/[webpac2]/trunk/lib/WebPAC/Manual.pod
This is repository of my old source code which isn't updated any more. Go to git.rot13.org for current projects!
ViewVC logotype

Annotation of /trunk/lib/WebPAC/Manual.pod

Parent Directory Parent Directory | Revision Log Revision Log


Revision 1 - (hide annotations)
Sat Jun 25 20:23:23 2005 UTC (18 years, 9 months ago) by dpavlin
File size: 4173 byte(s)
initial import of some documentation and module structure

1 dpavlin 1 =head1 WebPAC - Search engine or data-warehouse manual
2    
3     It's quite hard to explain conceisly what webpac is. It's a mix between
4     search engine and data warehousing application. Let's see that in detail...
5    
6     WebPAC was originally written to search CDS/ISIS records using C<swish-e>.
7     Since then it has, however, adopted different other input formats and added
8     support for alphabetical lists (earlier described as indexes).
9    
10     With evolution of this concept, we decided to produce following work-flow
11     of your data:
12    
13     step
14    
15     source file CDS/ISIS, MARC, Excel, robots, ...
16     |
17     1 | apply import normalisation rules (xml)
18     V
19     intermidiate this data is re-formatted source data converted
20     data to chunks based on tag names from import_xml
21     |
22     2 | apply output filter (TT2)
23     V
24     data search engine, HTML, OAI, RDBMS
25     |
26     3 | filter using query in REST format
27     4 | apply output filter (TT2)
28     V
29     client Web browser, SOAP
30    
31     =head2 Normalisation and Intermidiate data
32    
33     This is first step in working with your data.
34    
35     You are creating mappings, one-to-one from source data records to documents
36     in webpac. You can split or merge data from input records, apply filters
37     (perl subroutines), use lookups within same source file or do simple
38     evaluations while producing output.
39    
40     All that is controlled with C<import_xml> configuration file. You will want
41     to create fine-grained chunks of data (like separate first and last name),
42     which will later be used to produce output. You can think of conversation
43     process as application of C<import_xml> recepie on every input record.
44    
45     Each tag within recepie is creating one new records as long as there are
46     fields in input format (which can be repeatable) that satisfy at least one
47     field within tag.
48    
49     Users of older webpac should note that this file doesn't contain any more
50     formatting or specification of output type and that granularity of each tag
51     has increased.
52    
53     =head2 Output filter
54    
55     Now that we have normalized record, we can create some output. You can create
56     html from it, data files for search engine or insert them into RDBMS.
57    
58     The twist is that application of output filters can be recursive, allowing
59     you to query data generated in previous step. This enables to you represent
60     lists or trees from source data that have structure. This also requires to
61     produce structured data in step 2 which can be filtered and queried in steps
62     3 and 4 to produce final output.
63    
64     You should note that you can query intermidiate data in step 4 also, not
65     just data produced in step 2.
66    
67     Output filter use Template Toolkit 2, so you have full power of simple
68     procedural language (loops, conditions) and handy built-in functions to
69     produce output.
70    
71     =head2 REST Query Format
72    
73     Design decision is to use REST query format. This has benefit of simplicity
74     and ability to create unique URLs to all content within webpac. Simple query
75     format is:
76    
77     http://webpac/search/html/personal_name/Joe%20Doe/AND/year/LT%201995
78    
79     This REST query can be broken down to:
80    
81     =over
82    
83     =item http://webpac
84    
85     Hostname on which service is running. Not required if doing lookups, just
86     for browser usage.
87    
88     =item search
89    
90     Name of output filtering methods. This will specify search engine.
91    
92     =item html
93    
94     Specified template that will be used to produce output.
95    
96     =item perlsonal_name/Joe%20Doe...
97    
98     URL encoded query string. It is specific to filtering method used.
99    
100     =back
101    
102     You can easily produce RSS feed for same query using follwing REST url:
103    
104     http://webpac/search/rss/personal_name/Joe%20Doe/AND/year/LT%201995
105    
106     Yes, it really is that simple. As it should be.
107    
108     =head1 Tehnical stuff
109    
110     Following text will be more hard-code tehnical stuff about how is webpac
111     implemented and why.
112    
113     =head2 Search Engine
114    
115     We are using Hyper Estraier search engine using pgestraier PostgreSQL bindings
116     for it.
117    
118     It should be relativly easy to plugin another one if need arise.
119    
120     =head2 Data Warehouse
121    
122     In a nutshell, webpac has evolved to support hybrid data as input. That
123     means it has become kind of data-warehouse application. It doesn't support
124     directly roll-up and roll-down operations, but they can be emulated using
125     intermidiate data step or output step.
126    

  ViewVC Help
Powered by ViewVC 1.1.26