Part Four—Wrapping up: Two Case Studies
by Don Henderson, Paul Dorfman
Data Management Solutions Using SAS Hash Table Operations
- About This Book
- About These Authors
- Acknowledgments
- Part One—The HOW of the SAS Hash Object
- Chapter 1: Hash Object Essentials
- Chapter 2: Table-Level Operations
- 2.1 Introduction
- 2.2 CREATE Operation
- 2.2.1 Declaring a Hash Object
- 2.2.2 Creating a Hash Object Instance
- 2.2.3 Combining Declaration and Instantiation
- 2.2.4 Defining Hash Table Variables
- 2.2.5 Omitting the DEFINEDATA Method
- 2.2.6 Wrapping Up the Create Operation
- 2.2.7 PDV Host Variables and Parameter Type Matching
- 2.2.8 Other Ways of Hard-Coded Parameter Type Matching
- 2.2.9 Dynamic Parameter Type Matching via File Reference
- 2.2.10 Parameter Type Matching by Forced File Reference
- 2.2.11 Parameter Type Matching by Default File Reference
- 2.2.12 Defining Multiple Hash Variables
- 2.2.13 Defining Hash Variables as Non-Literal Expressions
- 2.2.14 Defining Hash Variables Dynamically One at a Time
- 2.2.15 Defining Hash Variables Using Metadata
- 2.2.16 Multiple Instances Issue
- 2.2.17 Ensuring Single Instance Usage
- 2.2.18 Handling Multiple Instances
- 2.2.19 Create Operation Hash Tools
- 2.3 DELETE (Table) Operation
- 2.4 CLEAR Operation
- 2.5 OUTPUT Operation
- 2.5.1 The OUTPUT Method
- 2.5.2 Open-Write-Close Cycle
- 2.5.3 Open-Write-Close Cycle Encapsulation
- 2.5.4 Avoiding Open File Conflicts
- 2.5.5 Output Data Set Member Types
- 2.5.6 Creating and Overwriting Output Data Set
- 2.5.7 Using Output Data Set Options
- 2.5.8 DATASET Argument as Non-Literal Expression
- 2.5.9 Output Data Order
- 2.5.10 Output Operation Hash Tools
- 2.6 DESCRIBE Operation
- Chapter 3: Item-Level Operations: Direct Access
- 3.1 Introduction
- 3.2 SEARCH (Pure LookUp) Operation
- 3.3 INSERT Operation
- 3.3.1 Dynamic Memory Acquisition
- 3.3.2 Implicit INSERT
- 3.3.3 Implicit INSERT: Method Call Mode
- 3.3.4 Implicit INSERT: Methods Other Than ADD
- 3.3.5 Implicit INSERT: Argument Tag Mode
- 3.3.6 Explicit INSERT
- 3.3.7 Explicit INSERT Rules
- 3.3.8 Implicit vs Explicit INSERT
- 3.3.9 Unique Key and Duplicate Key INSERT
- 3.3.10 Unique INSERT
- 3.3.11 Duplicate INSERT
- 3.3.12 Insertion Order
- 3.3.13 Insert Operation Hash Tools
- 3.3.14 INSERT Operation Hash-PDV Interaction
- 3.4 DELETE ALL Operation
- 3.5 RETRIEVE Operation
- 3.6 UPDATE ALL Operation
- 3.7 ORDER Operation
- 3.7.1 ORDER Operation Invocation
- 3.7.2 ORDERED Argument Tag Plasticity
- 3.7.3 Hash Items vs Hash Item Groups
- 3.7.4 OUTPUT Operation Effects
- 3.7.5 General Hash Table Order Principle
- 3.7.6 Ordering by Composite Keys
- 3.7.7 Setting the SORTEDBY= Option
- 3.7.8 ORDER Operation Hash Tools
- 3.7.9 ORDER Operation Hash-PDV Interaction
- Chapter 4: Item-Level Operations: Enumeration
- 4.1 Introduction
- 4.2 Enumeration: Basics and Classification
- 4.3 KEYNUMERATE Operation
- 4.3.1 KeyNumerate Operation Mechanics
- 4.3.2 FIND_NEXT: Implicit vs Explicit
- 4.3.3 Other KeyNumerate Coding Styles
- 4.3.4 Version 9.4 Add-On: DO_OVER
- 4.3.5 Forward and Backward, In and Out
- 4.3.6 Staying within the Item List (Keeping It Set)
- 4.3.7 HAS_NEXT and HAS_PREV Peculiarities
- 4.3.8 Harvesting Hash Items
- 4.3.9 Harvesting Hash Items via Explicit Calls
- 4.3.10 Selective DELETE and UPDATE Operations
- 4.3.11 Selective DELETE: Single Item
- 4.3.12 Selective Delete: Multiple Items
- 4.3.13 Selective UPDATE
- 4.3.14 Selective DELETE vs Selective UPDATE
- 4.3.15 KeyNumerate Operation Hash Tools
- 4.3.16 KeyNumerate Operation Hash-PDV Interaction
- 4.4 ENUMERATE ALL Operation
- 4.4.1 The Hash Iterator Object
- 4.4.2 Creating and Linking the Iterator Object
- 4.4.3 Hash Iterator Pointer
- 4.4.4 Direct Iterator Access: First Item
- 4.4.5 Direct Iterator Access: Last Item
- 4.4.6 Direct Iterator Access: Key-Item
- 4.4.7 Sequential Access
- 4.4.8 Enumerating from the End Points
- 4.4.9 Iterator Priming Using NEXT and PREV
- 4.4.10 FIRST/LAST vs NEXT/PREV
- 4.4.11 Keeping the Iterator in the Table
- 4.4.12 Enumerating Sequentially from a Key-Item
- 4.4.13 Harvesting Same-Key Items from a Key-Item
- 4.4.14 The Hash Iterator and Item Locking
- 4.4.15 Locking and Unlocking
- 4.4.16 Locking Same-Key Item Groups
- 4.4.17 Locking the Entire Hash Table
- 4.4.18 ENUMERATE ALL Operation Hash Tools
- 4.4.19 Hash-PDV Interaction
- Part Two—The WHAT and the WHY of the SAS Hash Object
- Chapter 5: Bizarro Ball Sample Data
- Chapter 6: Data Tasks Using Hash Table Operations
- Chapter 7: Supporting Data Warehouse Star Schemas
- Chapter 8: Creating Data Aggregates and Metrics
- Part Three—Expanding the WHAT and the WHY, along with the HOW of the SAS Hash Object
- Chapter 9: Hash of Hashes – Looping Through SAS Hash Objects
- Chapter 10: The Hash Object as a Dynamic Data Structure
- Chapter 11: Hash Object Memory Management
- 11.1 Introduction
- 11.2 Memory vs. Disk Trade-Off
- 11.3 Making Use of Existing Key Order
- 11.4 MD5 Hash Key Reduction
- 11.5 Data Portion Offload (Hash Index)
- 11.6 Uniform Input Split
- 11.7 Uniform MD5 Split On the Fly
- 11.8 Uniform Split Using a SAS Index
- 11.9 Combining Hash Memory-Saving Techniques
- 11.10 MD5 Argument Concatenation Ins and Outs
- 11.11 Summary
- Part Four—Wrapping up: Two Case Studies
- Index
Part Four—Wrapping up: Two Case Studies
It is always nice to finish a project on a positive note, which is the point of Part Four. This part concludes with several samples in the hope and expectation that the headquarters office of Bizarro Ball decides to use SAS, and specifically, the SAS hash object as a key component of their data management and reporting requirements.
Part Four contains two chapters that represent two different use cases—addressing ad-hoc questions and research needs.
1. Chapter 12 builds on previous examples that calculated metrics to highlight what can be done with the hash object given that it has the rich set of DATA step programming facilities. This case study provides an example of researching and comparing alternative metrics.
2. Chapter 13 provides an example of how existing programs could be repackaged and re-used to address the typical ad-hoc questions that come up in virtually every project.
A secondary goal of Part Four is to remind the users of our Proof of Concept that it is just that - a proof of concept. In other words, SAS has many tools that can be leveraged in order to build a series of tools that can be packaged as needed to address a given question.
-
No Comment
..................Content has been hidden....................
You can't read the all page of ebook, please click here login for view all page.