IFilterShop PDF+ IFilter Server Edition Release 3.0 README

CONTENT OF README FILE

• General Information
• System Requirements
• Information Retrieval
• Support for custom XMP schemas
• Indexing of XMP sidecar files
• Office Document Properties
• Installation Instructions
• Basic Installation Steps
• Text Extractor Setup
• Multiple Properties Output
• Additional Setup Steps
• How to Uninstall
• Known Issues
• Additional Information
• What's new in this version
• Contact Information

General Information

PDF+ IFilter is an enhanced IFilter for Adobe PDF files. It extends Adobe PDF IFilter to extract text and XMP metadata from PDF files. It may also work without Adobe PDF IFilter, in which case only XMP metadata will be indexed. PDF+ IFilter supports Dublin Core, XMP Basic, Adobe PDF and custom XMP schemas. PDF+ IFilter is easily extensible and can support other XMP core schemas such as Rights Management or Media Management. If your metadata needs are not covered by the core schemas, you may add custom schemas as extensions. Please refer to "Support for custom XMP schemas" section for more information. For better integration with Microsoft applications PDF+ IFilter also outputs common office document properties such as 'DocAuthor', 'DocKeywords' and others. See "Office Document Properties" sections below for more information.

PDF+ IFilter supports Indexing Service, SharePoint Portal Server, SQL Server Full-Text Search, Window Search Service and all other products based on Microsoft Search technology.

System Requirements

Information Retrieval

PDF+ IFilter extracts the following XMP Dublin Core metadata:

XMP Dublin Core Metadata	Property Name	Property Type	Description
dc:contributor	contributor	VT_LPWSTR	Contributors to the resource (other than the authors)
dc:coverage	coverage	VT_LPWSTR	The extent or scope of the resource
dc:creator	creator	VT_LPWSTR	The authors of the resource (listed in order of precedence, if significant)
dc:date	date	VT_FILETIME	Date(s) that something interesting happened to the resource
dc:description	description	VT_LPWSTR	A textual description of the content of the resource
dc:format	format	VT_LPWSTR	The file format used when saving the resource
dc:identifier	identifier	VT_LPWSTR	Unique identifier of the resource
dc:language	language	VT_LPWSTR	Language of the document
dc:publisher	publisher	VT_LPWSTR	Publishers
dc:relation	relation	VT_LPWSTR	How the content relates to other resources
dc:rights	rights	VT_LPWSTR	Informal rights statement
dc:source	source	VT_LPWSTR	Unique identifier of the work from which this resource was derived
dc:subject	subject	VT_LPWSTR	An unordered array of descriptive phrases or keywords that specify the topic of the content of the resource
dc:title	title	VT_LPWSTR	The title of the document, or the name given to the resource
dc:type	type	VT_LPWSTR	A document type; for example, novel, poem, or working paper

In accordance with Microsoft IFilter specification, PSD+ IFilter defines each metadata as combination of Property Set and Property Name. All XMP Dublin Core metadata belong to {DC099694-64F5-4371-9AA9-868846A5657E} Property Set GUID.

XMP Basic Schema support

The XMP Basic Schema contains properties that provide basic descriptive information.

• The schema namespace URI is http://ns.adobe.com/xap/1.0/
• The preferred schema namespace prefix is xap

• Open registry key "HKEY_LOCAL_MACHINE\SOFTWARE\IFilterShop\PdfPlusFilter\DefaultSchemas"
• Change "XAP" DWORD registry value from 1 to 0

PDF+ IFilter extracts the following XMP Basic metadata:

XMP Basic Metadata	Property Name	Property Type	Description
xap:Advisory	Advisory	VT_LPWSTR	An unordered array specifying properties that were edited outside the authoring application
xap:BaseURL	BaseURL	VT_LPWSTR	The base URL for relative URLs in the document content
xap:CreateDate	CreateDate	VT_FILETIME	The date and time the resource was originally created
xap:CreatorTool	CreatorTool	VT_LPWSTR	The name of the first known tool used to create the resource
xap:Identifier	Identifier	VT_LPWSTR	An unordered array of text strings that unambiguously identify the resource within a given context
xap:MetadataDate	MetadataDate	VT_FILETIME	The date and time that any metadata for this resource was last changed. It should be the same as or more recent than xap:ModifyDate
xap:ModifyDate	ModifyDate	VT_FILETIME	The date and time the resource was last modified
xap:Nickname	Nickname	VT_LPWSTR	A short informal name for the resource

All XMP Basic metadata belong to {BA64F93D-FBA6-4b75-8F7F-37FC8B493176} Property Set GUID.

Adobe PDF Schema support

Adobe PDF Schema specifies properties used with Adobe PDF files.

• The schema namespace URI is http://ns.adobe.com/pdf/1.3/
• The preferred schema namespace prefix is pdf

• Open registry key "HKEY_LOCAL_MACHINE\SOFTWARE\IFilterShop\PdfPlusFilter\DefaultSchemas"
• Change "PDF" DWORD registry value from 1 to 0

PDF+ IFilter extracts the following XMP Adobe PDF metadata:

XMP Adobe PDF Metadata	Property Name	Property Type	Description
pdf:Keywords	Keywords	VT_LPWSTR	External Keywords
pdf:PDFVersion	PDFVersion	VT_LPWSTR	PDF file version
pdf:Producer	Producer	VT_LPWSTR	Name of tool that created PDF document

All XMP Adobe PDF metadata belong to {A2BAC514-218A-43E8-A3EF-7598A66B19BE} Property Set GUID.

Support for custom XMP schemas

PDF+ IFilter is easily configurable for additional XMP core schemas and custom XMP schemas. To make your custom XMP schema searchable:

1. Open registry key "HKEY_LOCAL_MACHINE\SOFTWARE\IFilterShop\PdfPlusFilter\CustomSchemas"
2. Create a new key with custom XMP schema name. For example, for PDFx Schema the new entry can be:
   "HKEY_LOCAL_MACHINE\SOFTWARE\IFilterShop\PdfPlusFilter\CustomSchemas\PDFx Schema"
3. Under the registry key created add the following String values:

   Registry value	Description	Example for PDFx Schema
   NameSpace	URI for custom XMP schema	http://ns.adobe.com/pdfx/1.0/
   GUID	Property Set GUID that will be used by Indexing Service *	{2C443B1E-F1E2-404F-974D-E21FEF8E72AA}
   FileName	Full path to the text file with custom XMP schema properties mapping **	C:\IFilterShop\PdfPlusFilter\PDFxSchema.txt

   * GUID shall be a newly generated GUID

   ** FileName value is optional. If this value is missing then all properties within the schema will be indexed

   Each line in the text file referred by FileName value shall have the following structure:
   <XMP Metadata>;<Property Name>;<Property Type>, where

   • <XMP Metadata> is a name of property (or path in general case) in custom XMP schema to be extracted by PDF+ IFilter. <XMP Metadata> shall not include namespace prefix.
   • <Property Name> defines property name for Search service
   • <Property Type> defines property type for Search service. This parameter is optional; all properties have VT_LPWSTR type by default. Other currently supported property types are VT_FILETIME and VT_INT
   For example, PDFx Schema property setup file can be defined as:

   CustomProp1;ProjectName
   CustomProp2;ProjetNum;VT_INT
   CustomProp3;ProjetStartDate;VT_FILETIME
   

4. Close registry editor and restart all appropriate Search services

Indexing of XMP sidecar files

Office Document Properties

Installation Instructions

Text Extractor Setup

Multiple Properties Output

Additional Setup Steps

How to Uninstall

Known Issues

PDF+ IFilter uses PDF text extractor to index text of PDF files. When installed standalone, PDF+ IFilter extracts XMP metadata only. You have to install Adobe PDF IFilter or other PDF text indexer in addition to PDF+ IFilter in order to search both metadata and content of PDF files. Please refer to "Text Extractor Setup" section for more information.

PLEASE NOTE (for Indexing Service only). When both IFilters are installed Indexing Service relies on registration order to choose which one to use. Each time you start Indexing Service it looks at the list of DLLs in the "HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Control\ContentIndex\DLLsToRegister" registry value and registers each of the DLLs in that order. To resolve this issue, move the registration for PDF+ IFilter DLL (PdfPlusFilter.dll) to the end of the list that is maintained by Indexing Service:

• Stop Microsoft Indexing Service
• Start Registry Editor (Regedt32.exe)
• Locate the DLLsToRegister value under the registry key "HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Control\ContentIndex"
• On the Edit menu, click Multi String, and then move the line that references PdfPlusFilter.dll to the end of the list
• Click OK, and then quit Registry Editor
• Start Microsoft Indexing Service

PDF+ IFilter does not extract text of PDF files when integrated with Adobe PDF IFilter versions 5.0 or 6.0.

Adobe PDF IFilter (ver. 5.0 and 6.0) is an apartment threaded IFilter. Apartment threaded IFilters behave abnormally on some server platforms where indexing process is multithreaded. Follow the steps below to make Adobe PDF IFilter work on these platforms.

1. Change the registry value "HKEY_CLASSES_ROOT\CLSID\{4C904448-74A9-11d0-AF6E-00C04FD8DC02}\InprocServer32\ThreadingModel" from Apartment to Both
2. Double click the registry value "HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Control\ContentIndex\DLLsToRegister" value, and then click Modify
3. Remove the entry corresponding to pdffilt.dll from the list. This will prevent Adobe PDF IFilter from re-registering itself each time you restart Indexing Service (in case Indexing Service is not disabled)
4. Restart all appropriate Search services
5. Perform a full rescan

Additional Information

Adobe eXtensible Metadata Platform enables various types of content with open standards metadata. It works by embedding metadata packets into binary data file. XMP metadata can currently be embedded into various image files (GIF, PNG, JPEG, TIFF) and document files such as PDF, PostScript, Adobe Illustrator, Adobe FrameMaker. Metadata packets are specifically designed to preserve consistency of the file, so that other applications would not be affected. XMP metadata is extremely rich in nature and suits a large variety of tasks. More information about Adobe XMP can be found at http://www.adobe.com/products/xmp.

What is Dublin Core?

Dublin Core is an initiative to create digital library metadata for the Web. Dublin Core is made up of 15 metadata (data that describes data) elements that offer expanded cataloging information and improved document indexing for search engine programs. Two forms of Dublin Core exist: Simple Dublin Core and Qualified Dublin Core. Simple Dublin Core expresses elements as attribute-value pairs using just the 15 metadata elements from the Dublin Core Metadata Element Set. Qualified Dublin Core increases the specificity of metadata by adding information about encoding schemes, enumerated lists of values, or other processing clues. While enabling searches to be more specific, qualifiers are also more complex and can pose challenges to interoperability. More information about Dublin Core may be found at http://www.dublincore.org.

PLEASE NOTE. Adobe XMP and PDF+ IFilter use Dublin Core version 1.1. Lately Dublin Core Metadata Initiative board extended the set with more elements thus making the previous specification obsolete. This should not affect the performance of Adobe XMP and PDF+ IFilter. Current specification is fully backwards compatible with version 1.1.

What's new in this version

Contact Information

WWW:

http://www.ifiltershop.com

E-mail:

support@ifiltershop.com