pdf2json 0.1.0

data/pdf2json-0.52-source/xpdf/pdftotext.cc

@@ -0,0 +1,333 @@
+//========================================================================
+//
+// pdftotext.cc
+//
+// Copyright 1997-2003 Glyph & Cog, LLC
+//
+//========================================================================
+
+#include <aconf.h>
+#include <stdio.h>
+#include <stdlib.h>
+#include <stddef.h>
+#include <string.h>
+#include "parseargs.h"
+#include "GString.h"
+#include "gmem.h"
+#include "GlobalParams.h"
+#include "Object.h"
+#include "Stream.h"
+#include "Array.h"
+#include "Dict.h"
+#include "XRef.h"
+#include "Catalog.h"
+#include "Page.h"
+#include "PDFDoc.h"
+#include "TextOutputDev.h"
+#include "CharTypes.h"
+#include "UnicodeMap.h"
+#include "Error.h"
+#include "config.h"
+
+static void printInfoString(FILE *f, Dict *infoDict, char *key,
+			    char *text1, char *text2, UnicodeMap *uMap);
+static void printInfoDate(FILE *f, Dict *infoDict, char *key, char *fmt);
+
+static int firstPage = 1;
+static int lastPage = 0;
+static GBool physLayout = gFalse;
+static GBool rawOrder = gFalse;
+static GBool htmlMeta = gFalse;
+static char textEncName[128] = "";
+static char textEOL[16] = "";
+static GBool noPageBreaks = gFalse;
+static char ownerPassword[33] = "\001";
+static char userPassword[33] = "\001";
+static GBool quiet = gFalse;
+static char cfgFileName[256] = "";
+static GBool printVersion = gFalse;
+static GBool printHelp = gFalse;
+
+static ArgDesc argDesc[] = {
+  {"-f",       argInt,      &firstPage,     0,
+   "first page to convert"},
+  {"-l",       argInt,      &lastPage,      0,
+   "last page to convert"},
+  {"-layout",  argFlag,     &physLayout,    0,
+   "maintain original physical layout"},
+  {"-raw",     argFlag,     &rawOrder,      0,
+   "keep strings in content stream order"},
+  {"-htmlmeta", argFlag,   &htmlMeta,       0,
+   "generate a simple HTML file, including the meta information"},
+  {"-enc",     argString,   textEncName,    sizeof(textEncName),
+   "output text encoding name"},
+  {"-eol",     argString,   textEOL,        sizeof(textEOL),
+   "output end-of-line convention (unix, dos, or mac)"},
+  {"-nopgbrk", argFlag,     &noPageBreaks,  0,
+   "don't insert page breaks between pages"},
+  {"-opw",     argString,   ownerPassword,  sizeof(ownerPassword),
+   "owner password (for encrypted files)"},
+  {"-upw",     argString,   userPassword,   sizeof(userPassword),
+   "user password (for encrypted files)"},
+  {"-q",       argFlag,     &quiet,         0,
+   "don't print any messages or errors"},
+  {"-cfg",     argString,   cfgFileName,    sizeof(cfgFileName),
+   "configuration file to use in place of .xpdfrc"},
+  {"-v",       argFlag,     &printVersion,  0,
+   "print copyright and version info"},
+  {"-h",       argFlag,     &printHelp,     0,
+   "print usage information"},
+  {"-help",    argFlag,     &printHelp,     0,
+   "print usage information"},
+  {"--help",   argFlag,     &printHelp,     0,
+   "print usage information"},
+  {"-?",       argFlag,     &printHelp,     0,
+   "print usage information"},
+  {NULL}
+};
+
+int main(int argc, char *argv[]) {
+  PDFDoc *doc;
+  GString *fileName;
+  GString *textFileName;
+  GString *ownerPW, *userPW;
+  TextOutputDev *textOut;
+  FILE *f;
+  UnicodeMap *uMap;
+  Object info;
+  GBool ok;
+  char *p;
+  int exitCode;
+
+  exitCode = 99;
+
+  // parse args
+  ok = parseArgs(argDesc, &argc, argv);
+  if (!ok || argc < 2 || argc > 3 || printVersion || printHelp) {
+    fprintf(stderr, "pdftotext version %s\n", xpdfVersion);
+    fprintf(stderr, "%s\n", xpdfCopyright);
+    if (!printVersion) {
+      printUsage("pdftotext", "<PDF-file> [<text-file>]", argDesc);
+    }
+    goto err0;
+  }
+  fileName = new GString(argv[1]);
+
+  // read config file
+  globalParams = new GlobalParams(cfgFileName);
+  if (textEncName[0]) {
+    globalParams->setTextEncoding(textEncName);
+  }
+  if (textEOL[0]) {
+    if (!globalParams->setTextEOL(textEOL)) {
+      fprintf(stderr, "Bad '-eol' value on command line\n");
+    }
+  }
+  if (noPageBreaks) {
+    globalParams->setTextPageBreaks(gFalse);
+  }
+  if (quiet) {
+    globalParams->setErrQuiet(quiet);
+  }
+
+  // get mapping to output encoding
+  if (!(uMap = globalParams->getTextEncoding())) {
+    error(-1, "Couldn't get text encoding");
+    delete fileName;
+    goto err1;
+  }
+
+  // open PDF file
+  if (ownerPassword[0] != '\001') {
+    ownerPW = new GString(ownerPassword);
+  } else {
+    ownerPW = NULL;
+  }
+  if (userPassword[0] != '\001') {
+    userPW = new GString(userPassword);
+  } else {
+    userPW = NULL;
+  }
+  doc = new PDFDoc(fileName, ownerPW, userPW);
+  if (userPW) {
+    delete userPW;
+  }
+  if (ownerPW) {
+    delete ownerPW;
+  }
+  if (!doc->isOk()) {
+    exitCode = 1;
+    goto err2;
+  }
+
+  // check for copy permission
+  if (!doc->okToCopy()) {
+    error(-1, "Copying of text from this document is not allowed.");
+    exitCode = 3;
+    goto err2;
+  }
+
+  // construct text file name
+  if (argc == 3) {
+    textFileName = new GString(argv[2]);
+  } else {
+    p = fileName->getCString() + fileName->getLength() - 4;
+    if (!strcmp(p, ".pdf") || !strcmp(p, ".PDF")) {
+      textFileName = new GString(fileName->getCString(),
+				 fileName->getLength() - 4);
+    } else {
+      textFileName = fileName->copy();
+    }
+    textFileName->append(htmlMeta ? ".html" : ".txt");
+  }
+
+  // get page range
+  if (firstPage < 1) {
+    firstPage = 1;
+  }
+  if (lastPage < 1 || lastPage > doc->getNumPages()) {
+    lastPage = doc->getNumPages();
+  }
+
+  // write HTML header
+  if (htmlMeta) {
+    if (!textFileName->cmp("-")) {
+      f = stdout;
+    } else {
+      if (!(f = fopen(textFileName->getCString(), "wb"))) {
+	error(-1, "Couldn't open text file '%s'", textFileName->getCString());
+	exitCode = 2;
+	goto err3;
+      }
+    }
+    fputs("<html>\n", f);
+    fputs("<head>\n", f);
+    doc->getDocInfo(&info);
+    if (info.isDict()) {
+      printInfoString(f, info.getDict(), "Title", "<title>", "</title>\n",
+		      uMap);
+      printInfoString(f, info.getDict(), "Subject",
+		      "<meta name=\"Subject\" content=\"", "\">\n", uMap);
+      printInfoString(f, info.getDict(), "Keywords",
+		      "<meta name=\"Keywords\" content=\"", "\">\n", uMap);
+      printInfoString(f, info.getDict(), "Author",
+		      "<meta name=\"Author\" content=\"", "\">\n", uMap);
+      printInfoString(f, info.getDict(), "Creator",
+		      "<meta name=\"Creator\" content=\"", "\">\n", uMap);
+      printInfoString(f, info.getDict(), "Producer",
+		      "<meta name=\"Producer\" content=\"", "\">\n", uMap);
+      printInfoDate(f, info.getDict(), "CreationDate",
+		    "<meta name=\"CreationDate\" content=\"%s\">\n");
+      printInfoDate(f, info.getDict(), "LastModifiedDate",
+		    "<meta name=\"ModDate\" content=\"%s\">\n");
+    }
+    info.free();
+    fputs("</head>\n", f);
+    fputs("<body>\n", f);
+    fputs("<pre>\n", f);
+    if (f != stdout) {
+      fclose(f);
+    }
+  }
+
+  // write text file
+  textOut = new TextOutputDev(textFileName->getCString(),
+			      physLayout, rawOrder, htmlMeta);
+  if (textOut->isOk()) {
+    doc->displayPages(textOut, firstPage, lastPage, 72, 72, 0,
+		      gFalse, gTrue, gFalse);
+  } else {
+    delete textOut;
+    exitCode = 2;
+    goto err3;
+  }
+  delete textOut;
+
+  // write end of HTML file
+  if (htmlMeta) {
+    if (!textFileName->cmp("-")) {
+      f = stdout;
+    } else {
+      if (!(f = fopen(textFileName->getCString(), "ab"))) {
+	error(-1, "Couldn't open text file '%s'", textFileName->getCString());
+	exitCode = 2;
+	goto err3;
+      }
+    }
+    fputs("</pre>\n", f);
+    fputs("</body>\n", f);
+    fputs("</html>\n", f);
+    if (f != stdout) {
+      fclose(f);
+    }
+  }
+
+  exitCode = 0;
+
+  // clean up
+ err3:
+  delete textFileName;
+ err2:
+  delete doc;
+  uMap->decRefCnt();
+ err1:
+  delete globalParams;
+ err0:
+
+  // check for memory leaks
+  Object::memCheck(stderr);
+  gMemReport(stderr);
+
+  return exitCode;
+}
+
+static void printInfoString(FILE *f, Dict *infoDict, char *key,
+			    char *text1, char *text2, UnicodeMap *uMap) {
+  Object obj;
+  GString *s1;
+  GBool isUnicode;
+  Unicode u;
+  char buf[8];
+  int i, n;
+
+  if (infoDict->lookup(key, &obj)->isString()) {
+    fputs(text1, f);
+    s1 = obj.getString();
+    if ((s1->getChar(0) & 0xff) == 0xfe &&
+	(s1->getChar(1) & 0xff) == 0xff) {
+      isUnicode = gTrue;
+      i = 2;
+    } else {
+      isUnicode = gFalse;
+      i = 0;
+    }
+    while (i < obj.getString()->getLength()) {
+      if (isUnicode) {
+	u = ((s1->getChar(i) & 0xff) << 8) |
+	    (s1->getChar(i+1) & 0xff);
+	i += 2;
+      } else {
+	u = s1->getChar(i) & 0xff;
+	++i;
+      }
+      n = uMap->mapUnicode(u, buf, sizeof(buf));
+      fwrite(buf, 1, n, f);
+    }
+    fputs(text2, f);
+  }
+  obj.free();
+}
+
+static void printInfoDate(FILE *f, Dict *infoDict, char *key, char *fmt) {
+  Object obj;
+  char *s;
+
+  if (infoDict->lookup(key, &obj)->isString()) {
+    s = obj.getString()->getCString();
+    if (s[0] == 'D' && s[1] == ':') {
+      s += 2;
+    }
+    fprintf(f, fmt, s);
+  }
+  obj.free();
+}

data/pdf2json-0.52-source/xpdf/print.xbm

@@ -0,0 +1,6 @@
+#define print_width 15
+#define print_height 15
+static unsigned char print_bits[] = {
+   0xf0, 0x7f, 0x10, 0x40, 0x10, 0x40, 0xc8, 0x23, 0x08, 0x20, 0x68, 0x23,
+   0x04, 0x10, 0x34, 0x10, 0x04, 0x10, 0xff, 0x7f, 0x55, 0x55, 0xab, 0x6a,
+   0x55, 0x55, 0xab, 0x6a, 0xfe, 0x3f};

data/pdf2json-0.52-source/xpdf/printDis.xbm

@@ -0,0 +1,6 @@
+#define printDis_width 15
+#define printDis_height 15
+static unsigned char printDis_bits[] = {
+   0xa0, 0x2a, 0x10, 0x40, 0x00, 0x00, 0x40, 0x01, 0x08, 0x20, 0x40, 0x01,
+   0x00, 0x00, 0x14, 0x10, 0x00, 0x00, 0x55, 0x55, 0x00, 0x00, 0x01, 0x40,
+   0x00, 0x00, 0x01, 0x40, 0xaa, 0x2a};

data/pdf2json-0.52-source/xpdf/rightArrow.xbm

@@ -0,0 +1,5 @@
+#define rightArrow_width 8
+#define rightArrow_height 15
+static unsigned char rightArrow_bits[] = {
+   0x01, 0x03, 0x07, 0x0f, 0x1f, 0x3f, 0x7f, 0xff, 0x7f, 0x3f, 0x1f, 0x0f,
+   0x07, 0x03, 0x01};

data/pdf2json-0.52-source/xpdf/rightArrowDis.xbm

@@ -0,0 +1,5 @@
+#define rightArrowDis_width 8
+#define rightArrowDis_height 15
+static unsigned char rightArrowDis_bits[] = {
+   0x01, 0x02, 0x05, 0x0a, 0x15, 0x2a, 0x55, 0xaa, 0x55, 0x2a, 0x15, 0x0a,
+   0x05, 0x02, 0x01};

data/pdf2json-0.52-source/xpdf/vms_make.com

@@ -0,0 +1,129 @@
+$!========================================================================
+$!
+$! Xpdf compile script for VMS.
+$!
+$! Written by Patrick Moreau, Martin P.J. Zinser.
+$!
+$! Copyright 1996-2003 Glyph & Cog, LLC
+$!
+$!========================================================================
+$!
+$ i = 0
+$ j = 0
+$ APPS = "XPDF,PDFTOPS,PDFTOTEXT,PDFINFO,PDFTOPBM,PDFIMAGES,PDFFONTS"
+$ if f$search("COMMON.OLB").eqs."" then lib/create common.olb
+$!
+$ COMMON_OBJS = "Annot.obj,Array.obj,BuiltinFont.obj," + - 
+                "BuiltinFontTables.obj,Catalog.obj,CharCodeToUnicode.obj," + - 
+                "CMap.obj,Decrypt.obj,Dict.obj,Error.obj," + -
+                "FontEncodingTables.obj,FontFile.obj," + -
+                "Function.obj,Gfx.obj,GfxFont.obj,GfxState.obj,"+ - 
+                "GlobalParams.obj,JArithmeticDecoder.obj,JBIG2Stream.obj,"+ -
+                "Lexer.obj,Link.obj,NameToCharCode.obj,Object.obj,"+ -
+                "Outline.obj,OutputDev.obj,Page.obj,Parser.obj,PDFdoc.obj," + -
+                "PDFDocEncoding.obj,PSTokenizer.obj,Stream.obj," + -
+                "UnicodeMap.obj,UnicodeTypeTable.obj,XRef.obj"
+$ COMMON_LIBS = "[]common.olb/lib,[-.goo]libgoo.olb/lib"
+$!
+$ XPDF_OBJS = "xpdf.obj,FTFont.obj,PSOutputDev.obj," + -
+              "SFont.obj,T1Font.obj,TextOutputDev.obj,TTFont.obj," + -
+              "XOutputDev.obj,XPDFApp.obj,XPDFCore.obj,XPDFTree.obj," + -
+              "XPDFViewer.obj,XPixmapOutputDev.obj"
+$ XPDF_LIBS = ""
+$!
+$ PDFTOPS_OBJS   = "pdftops.obj,PSOutputDev.obj" 
+$ PDFTOPS_LIBS   = ""
+$!
+$ PDFTOTEXT_OBJS = "pdftotext.obj,TextOutputDev.obj"
+$ PDFTOTEXT_LIBS = ""
+$!
+$ PDFINFO_OBJS   =  "pdfinfo.obj"
+$ PDFINFO_LIBS   = ""
+$!
+$ PDFTOPBM_OBJS  = "pdftopbm.obj,FTFont.obj,PBMOutputDev.obj,SFont.obj," + -
+                   "T1Font.obj,TextOutputDev.obj,TTFont.obj,XOutputDev.obj"
+$ PDFTOPBM_LIBS  = ""
+$!
+$ PDFIMAGES_OBJS = "pdfimages.obj,ImageOutputDev.obj"
+$ PDFIMAGES_LIBS = ""
+$!
+$ PDFFONTS_OBJS  = "pdffonts.obj"
+$ PDFFONTS_LIBS  = ""
+$!
+$COMPILE_CXX_LOOP:
+$ file = f$element(i, ",",COMMON_OBJS)
+$ if file .eqs. "," then goto BUILD_APPS
+$ i = i + 1
+$ name = f$parse(file,,,"NAME")
+$ call make 'file "CXXCOMP ''name'.cc" - 
+       'name'.cc
+$ call make common.olb "lib/replace common.olb ''name'.obj" - 
+       'name'.obj
+$ goto COMPILE_CXX_LOOP
+$!
+$BUILD_APPS:
+$ curr_app = f$element(j,",",APPS)
+$ if curr_app .eqs. "," then exit
+$ j = j + 1
+$ i = 0
+$COMPILE_APP:
+$ file = f$element(i,",",'curr_app'_OBJS)
+$ if file .eqs. "," then goto LINK_APP
+$ i = i + 1
+$ name = f$parse(file,,,"NAME")
+$ call make 'file "CXXCOMP ''name'.cc" - 
+       'name'.cc
+$ goto COMPILE_APP
+$LINK_APP:
+$ if 'curr_app'_LIBS .nes. "" 
+$ then 
+$   LIBS = 'curr_app'_LIBS + "," + COMMON_LIBS
+$ else
+$   LIBS = COMMON_LIBS 
+$ endif
+$ OBJS = 'curr_app'_OBJS
+$ write sys$output "Linking ''curr_app'..."
+$ xpdf_link/exe='curr_app'.exe 'OBJS','libs',[-]xpdf.opt/opt
+$!  
+$ goto BUILD_APPS
+$ exit
+$!
+$MAKE: SUBROUTINE   !SUBROUTINE TO CHECK DEPENDENCIES
+$ V = 'F$Verify(0)
+$! P1 = What we are trying to make
+$! P2 = Command to make it
+$! P3 - P8  What it depends on
+$
+$ If F$Search(P1) .Eqs. "" Then Goto Makeit
+$ Time = F$CvTime(F$File(P1,"RDT"))
+$arg=3
+$Loop:
+$       Argument = P'arg
+$       If Argument .Eqs. "" Then Goto Exit
+$       El=0
+$Loop2:
+$       File = F$Element(El," ",Argument)
+$       If File .Eqs. " " Then Goto Endl
+$       AFile = ""
+$Loop3:
+$       OFile = AFile
+$       AFile = F$Search(File)
+$       If AFile .Eqs. "" .Or. AFile .Eqs. OFile Then Goto NextEl
+$       If F$CvTime(F$File(AFile,"RDT")) .Ges. Time Then Goto Makeit
+$       Goto Loop3
+$NextEL:
+$       El = El + 1
+$       Goto Loop2
+$EndL:
+$ arg=arg+1
+$ If arg .Le. 8 Then Goto Loop
+$ Goto Exit
+$
+$Makeit:
+$ VV=F$VERIFY(0)
+$ write sys$output P2
+$ 'P2
+$ VV='F$Verify(VV)
+$Exit:
+$ If V Then Set Verify
+$ENDSUBROUTINE