/**
* Extract text from a document for which a TET handle is already available.
*
* @param tet
* The TET object
* @param doc
* A valid TET document handle
* @param outfp
* Output file handle
*
* @throws TETException
* @throws IOException
*/
static void extract_text(TET tet, int doc, BinaryWriter outfp)
{
UnicodeEncoding unicode = new UnicodeEncoding(false, true);
/*
* Get number of pages in the document.
*/
int n_pages = (int)tet.pcos_get_number(doc, "length:pages");
/* loop over pages */
for (int pageno = 1; pageno <= n_pages; ++pageno)
{
String text;
int page;
page = tet.open_page(doc, pageno, pageoptlist);
if (page == -1)
{
Console.WriteLine("Error " + tet.get_errnum() + " in "
+ tet.get_apiname() + "() on page " + pageno + ": "
+ tet.get_errmsg());
continue; /* try next page */
}
/*
* Retrieve all text fragments; This loop is actually not required
* for granularity=page, but must be used for other granularities.
*/
while ((text = tet.get_text(page)) != null)
{
outfp.Write(unicode.GetBytes(text)); // print the retrieved text
/* print a separator between chunks of text */
outfp.Write(unicode.GetBytes(separator));
}
if (tet.get_errnum() != 0)
{
Console.WriteLine("Error " + tet.get_errnum() + " in "
+ tet.get_apiname() + "() on page " + pageno + ": "
+ tet.get_errmsg());
}
tet.close_page(page);
}
}
static int Main(string[] args)
{
/* global option list */
string globaloptlist = "searchpath={{../data} {../../data}}";
/* document-specific option list */
string docoptlist = "";
/* page-specific option list e.g
* "imageanalysis={merge={gap=1} smallimages={maxwidth=20}}"
*/
string pageoptlist = "";
TET tet;
int pageno = 0;
string outfilebase;
if (args.Length != 1)
{
Console.WriteLine("usage: image_resources <filename>");
return(2);
}
outfilebase = args.GetValue(0).ToString();
if ((outfilebase.Length > 4) && (outfilebase.Substring(outfilebase.Length - 4).Equals(".pdf")) || (outfilebase.Substring(outfilebase.Length - 4).Equals(".PDF")))
{
outfilebase = outfilebase.Substring(0, outfilebase.Length - 4);
}
tet = new TET();
try
{
int n_pages;
tet.set_option(globaloptlist);
int doc = tet.open_document(args.GetValue(0).ToString(), docoptlist);
if (doc == -1)
{
Console.WriteLine("Error {0} in {1}(): {2}",
tet.get_errnum(), tet.get_apiname(), tet.get_errmsg());
return(2);
}
/* Images will only be merged upon opening a page.
* In order to enumerate all merged image resources
* we open all pages before extracting the images.
*/
/* get number of pages in the document */
n_pages = (int)tet.pcos_get_number(doc, "length:pages");
/* Loop over all pages to trigger image merging */
for (pageno = 1; pageno <= n_pages; ++pageno)
{
string text;
int page;
page = tet.open_page(doc, pageno, pageoptlist);
if (page == -1)
{
Console.WriteLine("Error {0} in {1}() on page {2}: {3}",
tet.get_errnum(), tet.get_apiname(), pageno,
tet.get_errmsg());
continue; /* process next page */
}
if (tet.get_errnum() != 0)
{
Console.WriteLine("Error {0} in {1}() on page {2}: {3}",
tet.get_errnum(), tet.get_apiname(), pageno,
tet.get_errmsg());
}
tet.close_page(page);
}
int imageid, n_images;
/* Get the number of images in the document */
n_images = (int)tet.pcos_get_number(doc, "length:images");
/* Loop over image resources in the document */
for (imageid = 0; imageid < n_images; ++imageid)
{
string imageoptlist;
/* Skiop images which have been consumed by merging */
int mergetype = (int)tet.pcos_get_number(doc,
"images[" + imageid + "]/mergetype");
if (mergetype == 2)
{
continue;
}
/* Skip small images (see "smallimages" option) */
if (tet.pcos_get_number(doc, "images[" + imageid + "]/small") > 0)
{
continue;
}
/* Report image details: pixel geometry, color space etc . */
report_image_info(tet, doc, imageid);
/* Write image data to file */
imageoptlist = " filename={" + outfilebase + "_I" + imageid + "}";
if (tet.write_image_file(doc, imageid, imageoptlist) == -1)
{
Console.WriteLine(
"Error {0} in {1}(): {2}",
tet.get_errnum(), tet.get_apiname(), tet.get_errmsg());
continue; /* process next image */
}
}
tet.close_document(doc);
}
catch (TETException e) {
/* caught exception thrown by TET */
Console.WriteLine("Error {0} in {1}(): {2}",
e.get_errnum(), e.get_apiname(), e.get_errmsg());
return(2);
}
catch (Exception e)
{
Console.WriteLine("General Exception: " + e.ToString());
return(2);
}
finally
{
if (tet != null)
{
tet.Dispose();
}
}
return(0);
}
/**
* Open a named physical or virtual file, extract the text from it, search
* for document or page attachments, and process these recursively. Either
* filename must be supplied for physical files, or data+length from which a
* virtual file will be created. The caller cannot create the PVF file since
* we create a new TET object here in case an exception happens with the
* embedded document - the caller can happily continue with his TET object
* even in case of an exception here.
*
* @param outfp
* @param filename
* @param realname
* @param data
*
* @return 0 if successful, otherwise a non-null code to be used as exit
* status
*/
static int process_document(BinaryWriter outfp, String filename, String realname,
byte[] data)
{
int retval = 0;
TET tet = null;
try
{
String pvfname = "/pvf/attachment";
tet = new TET();
/*
* Construct a PVF file if data instead of a filename was provided
*/
if (filename == null || filename.Length == 0)
{
tet.create_pvf(pvfname, data, "");
filename = pvfname;
}
tet.set_option(globaloptlist);
int doc = tet.open_document(filename, docoptlist);
if (doc == -1)
{
Console.WriteLine("Error " + tet.get_errnum() + " in "
+ tet.get_apiname() + "() (source: attachment '"
+ realname + "'): " + tet.get_errmsg());
retval = 5;
}
else
{
process_document(outfp, tet, doc);
}
/*
* If there was no PVF file deleting it won't do any harm
*/
tet.delete_pvf(pvfname);
}
catch (TETException e)
{
Console.WriteLine("Error " + e.get_errnum() + " in "
+ e.get_apiname() + "() (source: attachment '" + realname
+ "'): " + e.get_errmsg());
retval = 1;
}
catch (Exception e)
{
Console.WriteLine("General Exception: " + e.ToString());
retval = 1;
}
finally
{
if (tet != null)
{
tet.Dispose();
}
}
return(retval);
}
static int Main(string[] args)
{
/* global option list */
string globaloptlist = "searchpath={{../data} {../../data}}";
/* document-specific option list */
string docoptlist = "";
/* page-specific option list e.g
* "imageanalysis={merge={gap=1} smallimages={maxwidth=20}}"
*/
string pageoptlist = "";
TET tet;
int pageno = 0;
string outfilebase;
if (args.Length != 1)
{
Console.WriteLine("usage: image_resources <filename>");
return(2);
}
outfilebase = args.GetValue(0).ToString();
if ((outfilebase.Length > 4) && (outfilebase.Substring(outfilebase.Length - 4).Equals(".pdf")) || (outfilebase.Substring(outfilebase.Length - 4).Equals(".PDF")))
{
outfilebase = outfilebase.Substring(0, outfilebase.Length - 4);
}
tet = new TET();
try
{
int n_pages;
tet.set_option(globaloptlist);
int doc = tet.open_document(args.GetValue(0).ToString(), docoptlist);
if (doc == -1)
{
Console.WriteLine("Error {0} in {1}(): {2}",
tet.get_errnum(), tet.get_apiname(), tet.get_errmsg());
return(2);
}
/* Get number of pages in the document */
n_pages = (int)tet.pcos_get_number(doc, "length:pages");
/* Loop over pages and extract images */
for (pageno = 1; pageno <= n_pages; ++pageno)
{
int page;
int imagecount = 0;
page = tet.open_page(doc, pageno, pageoptlist);
if (page == -1)
{
Console.WriteLine("Error {0} in {1}() on page {2}: {3}",
tet.get_errnum(), tet.get_apiname(), pageno, tet.get_errmsg());
continue; /* try next page */
}
/*
* Retrieve all images on the page
*/
while ((tet.get_image_info(page)) == 1)
{
String imageoptlist;
int maskid;
imagecount++;
/* Report image details: pixel geometry, color space etc. */
report_image_info(tet, doc, tet.imageid);
/* Report placement geometry */
Console.WriteLine(" placed on page " + pageno +
" at position (" + tet.x.ToString("f2") + ", " + tet.y.ToString("f2") + "): " +
(int)tet.width + "x" + (int)tet.height + "pt, alpha=" + tet.alpha + ", beta=" +
tet.beta);
/* Write image data to file */
imageoptlist = "filename={" + outfilebase + "_p" + pageno + "_" + imagecount + "_I" + tet.imageid + "}";
if (tet.write_image_file(doc, tet.imageid, imageoptlist) == -1)
{
Console.WriteLine("\nError [" + tet.get_errnum() +
" in " + tet.get_apiname() + "(): " + tet.get_errmsg());
continue; /* try next image */
}
/* Check whether the image has a mask attached... */
maskid = (int)tet.pcos_get_number(doc,
"images[" + tet.imageid + "]/maskid");
/* and retrieve it if present */
if (maskid != -1)
{
Console.WriteLine(" masked with ");
report_image_info(tet, doc, maskid);
imageoptlist = "filename={" + outfilebase + "_p" + pageno + "_" + imagecount + "_I" + tet.imageid + "mask_I" + maskid + "}";
if (tet.write_image_file(doc, tet.imageid, imageoptlist) == -1)
{
Console.WriteLine("\nError [" + tet.get_errnum() +
" in " + tet.get_apiname() +
"() for mask image: " + tet.get_errmsg());
continue; /* try next image */
}
}
if (tet.get_errnum() != 0)
{
Console.WriteLine("Error {0} in {1}() on page {2}: {3}",
tet.get_errnum(), tet.get_apiname(), pageno, tet.get_errmsg());
}
}
tet.close_page(page);
}
tet.close_document(doc);
}
catch (TETException e)
{
/* caught exception thrown by TET */
Console.WriteLine("Error {0} in {1}(): {2}",
e.get_errnum(), e.get_apiname(), e.get_errmsg());
return(2);
}
catch (Exception e)
{
Console.WriteLine("General Exception: " + e.ToString());
return(2);
}
finally
{
if (tet != null)
{
tet.Dispose();
}
}
return(0);
}
static int Main(string[] args)
{
/* global option list */
string globaloptlist = "searchpath={{../data} {../../data}}";
/* document-specific option list */
string docoptlist = "";
/* page-specific option list */
string pageoptlist = "granularity=page";
/* separator to emit after each chunk of text. This depends on the
* applications needs; for granularity=word a space character may be useful.
*/
string separator = "\n";
TET tet;
FileStream outfile;
BinaryWriter w;
int pageno = 0;
UnicodeEncoding unicode = new UnicodeEncoding(false, true);
Byte[] byteOrderMark = unicode.GetPreamble();
if (args.Length != 2)
{
Console.WriteLine("usage: extractor <infilename> <outfilename>");
return(2);
}
outfile = File.Create(args.GetValue(1).ToString());
w = new BinaryWriter(outfile);
w.Write(byteOrderMark);
tet = new TET();
try
{
int n_pages;
tet.set_option(globaloptlist);
int doc = tet.open_document(args.GetValue(0).ToString(), docoptlist);
if (doc == -1)
{
Console.WriteLine("Error {0} in {1}(): {2}",
tet.get_errnum(), tet.get_apiname(), tet.get_errmsg());
return(2);
}
/* get number of pages in the document */
n_pages = (int)tet.pcos_get_number(doc, "length:pages");
/* loop over pages in the document */
for (pageno = 1; pageno <= n_pages; ++pageno)
{
string text;
int page;
page = tet.open_page(doc, pageno, pageoptlist);
if (page == -1)
{
Console.WriteLine("Error {0} in {1}() on page {2}: {3}",
tet.get_errnum(), tet.get_apiname(), pageno,
tet.get_errmsg());
continue; /* try next page */
}
/* Retrieve all text fragments; This is actually not required
* for granularity=page, but must be used for other
* granularities.
*/
while ((text = tet.get_text(page)) != null)
{
/* print the retrieved text */
w.Write(unicode.GetBytes(text));
/* print a separator between chunks of text */
w.Write(unicode.GetBytes(separator));
}
if (tet.get_errnum() != 0)
{
Console.WriteLine("Error {0} in {1}(): {3}",
tet.get_errnum(), tet.get_apiname(), tet.get_errmsg());
}
tet.close_page(page);
}
tet.close_document(doc);
}
catch (TETException e) {
/* caught exception thrown by TET */
Console.WriteLine("Error {0} in {1}(): {2}",
e.get_errnum(), e.get_apiname(), e.get_errmsg());
}
catch (Exception e)
{
Console.WriteLine("General Exception: " + e.ToString());
return(2);
}
finally
{
outfile.Close();
if (tet != null)
{
tet.Dispose();
}
}
return(0);
}
public static void Main(String[] args)
{
/* Global option list. */
string globaloptlist = "searchpath={{../data} {../../data}}";
/* Document specific option list. */
string docoptlist = "";
/* Page-specific option list. */
string pageoptlist = "granularity=word";
FileStream outfile;
StreamWriter outfp;
if (args.Length != 2)
{
Console.WriteLine("usage: glyphinfo <infilename> <outfilename>");
return;
}
outfile = File.Create(args.GetValue(1).ToString());
outfp = new StreamWriter(outfile, System.Text.Encoding.UTF8);
TET tet = null;
try
{
tet = new TET();
tet.set_option(globaloptlist);
int doc = tet.open_document(args[0], docoptlist);
if (doc == -1)
{
Console.WriteLine("Error " + tet.get_errnum() + " in "
+ tet.get_apiname() + "(): " + tet.get_errmsg());
return;
}
/* get number of pages in the document */
int n_pages = (int)tet.pcos_get_number(doc, "length:pages");
/* Loop over pages in the document */
for (int pageno = 1; pageno <= n_pages; ++pageno)
{
string text;
int page;
int previouscolor = -1;
page = tet.open_page(doc, pageno, pageoptlist);
if (page == -1)
{
Console.WriteLine("Error " + tet.get_errnum() + " in "
+ tet.get_apiname() + "() on page "
+ pageno + ": " + tet.get_errmsg());
continue; /* try next page */
}
/* Administrative information */
outfp.WriteLine("[ Document: '" +
tet.pcos_get_string(doc, "filename") + "' ]");
outfp.WriteLine("[ Document options: '" + docoptlist + "' ]");
outfp.WriteLine("[ Page options: '" + pageoptlist + "' ]");
outfp.WriteLine("[ ----- Page " + pageno + " ----- ]");
/* Retrieve all text fragments */
while ((text = tet.get_text(page)) != null)
{
/* print the retrieved text */
outfp.WriteLine("[" + text + "]");
/* Loop over all glyphs and print their details */
while (tet.get_char_info(page) != -1)
{
string str;
string fontname;
/* Fetch the font name with pCOS (based on its ID) */
fontname = tet.pcos_get_string(doc,
"fonts[" + tet.fontid + "]/name");
/* Print the character */
str = String.Format("U+{0}", tet.uv.ToString("X4"));
/* ...and its UTF8 representation */
str = str + String.Format(" '" + (char)(tet.uv) + "'");
/* Print font name, size, and position */
str = str + String.Format(" {0} size={1} x={2} y={3}",
fontname, tet.fontsize.ToString("f2"),
tet.x.ToString("f2"), tet.y.ToString("f2"));
/* Print the color id */
str = str + String.Format(" colorid={0}", tet.colorid);
/* check wheather the text color changes */
if (tet.colorid != previouscolor)
{
str = print_color_value(str, tet, doc, tet.colorid);
previouscolor = tet.colorid;
}
/* Examine the "type" member */
if (tet.type == 1)
{
str = str + " ligature_start";
}
else if (tet.type == 10)
{
str = str + " ligature_cont";
}
/* Separators are only inserted for granularity > word*/
else if (tet.type == 12)
{
str = str + " inserted";
}
/* Examine the bit flags in the "attributes" member */
const int ATTR_NONE = 0;
const int ATTR_SUB = 1;
const int ATTR_SUP = 2;
const int ATTR_DROPCAP = 4;
const int ATTR_SHADOW = 8;
const int ATTR_DH_PRE = 16;
const int ATTR_DH_ARTF = 32;
const int ATTR_DH_POST = 64;
if (tet.attributes != ATTR_NONE)
{
if ((tet.attributes & ATTR_SUB) == ATTR_SUB)
{
str = str + "/sub";
}
if ((tet.attributes & ATTR_SUP) == ATTR_SUP)
{
str = str + "/sup";
}
if ((tet.attributes & ATTR_DROPCAP) == ATTR_DROPCAP)
{
str = str + "/dropcap";
}
if ((tet.attributes & ATTR_SHADOW) == ATTR_SHADOW)
{
str = str + "/shadow";
}
if ((tet.attributes & ATTR_DH_PRE) == ATTR_DH_PRE)
{
str = str + "/dehyphenation_pre";
}
if ((tet.attributes & ATTR_DH_ARTF) == ATTR_DH_ARTF)
{
str = str + "/dehyphenation_artifact";
}
if ((tet.attributes & ATTR_DH_POST) == ATTR_DH_POST)
{
str = str + "/dehyphenation_post";
}
}
outfp.WriteLine(str);
}
outfp.WriteLine("");
}
if (tet.get_errnum() != 0)
{
Console.WriteLine("Error " + tet.get_errnum() + " in "
+ tet.get_apiname() + "() on page "
+ pageno + ": " + tet.get_errmsg());
}
tet.close_page(page);
}
tet.close_document(doc);
}
catch (TETException e)
{
Console.WriteLine("Error " + e.get_errnum() + " in "
+ e.get_apiname() + "(): " + e.get_errmsg());
}
catch (Exception e)
{
Console.WriteLine("General Exception: " + e.ToString());
}
finally
{
if (tet != null)
{
tet.Dispose();
}
}
}
public static void Main(String[] args)
{
/* Global option list. */
string globaloptlist = "searchpath={{../data} {../../data}}";
/* Document specific option list. */
string docoptlist = "";
/* Page-specific option list. */
string pageoptlist = "granularity=line";
/* Search text with at least this size (use 0 to catch all sizes). */
double fontsizetrigger = 10;
/* Catch text where the font name contains this string (use empty string
* to catch all font names).
*/
String fontnametrigger = "Bold";
TET tet = null;
int pageno = 0;
if (args.Length != 1)
{
Console.WriteLine("usage: fontfilter <infilename>");
return;
}
try
{
tet = new TET();
tet.set_option(globaloptlist);
int doc = tet.open_document(args[0], docoptlist);
if (doc == -1)
{
Console.WriteLine("Error " + tet.get_errnum() + " in "
+ tet.get_apiname() + "(): " + tet.get_errmsg());
return;
}
/* Loop over pages in the document */
int n_pages = (int)tet.pcos_get_number(doc, "length:pages");
for (pageno = 1; pageno <= n_pages; ++pageno)
{
int page = tet.open_page(doc, pageno, pageoptlist);
if (page == -1)
{
Console.WriteLine("Error " + tet.get_errnum() + " in "
+ tet.get_apiname() + "(): " + tet.get_errmsg());
return; /* try next page */
}
/* Retrieve all text fragments for the page */
String text;
while ((text = tet.get_text(page)) != null)
{
/* Loop over all characters */
int ci;
while ((ci = tet.get_char_info(page)) != -1)
{
/* We need only the font name and size; the text
* position could be fetched from tet.x and tet.y.
*/
String fontname = tet.pcos_get_string(doc,
"fonts[" + tet.fontid + "]/name");
/* Check whether we found a match */
if (tet.fontsize >= fontsizetrigger &&
fontname.IndexOf(fontnametrigger) != -1)
{
/* print the retrieved font name, size, and text */
Console.WriteLine("[{0} {1:0.00}] {2}", fontname,
tet.fontsize, text);
}
/* In this sample we check only the first character of
* each fragment.
*/
break;
}
}
if (tet.get_errnum() != 0)
{
Console.WriteLine("Error " + tet.get_errnum() + " in "
+ tet.get_apiname() + "(): " + tet.get_errmsg());
}
tet.close_page(page);
}
tet.close_document(doc);
}
catch (TETException e)
{
if (pageno == 0)
{
Console.WriteLine("Error " + e.get_errnum() + " in "
+ e.get_apiname() + "(): " + e.get_errmsg() + "\n");
}
else
{
Console.WriteLine("Error " + e.get_errnum() + " in "
+ e.get_apiname() + "() on page " + pageno + ": "
+ e.get_errmsg() + "\n");
}
}
catch (Exception e)
{
Console.WriteLine("General Exception: " + e.ToString());
}
finally
{
tet.Dispose();
}
}
public static void Main(String[] args)
{
/* Global option list. */
string globaloptlist = "searchpath={{../data} {../../data}}";
/* Document specific option list. */
string basedocoptlist = "";
/* Page-specific option list. */
/* Remove the tetml= option if you don't need font and geometry
* information */
string pageoptlist = "granularity=word tetml={glyphdetails={all}}";
/* set this to true to generate TETML output in memory */
bool inmemory = false;
if (args.Length != 2)
{
Console.WriteLine("usage: tetml <pdffilename> <xmlfilename>");
return;
}
TET tet = null;
try
{
String docoptlist;
tet = new TET();
tet.set_option(globaloptlist);
if (inmemory)
{
/*
* This program fetches the TETML data encoded in UTF-8.
* Subsequently the data is converted to a VisualBasic String,
* which is encoded in UTF-16.
* While it is not strictly necessary in case of this program, it
* is more clean to instruct TET to put 'encoding="UTF-16"' into
* the XML header.
*/
docoptlist = "tetml={encodingname=UTF-16} " + basedocoptlist;
}
else
{
docoptlist = "tetml={filename={" + args[1] + "}} "
+ basedocoptlist;
}
int doc = tet.open_document(args[0], docoptlist);
if (doc == -1)
{
Console.WriteLine("Error " + tet.get_errnum() + " in "
+ tet.get_apiname() + "(): " + tet.get_errmsg());
return;
}
int n_pages = (int)tet.pcos_get_number(doc, "length:pages");
/* Loop over pages in the document */
for (int pageno = 1; pageno <= n_pages; ++pageno)
{
tet.process_page(doc, pageno, pageoptlist);
}
/* This could be combined with the last page-related call. */
tet.process_page(doc, 0, "tetml={trailer}");
if (inmemory)
{
/* Get the XML document as a byte array. */
byte[] tetml = tet.get_tetml(doc, "");
if (tetml == null)
{
Console.WriteLine("tetml: couldn't retrieve XML data");
return;
}
/* Process the in-memory XML document to print out some
* information that is extracted with the sax_handler class.
*/
XmlDocument xmldoc = new XmlDocument();
UTF8Encoding utf8_enc = new UTF8Encoding();
String stetml = utf8_enc.GetString(tetml);
xmldoc.LoadXml(stetml);
XmlNodeList nodeList;
XmlElement root = xmldoc.DocumentElement;
/* Create an XmlNamespaceManager for resolving namespaces. */
XmlNamespaceManager nsmgr =
new XmlNamespaceManager(xmldoc.NameTable);
nsmgr.AddNamespace("tet",
"http://www.pdflib.com/XML/TET5/TET-5.0");
nodeList = root.SelectNodes("//tet:Font", nsmgr);
IEnumerator ienum = nodeList.GetEnumerator();
while (ienum.MoveNext())
{
XmlNode font = (XmlNode)ienum.Current;
XmlAttributeCollection attrColl = font.Attributes;
XmlAttribute name_attr =
(XmlAttribute)attrColl.GetNamedItem("name");
XmlAttribute type_attr =
(XmlAttribute)attrColl.GetNamedItem("type");
Console.WriteLine("Font " + name_attr.Value + " "
+ type_attr.Value);
}
nodeList = root.SelectNodes("//tet:Word", nsmgr);
Console.WriteLine("Found " + nodeList.Count
+ " words in document");
}
tet.close_document(doc);
}
catch (TETException e)
{
Console.WriteLine("Error " + e.get_errnum() + " in "
+ e.get_apiname() + "(): " + e.get_errmsg());
}
catch (Exception e)
{
Console.WriteLine("General Exception: " + e.ToString());
}
finally
{
if (tet != null)
{
tet.Dispose();
}
}
}