Contents of /tags/WAIT_1_900/README

                             WAIT 1.6

                  Copyright (c) 1996, Ulrich Pfeifer

------------------------------------------------------------------------
    This program is free software; you can redistribute it and/or
    modify it under the same terms than Perl itself.

    This program is distributed in the hope that it will be useful,
    but WITHOUT ANY WARRANTY; without even the implied warranty of
    MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. 
------------------------------------------------------------------------

This software is not actively maintained by it's author.

For more two years now I tried to steal some time to clean this up
without any luck. So I decided to pass the baton on. I consider the
input part pretty satisfying. The query part - despite being operable
and useful - needs a major overhaul. To provide a forum for further
discussions an to coordinate further developement, I did setup a
mailinglist.  Drop me a line if you want to participate.

Ulrich Pfeifer <upf@wait.de> 

------------------------------------------------------------------------
NAME
    WAIT - a rewrite of the freeWAIS-sf engine in Perl

Status of this document
    I started writing down some information about the implementation
    before I forget them in my spare time. The stuff is incomplete
    at least. Any additions, corrections, ... welcome.

PURPOSE
    As you might know, I developed and maintained freeWAIS-sf (with
    the help of many people in The Net). FreeWAIS-sf is based on
    freeWAIS maintained by the Clearing House for Network
    Information Retrieval (CNIDR) which in turn is based on wais-8-
    b5 implemented by Thinking Machine et al. During this long
    history - implementation started about 1989 - many people
    contributed to the distribution and added features not foreseen
    by the original design. While the system fulfills its task now,
    the code has reached a state where adding new features is nearly
    impossible and even fixing longstanding bugs and removing
    limitations has become a very time consuming task.

    Therefore I decided to pass the maintenance to WSC Inc. and
    built a new system from scratch. For obvious reasons I choosed
    Perl as implementation language.

DESCRIPTION
    The central idea of the system is to provide a framework and the
    building blocks for any indexing and search system the users
    might want to build. Obviously the framework limits the class of
    system which can be build.

           +------+     +-----+     +------+
       ==> |Access| ==> |Parse| ==> |      |
           +------+     +-----+     |      |
                           ||       |      |     +-----+
                           ||       |Filter| ==> |Index|
                           \/       |      |     +-----+
          +-------+     +-----+     |      |
       <= |Display| <== |Query| <-> |      |
          +-------+     +-----+     +------+

    A collection (aka table) is defined by the instances of the
    access and parse module together with the filter definitions. At
    query time in addition a query and a display module must be
    choosen.

  Access

    The access module defines which documents where members of a
    database. Usually an access module is a tied hash, whose keys
    are the Ids of the documents (did = document id) and whose
    values are the documents themselves. The indexing process loops
    over the keys using `FIRSTKEY' and `NEXTKEY'. Documents are
    retrieved with `FETCH'.

    By convention access modules should be members of the
    `WAIT::Document' hierarchy. Have a look at the
    `WAIT::Document::Split' module to get the idea.

  Parse

    The task parse module is to split the documents into logical
    parts via the `split' method. E.g. the `WAIT::Parse::Nroff'
    splits manuals piped through nroff(1) into the sections *name*,
    *synopsis*, *options*, *description*, *author*, *example*,
    *bugs*, *text*, *see*, and *environment*. Here is the
    implementation of `WAIT::Parse::Base' which handes documents
    with a pretty simple tagged format:

      AU: Pfeifer, U.; Fuhr, N.; Huynh, T.
      TI: Searching Structured Documents with the Enhanced Retrieval
          Functionality of freeWAIS-sf and SFgate
      ER: D. Kroemker
      BT: Computer Networks and ISDN Systems; Proceedings of the third
          International World-Wide Web Conference
      PN: Elsevier
      PA: Amsterdam - Lausanne - New York - Oxford - Shannon - Tokyo
      PP: 1027-1036
      PY: 1995

      sub split {                     # called as method
        my %result;
        my $fld;
      
        for (split /\n/, $_[1]) {
          if (s/^(\S+):\s*//) {
            $fld = lc $1;
          }
          $result{$fld} .= $_ if defined $fld;
        }
        return \%result;
      } 

    Since the original document cannot be reconstructed from its
    attributes, we need a second method (*tag*) which marks the
    regions of the document with tags for the different attributes.
    This tagged form is used by the display module to hilight search
    terms in the documents. Besides the tags for the attributes, the
    method might assign the special tags `_b' and `_i' for
    indicating bold and italic regions.

      sub tag {
        my @result;
        my $tag;
        
        for (split /\n/, $_[1]) {
          next if /^\w\w:\s*$/;
          if (s/^(\S+)://) {
            push @result, {_b => 1}, "$1:";
            $tag = lc $1;
          }
          if (defined $tag) {
            push @result, {$tag => 1}, "$_\n";
          } else {
            push @result, {}, "$_\n";
          }
        }
        return @result;               # we don't go for speed
      } 

    Obviously one could implement `split' via `tag'. The reason for
    having two functions is speed. We need to call `split' for each
    document when indexing a collection. Therefore speed is
    essential. On the other hand, `tag' is called in order to
    display a single document and may be a little slower. It may
    care about tagging bold and italic regions. See
    `WAIT::Parse::Nroff' how this might decrease performance.

  Filter definition

    From the Information Retrieval perspective, the hardest part of
    the system is the filter module. The database administrator
    defines for each attribute, how the contents should be processed
    before it is stored in the index. Usually the processing
    contains steps to restrict the character set, case
    transformation, splitting to words and transforming to word
    stems. In WAIT these steps are defined naturally as a pipeline
    of processing steps. The pipelines are made up by functions in
    the package WAIT::Filter which is pre-populated by the most
    common functions but may be extended any time.

    The equivalent for a typical freeWAIS-sf processing would be
    this pipeline:

            [ 'isotr', 'isolc', 'split2', 'stop', 'Stem']

    The function `isotr' replaces unknown characters by blanks.
    `isolc' transforms to lower case. `split2' splits into words and
    removes words shorter than two characters. `stop' removes the
    freeWAIS-sf stopwords and `Stem' applies the Porter algorithm
    for computing the stem of the words.

    The filter definition for a collection defines a set of piplines
    for the attributes and modifies the pipelines which should be
    used for prefix and interval searches.

    Here is a complete example:

      my $stem  = [{
                    'prefix'    => ['unroff', 'isotr', 'isolc'],
                    'intervall' => ['unroff', 'isotr', 'isolc'],
                   },'unroff', 'isotr', 'isolc', 'split2', 'stop', 'Stem'];
      my $text  = [{
                    'prefix'    => ['unroff', 'isotr', 'isolc'],
                    'intervall' => ['unroff', 'isotr', 'isolc'],
                   },
                    'unroff', 'isotr', 'isolc', 'split2', 'stop'];
      my $sound = ['unroff', 'isotr', 'isolc', 'split2', 'Soundex'];
      
      my $spec  = [
          'name'         => $stem,
          'synopsis'     => $stem,
          'bugs'         => $stem,
          'description'  => $stem,
          'text'         => $stem,
          'environment'  => $text,
          'example'      => $text,  'example' => $stem,
          'author'       => $sound, 'author'  => $stem,
         ]

1	WAIT 1.6
2
3	Copyright (c) 1996, Ulrich Pfeifer
4
5	------------------------------------------------------------------------
6	This program is free software; you can redistribute it and/or
7	modify it under the same terms than Perl itself.
8
9	This program is distributed in the hope that it will be useful,
10	but WITHOUT ANY WARRANTY; without even the implied warranty of
11	MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.
12	------------------------------------------------------------------------
13
14	This software is not actively maintained by it's author.
15
16	For more two years now I tried to steal some time to clean this up
17	without any luck. So I decided to pass the baton on. I consider the
18	input part pretty satisfying. The query part - despite being operable
19	and useful - needs a major overhaul. To provide a forum for further
20	discussions an to coordinate further developement, I did setup a
21	mailinglist. Drop me a line if you want to participate.
22
23	Ulrich Pfeifer <upf@wait.de>
24
25	------------------------------------------------------------------------
26	NAME
27	WAIT - a rewrite of the freeWAIS-sf engine in Perl
28
29	Status of this document
30	I started writing down some information about the implementation
31	before I forget them in my spare time. The stuff is incomplete
32	at least. Any additions, corrections, ... welcome.
33
34	PURPOSE
35	As you might know, I developed and maintained freeWAIS-sf (with
36	the help of many people in The Net). FreeWAIS-sf is based on
37	freeWAIS maintained by the Clearing House for Network
38	Information Retrieval (CNIDR) which in turn is based on wais-8-
39	b5 implemented by Thinking Machine et al. During this long
40	history - implementation started about 1989 - many people
41	contributed to the distribution and added features not foreseen
42	by the original design. While the system fulfills its task now,
43	the code has reached a state where adding new features is nearly
44	impossible and even fixing longstanding bugs and removing
45	limitations has become a very time consuming task.
46
47	Therefore I decided to pass the maintenance to WSC Inc. and
48	built a new system from scratch. For obvious reasons I choosed
49	Perl as implementation language.
50
51	DESCRIPTION
52	The central idea of the system is to provide a framework and the
53	building blocks for any indexing and search system the users
54	might want to build. Obviously the framework limits the class of
55	system which can be build.
56
57	+------+ +-----+ +------+
58	==> \|Access\| ==> \|Parse\| ==> \| \|
59	+------+ +-----+ \| \|
60	\|\| \| \| +-----+
61	\|\| \|Filter\| ==> \|Index\|
62	\/ \| \| +-----+
63	+-------+ +-----+ \| \|
64	<= \|Display\| <== \|Query\| <-> \| \|
65	+-------+ +-----+ +------+
66
67	A collection (aka table) is defined by the instances of the
68	access and parse module together with the filter definitions. At
69	query time in addition a query and a display module must be
70	choosen.
71
72	Access
73
74	The access module defines which documents where members of a
75	database. Usually an access module is a tied hash, whose keys
76	are the Ids of the documents (did = document id) and whose
77	values are the documents themselves. The indexing process loops
78	over the keys using `FIRSTKEY' and `NEXTKEY'. Documents are
79	retrieved with `FETCH'.
80
81	By convention access modules should be members of the
82	`WAIT::Document' hierarchy. Have a look at the
83	`WAIT::Document::Split' module to get the idea.
84
85	Parse
86
87	The task parse module is to split the documents into logical
88	parts via the `split' method. E.g. the `WAIT::Parse::Nroff'
89	splits manuals piped through nroff(1) into the sections name,
90	synopsis, options, description, author, example,
91	bugs, text, see, and environment. Here is the
92	implementation of `WAIT::Parse::Base' which handes documents
93	with a pretty simple tagged format:
94
95	AU: Pfeifer, U.; Fuhr, N.; Huynh, T.
96	TI: Searching Structured Documents with the Enhanced Retrieval
97	Functionality of freeWAIS-sf and SFgate
98	ER: D. Kroemker
99	BT: Computer Networks and ISDN Systems; Proceedings of the third
100	International World-Wide Web Conference
101	PN: Elsevier
102	PA: Amsterdam - Lausanne - New York - Oxford - Shannon - Tokyo
103	PP: 1027-1036
104	PY: 1995
105
106	sub split { # called as method
107	my %result;
108	my $fld;
109
110	for (split /\n/, $_[1]) {
111	if (s/^(\S+):\s*//) {
112	$fld = lc $1;
113	}
114	$result{$fld} .= $_ if defined $fld;
115	}
116	return \%result;
117	}
118
119	Since the original document cannot be reconstructed from its
120	attributes, we need a second method (tag) which marks the
121	regions of the document with tags for the different attributes.
122	This tagged form is used by the display module to hilight search
123	terms in the documents. Besides the tags for the attributes, the
124	method might assign the special tags `_b' and `_i' for
125	indicating bold and italic regions.
126
127	sub tag {
128	my @result;
129	my $tag;
130
131	for (split /\n/, $_[1]) {
132	next if /^\w\w:\s*$/;
133	if (s/^(\S+)://) {
134	push @result, {_b => 1}, "$1:";
135	$tag = lc $1;
136	}
137	if (defined $tag) {
138	push @result, {$tag => 1}, "$_\n";
139	} else {
140	push @result, {}, "$_\n";
141	}
142	}
143	return @result; # we don't go for speed
144	}
145
146	Obviously one could implement `split' via `tag'. The reason for
147	having two functions is speed. We need to call `split' for each
148	document when indexing a collection. Therefore speed is
149	essential. On the other hand, `tag' is called in order to
150	display a single document and may be a little slower. It may
151	care about tagging bold and italic regions. See
152	`WAIT::Parse::Nroff' how this might decrease performance.
153
154	Filter definition
155
156	From the Information Retrieval perspective, the hardest part of
157	the system is the filter module. The database administrator
158	defines for each attribute, how the contents should be processed
159	before it is stored in the index. Usually the processing
160	contains steps to restrict the character set, case
161	transformation, splitting to words and transforming to word
162	stems. In WAIT these steps are defined naturally as a pipeline
163	of processing steps. The pipelines are made up by functions in
164	the package WAIT::Filter which is pre-populated by the most
165	common functions but may be extended any time.
166
167	The equivalent for a typical freeWAIS-sf processing would be
168	this pipeline:
169
170	[ 'isotr', 'isolc', 'split2', 'stop', 'Stem']
171
172	The function `isotr' replaces unknown characters by blanks.
173	`isolc' transforms to lower case. `split2' splits into words and
174	removes words shorter than two characters. `stop' removes the
175	freeWAIS-sf stopwords and `Stem' applies the Porter algorithm
176	for computing the stem of the words.
177
178	The filter definition for a collection defines a set of piplines
179	for the attributes and modifies the pipelines which should be
180	used for prefix and interval searches.
181
182	Here is a complete example:
183
184	my $stem = [{
185	'prefix' => ['unroff', 'isotr', 'isolc'],
186	'intervall' => ['unroff', 'isotr', 'isolc'],
187	},'unroff', 'isotr', 'isolc', 'split2', 'stop', 'Stem'];
188	my $text = [{
189	'prefix' => ['unroff', 'isotr', 'isolc'],
190	'intervall' => ['unroff', 'isotr', 'isolc'],
191	},
192	'unroff', 'isotr', 'isolc', 'split2', 'stop'];
193	my $sound = ['unroff', 'isotr', 'isolc', 'split2', 'Soundex'];
194
195	my $spec = [
196	'name' => $stem,
197	'synopsis' => $stem,
198	'bugs' => $stem,
199	'description' => $stem,
200	'text' => $stem,
201	'environment' => $text,
202	'example' => $text, 'example' => $stem,
203	'author' => $sound, 'author' => $stem,
204	]
205