Software, Inc.
% All Rights Reserved.
%
% This software is provided AS-IS with no warranty, either express or
% implied.
%
% This software is distributed under license and may not be copied,
% modified or distributed except as expressly authorized under the terms
% of the license contained in the file LICENSE in this distribution.
%
% Refer to licensing information at http://www.artifex.com or contact
% Artifex Software, Inc., 1305 Grant Avenue - Suite 200, Novato,
% CA 94945, U.S.A., +1(415)492-9861, for further information.
%
% pdf_base.ps
% Basic parser for PDF reader.
% This handles basic parsing of the file (including the trailer
% and cross-reference table), as well as objects, object references,
% streams, and name/number trees; it doesn't include any facilities for
% making marks on the page.
/.setlanguagelevel where { pop 2 .setlanguagelevel } if
.currentglobal //true .setglobal
pdfdict begin
% Define the name interpretation dictionary for reading values.
/valueopdict mark
(<<) cvn { mark } bind executeonly % don't push an actual mark!
(>>) cvn { { .dicttomark } stopped {
( **** Error: File has an unbalanced >> \(close dictionary\).\n)
pdfformaterror
( Output may be incorrect.\n) pdfformaterror
} if
} bind executeonly
([) cvn { mark } bind executeonly % ditto
(]) cvn dup load
% /true true % see .pdfexectoken below
% /false false % ibid.
% /null null % ibid.
/F dup cvx % see Objects section below
/R dup cvx % see Objects section below
/stream dup cvx % see Streams section below
.dicttomark readonly def
% ------ Utilities ------ %
% Define a scratch string. The PDF language definition says that
% no line in a PDF file can exceed 255 characters, but this string
% is also used to search for %PDF-, which needs 1024 characters.
/pdfstring 1024 string def
% Handle the PDF 1.2 #nn escape convention when reading from a file.
% This should eventually be done in C.
/.pdffixname { % <execname> .pdffixname <execname'>
PDFversion 1.2 ge {
dup .namestring (#) search {
name#escape cvn exch pop
} {
pop
} ifelse
} if
} bind executeonly def
/name#escape % <post> <(#)> <pre> name#escape <string>
{ exch pop
1 index 2 () /SubFileDecode filter dup (x) readhexstring
% Stack: post pre stream char t/f
not { % tolerate, but complain about bad syntax
pop closefile (#) concatstrings exch
( **** Error: Invalid hex following '#' name escape, using literal '#' in name.\n)
pdfformaterror
( Output may be incorrect.\n) pdfformaterror
} {
exch closefile concatstrings
exch 2 1 index length 2 sub getinterval
} ifelse
(#) search { name#escape } if concatstrings
} bind executeonly def
/num-chars-dict mark (0123456789-.) {dup} forall .dicttomark readonly def
% Execute a file, interpreting its executable names in a given
% dictionary. The name procedures may do whatever they want
% to the operand stack.
/.pdftokenerror { % <count> <opdict> <errtoken> .pdftokenerror -
BXlevel 0 le {
( **** Error: Unknown operator: ') pdfformaterror
dup =string cvs pdfformaterror
(') pdfformaterror
% Attempt a retry scan of the element after changing to PDFScanInvNum
<< /PDFScanInvNum //true >> setuserparams
=string cvs
token pop exch pop dup type
dup /integertype eq exch /realtype eq or {
exch pop exch pop
(, processed as number, value: ) pdfformaterror
dup =string cvs pdfformaterror (\n) pdfformaterror
<< /PDFScanInvNum //null >> setuserparams % reset to default scanning rules
//false % suppress any stack cleanup
} {
% error was non-recoverable with modified scanning rules, continue.
dup type /nametype eq {
//true 1 index .namestring {
//num-chars-dict exch known and
} forall { % perhaps, it's a malformed number.
PDFSTOPONERROR {//true}{
pop pop pop 0 //false
( looks like a malformed number, replacing with 0.) pdfformaterror
} ifelse
} {
//true % punt
} ifelse
} {
//true % punt
} ifelse
(\n) pdfformaterror
} ifelse
} {
//true
} ifelse
{ % clean up the operand stack if this was non-recoverable
pop pop count exch sub { pop } repeat % pop all the operands
} if
( Output may be incorrect.\n) pdfformaterror
} bind executeonly def
currentdict /num-chars-dict .undef
/.pdfexectoken { % <count> <opdict> <exectoken> .pdfexectoken ?
PDFDEBUG {
//pdfdict /PDFSTEPcount known not { //pdfdict /PDFSTEPcount 1 .forceput } executeonly if
PDFSTEP {
//pdfdict /PDFtokencount 2 copy .knownget { 1 add } { 1 } ifelse .forceput
PDFSTEPcount 1 gt {
//pdfdict /PDFSTEPcount PDFSTEPcount 1 sub .forceput
} executeonly
{
dup ==only
( step # ) print PDFtokencount =only
( ? ) print flush 1 //false .outputpage
(%stdin) (r) file 255 string readline {
token {
exch pop //pdfdict /PDFSTEPcount 3 -1 roll .forceput
} executeonly
{
//pdfdict /PDFSTEPcount 1 .forceput
} executeonly ifelse % token
} {
pop /PDFSTEP //false def % EOF on stdin
} ifelse % readline
} ifelse % PDFSTEPcount > 1
} executeonly
{
dup ==only () = flush
} ifelse % PDFSTEP
} executeonly if % PDFDEBUG
2 copy .knownget {
exch pop exch pop exch pop exec
} {
% Normally, true, false, and null would appear in opdict
% and be treated as "operators". However, there is a
% special fast case in the PostScript interpreter for names
% that are defined in, and only in, systemdict and/or
% userdict: putting these three names in the PDF dictionaries
% destroys this property for them, slowing down their
% interpretation in all PostScript code. Therefore, we
% check for them explicitly here instead.
dup dup dup /true eq exch /false eq or exch /null eq or {
exch pop exch pop //systemdict exch get
} {
% Hackish fix to detect missing whitespace after "endobj". Yet another
% problem that (you guessed it!) Adobe Acrobat ignores silently
dup .namestring (endobj) anchorsearch {
( **** Error: Missing whitespace after 'endobj'.\n) pdfformaterror
( Output may be incorrect.\n) pdfformaterror
pop pop pop exch pop /endobj get exec
} {
%% First, lets try and see if this 'might' be a broken number
%% we look for 0-9 as well as '.' ',' and '-' to permit those
%% locales in which the separator is a comma, as well as negative
%% numbers. We've seena t least one tool replace '0' with '-' *sometimes*
true exch
dup length 1 sub 0 1 3 -1 roll {
1 index exch get
dup 44 lt {
pop exch pop false exch exit
}{
dup 57 gt {
pop exch pop false exch exit
}{
dup 47 eq {
pop exch pop false exch exit
}{
pop
} ifelse
} ifelse
} ifelse
} for
pop
{
%% If it looks like a number, try to deal with it as such
PDFSTOPONERROR {
dup .pdftokenerror
}{
dup {.pdftokenerror} stopped
} ifelse
{ pop
%% But if it fails, fall back to converting into a anem.
%% This will propagate through and cause different
%% fallback code to try and take care ot if.
cvlit exch pop exch pop
}{
exch pop
} ifelse
}{
%% Doesn't look like a number, convert it into a name
cvlit exch pop exch pop
} ifelse
} ifelse
} ifelse
} ifelse
} bind executeonly odef
/PDFScanRules_true << /PDFScanRules //true >> def
/PDFScanRules_null << /PDFScanRules //null >> def
/.pdfrun { % <file> <opdict> .pdfrun -
% Construct a procedure with the stack depth, file and opdict
% bound into it.
1 index cvlit % file <<>> file
count 2 sub % file <<>> file cnt
3 1 roll mark % file cnt <<>> file [
/PDFScanRules .getuserparam //null eq {
//PDFScanRules_true { setuserparams } 0 get % force PDF scanning mode
mark 7 4 roll
} {
mark 5 2 roll % file [ [ cnt <<>> file
} ifelse
{ % Stack: ..operands.. count opdict file
{ token } stopped {
dup type /filetype eq { pop } if
pop pop stop
} if {
dup type /nametype eq {
dup xcheck {
.pdfexectoken
} {
.pdffixname
exch pop exch pop PDFDEBUG {
PDFSTEPcount 1 le {
dup ==only ( ) print flush
} if
} if
} ifelse
} {
exch pop exch pop PDFDEBUG {
PDFSTEPcount 1 le {
dup ==only ( ) print flush
} if
} if
} ifelse
} {
pop pop exit
} ifelse
}
aload pop .packtomark cvx % file [ {cnt <<>> file ... }
{ loop } 0 get 2 packedarray cvx % file [ { {cnt <<>> file ... } loop }
PDFSTOPONERROR { {exec //false} } { {stopped} } ifelse
aload pop % file [ { {cnt <<>> file ... } loop } stopped
/PDFScanRules .getuserparam //null eq {
//PDFScanRules_null { setuserparams } 0 get % reset PDF scannig mode if it was off
} if
/PDFsource PDFsource % file [ { {cnt <<>> file ... } loop } stopped /PDFsource PDFsource
{ store {
/StreamRunAborted //true store
( **** Error reading a content stream. The page may be incomplete.\n)
pdfformaterror
( Output may be incorrect.\n) pdfformaterror
} if
} aload pop % file [ { {cnt <<>> file ... } loop } stopped /PDFsource PDFsource store {...} if
.packtomark cvx % file { { {cnt <<>> file ... } loop } stopped /PDFsource PDFsource store {...} if}
/PDFsource 3 -1 roll store % {...}
exec
} bind executeonly def
% Execute a file, like .pdfrun, for a marking context.
% This temporarily rebinds LocalResources and DefaultQstate.
/.pdfruncontext { % <resdict> <file> <opdict> .pdfruncontext -
/.pdfrun load LocalResources DefaultQstate
/LocalResources 7 -1 roll store
/DefaultQstate qstate store
3 .execn
/DefaultQstate exch store
/LocalResources exch store
} bind executeonly def
% Get the depth of the PDF operand stack. The caller sets pdfemptycount
% before calling .pdfrun or .pdfruncontext. It is initially set by
% pdf_main, and is also set by any routine which changes the operand
% stack depth (currently .pdfpaintproc, although there are other callers
% of .pdfrun{context} which have not been checked for opstack depth.
/.pdfcount { % - .pdfcount <count>
count pdfemptycount sub
} bind executeonly def
% Read a token, but simply return false (no token read) in the case of an
% error. This is messy because 'token' either may or may not pop its operand
% if an error occurs, and because the return values are different depending
% on whether the source is a file or a string. To avoid closing the file
% check for '{' before trying 'token'.
/token_nofail_dict mark
( ) { dup ( ) readstring pop pop } bind executeonly
(\t) 1 index
(\r) 1 index
(\n) 1 index
(\000) 1 index
({) { //null //true exit } bind executeonly
.dicttomark def
/token_nofail { % <file|string> token_nofail false
% <file> token_nofail <token> true
% <string> token_nofail <post> <token> true
dup type /filetype eq {
{ dup ( ) .peekstring not { ({) } if
//token_nofail_dict exch .knownget not {
//null 1 index { token } .internalstopped exit
} if
exec
} loop
{ % stack: source null [source]
//null ne { pop } if pop //false
} { % stack: source null ([post] token true | false)
{ 3 1 roll pop pop //true }
{ pop pop //false }
ifelse
} ifelse
} {
//null 1 index % stack: source null source
{ token } .internalstopped { % stack: source null [source]
//null ne { pop } if pop //false
} { % stack: source null ([post] token true | false)
{ 4 2 roll pop pop //true }
{ pop pop //false }
ifelse
} ifelse
} ifelse
} bind executeonly def
currentdict /token_nofail_dict .undef
% ================================ Objects ================================ %
% We keep track of PDF objects using the following PostScript variables:
%
% Generations (string): Generations[N] holds 1+ the current
% generation number for object number N. (As far as we can tell,
% this is needed only for error checking.) For free objects,
% Generations[N] is 0.
%
% Objects (array): If object N is loaded, Objects[N] is the actual
% object; otherwise, Objects[N] is an executable integer giving
% the file offset of the object's location in the file. If
% ObjectStream[N] is non-zero then Objects[N] contains the index
% into the object stream instead of the file offset of the object.
%
% ObjectStream (array): If object N is in an object stream then
% ObjectStream[N] holds the object number of the object stream.
% Otherwise ObjectStream[N] contains 0. If ObjectStream[N]
% is non-zero then Objects[N] contains the index into the object
% stream.
%
% GlobalObjects (dictionary): If object N has been resolved in
% global VM, GlobalObjects[N] is the same as Objects[N]
% (except that GlobalObjects itself is stored in global VM,
% so the entry will not be deleted at the end of the page).
%
% IsGlobal (string): IsGlobal[N] = 1 iff object N was resolved in
% global VM. This is an accelerator to avoid having to do a
% dictionary lookup in GlobalObjects when resolving every object.
% Initialize the PDF object tables.
/initPDFobjects { % - initPDFobjects -
/ObjectStream 0 array def
/Objects 0 array def
/Generations 0 string def
.currentglobal //true .setglobal
/GlobalObjects 20 dict def
.setglobal
/IsGlobal 0 string def
} bind executeonly def
% Grow the tables to a specified size.
/growPDFobjects { % <minsize> growPDFobjects -
dup ObjectStream length gt {
dup ObjectStream exch array dup 3 1 roll copy pop /ObjectStream exch def
} if
dup Objects length gt {
dup Objects exch array dup 3 1 roll copy pop /Objects exch def
} if
dup Generations length gt {
dup Generations exch string dup 3 1 roll copy pop /Generations exch def
} if
dup IsGlobal length gt {
dup IsGlobal exch string dup 3 1 roll copy pop /IsGlobal exch def
} if
pop
} bind executeonly def
% We represent an unresolved object reference by a procedure of the form
% {obj# gen# resolveR}. This is not a possible PDF object, because PDF has
% no way to represent procedures. Since PDF in fact has no way to represent
% any PostScript object that doesn't evaluate to itself, we can 'force'
% a possibly indirect object painlessly with 'exec'.
% Note that since we represent streams by executable dictionaries
% (see below), we need both an xcheck and a type check to determine
% whether an object has been resolved.
/resolved? { % <object#> resolved? <value> true
% <object#> resolved? false
Objects 1 index get dup xcheck { % Check if executable
dup type /integertype eq { % Check if an integer
% Check whether the object is in GlobalObjects.
pop IsGlobal 1 index get 0 eq { % 0 --> Not in GlabalObjects
pop //false % The object is not resolved
} { % The object is in GlobalObjects
% Update Objects from GlobalObjects
PDFDEBUG { (%Global=>local: ) print dup //== exec } if
GlobalObjects 1 index get dup Objects 4 1 roll put //true
} ifelse
} { % Else object is executable but not integer
exch pop //true % Therefore must be executable dict. (stream)
} ifelse
} { % Else object is not executable.
exch pop //true % Therefore it must have been resolved.
} ifelse
} bind executeonly def
/oforce /exec load def
/oget { % <array> <index> oget <object>
% <dict> <key> oget <object>
% Before release 6.20, this procedure stored the resolved
% object back into the referring slot. In order to support
% PDF linearization, we no longer do this.
get oforce
} bind executeonly def
/oforce_array { % <array> oforce_array <array>
[ exch { oforce } forall ]
} bind executeonly def
/oforce_elems { % <array> oforce_elems <first> ... <last>
{ oforce } forall
} bind executeonly def
%% Safe recursion maintain a dicitonary in which we record all the object numbers
%% of objects which we resolve. Whenever it resolves a new one, it checks all the
%% existing ones to see if its already present, in which case we assume recursion
%% has taken place. Whenever this procedure calls itself it copies the existing
%% dictionary, and throws it away afterwards. This means that we won't falsely
%% detect recursion if two elements at the same level indirect to the same
%% object.
/safe_recursive { % <<recursion dict>> <any> oforce_recursive <recursion dict> <any>
dup type dup /arraytype eq exch /packedarraytype eq or {
dup rcheck { % protect tint transform functions, etc.
dup length 3 eq {
dup 2 get /resolveR eq {
dup 0 get % <recursion dict> {x y resolveR} x
dup % <recursion dict> {x y resolveR} x x
3 index exch known {
( **** Error: detected circular reference in object number ) pdfformaterror pdfstring cvs pdfformaterror
(\n) pdfformaterror
pop //null
} {
dup 3 index % <recursion dict> {x y resolveR} x x <recursion dict>
3 1 roll put % <recursion dict> {x y resolveR}
} ifelse
} if
} if
oforce % but dereference {1 0 R}
dup type dup /arraytype eq exch /packedarraytype eq or {
[ exch 2 index <<>> copy exch { safe_recursive exch } forall pop]
} {
dup type /dicttype eq {
<< exch 2 index <<>> copy exch { 3 -1 roll exch safe_recursive 3 -1 roll oforce 3 1 roll exch } forall pop>>
} if
} ifelse
} if
} {
dup type /dicttype eq {
<< exch 2 index <<>> copy exch { 3 -1 roll exch safe_recursive 3 -1 roll oforce 3 1 roll exch } forall pop>>
} if
} ifelse
} bind executeonly def
/oforce_recursive {
<< >> exch safe_recursive exch pop
} bind executeonly def
% A null value in a dictionary is equivalent to an omitted key;
% we must check for this specially.
/knownoget { % <dict> <key> knownoget <value> true
% <dict> <key> knownoget false
% See oget above regarding this procedure.
.knownget {
oforce dup //null eq { pop //false } { //true } ifelse
} {
//false
} ifelse
} bind executeonly def
% See /knownoget above.
/oknown { % <dict> <key> oknown <bool>
.knownget { oforce //null ne } { //false } ifelse
} bind executeonly def
/knownogetdict { % <dict> <key> knownogetdict <dict> true
% <dict> <key> knownogetdict false
//knownoget exec dup {
1 index type /dicttype ne { pop pop //false } if
} if
} bind executeonly def
% PDF 1.1 defines a 'foreign file reference', but not its meaning.
% Per the specification, we convert these to nulls.
/F { % <file#> <object#> <generation#> F <object>
% Some PDF 1.1 files use F as a synonym for f!
.pdfcount 3 lt { f } { pop pop pop //null } ifelse
} bind executeonly def
% Verify the generation number for a specified object
% Note: The values in Generations is the generation number plus 1.
% If the value in Generations is zero then the object is free.
/checkgeneration { % <object#> <generation#> checkgeneration <object#> <OK>
Generations 2 index get 1 sub 1 index eq { % If generation # match ...
pop //true % Then return true
} { % Else not a match ...
QUIET not { % Create warning message if not QUIET
Generations 2 index get 0 eq { % Check if object is free ...
( **** Warning: reference to free object: )
2 index =string cvs concatstrings ( ) concatstrings % put obj #
exch =string cvs concatstrings ( R\n) concatstrings % put gen #
} {
( **** Warning: wrong generation: )
2 index =string cvs concatstrings ( ) concatstrings % put obj #
exch =string cvs concatstrings % put gen #
(, xref gen#: ) concatstrings 1 index Generations % put xref gen #
exch get 1 sub =string cvs concatstrings (\n) concatstrings
} ifelse
pdfformatwarning % Output warning message
} { % Else QUIET ...
pop % Pop generation number
} ifelse
% We should return false for an incorrect generation number, however
% we are simply printing a warning and then returning true. This makes
% Ghostscript tolerant of of bad generation numbers.
//true
} ifelse
} bind executeonly def
/R { % <object#> <generation#> R <object>
%% Parameter validation; this should not be required but we have found files
%% (Bug 697351) which are corrupted in ways that mean we see a 'R' in a stream
%% even though it is not an indirect object reference. Because a stream
%% may reference objetcs which have not been resolved, we cannot simply
%% 'undef' teh R operator, the only solution is to have it chekc the
%% types of its operands. This is poterntially slow of course, we may
%% need to remove this in future if the performance penalty is too great.
%% We cannot slow down all files significantly purely to work-around
%% files which are invalid.
1 index type /integertype eq 1 index type /integertype eq and {
/resolveR cvx 3 packedarray cvx
} {
( **** Error: indirect object reference \(R\) encountered with invalid arguments.) pdfformaterror
( Output may be incorrect.\n) pdfformaterror
} ifelse
} bind executeonly def
% If we encounter an object definition while reading sequentially,
% we just store it away and keep going.
/objopdict mark
valueopdict { } forall
/endobj dup cvx
.dicttomark readonly def
/obj { % <object#> <generation#> obj <object>
PDFfile objopdict .pdfrun
} bind executeonly def
/endobj { % <object#> <generation#> <object> endobj <object>
%% If we have a stream wioth a broken endstream we can get here without
%% an object definition on nthe stack. If we simply pop the 'extraneous'
%% data we will break the stream definition and throw an error. So we look
%% to see if this appears to be this specific case.
2 index type /marktype eq 2 index type /marktype eq or {
{
dup type /marktype eq {
exit
}{
pop
}if
}loop
}{
{
2 index type /integertype ne 2 index type /integertype ne or {
count 3 gt {
( **** Error: obj definition followed by multiple tokens, attempting to recover.\n) pdfformaterror
( Output may be incorrect.\n) pdfformaterror
pop
} {
//null
( **** Error: ignoring obj followed by multiple tokens.\n) pdfformaterror
( Output may be incorrect.\n) pdfformaterror
exit
} ifelse
} {
exit
} ifelse
} loop
3 1 roll
% Read the xref entry if we haven't yet done so.
% This is only needed for generation # checking.
1 index resolved? {
pop
} if
checkgeneration {
% The only global objects we bother to save are
% (resource) dictionaries.
1 index dup gcheck exch type /dicttype eq and {
PDFDEBUG { (%Local=>global: ) print dup //== exec } if
GlobalObjects 1 index 3 index put
IsGlobal 1 index 1 put
} if
Objects exch 2 index put
} {
pop pop //null
} ifelse
}ifelse
} bind executeonly def
% When resolving an object reference in an object stream, we stop at
% the end of file. Note: Objects in an object stream do not have either
% a starting 'obj' or and ending 'endobj'.
/resolveobjstreamopdict mark
valueopdict { } forall
(%%EOF) cvn { exit } bind executeonly
/endobj { % bug 689795
( **** Error: Objects in an object stream should not have 'endobj'.\n)
pdfformaterror
( Output may be incorrect.\n) pdfformaterror
} bind executeonly
.dicttomark readonly def
% Note: This version of this function is not currently being used.
% Resolve all objects in an object stream
/resolveobjectstream { % <object stream #> resolveobjectstream -
PDFDEBUG { (%Resolving object stream: ) print } if
0 resolveR % Get the objectstream dict, all objstrms use 0 as the gen #
dup /First get % Save location of first object onto the stack
1 index /N get % Save number of objects onto the stack
2 index //false resolvestream % Convert stream dict into a stream
/ReusableStreamDecode filter % We need to be able to position stream
% Objectstreams begin with list of object numbers and locations
% Create two arrays to hold object numbers and stream location
1 index array % Array for holding object number
2 index array % Array for holding stream object location
% Get the object numbers and locations.
0 1 5 index 1 sub { % Loop and collect obj # and locations
% Stack: objstreamdict First N objectstream [obj#] [loc] index
2 index 1 index % Setup to put obj# into object number array
5 index token pop put % Get stream, then get obj# and put into array
1 index 1 index % Setup to put object loc into location array
5 index token pop put % Get stream, get obj loc and put into array
pop % Remove loop index
} for
% Create a bytestring big enough for reading any object data
% Scan for the size of the largest object
0 0 % Init max object size and previous location
2 index { % Loop through all object locations
% Stack: ... maxsize prevloc currentloc
dup 4 1 roll % Save copy of object location into stack
exch sub % Object size = currentloc - prevloc
.max % Determine maximum object size
exch % Put max size under previous location
} forall
pop % Remove previous location
.bigstring % Create bytestring based upon max obj size
% Move to the start of the object data
3 index 6 index % Get objectstream and start of first object
setfileposition % Move to the start of the data
% Read the data for all objects except the last. We do
% not know the size of the last object so we need to treat
% it as a special case.
0 1 6 index 2 sub {
dup 4 index exch get % Get our current object number
% Stack: objstreamdict First N objectstream [obj#] [loc]
% bytestring loopindex object#
dup resolved? { % If we already have this object
pop pop % Remove object and object number
1 add 2 index exch get % Get location of next object
6 index add 6 index exch % Form location of next object and get stream
setfileposition % Move to the start of the next object data
} { % Else this is a new object ...
% We are going to create a string for reading the object
2 index 0 % use our working string
% Determine the size of the object
5 index 4 index 1 add get % Get location of the next object
6 index 5 index get % Get location of this object
sub % Size of object = next loc - this loc
getinterval % Create string for reading object
6 index exch readstring pop % Read object
/ReusableStreamDecode filter % Convert string into a stream
resolveobjstreamopdict .pdfrun % Get PDF object
Objects exch 2 index exch put % Put object into Objects array
pop pop % Remove object # and loop index
} ifelse
} for
pop pop % Remove our working string and loc array
% Now read the last object in the object stream. Since it
% is the last object, we can use the original stream and
% terminate when we hit the end of the stream
% Stack: objstreamdict First N objectstream [obj#]
2 index 1 sub get % Get our current object number
dup resolved? not { % If we do not already have this object
exch % Get our object stream
resolveobjstreamopdict .pdfrun % Get PDF object
Objects exch 2 index exch put % Put object into Objects array
} if
pop pop pop pop % Clear stack
} bind executeonly def
/no_debug_dict <<
/PDFDEBUG //false
>> readonly def
% Resolve all objects in an object stream
/resolveobjectstream { % <object stream #> resolveobjectstream -
PDFDEBUG { (%Resolving object stream: ) print } if
dup 0 resolveR % Get the objectstream dict, all objstrms use 0 as the gen #
dup /Type get /ObjStm ne { % Verify type is object stream
( **** Error: Incorrect Type in object stream dictionary.\n) pdfformaterror
( Output may be incorrect.\n) pdfformaterror
/resolveobjectstream cvx /typecheck signalerror
} if
dup /N get % Save number of objects onto the stack
1 index //false resolvestream % Convert stream dict into a stream
/ReusableStreamDecode filter % We need to be able to position stream
% Objectstreams begin with list of object numbers and locations
1 index array % Create array for holding object number
% Get the object numbers
0 1 4 index 1 sub { % Loop and collect obj numbers
% Stack: strm# objstreamdict N PDFDEBUG objectstream [obj#] loopindex
1 index 1 index % Setup to put obj# into object number array
4 index token pop put % Get stream, then get obj# and put into array
2 index token pop pop pop % Get stream, get obj loc and clear stack
} for
% Move to the start of the object data
1 index 4 index /First get % Get objectstream and start of first object
setfileposition % Move to the start of the data
% We disable PDFDEBUG while reading the data stream. We will
% print the data later
PDFDEBUG { //no_debug_dict begin } if
% Read the data for all objects. We check to see if we get
% the number of objects that we expect.
% Stack: strm# objstreamdict N objectstream [obj#] PDFDEBUG
mark 3 -1 roll % Get objectstream
count 4 index add % Determine stack depth with objects
3 1 roll
resolveobjstreamopdict .pdfrun % Get PDF objects
count counttomark 1 add index ne { % Check stack depth
( **** Error: Incorrect object count in object stream.\n) pdfformaterror
( Output may be incorrect.\n) pdfformaterror
/resolveobjectstream cvx /rangecheck signalerror
} if
% We have the object data
counttomark array astore % Put objects into an array
exch pop exch pop % Remove mark and count
currentdict //no_debug_dict eq { end } if % Restore debug context
% Save the objects into Objects
0 1 2 index length 1 sub { % Loop through all objects
% Stack: strm# objstreamdict N [obj#] [objects] loopindex
dup 3 index exch get % Get our current object number
% Stack: strm# objstreamdict N [obj#] [objects] loopindex obj#
dup ObjectStream exch get 7 index eq {
dup resolved? { % If we already have this object
pop pop % Remove object and object number
} { % Else if we do not have this object
PDFDEBUG { (%Resolving compressed object: [) print dup =only ( 0]) = } if
Objects exch 3 index % Put the object into Objects
3 index get
PDFDEBUG { dup === flush } if
put
} ifelse
} {
pop % Ignore old object; remove object number.
} ifelse
pop % Remove loop index
} for
pop pop pop pop pop % Remove strm# objstream, N, (obj#], and [objects]
} bind executeonly def
currentdict /no_debug_dict undef
% When resolving an object reference, we stop at the endobj or endstream.
/resolveopdict mark
valueopdict { } forall
/xref { % Bug 697761
( **** Error: Encountered 'xref' while expecting 'endobj'.\n) pdfformaterror
( Treating this as a missing 'endobj', output may be incorrect.\n) pdfformaterror
endobj exit
} bind executeonly
/endstream { endobj exit } bind
/endobj { endobj exit } bind
/endjobj { % Bug 689876.
( **** Error: Operator 'endobj' is misspelled as 'endjobj'.\n) pdfformaterror
( Output may be incorrect.\n) pdfformaterror
endobj exit
} bind executeonly
/enbobj { % Bug 690397.
( **** Error: Operator 'endobj' is misspelled as 'enbobj'.\n) pdfformaterror
( Output may be incorrect.\n) pdfformaterror
endobj exit
} bind executeonly
/obj {
% OmniForm generates PDF file with endobj missing in some
% objects. AR ignores this. So we have to do it too.
( **** Error: Encountered 'obj' while expecting 'endobj'.\n) pdfformaterror
( Treating this as a missing 'endobj', output may be incorrect.\n) pdfformaterror
pop pop endobj exit
} bind executeonly
.dicttomark readonly def
/resolveR { % <object#> <generation#> resolveR <object>
1 index 3 1 roll % copy the object for the code after pdf_run_resolve (may store it)
PDFDEBUG {
PDFSTEPcount 1 le {
(%Resolving: ) print 2 copy 2 array astore //== exec
} if
} if
1 index dup 0 le exch NumObjects ge or {
( **** Error: Considering object with an invalid number )
2 index 20 string cvs concatstrings
( as null.\n) concatstrings pdfformaterror
( Output may be incorrect.\n) pdfformaterror
pop pop //null
} {
1 index resolved? { % If object has already been resolved ...
exch pop exch pop % then clear stack and return object
} { % Else if not resolved ...
PDFfile fileposition 3 1 roll % Save current file position
1 index Objects exch get % Get location of object from xref
3 1 roll checkgeneration { % Verify the generation number
% Stack: savepos objpos obj#
ObjectStream 1 index get dup 0 eq { % Check if obj in not an objstream
pop exch PDFoffset add PDFfile exch setfileposition
PDFfile token pop 2 copy ne
{ ( **** Error: Unrecoverable error in xref!\n) pdfformaterror
( Output may be incorrect.\n) pdfformaterror
/resolveR cvx /rangecheck signalerror
}
if pop PDFfile token pop
PDFfile token pop /obj ne
{ ( **** Error: Unrecoverable error in xref!\n) pdfformaterror
( Output may be incorrect.\n) pdfformaterror
/resolveR cvx /rangecheck signalerror
}
if
pdf_run_resolve % PDFfile resolveopdict .pdfrun
} { % Else the object is in an ObjectStream
% Process an objectstream object. We are going to resolve all
% of the objects in sthe stream and place them into the Objects
% array.
% Stack: savepos objpos obj# objectstream#
resolveobjectstream
resolved? { % If object has already been resolved ...
exch pop % Remove object pos from stack.
} {
pop pop //null % Pop objpos and obj#, put null for object
} ifelse
} ifelse
} { % Else the generation number is wrong
% Don't cache if the generation # is wrong.
pop pop //null % Pop objpos and obj#, put null for object
} ifelse % ifelse generation number is correct
exch PDFfile exch setfileposition % Return to original file position
} ifelse
} ifelse
dup type /dicttype eq % For dictionaries only
{
dup /Type known % make sure the dictionary has a type
{
%% We must not add to *all* dictionaries, because some of them are handled by
%% executing all the key/value pairs, and adding a integer causes problems
%% Since we only use this number for font identification, or checking self-references,
%% we only need it for dictionaries with a Type, and these are always handled more carefully.
dup 3 -1 roll /.gs.pdfobj# exch put
}
{exch pop}
ifelse
}
{exch pop}
ifelse
} bind executeonly def
% ================================ Streams ================================ %
% We represent a stream by an executable dictionary that contains,
% in addition to the contents of the original stream dictionary:
% /File - the file or string where the stream contents are stored,
% if the stream is not an external one.
% /FilePosition - iff File is a file, the position in the file
% where the contents start.
% /StreamKey - the key used to decrypt this stream, if any.
% We do the real work of constructing the data stream only when the
% contents are needed.
% Construct a stream. The length is not reliable in the face of
% different end-of-line conventions, but it's all we've got.
%
% PDF files are inconsistent about what may fall between the 'stream' keyword
% and the actual stream data, and it appears that no one algorithm can
% detect this reliably. We used to try to guess whether the file included
% extraneous \r and/or \n characters, but we no longer attempt to do so,
% especially since the PDF 1.2 specification states flatly that the only
% legal terminators following the 'stream' keyword are \n or \r\n, both of
% which are properly skipped and discarded by the token operator.
% Unfortunately, this doesn't account for other whitespace characters that
% may have preceded the EOL, such as spaces or tabs. Thus we back up one
% character and scan until we find the \n terminator.
/stream { % <dict> stream <modified_dict>
dup /F known dup PDFsource PDFfile eq or {
not {
dup /File PDFfile put
% make sure that we are just past the EOL \n character
PDFfile dup fileposition 1 sub setfileposition % back up one
% Skip spaces till \n or a non-space character is found.
//false
{
PDFfile read pop
dup 32 eq { pop //true or } { exit } ifelse
} loop
dup 13 eq {
pop
PDFfile read pop
} if
10 ne {
PDFfile dup fileposition 1 sub setfileposition
//true or
} if {
( **** Error: stream operator isn't terminated by valid EOL.\n) pdfformaterror
( Output may be incorrect.\n) pdfformaterror
} if
dup /FilePosition PDFfile fileposition put
PDFDEBUG {
PDFSTEPcount 1 le {
(%FilePosition: ) print dup /FilePosition get //== exec
} if
} if
} if
% Some (bad) PDf files have invalid stream lengths. This causes problems
% if we reposition beyond the end of the file. So we compare the given
% length to number of bytes left in the file.
dup /Length knownoget {
dup PDFfile bytesavailable lt { % compare to to bytes left in file
PDFfile fileposition % reposition to the end of stream
add PDFfile exch setfileposition
} {
pop % bad stream length - do not reposition.
% This will force a length warning below
} ifelse
} if
} {
pop
% We're already reading from a stream, which we can't reposition.
% Capture the sub-stream contents in a string.
dup /Length oget string PDFsource exch readstring
not {
( **** Error: Unexpected EOF in stream!\n) pdfformaterror
( Output may be incorrect.\n) pdfformaterror
/stream cvx /rangecheck signalerror
} if
1 index exch /File exch put
} ifelse
PDFsource token_nofail not { //null } if
dup /endobj eq {
% Another case that Acrobat Reader handles -- 'endobj' without 'endstream'.
( **** Error: stream missing 'endstream'.\n) pdfformaterror
( Output may be incorrect.\n) pdfformaterror
pop /endstream % fake a valid endstream
} if
/endstream ne {
( **** Error: stream Length incorrect.\n) pdfformaterror
( Output may be incorrect.\n) pdfformaterror
dup /Length undef % prevent the use of the incorrect length.
%% Bug #696560 If we get a stream whose length is incorrect we cannot
%% safely *assume* it has an object definition as we can get here when
%% reading an XRef stream, and in that case we will not have the object
%% and generation numbers on the stack!
%% So before we call endobj and exit pdfrun, try and see if ths looks like
%% a regular object or not. If it isn't then don't try and ahndle it as
%% one, just return the object.
count 3 ge {
1 index type /integertype eq 2 index type /integertype eq and {
cvx endobj exit % exit from .pdfrun now.
} if
} if
} {
PDFsource (??) .peekstring pop (>>) eq { % Bug 690161, sample #1
( **** Error: Spurious '>>' after 'endstream' ignored.\n) pdfformaterror
( Output may be incorrect.\n) pdfformaterror
PDFsource (12) readstring pop pop
} if
} ifelse
cvx
} bind executeonly def
/endstream {
exit
} bind executeonly def
% Contrary to the published PDF (1.3) specification, Acrobat Reader
% accepts abbreviated filter names everywhere, not just for in-line images,
% and some applications (notably htmldoc) rely on this.
/unabbrevfilterdict mark
/AHx /ASCIIHexDecode /A85 /ASCII85Decode /CCF /CCITTFaxDecode
/DCT /DCTDecode /Fl /FlateDecode /LZW /LZWDecode /RL /RunLengthDecode
.dicttomark readonly def
% Extract and apply filters.
/filterparms { % <dict> <DPkey> <Fkey> filterparms
% <dict> <parms> <filternames>
2 index exch knownoget {
oforce_recursive
exch 2 index exch knownoget {
dup length 0 eq {
pop //false
} {
//true
} ifelse
} {
//false
} ifelse {
% Both filters and parameters.
oforce_recursive
exch dup type /nametype eq {
1 array astore exch
dup type /arraytype ne { 1 array astore } if exch
} if
} {
% Filters, but no parameters.
//null exch
dup type /nametype eq { 1 array astore } if
} ifelse
} {
% No filters: ignore parameters, if any.
pop //null { }
} ifelse
} bind executeonly def
/filtername { % <filtername> filtername <filtername'>
//unabbrevfilterdict 1 index .knownget { exch pop } if
dup /Filter resourcestatus { pop pop } {
Repaired exch % this error is not the creator's fault
RepiredAnError exch
( **** ERROR: Unable to process ) pdfformaterror
64 string cvs pdfformaterror
( data. Page will be missing data.\n) pdfformaterror
/RepairedAnError exch store
/Repaired exch store % restore the previous "Repaired" state
% provide a filter that returns EOF (no data)
/.EOFDecode
} ifelse
} bind executeonly def
/pdf_rules_dict << /PDFRules //true >> readonly def
% Add PDF option to ASCII85Decode filter
% <source> <name> add_A85_param <source> <dict'> <name>
% <source> <dict> <name> add_A85_param <source> <dict'> <name>
/add_A85_param {
dup /ASCII85Decode eq {
1 index type /dicttype eq {
exch dup length 1 add dict copy
dup /PDFRules //true put
exch
} {
//pdf_rules_dict exch
} ifelse
} if
} bind executeonly def
currentdict /pdf_rules_dict undef
/applyfilters { % <parms> <source> <filternames> applyfilters <stream>
2 index //null eq
{
{ filtername add_A85_param filter }
}
{
dup length 3 index length ne %% compare lengths of DecodeParmas and Filter arrays
{ %% if they aren't the same, ignore the decode params
( **** Error: ignoring stream /DecodeParams array as its length is different to the Filters array.\n) pdfformaterror
( Output may be incorrect.\n) pdfformaterror
3 -1 roll pop //null 3 1 roll
{ filtername add_A85_param filter }
}
{
{ % Stack: parms source filtername
2 index 0 oget dup type /dicttype ne
{ pop }
{
exch filtername dup /JBIG2Decode eq { exch jbig2cachectx exch } if
} ifelse
add_A85_param filter
exch dup length 1 sub 1 exch getinterval exch
}
} ifelse
} ifelse
forall exch pop
} bind executeonly def
% JBIG2 streams have an optional 'globals' stream obj for
% sharing redundant data between page images. Here we resolve
% that stream reference (if any) and run it through the decoder,
% creating a special -jbig2globalctx- postscript object our
% JBIG2Decode filter implementation looks for in the parm dict.
/jbig2cachectx { % <parmdict> jbig2cachectx <parmdict>
dup /JBIG2Globals knownoget {
% make global ctx
PDFfile fileposition exch % resolvestream is not reentrant
//true resolvestream % stack after: PDFfileposition -file-
% Read the data in a loop until EOF to so we can move the strings into a bytestring
[ { counttomark 1 add index 60000 string readstring not { exit } if } loop ]
exch pop 0 1 index { length add } forall % compute the total length
% now copy the data from the array of strings into a bytestring
.bytestring exch 0 exch { 3 copy putinterval length add } forall pop
.jbig2makeglobalctx
PDFfile 3 -1 roll setfileposition
1 index exch
/.jbig2globalctx exch put
} if
} bind executeonly def
% Resolve a stream dictionary to a PostScript stream.
% Streams with no filters require special handling:
% - Whether we are going to interpret the stream, or If we are just
% going to read data from them, we impose a SubFileDecode filter
% that reads just the requisite amount of data.
% Note that, in general, resolving a stream repositions PDFfile.
% Clients must save and restore the position of PDFfile themselves.
/resolvestream { % <streamdict> <readdata?> resolvestream <stream>
1 index /F knownoget {
% This stream is stored on an external file.
(r) file 3 -1 roll
/FDecodeParms /FFilter filterparms
% Stack: readdata? file dict parms filternames
4 -1 roll exch
pdf_decrypt_stream
applyfilters
} {
exch
dup /Length knownoget { 0 eq } { //false } ifelse {
() 0 () /SubFileDecode filter
} {
dup /FilePosition .knownget {
1 index /File get exch setfileposition
} if
% Stack: readdata? dict
/DecodeParms /Filter filterparms
% Stack: readdata? dict parms filternames
2 index /File .knownget not {
( **** Error: Stream object is missing the stream data.\n)
( Output may be incorrect.\n) pdfformaterror
pdfformaterror
()
} if exch
% Stack: readdata? dict parms file/string filternames
dup length 0 eq {
% All the PDF filters have EOD markers, but in this case
% there is no specified filter.
exch dup type /filetype eq 5 index or {
% Use length for any files or reading data from any source.
3 index /Length knownoget not { 0 } if
} {
0 % Otherwise length of 0 for whole string
} ifelse
4 index /IDFlag known { pop } { () /SubFileDecode filter } ifelse
exch
pdf_decrypt_stream % add decryption if needed
pop exch pop
} {
% Stack: readdata? dict parms source filternames
exch 3 index /Length knownoget {
() /SubFileDecode filter
} if exch
pdf_decrypt_stream % add decryption if needed
applyfilters
} ifelse
} ifelse
} ifelse
% Stack: readdata? dict file
exch pop exch pop
} bind executeonly def
% ============================ Name/number trees ============================ %
/nameoget { % <nametree> <key> nameoget <obj|null>
exch /Names exch .treeget
} bind executeonly def
/numoget { % <numtree> <key> numoget <obj|null>
exch /Nums exch .treeget
} bind executeonly def
/.treeget { % <key> <leafkey> <tree> .treeget <obj|null>
dup /Kids knownoget {
exch pop .branchget
} {
exch oget .leafget
} ifelse
} bind executeonly def
/.branchget { % <key> <leafkey> <kids> .branchget <obj|null>
dup length 0 eq {
pop pop pop //null
} {
dup length -1 bitshift 2 copy oget
% Stack: key leafkey kids mid kids[mid]
dup /Limits oget aload pop
% Stack: key leafkey kids mid kids[mid] min max
6 index lt {
pop pop
1 add 1 index length 1 index sub getinterval .branchget
} {
5 index gt {
pop
0 exch getinterval .branchget
} {
exch pop exch pop .treeget
} ifelse
} ifelse
} ifelse
} bind executeonly def
/.leafget { % <key> <pairs> .leafget <obj|null>
dup length 2 eq {
dup 0 get 2 index eq { 1 oget } { pop null } ifelse
exch pop
} {
dup length -1 bitshift -2 and 2 copy oget
% Stack: key pairs mid pairs[mid]
3 index gt { 0 exch } { 1 index length 1 index sub } ifelse
getinterval .leafget
} ifelse
} bind executeonly def
% The following variants return tree entry whose key is closest but
% less or equal to the given key.
/numogetle { % <numtree> <key> numogetle <key obj true|false>
exch /Nums exch .treegetle
} bind executeonly def
/.treegetle { % <key> <leafkey> <tree> .treegetle <key obj true|false>
dup /Kids knownoget {
exch pop .branchgetle
} {
exch oget .leafgetle
} ifelse
} bind executeonly def
/.branchgetle { % <key> <leafkey> <kids> .branchgetle <key obj true|false>
dup length 0 eq {
pop pop pop false
} {
dup length -1 bitshift 2 copy oget
dup /Limits oget aload pop
% Stack: key leafkeyb kids mid kids[mid] min max
6 index ge {
5 index le {
exch pop exch pop .treegetle
} {
pop 0 exch getinterval .branchgetle
} ifelse
} {
pop 2 index length 2 index sub 1 gt {
pop 1 index length 1 index sub getinterval .branchgetle
} {
exch pop exch pop .treegetle
} ifelse
} ifelse
} ifelse
} bind executeonly def
/.leafgetle { % <key> <pairs> .leafget <obj|null>
dup length 2 eq {
dup 0 get
2 index le { exch pop aload pop true } { pop pop false } ifelse
} {
dup length -1 bitshift -2 and 2 copy oget
% Stack: key pairs mid pairs[mid]
3 index gt { 0 exch } { 1 index length 1 index sub } ifelse
getinterval .leafgetle
} ifelse
} bind executeonly def
end % pdfdict
.setglobal
Mini Shell