[Bio] / Sprout / ERDB.pm Repository:
ViewVC logotype

Annotation of /Sprout/ERDB.pm

Parent Directory Parent Directory | Revision Log Revision Log


Revision 1.86 - (view) (download) (as text)

1 : parrello 1.1 package ERDB;
2 :    
3 : parrello 1.10 use strict;
4 :     use Tracer;
5 : olson 1.14 use DBrtns;
6 : parrello 1.10 use Data::Dumper;
7 :     use XML::Simple;
8 :     use DBQuery;
9 :     use DBObject;
10 :     use Stats;
11 :     use Time::HiRes qw(gettimeofday);
12 : parrello 1.42 use Digest::MD5 qw(md5_base64);
13 : parrello 1.19 use FIG;
14 : parrello 1.73 use CGI;
15 : parrello 1.1
16 :     =head1 Entity-Relationship Database Package
17 :    
18 :     =head2 Introduction
19 :    
20 :     The Entity-Relationship Database Package allows the client to create an easily-configurable
21 :     database of Entities connected by Relationships. Each entity is represented by one or more
22 :     relations in an underlying SQL database. Each relationship is represented by a single
23 :     relation that connects two entities.
24 :    
25 :     Although this package is designed for general use, all examples are derived from the
26 :     Sprout database, which is the first database implemented using this package.
27 :    
28 :     Each entity has at least one relation, the I<primary relation>, that has the same name as
29 :     the entity. The primary relation contains a field named C<id> that contains the unique
30 :     identifier of each entity instance. An entity may have additional relations that contain
31 :     fields which are optional or can occur more than once. For example, the B<FEATURE> entity
32 :     has a B<feature-type> attribute that occurs exactly once for each feature. This attribute
33 :     is implemented by a C<feature_type> column in the primary relation C<Feature>. In addition,
34 :     however, a feature may have zero or more aliases. These are implemented using a C<FeatureAlias>
35 :     relation that contains two fields-- the feature ID (C<id>) and the alias name (C<alias>).
36 :     The B<FEATURE> entity also contains an optional virulence number. This is implemented
37 :     as a separate relation C<FeatureVirulence> which contains an ID (C<id>) and a virulence number
38 : parrello 1.8 (C<virulence>). If the virulence of a feature I<ABC> is known to be 6, there will be one row in
39 :     the C<FeatureVirulence> relation possessing the value I<ABC> as its ID and 6 as its virulence
40 :     number. If the virulence of I<ABC> is not known, there will not be any rows for it in
41 :     C<FeatureVirulence>.
42 : parrello 1.1
43 :     Entities are connected by binary relationships implemented using single relations possessing the
44 :     same name as the relationship itself and that has an I<arity> of 1-to-1 (C<11>), 1-to-many (C<1M>),
45 :     or many-to-many (C<MM>). Each relationship's relation contains a C<from-link> field that contains the
46 :     ID of the source entity and a C<to-link> field that contains the ID of the target entity. The name
47 :     of the relationship is generally a verb phrase with the source entity as the subject and the
48 :     target entity as the object. So, for example, the B<ComesFrom> relationship connects the B<GENOME>
49 :     and B<SOURCE> entities, and indicates that a particular source organization participated in the
50 :     mapping of the genome. A source organization frequently participates in the mapping
51 :     of many genomes, and many source organizations can cooperate in the mapping of a single genome, so
52 :     this relationship has an arity of many-to-many (C<MM>). The relation that implements the B<ComesFrom>
53 :     relationship is called C<ComesFrom> and contains two fields-- C<from-link>, which contains a genome ID,
54 :     and C<to-link>, which contains a source ID.
55 :    
56 :     A relationship may itself have attributes. These attributes, known as I<intersection data attributes>,
57 :     are implemented as additional fields in the relationship's relation. So, for example, the
58 :     B<IsMadeUpOf> relationship connects the B<Contig> entity to the B<Sequence> entity, and is used
59 :     to determine which sequences make up a contig. The relationship has as an attribute the
60 :     B<start-position>, which indicates where in the contig that the sequence begins. This attribute
61 :     is implemented as the C<start_position> field in the C<IsMadeUpOf> relation.
62 :    
63 : parrello 1.73 The database itself is described by an XML file. In addition to all the data required to define
64 :     the entities, relationships, and attributes, the schema provides space for notes describing
65 :     the data and what it means. These notes are used by L</ShowMetaData> to generate documentation
66 :     for the database.
67 :    
68 :     Special support is provided for text searching. An entity field can be marked as <em>searchable</em>,
69 :     in which case it will be used to generate a text search index in which the user searches for words
70 :     in the field instead of a particular field value.
71 : parrello 1.1
72 :     Finally, every entity and relationship object has a flag indicating if it is new or old. The object
73 :     is considered I<old> if it was loaded by the L</LoadTables> method. It is considered I<new> if it
74 :     was inserted by the L</InsertObject> method.
75 :    
76 : parrello 1.8 =head2 XML Database Description
77 :    
78 :     =head3 Data Types
79 :    
80 :     The ERDB system supports the following data types. Note that there are numerous string
81 :     types depending on the maximum length. Some database packages limit the total number of
82 :     characters you have in an index key; to insure the database works in all environments,
83 :     the type of string should be the shortest one possible that supports all the known values.
84 :    
85 :     =over 4
86 :    
87 :     =item char
88 :    
89 :     single ASCII character
90 :    
91 :     =item int
92 :    
93 :     32-bit signed integer
94 :    
95 : parrello 1.60 =item counter
96 :    
97 :     32-bit unsigned integer
98 :    
99 : parrello 1.8 =item date
100 :    
101 :     64-bit unsigned integer, representing a PERL date/time value
102 :    
103 :     =item text
104 :    
105 :     long string; Text fields cannot be used in indexes or sorting and do not support the
106 :     normal syntax of filter clauses, but can be up to a billion character in length
107 :    
108 :     =item float
109 :    
110 :     double-precision floating-point number
111 :    
112 :     =item boolean
113 :    
114 :     single-bit numeric value; The value is stored as a 16-bit signed integer (for
115 :     compatability with certain database packages), but the only values supported are
116 :     0 and 1.
117 :    
118 : parrello 1.44 =item id-string
119 :    
120 :     variable-length string, maximum 25 characters
121 :    
122 : parrello 1.8 =item key-string
123 :    
124 :     variable-length string, maximum 40 characters
125 :    
126 :     =item name-string
127 :    
128 :     variable-length string, maximum 80 characters
129 :    
130 :     =item medium-string
131 :    
132 :     variable-length string, maximum 160 characters
133 :    
134 :     =item string
135 :    
136 :     variable-length string, maximum 255 characters
137 :    
138 : parrello 1.42 =item hash-string
139 :    
140 :     variable-length string, maximum 22 characters
141 :    
142 : parrello 1.8 =back
143 :    
144 : parrello 1.42 The hash-string data type has a special meaning. The actual key passed into the loader will
145 :     be a string, but it will be digested into a 22-character MD5 code to save space. Although the
146 :     MD5 algorithm is not perfect, it is extremely unlikely two strings will have the same
147 :     digest. Therefore, it is presumed the keys will be unique. When the database is actually
148 :     in use, the hashed keys will be presented rather than the original values. For this reason,
149 :     they should not be used for entities where the key is meaningful.
150 :    
151 : parrello 1.8 =head3 Global Tags
152 :    
153 :     The entire database definition must be inside a B<Database> tag. The display name of
154 :     the database is given by the text associated with the B<Title> tag. The display name
155 :     is only used in the automated documentation. It has no other effect. The entities and
156 :     relationships are listed inside the B<Entities> and B<Relationships> tags,
157 :     respectively. None of these tags have attributes.
158 :    
159 : parrello 1.10 <Database>
160 :     <Title>... display title here...</Title>
161 :     <Entities>
162 :     ... entity definitions here ...
163 :     </Entities>
164 :     <Relationships>
165 :     ... relationship definitions here...
166 :     </Relationships>
167 :     </Database>
168 : parrello 1.8
169 :     Entities, relationships, indexes, and fields all allow a text tag called B<Notes>.
170 :     The text inside the B<Notes> tag contains comments that will appear when the database
171 :     documentation is generated. Within a B<Notes> tag, you may use C<[i]> and C<[/i]> for
172 :     italics, C<[b]> and C<[/b]> for bold, and C<[p]> for a new paragraph.
173 :    
174 :     =head3 Fields
175 :    
176 :     Both entities and relationships have fields described by B<Field> tags. A B<Field>
177 :     tag can have B<Notes> associated with it. The complete set of B<Field> tags for an
178 :     object mus be inside B<Fields> tags.
179 :    
180 : parrello 1.10 <Entity ... >
181 :     <Fields>
182 :     ... Field tags ...
183 :     </Fields>
184 :     </Entity>
185 : parrello 1.8
186 :     The attributes for the B<Field> tag are as follows.
187 :    
188 :     =over 4
189 :    
190 :     =item name
191 :    
192 :     Name of the field. The field name should contain only letters, digits, and hyphens (C<->),
193 :     and the first character should be a letter. Most underlying databases are case-insensitive
194 : parrello 1.70 with the respect to field names, so a best practice is to use lower-case letters only. Finally,
195 :     the name C<search-relevance> has special meaning for full-text searches and should not be
196 :     used as a field name.
197 : parrello 1.8
198 :     =item type
199 :    
200 :     Data type of the field. The legal data types are given above.
201 :    
202 :     =item relation
203 :    
204 :     Name of the relation containing the field. This should only be specified for entity
205 :     fields. The ERDB system does not support optional fields or multi-occurring fields
206 :     in the primary relation of an entity. Instead, they are put into secondary relations.
207 :     So, for example, in the C<Genome> entity, the C<group-name> field indicates a special
208 :     grouping used to select a subset of the genomes. A given genome may not be in any
209 :     groups or may be in multiple groups. Therefore, C<group-name> specifies a relation
210 :     value. The relation name specified must be a valid table name. By convention, it is
211 :     usually the entity name followed by a qualifying word (e.g. C<GenomeGroup>). In an
212 :     entity, the fields without a relation attribute are said to belong to the
213 :     I<primary relation>. This relation has the same name as the entity itself.
214 :    
215 : parrello 1.70 =item searchable
216 :    
217 :     If specified, then the field is a candidate for full-text searching. A single full-text
218 :     index will be created for each relation with at least one searchable field in it.
219 :     For best results, this option should only be used for string or text fields.
220 :    
221 : parrello 1.73 =item special
222 :    
223 :     This attribute allows the subclass to assign special meaning for certain fields.
224 :     The interpretation is up to the subclass itself. Currently, only entity fields
225 :     can have this attribute.
226 :    
227 : parrello 1.8 =back
228 :    
229 :     =head3 Indexes
230 :    
231 : parrello 1.86 An entity can have multiple alternate indexes associated with it. The fields in an
232 :     index must all be from the same relation. The alternate indexes assist in searching
233 :     on fields other than the entity ID. A relationship has at least two indexes-- a I<to-index> and a
234 :     I<from-index> that order the results when crossing the relationship. For
235 : parrello 1.8 example, in the relationship C<HasContig> from C<Genome> to C<Contig>, the
236 :     from-index would order the contigs of a ganome, and the to-index would order
237 : parrello 1.86 the genomes of a contig. In addition, it can have zero or more alternate
238 :     indexes. A relationship's index must specify only fields in
239 : parrello 1.8 the relationship.
240 :    
241 : parrello 1.86 The alternate indexes for an entity or relationship must be listed inside the B<Indexes> tag.
242 :     The from-index of a relationship is specified using the B<FromIndex> tag; the to-index is
243 :     specified using the B<ToIndex> tag.
244 : parrello 1.8
245 :     Each index can contain a B<Notes> tag. In addition, it will have an B<IndexFields>
246 :     tag containing the B<IndexField> tags. These specify, in order, the fields used in
247 :     the index. The attributes of an B<IndexField> tag are as follows.
248 :    
249 :     =over 4
250 :    
251 :     =item name
252 :    
253 :     Name of the field.
254 :    
255 :     =item order
256 :    
257 :     Sort order of the field-- C<ascending> or C<descending>.
258 :    
259 :     =back
260 :    
261 : parrello 1.78 The B<FromIndex>, and B<ToIndex> tags have no attributes. The B<Index> tag can
262 :     have a B<Unique> attribute. If specified, the index will be generated as a unique
263 :     index.
264 : parrello 1.8
265 :     =head3 Object and Field Names
266 :    
267 :     By convention entity and relationship names use capital casing (e.g. C<Genome> or
268 :     C<HasRegionsIn>. Most underlying databases, however, are aggressively case-insensitive
269 :     with respect to relation names, converting them internally to all-upper case or
270 :     all-lower case.
271 :    
272 :     If syntax or parsing errors occur when you try to load or use an ERDB database, the
273 :     most likely reason is that one of your objects has an SQL reserved word as its name.
274 :     The list of SQL reserved words keeps increasing; however, most are unlikely to show
275 :     up as a noun or declarative verb phrase. The exceptions are C<Group>, C<User>,
276 :     C<Table>, C<Index>, C<Object>, C<Date>, C<Number>, C<Update>, C<Time>, C<Percent>,
277 :     C<Memo>, C<Order>, and C<Sum>. This problem can crop up in field names as well.
278 :    
279 :     Every entity has a field called C<id> that acts as its primary key. Every relationship
280 :     has fields called C<from-link> and C<to-link> that contain copies of the relevant
281 :     entity IDs. These are essentially ERDB's reserved words, and should not be used
282 :     for user-defined field names.
283 :    
284 :     =head3 Entities
285 :    
286 :     An entity is described by the B<Entity> tag. The entity can contain B<Notes>, an
287 :     B<Indexes> tag containing one or more secondary indexes, and a B<Fields> tag
288 :     containing one or more fields. The attributes of the B<Entity> tag are as follows.
289 :    
290 :     =over 4
291 :    
292 :     =item name
293 :    
294 :     Name of the entity. The entity name, by convention, uses capital casing (e.g. C<Genome>
295 :     or C<GroupBlock>) and should be a noun or noun phrase.
296 :    
297 :     =item keyType
298 :    
299 :     Data type of the primary key. The primary key is always named C<id>.
300 :    
301 :     =back
302 :    
303 :     =head3 Relationships
304 :    
305 :     A relationship is described by the C<Relationship> tag. Within a relationship,
306 :     there can be a C<Notes> tag, a C<Fields> tag containing the intersection data
307 : parrello 1.86 fields, a C<FromIndex> tag containing the from-index, a C<ToIndex> tag containing
308 :     the to-index, and an C<Indexes> tag containing the alternate indexes.
309 : parrello 1.8
310 :     The C<Relationship> tag has the following attributes.
311 :    
312 :     =over 4
313 :    
314 :     =item name
315 :    
316 :     Name of the relationship. The relationship name, by convention, uses capital casing
317 :     (e.g. C<ContainsRegionIn> or C<HasContig>), and should be a declarative verb
318 :     phrase, designed to fit between the from-entity and the to-entity (e.g.
319 :     Block C<ContainsRegionIn> Genome).
320 :    
321 :     =item from
322 :    
323 :     Name of the entity from which the relationship starts.
324 :    
325 :     =item to
326 :    
327 :     Name of the entity to which the relationship proceeds.
328 :    
329 :     =item arity
330 :    
331 :     Relationship type: C<1M> for one-to-many and C<MM> for many-to-many.
332 :    
333 :     =back
334 :    
335 : parrello 1.1 =cut
336 :    
337 :     # GLOBALS
338 :    
339 :     # Table of information about our datatypes. "sqlType" is the corresponding SQL datatype string.
340 :     # "maxLen" is the maximum permissible length of the incoming string data used to populate a field
341 : parrello 1.73 # of the specified type. "avgLen" is the average byte length for estimating
342 : parrello 1.75 # record sizes. "sort" is the key modifier for the sort command, "notes" is a type description,
343 :     # and "indexMod", if non-zero, is the number of characters to use when the field is specified in an
344 :     # index
345 : parrello 1.73 my %TypeTable = ( char => { sqlType => 'CHAR(1)', maxLen => 1, avgLen => 1, sort => "",
346 : parrello 1.75 indexMod => 0, notes => "single ASCII character"},
347 : parrello 1.73 int => { sqlType => 'INTEGER', maxLen => 20, avgLen => 4, sort => "n",
348 : parrello 1.75 indexMod => 0, notes => "signed 32-bit integer"},
349 : parrello 1.73 counter => { sqlType => 'INTEGER UNSIGNED', maxLen => 20, avgLen => 4, sort => "n",
350 : parrello 1.75 indexMod => 0, notes => "unsigned 32-bit integer"},
351 : parrello 1.73 string => { sqlType => 'VARCHAR(255)', maxLen => 255, avgLen => 100, sort => "",
352 : parrello 1.75 indexMod => 0, notes => "character string, 0 to 255 characters"},
353 : parrello 1.73 text => { sqlType => 'TEXT', maxLen => 1000000000, avgLen => 500, sort => "",
354 : parrello 1.75 indexMod => 255, notes => "character string, nearly unlimited length, only first 255 characters are indexed"},
355 : parrello 1.73 date => { sqlType => 'BIGINT', maxLen => 80, avgLen => 8, sort => "n",
356 : parrello 1.75 indexMod => 0, notes => "signed, 64-bit integer"},
357 : parrello 1.73 float => { sqlType => 'DOUBLE PRECISION', maxLen => 40, avgLen => 8, sort => "g",
358 : parrello 1.75 indexMod => 0, notes => "64-bit double precision floating-point number"},
359 : parrello 1.73 boolean => { sqlType => 'SMALLINT', maxLen => 1, avgLen => 1, sort => "n",
360 : parrello 1.75 indexMod => 0, notes => "boolean value: 0 if false, 1 if true"},
361 : parrello 1.42 'hash-string' =>
362 : parrello 1.73 { sqlType => 'VARCHAR(22)', maxLen => 22, avgLen => 22, sort => "",
363 : parrello 1.75 indexMod => 0, notes => "string stored in digested form, used for certain types of key fields"},
364 : parrello 1.44 'id-string' =>
365 : parrello 1.73 { sqlType => 'VARCHAR(25)', maxLen => 25, avgLen => 25, sort => "",
366 : parrello 1.75 indexMod => 0, notes => "character string, 0 to 25 characters"},
367 : parrello 1.10 'key-string' =>
368 : parrello 1.73 { sqlType => 'VARCHAR(40)', maxLen => 40, avgLen => 10, sort => "",
369 : parrello 1.75 indexMod => 0, notes => "character string, 0 to 40 characters"},
370 : parrello 1.10 'name-string' =>
371 : parrello 1.73 { sqlType => 'VARCHAR(80)', maxLen => 80, avgLen => 40, sort => "",
372 : parrello 1.75 indexMod => 0, notes => "character string, 0 to 80 characters"},
373 : parrello 1.10 'medium-string' =>
374 : parrello 1.73 { sqlType => 'VARCHAR(160)', maxLen => 160, avgLen => 40, sort => "",
375 : parrello 1.75 indexMod => 0, notes => "character string, 0 to 160 characters"},
376 : parrello 1.10 );
377 : parrello 1.1
378 :     # Table translating arities into natural language.
379 :     my %ArityTable = ( '11' => 'one-to-one',
380 : parrello 1.10 '1M' => 'one-to-many',
381 :     'MM' => 'many-to-many'
382 :     );
383 : parrello 1.1
384 : parrello 1.73 # Options for XML input and output.
385 :    
386 :     my %XmlOptions = (GroupTags => { Relationships => 'Relationship',
387 :     Entities => 'Entity',
388 :     Fields => 'Field',
389 :     Indexes => 'Index',
390 :     IndexFields => 'IndexField'
391 :     },
392 :     KeyAttr => { Relationship => 'name',
393 :     Entity => 'name',
394 :     Field => 'name'
395 :     },
396 :     SuppressEmpty => 1,
397 :     );
398 :    
399 :     my %XmlInOpts = (
400 : parrello 1.78 ForceArray => ['Field', 'Index', 'IndexField', 'Relationship', 'Entity'],
401 : parrello 1.73 ForceContent => 1,
402 :     NormalizeSpace => 2,
403 :     );
404 :     my %XmlOutOpts = (
405 :     RootName => 'Database',
406 :     XMLDecl => 1,
407 :     );
408 : parrello 1.1
409 :    
410 :     =head2 Public Methods
411 :    
412 :     =head3 new
413 :    
414 : parrello 1.5 C<< my $database = ERDB->new($dbh, $metaFileName); >>
415 : parrello 1.1
416 :     Create a new ERDB object.
417 :    
418 :     =over 4
419 :    
420 :     =item dbh
421 :    
422 :     DBKernel database object for the target database.
423 :    
424 :     =item metaFileName
425 :    
426 :     Name of the XML file containing the metadata.
427 :    
428 :     =back
429 :    
430 :     =cut
431 :    
432 :     sub new {
433 : parrello 1.10 # Get the parameters.
434 :     my ($class, $dbh, $metaFileName, $options) = @_;
435 :     # Load the meta-data.
436 :     my $metaData = _LoadMetaData($metaFileName);
437 :     # Create the object.
438 :     my $self = { _dbh => $dbh,
439 :     _metaData => $metaData
440 :     };
441 :     # Bless and return it.
442 :     bless $self, $class;
443 :     return $self;
444 : parrello 1.1 }
445 :    
446 :     =head3 ShowMetaData
447 :    
448 : parrello 1.18 C<< $erdb->ShowMetaData($fileName); >>
449 : parrello 1.1
450 :     This method outputs a description of the database. This description can be used to help users create
451 :     the data to be loaded into the relations.
452 :    
453 :     =over 4
454 :    
455 :     =item filename
456 :    
457 :     The name of the output file.
458 :    
459 :     =back
460 :    
461 :     =cut
462 :    
463 :     sub ShowMetaData {
464 : parrello 1.10 # Get the parameters.
465 :     my ($self, $filename) = @_;
466 :     # Get the metadata and the title string.
467 :     my $metadata = $self->{_metaData};
468 :     # Get the title string.
469 :     my $title = $metadata->{Title};
470 :     # Get the entity and relationship lists.
471 :     my $entityList = $metadata->{Entities};
472 :     my $relationshipList = $metadata->{Relationships};
473 :     # Open the output file.
474 :     open(HTMLOUT, ">$filename") || Confess("Could not open MetaData display file $filename: $!");
475 :     Trace("Building MetaData table of contents.") if T(4);
476 :     # Write the HTML heading stuff.
477 :     print HTMLOUT "<html>\n<head>\n<title>$title</title>\n";
478 :     print HTMLOUT "</head>\n<body>\n";
479 : parrello 1.45 # Write the documentation.
480 :     print HTMLOUT $self->DisplayMetaData();
481 :     # Close the document.
482 :     print HTMLOUT "</body>\n</html>\n";
483 :     # Close the file.
484 :     close HTMLOUT;
485 :     }
486 :    
487 :     =head3 DisplayMetaData
488 :    
489 :     C<< my $html = $erdb->DisplayMetaData(); >>
490 :    
491 :     Return an HTML description of the database. This description can be used to help users create
492 :     the data to be loaded into the relations and form queries. The output is raw includable HTML
493 :     without any HEAD or BODY tags.
494 :    
495 :     =over 4
496 :    
497 :     =item filename
498 :    
499 :     The name of the output file.
500 :    
501 :     =back
502 :    
503 :     =cut
504 :    
505 :     sub DisplayMetaData {
506 :     # Get the parameters.
507 :     my ($self) = @_;
508 :     # Get the metadata and the title string.
509 :     my $metadata = $self->{_metaData};
510 :     # Get the title string.
511 :     my $title = $metadata->{Title};
512 :     # Get the entity and relationship lists.
513 :     my $entityList = $metadata->{Entities};
514 :     my $relationshipList = $metadata->{Relationships};
515 :     # Declare the return variable.
516 :     my $retVal = "";
517 :     # Open the output file.
518 :     Trace("Building MetaData table of contents.") if T(4);
519 : parrello 1.10 # Here we do the table of contents. It starts as an unordered list of section names. Each
520 :     # section contains an ordered list of entity or relationship subsections.
521 : parrello 1.45 $retVal .= "<ul>\n<li><a href=\"#EntitiesSection\">Entities</a>\n<ol>\n";
522 : parrello 1.10 # Loop through the Entities, displaying a list item for each.
523 :     foreach my $key (sort keys %{$entityList}) {
524 :     # Display this item.
525 : parrello 1.45 $retVal .= "<li><a href=\"#$key\">$key</a></li>\n";
526 : parrello 1.10 }
527 :     # Close off the entity section and start the relationship section.
528 : parrello 1.45 $retVal .= "</ol></li>\n<li><a href=\"#RelationshipsSection\">Relationships</a>\n<ol>\n";
529 : parrello 1.10 # Loop through the Relationships.
530 :     foreach my $key (sort keys %{$relationshipList}) {
531 :     # Display this item.
532 :     my $relationshipTitle = _ComputeRelationshipSentence($key, $relationshipList->{$key});
533 : parrello 1.45 $retVal .= "<li><a href=\"#$key\">$relationshipTitle</a></li>\n";
534 : parrello 1.10 }
535 :     # Close off the relationship section and list the join table section.
536 : parrello 1.45 $retVal .= "</ol></li>\n<li><a href=\"#JoinTable\">Join Table</a></li>\n";
537 : parrello 1.10 # Close off the table of contents itself.
538 : parrello 1.45 $retVal .= "</ul>\n";
539 : parrello 1.10 # Now we start with the actual data. Denote we're starting the entity section.
540 : parrello 1.45 $retVal .= "<a name=\"EntitiesSection\"></a><h2>Entities</h2>\n";
541 : parrello 1.10 # Loop through the entities.
542 :     for my $key (sort keys %{$entityList}) {
543 :     Trace("Building MetaData entry for $key entity.") if T(4);
544 :     # Create the entity header. It contains a bookmark and the entity name.
545 : parrello 1.45 $retVal .= "<a name=\"$key\"></a><h3>$key</h3>\n";
546 : parrello 1.10 # Get the entity data.
547 :     my $entityData = $entityList->{$key};
548 :     # If there's descriptive text, display it.
549 :     if (my $notes = $entityData->{Notes}) {
550 : parrello 1.73 $retVal .= "<p>" . HTMLNote($notes->{content}) . "</p>\n";
551 : parrello 1.10 }
552 : parrello 1.83 # See if we need a list of the entity's relationships.
553 :     my $relCount = keys %{$relationshipList};
554 :     if ($relCount > 0) {
555 :     # First, we set up the relationship subsection.
556 :     $retVal .= "<h4>Relationships for <b>$key</b></h4>\n<ul>\n";
557 :     # Loop through the relationships.
558 :     for my $relationship (sort keys %{$relationshipList}) {
559 :     # Get the relationship data.
560 :     my $relationshipStructure = $relationshipList->{$relationship};
561 :     # Only use the relationship if if has this entity in its FROM or TO fields.
562 :     if ($relationshipStructure->{from} eq $key || $relationshipStructure->{to} eq $key) {
563 :     # Get the relationship sentence and append the arity.
564 :     my $relationshipDescription = _ComputeRelationshipSentence($relationship, $relationshipStructure);
565 :     # Display the relationship data.
566 :     $retVal .= "<li><a href=\"#$relationship\">$relationshipDescription</a></li>\n";
567 :     }
568 : parrello 1.10 }
569 : parrello 1.83 # Close off the relationship list.
570 :     $retVal .= "</ul>\n";
571 : parrello 1.10 }
572 :     # Get the entity's relations.
573 :     my $relationList = $entityData->{Relations};
574 :     # Create a header for the relation subsection.
575 : parrello 1.45 $retVal .= "<h4>Relations for <b>$key</b></h4>\n";
576 : parrello 1.10 # Loop through the relations, displaying them.
577 :     for my $relation (sort keys %{$relationList}) {
578 :     my $htmlString = _ShowRelationTable($relation, $relationList->{$relation});
579 : parrello 1.45 $retVal .= $htmlString;
580 : parrello 1.10 }
581 :     }
582 :     # Denote we're starting the relationship section.
583 : parrello 1.45 $retVal .= "<a name=\"RelationshipsSection\"></a><h2>Relationships</h2>\n";
584 : parrello 1.10 # Loop through the relationships.
585 :     for my $key (sort keys %{$relationshipList}) {
586 :     Trace("Building MetaData entry for $key relationship.") if T(4);
587 :     # Get the relationship's structure.
588 :     my $relationshipStructure = $relationshipList->{$key};
589 :     # Create the relationship header.
590 :     my $headerText = _ComputeRelationshipHeading($key, $relationshipStructure);
591 : parrello 1.45 $retVal .= "<h3><a name=\"$key\"></a>$headerText</h3>\n";
592 : parrello 1.10 # Get the entity names.
593 :     my $fromEntity = $relationshipStructure->{from};
594 :     my $toEntity = $relationshipStructure->{to};
595 :     # Describe the relationship arity. Note there's a bit of trickiness involving recursive
596 :     # many-to-many relationships. In a normal many-to-many we use two sentences to describe
597 :     # the arity (one for each direction). This is a bad idea for a recursive relationship,
598 :     # since both sentences will say the same thing.
599 :     my $arity = $relationshipStructure->{arity};
600 :     if ($arity eq "11") {
601 : parrello 1.45 $retVal .= "<p>Each <b>$fromEntity</b> relates to at most one <b>$toEntity</b>.\n";
602 : parrello 1.10 } else {
603 : parrello 1.45 $retVal .= "<p>Each <b>$fromEntity</b> relates to multiple <b>$toEntity</b>s.\n";
604 : parrello 1.10 if ($arity eq "MM" && $fromEntity ne $toEntity) {
605 : parrello 1.45 $retVal .= "Each <b>$toEntity</b> relates to multiple <b>$fromEntity</b>s.\n";
606 : parrello 1.10 }
607 :     }
608 : parrello 1.45 $retVal .= "</p>\n";
609 : parrello 1.10 # If there are notes on this relationship, display them.
610 :     if (my $notes = $relationshipStructure->{Notes}) {
611 : parrello 1.73 $retVal .= "<p>" . HTMLNote($notes->{content}) . "</p>\n";
612 : parrello 1.10 }
613 :     # Generate the relationship's relation table.
614 :     my $htmlString = _ShowRelationTable($key, $relationshipStructure->{Relations}->{$key});
615 : parrello 1.45 $retVal .= $htmlString;
616 : parrello 1.10 }
617 :     Trace("Building MetaData join table.") if T(4);
618 :     # Denote we're starting the join table.
619 : parrello 1.45 $retVal .= "<a name=\"JoinTable\"></a><h3>Join Table</h3>\n";
620 : parrello 1.10 # Create a table header.
621 : parrello 1.45 $retVal .= _OpenTable("Join Table", "Source", "Target", "Join Condition");
622 : parrello 1.10 # Loop through the joins.
623 :     my $joinTable = $metadata->{Joins};
624 :     my @joinKeys = keys %{$joinTable};
625 :     for my $joinKey (sort @joinKeys) {
626 :     # Separate out the source, the target, and the join clause.
627 :     $joinKey =~ m!^([^/]+)/(.+)$!;
628 :     my ($sourceRelation, $targetRelation) = ($1, $2);
629 : parrello 1.30 Trace("Join with key $joinKey is from $sourceRelation to $targetRelation.") if T(Joins => 4);
630 : parrello 1.10 my $source = $self->ComputeObjectSentence($sourceRelation);
631 :     my $target = $self->ComputeObjectSentence($targetRelation);
632 :     my $clause = $joinTable->{$joinKey};
633 :     # Display them in a table row.
634 : parrello 1.45 $retVal .= "<tr><td>$source</td><td>$target</td><td>$clause</td></tr>\n";
635 : parrello 1.10 }
636 :     # Close the table.
637 : parrello 1.45 $retVal .= _CloseTable();
638 :     Trace("Built MetaData HTML.") if T(3);
639 :     # Return the HTML.
640 :     return $retVal;
641 : parrello 1.1 }
642 :    
643 :     =head3 DumpMetaData
644 :    
645 : parrello 1.18 C<< $erdb->DumpMetaData(); >>
646 : parrello 1.1
647 :     Return a dump of the metadata structure.
648 :    
649 :     =cut
650 :    
651 :     sub DumpMetaData {
652 : parrello 1.10 # Get the parameters.
653 :     my ($self) = @_;
654 :     # Dump the meta-data.
655 :     return Data::Dumper::Dumper($self->{_metaData});
656 : parrello 1.1 }
657 :    
658 : parrello 1.74 =head3 FindIndexForEntity
659 :    
660 :     C<< my $indexFound = ERDB::FindIndexForEntity($xml, $entityName, $attributeName); >>
661 :    
662 :     This method locates the entry in an entity's index list that begins with the
663 :     specified attribute name. If the entity has no index list, one will be
664 :     created. This method works on raw XML, not a live ERDB object.
665 :    
666 :     =over 4
667 :    
668 :     =item xml
669 :    
670 :     The raw XML structure defining the database.
671 :    
672 :     =item entityName
673 :    
674 :     The name of the relevant entity.
675 :    
676 :     =item attributeName
677 :    
678 :     The name of the attribute relevant to the search.
679 :    
680 :     =item RETURN
681 :    
682 :     The numerical index in the index list of the index entry for the specified entity and
683 :     attribute, or C<undef> if no such index exists.
684 :    
685 :     =back
686 :    
687 :     =cut
688 :    
689 :     sub FindIndexForEntity {
690 :     # Get the parameters.
691 :     my ($xml, $entityName, $attributeName) = @_;
692 :     # Declare the return variable.
693 :     my $retVal;
694 :     # Get the named entity.
695 :     my $entityData = $xml->{Entities}->{$entityName};
696 :     if (! $entityData) {
697 :     Confess("Entity $entityName not found in DBD structure.");
698 :     } else {
699 :     # Insure it has an index list.
700 :     if (! exists $entityData->{Indexes}) {
701 :     $entityData->{Indexes} = [];
702 :     } else {
703 :     # Search for the desired index.
704 :     my $indexList = $entityData->{Indexes};
705 :     my $n = scalar @{$indexList};
706 :     Trace("Searching $n indexes in index list for $entityName.") if T(2);
707 :     # We use an indexed FOR here because we're returning an
708 :     # index number instead of an object. We do THAT so we can
709 :     # delete the index from the list if needed.
710 :     for (my $i = 0; $i < $n && !defined($retVal); $i++) {
711 :     my $index = $indexList->[$i];
712 :     my $fields = $index->{IndexFields};
713 :     # Technically this IF should be safe (that is, we are guaranteed
714 :     # the existence of a "$fields->[0]"), because when we load the XML
715 :     # we have SuppressEmpty specified.
716 :     if ($fields->[0]->{name} eq $attributeName) {
717 :     $retVal = $i;
718 :     }
719 :     }
720 :     }
721 :     }
722 :     Trace("Index for $attributeName of $entityName found at position $retVal.") if defined($retVal) && T(3);
723 :     Trace("Index for $attributeName not found in $entityName.") if !defined($retVal) && T(3);
724 :     # Return the result.
725 :     return $retVal;
726 :     }
727 :    
728 : parrello 1.1 =head3 CreateTables
729 :    
730 : parrello 1.18 C<< $erdb->CreateTables(); >>
731 : parrello 1.1
732 :     This method creates the tables for the database from the metadata structure loaded by the
733 :     constructor. It is expected this function will only be used on rare occasions, when the
734 : parrello 1.2 user needs to start with an empty database. Otherwise, the L</LoadTables> method can be
735 : parrello 1.1 used by itself with the truncate flag turned on.
736 :    
737 :     =cut
738 :    
739 :     sub CreateTables {
740 : parrello 1.10 # Get the parameters.
741 :     my ($self) = @_;
742 : parrello 1.23 # Get the relation names.
743 :     my @relNames = $self->GetTableNames();
744 :     # Loop through the relations.
745 :     for my $relationName (@relNames) {
746 :     # Create a table for this relation.
747 :     $self->CreateTable($relationName);
748 :     Trace("Relation $relationName created.") if T(2);
749 : parrello 1.10 }
750 : parrello 1.1 }
751 :    
752 :     =head3 CreateTable
753 :    
754 : parrello 1.18 C<< $erdb->CreateTable($tableName, $indexFlag, $estimatedRows); >>
755 : parrello 1.1
756 :     Create the table for a relation and optionally create its indexes.
757 :    
758 :     =over 4
759 :    
760 :     =item relationName
761 :    
762 :     Name of the relation (which will also be the table name).
763 :    
764 : parrello 1.18 =item indexFlag
765 : parrello 1.1
766 :     TRUE if the indexes for the relation should be created, else FALSE. If FALSE,
767 :     L</CreateIndexes> must be called later to bring the indexes into existence.
768 :    
769 : parrello 1.18 =item estimatedRows (optional)
770 :    
771 :     If specified, the estimated maximum number of rows for the relation. This
772 :     information allows the creation of tables using storage engines that are
773 :     faster but require size estimates, such as MyISAM.
774 :    
775 : parrello 1.1 =back
776 :    
777 :     =cut
778 :    
779 :     sub CreateTable {
780 : parrello 1.10 # Get the parameters.
781 : parrello 1.18 my ($self, $relationName, $indexFlag, $estimatedRows) = @_;
782 : parrello 1.10 # Get the database handle.
783 :     my $dbh = $self->{_dbh};
784 :     # Get the relation data and determine whether or not the relation is primary.
785 :     my $relationData = $self->_FindRelation($relationName);
786 :     my $rootFlag = $self->_IsPrimary($relationName);
787 :     # Create a list of the field data.
788 :     my @fieldList;
789 :     for my $fieldData (@{$relationData->{Fields}}) {
790 :     # Assemble the field name and type.
791 :     my $fieldName = _FixName($fieldData->{name});
792 :     my $fieldString = "$fieldName $TypeTable{$fieldData->{type}}->{sqlType} NOT NULL ";
793 :     # Push the result into the field list.
794 :     push @fieldList, $fieldString;
795 :     }
796 :     # If this is a root table, add the "new_record" flag. It defaults to 0, so
797 :     if ($rootFlag) {
798 :     push @fieldList, "new_record $TypeTable{boolean}->{sqlType} NOT NULL DEFAULT 0";
799 :     }
800 :     # Convert the field list into a comma-delimited string.
801 :     my $fieldThing = join(', ', @fieldList);
802 :     # Insure the table is not already there.
803 :     $dbh->drop_table(tbl => $relationName);
804 :     Trace("Table $relationName dropped.") if T(2);
805 : parrello 1.18 # If there are estimated rows, create an estimate so we can take advantage of
806 :     # faster DB technologies.
807 :     my $estimation = undef;
808 :     if ($estimatedRows) {
809 :     $estimation = [$self->EstimateRowSize($relationName), $estimatedRows];
810 :     }
811 : parrello 1.10 # Create the table.
812 :     Trace("Creating table $relationName: $fieldThing") if T(2);
813 : parrello 1.18 $dbh->create_table(tbl => $relationName, flds => $fieldThing, estimates => $estimation);
814 : parrello 1.10 Trace("Relation $relationName created in database.") if T(2);
815 : parrello 1.70 # If we want to build the indexes, we do it here. Note that the full-text search
816 :     # index will not be built until the table has been loaded.
817 : parrello 1.10 if ($indexFlag) {
818 :     $self->CreateIndex($relationName);
819 :     }
820 : parrello 1.1 }
821 :    
822 : parrello 1.31 =head3 VerifyFields
823 :    
824 :     C<< my $count = $erdb->VerifyFields($relName, \@fieldList); >>
825 :    
826 :     Run through the list of proposed field values, insuring that all the character fields are
827 :     below the maximum length. If any fields are too long, they will be truncated in place.
828 :    
829 :     =over 4
830 :    
831 :     =item relName
832 :    
833 :     Name of the relation for which the specified fields are destined.
834 :    
835 :     =item fieldList
836 :    
837 :     Reference to a list, in order, of the fields to be put into the relation.
838 :    
839 :     =item RETURN
840 :    
841 :     Returns the number of fields truncated.
842 :    
843 :     =back
844 :    
845 :     =cut
846 :    
847 :     sub VerifyFields {
848 :     # Get the parameters.
849 :     my ($self, $relName, $fieldList) = @_;
850 :     # Initialize the return value.
851 :     my $retVal = 0;
852 :     # Get the relation definition.
853 :     my $relData = $self->_FindRelation($relName);
854 :     # Get the list of field descriptors.
855 :     my $fieldTypes = $relData->{Fields};
856 :     my $fieldCount = scalar @{$fieldTypes};
857 :     # Loop through the two lists.
858 :     for (my $i = 0; $i < $fieldCount; $i++) {
859 :     # Get the type of the current field.
860 :     my $fieldType = $fieldTypes->[$i]->{type};
861 :     # If it's a character field, verify the length.
862 :     if ($fieldType =~ /string/) {
863 :     my $maxLen = $TypeTable{$fieldType}->{maxLen};
864 :     my $oldString = $fieldList->[$i];
865 :     if (length($oldString) > $maxLen) {
866 :     # Here it's too big, so we truncate it.
867 :     Trace("Truncating field $i in relation $relName to $maxLen characters from \"$oldString\".") if T(1);
868 :     $fieldList->[$i] = substr $oldString, 0, $maxLen;
869 :     $retVal++;
870 :     }
871 :     }
872 :     }
873 :     # Return the truncation count.
874 :     return $retVal;
875 :     }
876 :    
877 : parrello 1.42 =head3 DigestFields
878 :    
879 :     C<< $erdb->DigestFields($relName, $fieldList); >>
880 :    
881 :     Digest the strings in the field list that correspond to data type C<hash-string> in the
882 :     specified relation.
883 :    
884 :     =over 4
885 :    
886 :     =item relName
887 :    
888 :     Name of the relation to which the fields belong.
889 :    
890 :     =item fieldList
891 :    
892 :     List of field contents to be loaded into the relation.
893 :    
894 :     =back
895 :    
896 :     =cut
897 :     #: Return Type ;
898 :     sub DigestFields {
899 :     # Get the parameters.
900 :     my ($self, $relName, $fieldList) = @_;
901 :     # Get the relation definition.
902 :     my $relData = $self->_FindRelation($relName);
903 :     # Get the list of field descriptors.
904 :     my $fieldTypes = $relData->{Fields};
905 :     my $fieldCount = scalar @{$fieldTypes};
906 :     # Loop through the two lists.
907 :     for (my $i = 0; $i < $fieldCount; $i++) {
908 :     # Get the type of the current field.
909 :     my $fieldType = $fieldTypes->[$i]->{type};
910 :     # If it's a hash string, digest it in place.
911 :     if ($fieldType eq 'hash-string') {
912 : parrello 1.46 $fieldList->[$i] = $self->DigestKey($fieldList->[$i]);
913 : parrello 1.42 }
914 :     }
915 :     }
916 :    
917 : parrello 1.46 =head3 DigestKey
918 :    
919 :     C<< my $digested = $erdb->DigestKey($keyValue); >>
920 :    
921 :     Return the digested value of a symbolic key. The digested value can then be plugged into a
922 :     key-based search into a table with key-type hash-string.
923 :    
924 :     Currently the digesting process is independent of the database structure, but that may not
925 :     always be the case, so this is an instance method instead of a static method.
926 :    
927 :     =over 4
928 :    
929 :     =item keyValue
930 :    
931 :     Key value to digest.
932 :    
933 :     =item RETURN
934 :    
935 : parrello 1.56 Digested value of the key.
936 : parrello 1.46
937 :     =back
938 :    
939 :     =cut
940 :    
941 :     sub DigestKey {
942 :     # Get the parameters.
943 :     my ($self, $keyValue) = @_;
944 :     # Compute the digest.
945 :     my $retVal = md5_base64($keyValue);
946 :     # Return the result.
947 :     return $retVal;
948 :     }
949 :    
950 : parrello 1.1 =head3 CreateIndex
951 :    
952 : parrello 1.18 C<< $erdb->CreateIndex($relationName); >>
953 : parrello 1.1
954 :     Create the indexes for a relation. If a table is being loaded from a large source file (as
955 : parrello 1.12 is the case in L</LoadTable>), it is sometimes best to create the indexes after the load.
956 :     If that is the case, then L</CreateTable> should be called with the index flag set to
957 :     FALSE, and this method used after the load to create the indexes for the table.
958 : parrello 1.1
959 :     =cut
960 :    
961 :     sub CreateIndex {
962 : parrello 1.10 # Get the parameters.
963 :     my ($self, $relationName) = @_;
964 :     # Get the relation's descriptor.
965 :     my $relationData = $self->_FindRelation($relationName);
966 :     # Get the database handle.
967 :     my $dbh = $self->{_dbh};
968 :     # Now we need to create this relation's indexes. We do this by looping through its index table.
969 :     my $indexHash = $relationData->{Indexes};
970 :     for my $indexName (keys %{$indexHash}) {
971 :     my $indexData = $indexHash->{$indexName};
972 :     # Get the index's field list.
973 : parrello 1.75 my @rawFields = @{$indexData->{IndexFields}};
974 :     # Get a hash of the relation's field types.
975 :     my %types = map { $_->{name} => $_->{type} } @{$relationData->{Fields}};
976 : parrello 1.78 # We need to check for text fields so we can append a length limitation for them. To do
977 : parrello 1.75 # that, we need the relation's field list.
978 :     my $relFields = $relationData->{Fields};
979 :     for (my $i = 0; $i <= $#rawFields; $i++) {
980 :     # Get the field type.
981 :     my $field = $rawFields[$i];
982 :     my $type = $types{$field};
983 :     # Ask if it requires using prefix notation for the index.
984 :     my $mod = $TypeTable{$type}->{indexMod};
985 :     Trace("Field $field ($i) in $relationName has type $type and indexMod $mod.") if T(3);
986 :     if ($mod) {
987 :     # Append the prefix length to the field name,
988 :     $rawFields[$i] .= "($mod)";
989 :     }
990 :     }
991 :     my @fieldList = _FixNames(@rawFields);
992 : parrello 1.10 my $flds = join(', ', @fieldList);
993 :     # Get the index's uniqueness flag.
994 : parrello 1.70 my $unique = (exists $indexData->{Unique} ? 'unique' : undef);
995 : parrello 1.10 # Create the index.
996 : parrello 1.24 my $rv = $dbh->create_index(idx => $indexName, tbl => $relationName,
997 : parrello 1.70 flds => $flds, kind => $unique);
998 : parrello 1.24 if ($rv) {
999 :     Trace("Index created: $indexName for $relationName ($flds)") if T(1);
1000 :     } else {
1001 :     Confess("Error creating index $indexName for $relationName using ($flds): " . $dbh->error_message());
1002 :     }
1003 : parrello 1.10 }
1004 : parrello 1.1 }
1005 :    
1006 : parrello 1.75 =head3 GetSecondaryFields
1007 :    
1008 :     C<< my %fieldTuples = $erdb->GetSecondaryFields($entityName); >>
1009 :    
1010 :     This method will return a list of the name and type of each of the secondary
1011 :     fields for a specified entity. Secondary fields are stored in two-column tables
1012 :     in addition to the primary entity table. This enables the field to have no value
1013 :     or to have multiple values.
1014 :    
1015 :     =over 4
1016 :    
1017 :     =item entityName
1018 :    
1019 :     Name of the entity whose secondary fields are desired.
1020 :    
1021 :     =item RETURN
1022 :    
1023 :     Returns a hash mapping the field names to their field types.
1024 :    
1025 :     =back
1026 :    
1027 :     =cut
1028 :    
1029 :     sub GetSecondaryFields {
1030 :     # Get the parameters.
1031 :     my ($self, $entityName) = @_;
1032 :     # Declare the return variable.
1033 :     my %retVal = ();
1034 :     # Look for the entity.
1035 :     my $table = $self->GetFieldTable($entityName);
1036 :     # Loop through the fields, pulling out the secondaries.
1037 :     for my $field (sort keys %{$table}) {
1038 :     if ($table->{$field}->{relation} ne $entityName) {
1039 :     # Here we have a secondary field.
1040 :     $retVal{$field} = $table->{$field}->{type};
1041 :     }
1042 :     }
1043 :     # Return the result.
1044 :     return %retVal;
1045 :     }
1046 :    
1047 :     =head3 GetFieldRelationName
1048 :    
1049 :     C<< my $name = $erdb->GetFieldRelationName($objectName, $fieldName); >>
1050 :    
1051 :     Return the name of the relation containing a specified field.
1052 :    
1053 :     =over 4
1054 :    
1055 :     =item objectName
1056 :    
1057 :     Name of the entity or relationship containing the field.
1058 :    
1059 :     =item fieldName
1060 :    
1061 :     Name of the relevant field in that entity or relationship.
1062 :    
1063 :     =item RETURN
1064 :    
1065 :     Returns the name of the database relation containing the field, or C<undef> if
1066 :     the field does not exist.
1067 :    
1068 :     =back
1069 :    
1070 :     =cut
1071 :    
1072 :     sub GetFieldRelationName {
1073 :     # Get the parameters.
1074 :     my ($self, $objectName, $fieldName) = @_;
1075 :     # Declare the return variable.
1076 :     my $retVal;
1077 :     # Get the object field table.
1078 :     my $table = $self->GetFieldTable($objectName);
1079 :     # Only proceed if the field exists.
1080 :     if (exists $table->{$fieldName}) {
1081 :     # Determine the name of the relation that contains this field.
1082 :     $retVal = $table->{$fieldName}->{relation};
1083 :     }
1084 :     # Return the result.
1085 :     return $retVal;
1086 :     }
1087 :    
1088 :     =head3 DeleteValue
1089 :    
1090 :     C<< my $numDeleted = $erdb->DeleteValue($entityName, $id, $fieldName, $fieldValue); >>
1091 :    
1092 :     Delete secondary field values from the database. This method can be used to delete all
1093 :     values of a specified field for a particular entity instance, or only a single value.
1094 :    
1095 :     Secondary fields are stored in two-column relations separate from an entity's primary
1096 :     table, and as a result a secondary field can legitimately have no value or multiple
1097 :     values. Therefore, it makes sense to talk about deleting secondary fields where it
1098 :     would not make sense for primary fields.
1099 :    
1100 :     =over 4
1101 :    
1102 :     =item entityName
1103 :    
1104 :     Name of the entity from which the fields are to be deleted.
1105 :    
1106 :     =item id
1107 :    
1108 :     ID of the entity instance to be processed. If the instance is not found, this
1109 : parrello 1.76 method will have no effect. If C<undef> is specified, all values for all of
1110 :     the entity instances will be deleted.
1111 : parrello 1.75
1112 :     =item fieldName
1113 :    
1114 :     Name of the field whose values are to be deleted.
1115 :    
1116 :     =item fieldValue (optional)
1117 :    
1118 :     Value to be deleted. If not specified, then all values of the specified field
1119 :     will be deleted for the entity instance. If specified, then only the values which
1120 :     match this parameter will be deleted.
1121 :    
1122 :     =item RETURN
1123 :    
1124 :     Returns the number of rows deleted.
1125 :    
1126 :     =back
1127 :    
1128 :     =cut
1129 :    
1130 :     sub DeleteValue {
1131 :     # Get the parameters.
1132 :     my ($self, $entityName, $id, $fieldName, $fieldValue) = @_;
1133 :     # Declare the return value.
1134 :     my $retVal = 0;
1135 :     # We need to set up an SQL command to do the deletion. First, we
1136 :     # find the name of the field's relation.
1137 :     my $table = $self->GetFieldTable($entityName);
1138 :     my $field = $table->{$fieldName};
1139 :     my $relation = $field->{relation};
1140 :     # Make sure this is a secondary field.
1141 :     if ($relation eq $entityName) {
1142 :     Confess("Cannot delete values of $fieldName for $entityName.");
1143 :     } else {
1144 :     # Set up the SQL command to delete all values.
1145 : parrello 1.76 my $sql = "DELETE FROM $relation";
1146 :     # Build the filter.
1147 :     my @filters = ();
1148 :     my @parms = ();
1149 :     # Check for a filter by ID.
1150 :     if (defined $id) {
1151 :     push @filters, "id = ?";
1152 :     push @parms, $id;
1153 :     }
1154 :     # Check for a filter by value.
1155 : parrello 1.75 if (defined $fieldValue) {
1156 : parrello 1.76 push @filters, "$fieldName = ?";
1157 : parrello 1.75 push @parms, $fieldValue;
1158 :     }
1159 : parrello 1.76 # Append the filters to the command.
1160 :     if (@filters) {
1161 :     $sql .= " WHERE " . join(" AND ", @filters);
1162 :     }
1163 : parrello 1.75 # Execute the command.
1164 :     my $dbh = $self->{_dbh};
1165 :     $retVal = $dbh->SQL($sql, 0, @parms);
1166 :     }
1167 :     # Return the result.
1168 :     return $retVal;
1169 :     }
1170 :    
1171 : parrello 1.1 =head3 LoadTables
1172 :    
1173 : parrello 1.18 C<< my $stats = $erdb->LoadTables($directoryName, $rebuild); >>
1174 : parrello 1.1
1175 :     This method will load the database tables from a directory. The tables must already have been created
1176 :     in the database. (This can be done by calling L</CreateTables>.) The caller passes in a directory name;
1177 :     all of the relations to be loaded must have a file in the directory with the same name as the relation
1178 :     (optionally with a suffix of C<.dtx>). Each file must be a tab-delimited table of field values. Each
1179 :     line of the file will be loaded as a row of the target relation table. The field values should be in
1180 :     the same order as the fields in the relation tables generated by L</ShowMetaData>. The old data is
1181 :     erased before the new data is loaded in.
1182 :    
1183 :     A certain amount of translation automatically takes place. Ctrl-M characters are deleted, and
1184 :     tab and new-line characters inside a field are escaped as C<\t> and C<\n>, respectively. Dates must
1185 :     be entered as a Unix timestamp, that is, as an integer number of seconds since the base epoch.
1186 :    
1187 :     =over 4
1188 :    
1189 :     =item directoryName
1190 :    
1191 :     Name of the directory containing the relation files to be loaded.
1192 :    
1193 :     =item rebuild
1194 :    
1195 :     TRUE if the tables should be dropped and rebuilt, else FALSE. This is, unfortunately, the
1196 :     only way to erase existing data in the tables, since the TRUNCATE command is not supported
1197 :     by all of the DB engines we use.
1198 :    
1199 :     =item RETURN
1200 :    
1201 :     Returns a statistical object describing the number of records read and a list of the error messages.
1202 :    
1203 :     =back
1204 :    
1205 :     =cut
1206 :    
1207 :     sub LoadTables {
1208 : parrello 1.10 # Get the parameters.
1209 :     my ($self, $directoryName, $rebuild) = @_;
1210 :     # Start the timer.
1211 :     my $startTime = gettimeofday;
1212 :     # Clean any trailing slash from the directory name.
1213 :     $directoryName =~ s!/\\$!!;
1214 :     # Declare the return variable.
1215 :     my $retVal = Stats->new();
1216 : parrello 1.23 # Get the relation names.
1217 :     my @relNames = $self->GetTableNames();
1218 :     for my $relationName (@relNames) {
1219 :     # Try to load this relation.
1220 :     my $result = $self->_LoadRelation($directoryName, $relationName, $rebuild);
1221 : parrello 1.10 # Accumulate the statistics.
1222 :     $retVal->Accumulate($result);
1223 :     }
1224 :     # Add the duration of the load to the statistical object.
1225 :     $retVal->Add('duration', gettimeofday - $startTime);
1226 :     # Return the accumulated statistics.
1227 :     return $retVal;
1228 : parrello 1.1 }
1229 :    
1230 : parrello 1.23
1231 : parrello 1.1 =head3 GetTableNames
1232 :    
1233 : parrello 1.18 C<< my @names = $erdb->GetTableNames; >>
1234 : parrello 1.1
1235 :     Return a list of the relations required to implement this database.
1236 :    
1237 :     =cut
1238 :    
1239 :     sub GetTableNames {
1240 : parrello 1.10 # Get the parameters.
1241 :     my ($self) = @_;
1242 :     # Get the relation list from the metadata.
1243 :     my $relationTable = $self->{_metaData}->{RelationTable};
1244 :     # Return the relation names.
1245 :     return keys %{$relationTable};
1246 : parrello 1.1 }
1247 :    
1248 :     =head3 GetEntityTypes
1249 :    
1250 : parrello 1.18 C<< my @names = $erdb->GetEntityTypes; >>
1251 : parrello 1.1
1252 :     Return a list of the entity type names.
1253 :    
1254 :     =cut
1255 :    
1256 :     sub GetEntityTypes {
1257 : parrello 1.10 # Get the database object.
1258 :     my ($self) = @_;
1259 :     # Get the entity list from the metadata object.
1260 :     my $entityList = $self->{_metaData}->{Entities};
1261 :     # Return the list of entity names in alphabetical order.
1262 :     return sort keys %{$entityList};
1263 : parrello 1.1 }
1264 :    
1265 : parrello 1.73 =head3 GetDataTypes
1266 :    
1267 :     C<< my %types = ERDB::GetDataTypes(); >>
1268 :    
1269 :     Return a table of ERDB data types. The table returned is a hash of hashes.
1270 :     The keys of the big hash are the datatypes. Each smaller hash has several
1271 :     values used to manage the data. The most interesting is the SQL type (key
1272 :     C<sqlType>) and the descriptive node (key C<notes>).
1273 :    
1274 :     Note that changing the values in the smaller hashes will seriously break
1275 :     things, so this data should be treated as read-only.
1276 :    
1277 :     =cut
1278 :    
1279 :     sub GetDataTypes {
1280 :     return %TypeTable;
1281 :     }
1282 :    
1283 :    
1284 : parrello 1.20 =head3 IsEntity
1285 :    
1286 :     C<< my $flag = $erdb->IsEntity($entityName); >>
1287 :    
1288 :     Return TRUE if the parameter is an entity name, else FALSE.
1289 :    
1290 :     =over 4
1291 :    
1292 :     =item entityName
1293 :    
1294 :     Object name to be tested.
1295 :    
1296 :     =item RETURN
1297 :    
1298 :     Returns TRUE if the specified string is an entity name, else FALSE.
1299 :    
1300 :     =back
1301 :    
1302 :     =cut
1303 :    
1304 :     sub IsEntity {
1305 :     # Get the parameters.
1306 :     my ($self, $entityName) = @_;
1307 :     # Test to see if it's an entity.
1308 :     return exists $self->{_metaData}->{Entities}->{$entityName};
1309 :     }
1310 :    
1311 : parrello 1.1 =head3 Get
1312 :    
1313 : parrello 1.45 C<< my $query = $erdb->Get(\@objectNames, $filterClause, \@params); >>
1314 : parrello 1.1
1315 :     This method returns a query object for entities of a specified type using a specified filter.
1316 :     The filter is a standard WHERE/ORDER BY clause with question marks as parameter markers and each
1317 :     field name represented in the form B<I<objectName>(I<fieldName>)>. For example, the
1318 :     following call requests all B<Genome> objects for the genus specified in the variable
1319 :     $genus.
1320 :    
1321 : parrello 1.45 C<< $query = $erdb->Get(['Genome'], "Genome(genus) = ?", [$genus]); >>
1322 : parrello 1.1
1323 :     The WHERE clause contains a single question mark, so there is a single additional
1324 :     parameter representing the parameter value. It would also be possible to code
1325 :    
1326 : parrello 1.18 C<< $query = $erdb->Get(['Genome'], "Genome(genus) = \'$genus\'"); >>
1327 : parrello 1.1
1328 :     however, this version of the call would generate a syntax error if there were any quote
1329 :     characters inside the variable C<$genus>.
1330 :    
1331 :     The use of the strange parenthesized notation for field names enables us to distinguish
1332 :     hyphens contained within field names from minus signs that participate in the computation
1333 :     of the WHERE clause. All of the methods that manipulate fields will use this same notation.
1334 :    
1335 :     It is possible to specify multiple entity and relationship names in order to retrieve more than
1336 :     one object's data at the same time, which allows highly complex joined queries. For example,
1337 :    
1338 : parrello 1.45 C<< $query = $erdb->Get(['Genome', 'ComesFrom', 'Source'], "Genome(genus) = ?", [$genus]); >>
1339 : parrello 1.1
1340 :     If multiple names are specified, then the query processor will automatically determine a
1341 :     join path between the entities and relationships. The algorithm used is very simplistic.
1342 : parrello 1.39 In particular, if a relationship is recursive, the path is determined by the order in which
1343 :     the entity and the relationship appear. For example, consider a recursive relationship
1344 :     B<IsParentOf> which relates B<People> objects to other B<People> objects. If the join path is
1345 : parrello 1.1 coded as C<['People', 'IsParentOf']>, then the people returned will be parents. If, however,
1346 :     the join path is C<['IsParentOf', 'People']>, then the people returned will be children.
1347 :    
1348 : parrello 1.39 If an entity or relationship is mentioned twice, the name for the second occurrence will
1349 :     be suffixed with C<2>, the third occurrence will be suffixed with C<3>, and so forth. So,
1350 :     for example, if we have C<['Feature', 'HasContig', 'Contig', 'HasContig']>, then the
1351 :     B<to-link> field of the first B<HasContig> is specified as C<HasContig(to-link)>, while
1352 :     the B<to-link> field of the second B<HasContig> is specified as C<HasContig2(to-link)>.
1353 :    
1354 : parrello 1.1 =over 4
1355 :    
1356 :     =item objectNames
1357 :    
1358 :     List containing the names of the entity and relationship objects to be retrieved.
1359 :    
1360 :     =item filterClause
1361 :    
1362 :     WHERE clause (without the WHERE) to be used to filter and sort the query. The WHERE clause can
1363 :     be parameterized with parameter markers (C<?>). Each field used in the WHERE clause must be
1364 :     specified in the standard form B<I<objectName>(I<fieldName>)>. Any parameters specified
1365 :     in the filter clause should be added to the parameter list as additional parameters. The
1366 :     fields in a filter clause can come from primary entity relations, relationship relations,
1367 :     or secondary entity relations; however, all of the entities and relationships involved must
1368 :     be included in the list of object names.
1369 :    
1370 :     The filter clause can also specify a sort order. To do this, simply follow the filter string
1371 :     with an ORDER BY clause. For example, the following filter string gets all genomes for a
1372 :     particular genus and sorts them by species name.
1373 :    
1374 :     C<< "Genome(genus) = ? ORDER BY Genome(species)" >>
1375 :    
1376 : parrello 1.30 Note that the case is important. Only an uppercase "ORDER BY" with a single space will
1377 :     be processed. The idea is to make it less likely to find the verb by accident.
1378 :    
1379 : parrello 1.1 The rules for field references in a sort order are the same as those for field references in the
1380 :     filter clause in general; however, odd things may happen if a sort field is from a secondary
1381 :     relation.
1382 :    
1383 : parrello 1.39 Finally, you can limit the number of rows returned by adding a LIMIT clause. The LIMIT must
1384 :     be the last thing in the filter clause, and it contains only the word "LIMIT" followed by
1385 :     a positive number. So, for example
1386 :    
1387 :     C<< "Genome(genus) = ? ORDER BY Genome(species) LIMIT 10" >>
1388 :    
1389 :     will only return the first ten genomes for the specified genus. The ORDER BY clause is not
1390 :     required. For example, to just get the first 10 genomes in the B<Genome> table, you could
1391 :     use
1392 :    
1393 :     C<< "LIMIT 10" >>
1394 :    
1395 : parrello 1.45 =item params
1396 : parrello 1.1
1397 : parrello 1.45 Reference to a list of parameter values to be substituted into the filter clause.
1398 : parrello 1.1
1399 :     =item RETURN
1400 :    
1401 :     Returns a B<DBQuery> that can be used to iterate through all of the results.
1402 :    
1403 :     =back
1404 :    
1405 :     =cut
1406 :    
1407 :     sub Get {
1408 : parrello 1.10 # Get the parameters.
1409 : parrello 1.45 my ($self, $objectNames, $filterClause, $params) = @_;
1410 :     # Process the SQL stuff.
1411 :     my ($suffix, $mappedNameListRef, $mappedNameHashRef) =
1412 :     $self->_SetupSQL($objectNames, $filterClause);
1413 :     # Create the query.
1414 :     my $command = "SELECT DISTINCT " . join(".*, ", @{$mappedNameListRef}) .
1415 :     ".* $suffix";
1416 :     my $sth = $self->_GetStatementHandle($command, $params);
1417 : parrello 1.39 # Now we create the relation map, which enables DBQuery to determine the order, name
1418 :     # and mapped name for each object in the query.
1419 :     my @relationMap = ();
1420 : parrello 1.45 for my $mappedName (@{$mappedNameListRef}) {
1421 :     push @relationMap, [$mappedName, $mappedNameHashRef->{$mappedName}];
1422 : parrello 1.39 }
1423 : parrello 1.10 # Return the statement object.
1424 : parrello 1.39 my $retVal = DBQuery::_new($self, $sth, \@relationMap);
1425 : parrello 1.10 return $retVal;
1426 : parrello 1.1 }
1427 :    
1428 : parrello 1.78
1429 :    
1430 : parrello 1.70 =head3 Search
1431 :    
1432 :     C<< my $query = $erdb->Search($searchExpression, $idx, \@objectNames, $filterClause, \@params); >>
1433 :    
1434 :     Perform a full text search with filtering. The search will be against a specified object
1435 :     in the object name list. That object will get an extra field containing the search
1436 :     relevance. Note that except for the search expression, the parameters of this method are
1437 :     the same as those for L</Get> and follow the same rules.
1438 :    
1439 :     =over 4
1440 :    
1441 :     =item searchExpression
1442 :    
1443 : parrello 1.75 Boolean search expression for the text fields of the target object. The default mode for
1444 :     a Boolean search expression is OR, but we want the default to be AND, so we will
1445 :     add a C<+> operator to each word with no other operator before it.
1446 : parrello 1.70
1447 :     =item idx
1448 :    
1449 :     Index in the I<$objectNames> list of the table to be searched in full-text mode.
1450 :    
1451 :     =item objectNames
1452 :    
1453 :     List containing the names of the entity and relationship objects to be retrieved.
1454 :    
1455 :     =item filterClause
1456 :    
1457 :     WHERE clause (without the WHERE) to be used to filter and sort the query. The WHERE clause can
1458 :     be parameterized with parameter markers (C<?>). Each field used in the WHERE clause must be
1459 :     specified in the standard form B<I<objectName>(I<fieldName>)>. Any parameters specified
1460 :     in the filter clause should be added to the parameter list as additional parameters. The
1461 :     fields in a filter clause can come from primary entity relations, relationship relations,
1462 :     or secondary entity relations; however, all of the entities and relationships involved must
1463 :     be included in the list of object names.
1464 :    
1465 :     =item params
1466 :    
1467 :     Reference to a list of parameter values to be substituted into the filter clause.
1468 :    
1469 :     =item RETURN
1470 :    
1471 :     Returns a query object for the specified search.
1472 :    
1473 :     =back
1474 :    
1475 :     =cut
1476 :    
1477 :     sub Search {
1478 :     # Get the parameters.
1479 :     my ($self, $searchExpression, $idx, $objectNames, $filterClause, $params) = @_;
1480 :     # Declare the return variable.
1481 :     my $retVal;
1482 : parrello 1.72 # Create a safety copy of the parameter list. Note we have to be careful to insure
1483 :     # a parameter list exists before we copy it.
1484 :     my @myParams = ();
1485 :     if (defined $params) {
1486 :     @myParams = @{$params};
1487 :     }
1488 : parrello 1.70 # Get the first object's structure so we have access to the searchable fields.
1489 :     my $object1Name = $objectNames->[$idx];
1490 :     my $object1Structure = $self->_GetStructure($object1Name);
1491 :     # Get the field list.
1492 :     if (! exists $object1Structure->{searchFields}) {
1493 :     Confess("No searchable index for $object1Name.");
1494 :     } else {
1495 :     # Get the field list.
1496 :     my @fields = @{$object1Structure->{searchFields}};
1497 : parrello 1.71 # Clean the search expression.
1498 :     my $actualKeywords = $self->CleanKeywords($searchExpression);
1499 : parrello 1.75 # Prefix a "+" to each uncontrolled word. This converts the default
1500 :     # search mode from OR to AND.
1501 : parrello 1.80 $actualKeywords =~ s/(^|\s)(\w|")/$1\+$2/g;
1502 : parrello 1.72 Trace("Actual keywords for search are\n$actualKeywords") if T(3);
1503 : parrello 1.70 # We need two match expressions, one for the filter clause and one in the
1504 :     # query itself. Both will use a parameter mark, so we need to push the
1505 :     # search expression onto the front of the parameter list twice.
1506 : parrello 1.71 unshift @myParams, $actualKeywords, $actualKeywords;
1507 : parrello 1.70 # Build the match expression.
1508 :     my @matchFilterFields = map { "$object1Name." . _FixName($_) } @fields;
1509 :     my $matchClause = "MATCH (" . join(", ", @matchFilterFields) . ") AGAINST (? IN BOOLEAN MODE)";
1510 :     # Process the SQL stuff.
1511 :     my ($suffix, $mappedNameListRef, $mappedNameHashRef) =
1512 :     $self->_SetupSQL($objectNames, $filterClause, $matchClause);
1513 :     # Create the query. Note that the match clause is inserted at the front of
1514 :     # the select fields.
1515 :     my $command = "SELECT DISTINCT $matchClause, " . join(".*, ", @{$mappedNameListRef}) .
1516 :     ".* $suffix";
1517 :     my $sth = $self->_GetStatementHandle($command, \@myParams);
1518 :     # Now we create the relation map, which enables DBQuery to determine the order, name
1519 :     # and mapped name for each object in the query.
1520 :     my @relationMap = _RelationMap($mappedNameHashRef, $mappedNameListRef);
1521 :     # Return the statement object.
1522 :     $retVal = DBQuery::_new($self, $sth, \@relationMap, $object1Name);
1523 :     }
1524 :     return $retVal;
1525 :     }
1526 :    
1527 : parrello 1.45 =head3 GetFlat
1528 :    
1529 :     C<< my @list = $erdb->GetFlat(\@objectNames, $filterClause, \@parameterList, $field); >>
1530 :    
1531 :     This is a variation of L</GetAll> that asks for only a single field per record and
1532 :     returns a single flattened list.
1533 :    
1534 :     =over 4
1535 :    
1536 :     =item objectNames
1537 :    
1538 :     List containing the names of the entity and relationship objects to be retrieved.
1539 :    
1540 :     =item filterClause
1541 :    
1542 :     WHERE/ORDER BY clause (without the WHERE) to be used to filter and sort the query. The WHERE clause can
1543 :     be parameterized with parameter markers (C<?>). Each field used must be specified in the standard form
1544 :     B<I<objectName>(I<fieldName>)>. Any parameters specified in the filter clause should be added to the
1545 :     parameter list as additional parameters. The fields in a filter clause can come from primary
1546 :     entity relations, relationship relations, or secondary entity relations; however, all of the
1547 :     entities and relationships involved must be included in the list of object names.
1548 :    
1549 :     =item parameterList
1550 :    
1551 :     List of the parameters to be substituted in for the parameters marks in the filter clause.
1552 :    
1553 :     =item field
1554 :    
1555 :     Name of the field to be used to get the elements of the list returned.
1556 :    
1557 :     =item RETURN
1558 :    
1559 :     Returns a list of values.
1560 :    
1561 :     =back
1562 :    
1563 :     =cut
1564 :     #: Return Type @;
1565 :     sub GetFlat {
1566 :     # Get the parameters.
1567 :     my ($self, $objectNames, $filterClause, $parameterList, $field) = @_;
1568 :     # Construct the query.
1569 :     my $query = $self->Get($objectNames, $filterClause, $parameterList);
1570 :     # Create the result list.
1571 :     my @retVal = ();
1572 :     # Loop through the records, adding the field values found to the result list.
1573 :     while (my $row = $query->Fetch()) {
1574 :     push @retVal, $row->Value($field);
1575 :     }
1576 :     # Return the list created.
1577 :     return @retVal;
1578 :     }
1579 :    
1580 : parrello 1.73 =head3 SpecialFields
1581 :    
1582 :     C<< my %specials = $erdb->SpecialFields($entityName); >>
1583 :    
1584 :     Return a hash mapping special fields in the specified entity to the value of their
1585 :     C<special> attribute. This enables the subclass to get access to the special field
1586 :     attributes without needed to plumb the internal ERDB data structures.
1587 :    
1588 :     =over 4
1589 :    
1590 :     =item entityName
1591 :    
1592 :     Name of the entity whose special fields are desired.
1593 :    
1594 :     =item RETURN
1595 :    
1596 :     Returns a hash. The keys of the hash are the special field names, and the values
1597 :     are the values from each special field's C<special> attribute.
1598 :    
1599 :     =back
1600 :    
1601 :     =cut
1602 :    
1603 :     sub SpecialFields {
1604 :     # Get the parameters.
1605 :     my ($self, $entityName) = @_;
1606 :     # Declare the return variable.
1607 :     my %retVal = ();
1608 :     # Find the entity's data structure.
1609 : parrello 1.75 my $entityData = $self->{_metaData}->{Entities}->{$entityName};
1610 : parrello 1.73 # Loop through its fields, adding each special field to the return hash.
1611 :     my $fieldHash = $entityData->{Fields};
1612 :     for my $fieldName (keys %{$fieldHash}) {
1613 :     my $fieldData = $fieldHash->{$fieldName};
1614 :     if (exists $fieldData->{special}) {
1615 :     $retVal{$fieldName} = $fieldData->{special};
1616 :     }
1617 :     }
1618 :     # Return the result.
1619 :     return %retVal;
1620 :     }
1621 :    
1622 : parrello 1.32 =head3 Delete
1623 :    
1624 : parrello 1.78 C<< my $stats = $erdb->Delete($entityName, $objectID, %options); >>
1625 : parrello 1.32
1626 :     Delete an entity instance from the database. The instance is deleted along with all entity and
1627 : parrello 1.77 relationship instances dependent on it. The definition of I<dependence> is recursive.
1628 :    
1629 :     An object is always dependent on itself. An object is dependent if it is a 1-to-many or many-to-many
1630 :     relationship connected to a dependent entity or if it is the "to" entity connected to a 1-to-many
1631 : parrello 1.32 dependent relationship.
1632 :    
1633 :     =over 4
1634 :    
1635 :     =item entityName
1636 :    
1637 :     Name of the entity type for the instance being deleted.
1638 :    
1639 :     =item objectID
1640 :    
1641 :     ID of the entity instance to be deleted. If the ID contains a wild card character (C<%>),
1642 :     then it is presumed to by a LIKE pattern.
1643 :    
1644 : parrello 1.78 =item options
1645 : parrello 1.32
1646 : parrello 1.78 A hash detailing the options for this delete operation.
1647 : parrello 1.32
1648 :     =item RETURN
1649 :    
1650 :     Returns a statistics object indicating how many records of each particular table were
1651 :     deleted.
1652 :    
1653 :     =back
1654 :    
1655 : parrello 1.78 The permissible options for this method are as follows.
1656 :    
1657 :     =over 4
1658 :    
1659 :     =item testMode
1660 :    
1661 :     If TRUE, then the delete statements will be traced, but no changes will be made to the database.
1662 :    
1663 :     =item keepRoot
1664 :    
1665 :     If TRUE, then the entity instances will not be deleted, only the dependent records.
1666 :    
1667 :     =back
1668 :    
1669 : parrello 1.32 =cut
1670 :     #: Return Type $%;
1671 :     sub Delete {
1672 :     # Get the parameters.
1673 : parrello 1.78 my ($self, $entityName, $objectID, %options) = @_;
1674 : parrello 1.32 # Declare the return variable.
1675 :     my $retVal = Stats->new();
1676 :     # Get the DBKernel object.
1677 :     my $db = $self->{_dbh};
1678 :     # We're going to generate all the paths branching out from the starting entity. One of
1679 :     # the things we have to be careful about is preventing loops. We'll use a hash to
1680 :     # determine if we've hit a loop.
1681 :     my %alreadyFound = ();
1682 : parrello 1.33 # These next lists will serve as our result stack. We start by pushing object lists onto
1683 : parrello 1.32 # the stack, and then popping them off to do the deletes. This means the deletes will
1684 :     # start with the longer paths before getting to the shorter ones. That, in turn, makes
1685 :     # sure we don't delete records that might be needed to forge relationships back to the
1686 : parrello 1.33 # original item. We have two lists-- one for TO-relationships, and one for
1687 :     # FROM-relationships and entities.
1688 :     my @fromPathList = ();
1689 :     my @toPathList = ();
1690 : parrello 1.78 # This final list is used to remember what work still needs to be done. We push paths
1691 : parrello 1.32 # onto the list, then pop them off to extend the paths. We prime it with the starting
1692 :     # point. Note that we will work hard to insure that the last item on a path in the
1693 : parrello 1.75 # to-do list is always an entity.
1694 : parrello 1.32 my @todoList = ([$entityName]);
1695 :     while (@todoList) {
1696 :     # Get the current path.
1697 :     my $current = pop @todoList;
1698 :     # Copy it into a list.
1699 :     my @stackedPath = @{$current};
1700 :     # Pull off the last item on the path. It will always be an entity.
1701 : parrello 1.78 my $myEntityName = pop @stackedPath;
1702 : parrello 1.32 # Add it to the alreadyFound list.
1703 : parrello 1.78 $alreadyFound{$myEntityName} = 1;
1704 :     # Figure out if we need to delete this entity.
1705 :     if ($myEntityName ne $entityName || ! $options{keepRoot}) {
1706 :     # Get the entity data.
1707 :     my $entityData = $self->_GetStructure($myEntityName);
1708 :     # Loop through the entity's relations. A DELETE command will be needed for each of them.
1709 :     my $relations = $entityData->{Relations};
1710 :     for my $relation (keys %{$relations}) {
1711 :     my @augmentedList = (@stackedPath, $relation);
1712 :     push @fromPathList, \@augmentedList;
1713 :     }
1714 : parrello 1.32 }
1715 :     # Now we need to look for relationships connected to this entity.
1716 :     my $relationshipList = $self->{_metaData}->{Relationships};
1717 :     for my $relationshipName (keys %{$relationshipList}) {
1718 :     my $relationship = $relationshipList->{$relationshipName};
1719 :     # Check the FROM field. We're only interested if it's us.
1720 : parrello 1.78 if ($relationship->{from} eq $myEntityName) {
1721 : parrello 1.32 # Add the path to this relationship.
1722 : parrello 1.78 my @augmentedList = (@stackedPath, $myEntityName, $relationshipName);
1723 : parrello 1.33 push @fromPathList, \@augmentedList;
1724 : parrello 1.32 # Check the arity. If it's MM we're done. If it's 1M
1725 :     # and the target hasn't been seen yet, we want to
1726 :     # stack the entity for future processing.
1727 :     if ($relationship->{arity} eq '1M') {
1728 :     my $toEntity = $relationship->{to};
1729 :     if (! exists $alreadyFound{$toEntity}) {
1730 :     # Here we have a new entity that's dependent on
1731 :     # the current entity, so we need to stack it.
1732 :     my @stackList = (@augmentedList, $toEntity);
1733 : parrello 1.33 push @fromPathList, \@stackList;
1734 : parrello 1.34 } else {
1735 :     Trace("$toEntity ignored because it occurred previously.") if T(4);
1736 : parrello 1.32 }
1737 :     }
1738 :     }
1739 :     # Now check the TO field. In this case only the relationship needs
1740 : parrello 1.33 # deletion.
1741 : parrello 1.78 if ($relationship->{to} eq $myEntityName) {
1742 :     my @augmentedList = (@stackedPath, $myEntityName, $relationshipName);
1743 : parrello 1.33 push @toPathList, \@augmentedList;
1744 : parrello 1.32 }
1745 :     }
1746 :     }
1747 :     # Create the first qualifier for the WHERE clause. This selects the
1748 :     # keys of the primary entity records to be deleted. When we're deleting
1749 : parrello 1.78 # from a dependent table, we construct a join path from the first qualifier
1750 : parrello 1.32 # to the table containing the dependent records to delete.
1751 :     my $qualifier = ($objectID =~ /%/ ? "LIKE ?" : "= ?");
1752 : parrello 1.33 # We need to make two passes. The first is through the to-list, and
1753 :     # the second through the from-list. The from-list is second because
1754 :     # the to-list may need to pass through some of the entities the
1755 :     # from-list would delete.
1756 :     my %stackList = ( from_link => \@fromPathList, to_link => \@toPathList );
1757 :     # Now it's time to do the deletes. We do it in two passes.
1758 :     for my $keyName ('to_link', 'from_link') {
1759 :     # Get the list for this key.
1760 :     my @pathList = @{$stackList{$keyName}};
1761 : parrello 1.34 Trace(scalar(@pathList) . " entries in path list for $keyName.") if T(3);
1762 : parrello 1.33 # Loop through this list.
1763 :     while (my $path = pop @pathList) {
1764 :     # Get the table whose rows are to be deleted.
1765 :     my @pathTables = @{$path};
1766 : parrello 1.37 # Start the DELETE statement. We need to call DBKernel because the
1767 :     # syntax of a DELETE-USING varies among DBMSs.
1768 : parrello 1.33 my $target = $pathTables[$#pathTables];
1769 : parrello 1.37 my $stmt = $db->SetUsing(@pathTables);
1770 : parrello 1.33 # Now start the WHERE. The first thing is the ID field from the starting table. That
1771 :     # starting table will either be the entity relation or one of the entity's
1772 :     # sub-relations.
1773 :     $stmt .= " WHERE $pathTables[0].id $qualifier";
1774 :     # Now we run through the remaining entities in the path, connecting them up.
1775 :     for (my $i = 1; $i <= $#pathTables; $i += 2) {
1776 :     # Connect the current relationship to the preceding entity.
1777 :     my ($entity, $rel) = @pathTables[$i-1,$i];
1778 :     # The style of connection depends on the direction of the relationship.
1779 : parrello 1.35 $stmt .= " AND $entity.id = $rel.$keyName";
1780 : parrello 1.32 if ($i + 1 <= $#pathTables) {
1781 :     # Here there's a next entity, so connect that to the relationship's
1782 :     # to-link.
1783 :     my $entity2 = $pathTables[$i+1];
1784 : parrello 1.35 $stmt .= " AND $rel.to_link = $entity2.id";
1785 : parrello 1.32 }
1786 :     }
1787 : parrello 1.33 # Now we have our desired DELETE statement.
1788 : parrello 1.78 if ($options{testMode}) {
1789 : parrello 1.33 # Here the user wants to trace without executing.
1790 :     Trace($stmt) if T(0);
1791 :     } else {
1792 : parrello 1.77 # Here we can delete. Note that the SQL method dies with a confession
1793 : parrello 1.33 # if an error occurs, so we just go ahead and do it.
1794 : parrello 1.36 Trace("Executing delete from $target using '$objectID'.") if T(3);
1795 :     my $rv = $db->SQL($stmt, 0, $objectID);
1796 : parrello 1.33 # Accumulate the statistics for this delete. The only rows deleted
1797 :     # are from the target table, so we use its name to record the
1798 :     # statistic.
1799 :     $retVal->Add($target, $rv);
1800 :     }
1801 : parrello 1.32 }
1802 :     }
1803 :     # Return the result.
1804 :     return $retVal;
1805 :     }
1806 :    
1807 : parrello 1.78 =head3 Disconnect
1808 :    
1809 :     C<< $erdb->Disconnect($relationshipName, $originEntityName, $originEntityID); >>
1810 :    
1811 :     Disconnect an entity instance from all the objects to which it is related. This
1812 :     will delete each relationship instance that connects to the specified entity.
1813 :    
1814 :     =over 4
1815 :    
1816 :     =item relationshipName
1817 :    
1818 :     Name of the relationship whose instances are to be deleted.
1819 :    
1820 :     =item originEntityName
1821 :    
1822 :     Name of the entity that is to be disconnected.
1823 :    
1824 :     =item originEntityID
1825 :    
1826 :     ID of the entity that is to be disconnected.
1827 :    
1828 :     =back
1829 :    
1830 :     =cut
1831 :    
1832 :     sub Disconnect {
1833 :     # Get the parameters.
1834 :     my ($self, $relationshipName, $originEntityName, $originEntityID) = @_;
1835 :     # Get the relationship descriptor.
1836 :     my $structure = $self->_GetStructure($relationshipName);
1837 :     # Insure we have a relationship.
1838 :     if (! exists $structure->{from}) {
1839 :     Confess("$relationshipName is not a relationship in the database.");
1840 :     } else {
1841 :     # Get the database handle.
1842 :     my $dbh = $self->{_dbh};
1843 :     # We'll set this value to 1 if we find our entity.
1844 :     my $found = 0;
1845 :     # Loop through the ends of the relationship.
1846 :     for my $dir ('from', 'to') {
1847 :     if ($structure->{$dir} eq $originEntityName) {
1848 :     # Delete all relationship instances on this side of the entity instance.
1849 : parrello 1.84 Trace("Disconnecting in $dir direction with ID \"$originEntityID\".");
1850 : parrello 1.78 $dbh->SQL("DELETE FROM $relationshipName WHERE ${dir}_link = ?", 0, $originEntityID);
1851 :     $found = 1;
1852 :     }
1853 :     }
1854 :     # Insure we found the entity on at least one end.
1855 :     if (! $found) {
1856 :     Confess("Entity \"$originEntityName\" does not use $relationshipName.");
1857 :     }
1858 :     }
1859 :     }
1860 :    
1861 : parrello 1.79 =head3 DeleteRow
1862 :    
1863 :     C<< $erdb->DeleteRow($relationshipName, $fromLink, $toLink, \%values); >>
1864 :    
1865 :     Delete a row from a relationship. In most cases, only the from-link and to-link are
1866 :     needed; however, for relationships with intersection data values can be specified
1867 :     for the other fields using a hash.
1868 :    
1869 :     =over 4
1870 :    
1871 :     =item relationshipName
1872 :    
1873 :     Name of the relationship from which the row is to be deleted.
1874 :    
1875 :     =item fromLink
1876 :    
1877 :     ID of the entity instance in the From direction.
1878 :    
1879 :     =item toLink
1880 :    
1881 :     ID of the entity instance in the To direction.
1882 :    
1883 :     =item values
1884 :    
1885 :     Reference to a hash of other values to be used for filtering the delete.
1886 :    
1887 :     =back
1888 :    
1889 :     =cut
1890 :    
1891 :     sub DeleteRow {
1892 :     # Get the parameters.
1893 :     my ($self, $relationshipName, $fromLink, $toLink, $values) = @_;
1894 :     # Create a hash of all the filter information.
1895 :     my %filter = ('from-link' => $fromLink, 'to-link' => $toLink);
1896 :     if (defined $values) {
1897 :     for my $key (keys %{$values}) {
1898 :     $filter{$key} = $values->{$key};
1899 :     }
1900 :     }
1901 :     # Build an SQL statement out of the hash.
1902 :     my @filters = ();
1903 :     my @parms = ();
1904 :     for my $key (keys %filter) {
1905 :     push @filters, _FixName($key) . " = ?";
1906 :     push @parms, $filter{$key};
1907 :     }
1908 :     Trace("Parms for delete row are " . join(", ", map { "\"$_\"" } @parms) . ".") if T(SQL => 4);
1909 :     my $command = "DELETE FROM $relationshipName WHERE " .
1910 :     join(" AND ", @filters);
1911 :     # Execute it.
1912 :     my $dbh = $self->{_dbh};
1913 :     $dbh->SQL($command, undef, @parms);
1914 :     }
1915 :    
1916 : parrello 1.70 =head3 SortNeeded
1917 :    
1918 :     C<< my $parms = $erdb->SortNeeded($relationName); >>
1919 :    
1920 :     Return the pipe command for the sort that should be applied to the specified
1921 :     relation when creating the load file.
1922 :    
1923 :     For example, if the load file should be sorted ascending by the first
1924 :     field, this method would return
1925 :    
1926 :     sort -k1 -t"\t"
1927 :    
1928 :     If the first field is numeric, the method would return
1929 :    
1930 :     sort -k1n -t"\t"
1931 :    
1932 :     Unfortunately, due to a bug in the C<sort> command, we cannot eliminate duplicate
1933 :     keys using a sort.
1934 :    
1935 :     =over 4
1936 :    
1937 :     =item relationName
1938 :    
1939 :     Name of the relation to be examined.
1940 :    
1941 :     =item
1942 :    
1943 :     Returns the sort command to use for sorting the relation, suitable for piping.
1944 :    
1945 :     =back
1946 :    
1947 :     =cut
1948 :     #: Return Type $;
1949 :     sub SortNeeded {
1950 :     # Get the parameters.
1951 :     my ($self, $relationName) = @_;
1952 :     # Declare a descriptor to hold the names of the key fields.
1953 :     my @keyNames = ();
1954 :     # Get the relation structure.
1955 :     my $relationData = $self->_FindRelation($relationName);
1956 :     # Find out if the relation is a primary entity relation,
1957 :     # a relationship relation, or a secondary entity relation.
1958 :     my $entityTable = $self->{_metaData}->{Entities};
1959 :     my $relationshipTable = $self->{_metaData}->{Relationships};
1960 :     if (exists $entityTable->{$relationName}) {
1961 :     # Here we have a primary entity relation.
1962 :     push @keyNames, "id";
1963 :     } elsif (exists $relationshipTable->{$relationName}) {
1964 :     # Here we have a relationship. We sort using the FROM index.
1965 :     my $relationshipData = $relationshipTable->{$relationName};
1966 : parrello 1.75 my $index = $relationData->{Indexes}->{idxFrom};
1967 : parrello 1.70 push @keyNames, @{$index->{IndexFields}};
1968 :     } else {
1969 :     # Here we have a secondary entity relation, so we have a sort on the ID field.
1970 :     push @keyNames, "id";
1971 :     }
1972 :     # Now we parse the key names into sort parameters. First, we prime the return
1973 :     # string.
1974 :     my $retVal = "sort -t\"\t\" ";
1975 :     # Get the relation's field list.
1976 :     my @fields = @{$relationData->{Fields}};
1977 :     # Loop through the keys.
1978 :     for my $keyData (@keyNames) {
1979 :     # Get the key and the ordering.
1980 :     my ($keyName, $ordering);
1981 :     if ($keyData =~ /^([^ ]+) DESC/) {
1982 :     ($keyName, $ordering) = ($1, "descending");
1983 :     } else {
1984 :     ($keyName, $ordering) = ($keyData, "ascending");
1985 :     }
1986 :     # Find the key's position and type.
1987 :     my $fieldSpec;
1988 :     for (my $i = 0; $i <= $#fields && ! $fieldSpec; $i++) {
1989 :     my $thisField = $fields[$i];
1990 :     if ($thisField->{name} eq $keyName) {
1991 :     # Get the sort modifier for this field type. The modifier
1992 :     # decides whether we're using a character, numeric, or
1993 :     # floating-point sort.
1994 :     my $modifier = $TypeTable{$thisField->{type}}->{sort};
1995 :     # If the index is descending for this field, denote we want
1996 :     # to reverse the sort order on this field.
1997 :     if ($ordering eq 'descending') {
1998 :     $modifier .= "r";
1999 :     }
2000 :     # Store the position and modifier into the field spec, which
2001 :     # will stop the inner loop. Note that the field number is
2002 :     # 1-based in the sort command, so we have to increment the
2003 :     # index.
2004 :     $fieldSpec = ($i + 1) . $modifier;
2005 :     }
2006 :     }
2007 :     # Add this field to the sort command.
2008 :     $retVal .= " -k$fieldSpec";
2009 :     }
2010 :     # Return the result.
2011 :     return $retVal;
2012 :     }
2013 :    
2014 : parrello 1.6 =head3 GetList
2015 :    
2016 : parrello 1.45 C<< my @dbObjects = $erdb->GetList(\@objectNames, $filterClause, \@params); >>
2017 : parrello 1.6
2018 :     Return a list of object descriptors for the specified objects as determined by the
2019 :     specified filter clause.
2020 :    
2021 :     This method is essentially the same as L</Get> except it returns a list of objects rather
2022 : parrello 1.7 than a query object that can be used to get the results one record at a time.
2023 : parrello 1.6
2024 :     =over 4
2025 :    
2026 :     =item objectNames
2027 :    
2028 :     List containing the names of the entity and relationship objects to be retrieved.
2029 :    
2030 :     =item filterClause
2031 :    
2032 :     WHERE clause (without the WHERE) to be used to filter and sort the query. The WHERE clause can
2033 :     be parameterized with parameter markers (C<?>). Each field used in the WHERE clause must be
2034 :     specified in the standard form B<I<objectName>(I<fieldName>)>. Any parameters specified
2035 :     in the filter clause should be added to the parameter list as additional parameters. The
2036 :     fields in a filter clause can come from primary entity relations, relationship relations,
2037 :     or secondary entity relations; however, all of the entities and relationships involved must
2038 :     be included in the list of object names.
2039 :    
2040 :     The filter clause can also specify a sort order. To do this, simply follow the filter string
2041 :     with an ORDER BY clause. For example, the following filter string gets all genomes for a
2042 :     particular genus and sorts them by species name.
2043 :    
2044 :     C<< "Genome(genus) = ? ORDER BY Genome(species)" >>
2045 :    
2046 :     The rules for field references in a sort order are the same as those for field references in the
2047 :     filter clause in general; however, odd things may happen if a sort field is from a secondary
2048 :     relation.
2049 :    
2050 : parrello 1.45 =item params
2051 : parrello 1.6
2052 : parrello 1.45 Reference to a list of parameter values to be substituted into the filter clause.
2053 : parrello 1.6
2054 :     =item RETURN
2055 :    
2056 :     Returns a list of B<DBObject>s that satisfy the query conditions.
2057 :    
2058 :     =back
2059 :    
2060 :     =cut
2061 :     #: Return Type @%
2062 :     sub GetList {
2063 :     # Get the parameters.
2064 : parrello 1.45 my ($self, $objectNames, $filterClause, $params) = @_;
2065 : parrello 1.10 # Declare the return variable.
2066 :     my @retVal = ();
2067 :     # Perform the query.
2068 : parrello 1.45 my $query = $self->Get($objectNames, $filterClause, $params);
2069 : parrello 1.10 # Loop through the results.
2070 :     while (my $object = $query->Fetch) {
2071 :     push @retVal, $object;
2072 :     }
2073 : parrello 1.6 # Return the result.
2074 :     return @retVal;
2075 :     }
2076 :    
2077 : parrello 1.45 =head3 GetCount
2078 :    
2079 :     C<< my $count = $erdb->GetCount(\@objectNames, $filter, \@params); >>
2080 :    
2081 :     Return the number of rows found by a specified query. This method would
2082 :     normally be used to count the records in a single table. For example, in a
2083 :     genetics database
2084 :    
2085 :     my $count = $erdb->GetCount(['Genome'], 'Genome(genus-species) LIKE ?', ['homo %']);
2086 :    
2087 :     would return the number of genomes for the genus I<homo>. It is conceivable, however,
2088 :     to use it to return records based on a join. For example,
2089 :    
2090 : parrello 1.47 my $count = $erdb->GetCount(['HasFeature', 'Genome'], 'Genome(genus-species) LIKE ?',
2091 : parrello 1.45 ['homo %']);
2092 :    
2093 :     would return the number of features for genomes in the genus I<homo>. Note that
2094 :     only the rows from the first table are counted. If the above command were
2095 :    
2096 :     my $count = $erdb->GetCount(['Genome', 'Feature'], 'Genome(genus-species) LIKE ?',
2097 :     ['homo %']);
2098 :    
2099 :     it would return the number of genomes, not the number of genome/feature pairs.
2100 :    
2101 :     =over 4
2102 :    
2103 :     =item objectNames
2104 :    
2105 :     Reference to a list of the objects (entities and relationships) included in the
2106 :     query.
2107 :    
2108 :     =item filter
2109 :    
2110 :     A filter clause for restricting the query. The rules are the same as for the L</Get>
2111 :     method.
2112 :    
2113 :     =item params
2114 :    
2115 :     Reference to a list of the parameter values to be substituted for the parameter marks
2116 :     in the filter.
2117 :    
2118 :     =item RETURN
2119 :    
2120 :     Returns a count of the number of records in the first table that would satisfy
2121 :     the query.
2122 :    
2123 :     =back
2124 :    
2125 :     =cut
2126 :    
2127 :     sub GetCount {
2128 :     # Get the parameters.
2129 :     my ($self, $objectNames, $filter, $params) = @_;
2130 : parrello 1.66 # Insure the params argument is an array reference if the caller left it off.
2131 :     if (! defined($params)) {
2132 :     $params = [];
2133 :     }
2134 : parrello 1.45 # Declare the return variable.
2135 :     my $retVal;
2136 : parrello 1.47 # Find out if we're counting an entity or a relationship.
2137 :     my $countedField;
2138 :     if ($self->IsEntity($objectNames->[0])) {
2139 :     $countedField = "id";
2140 :     } else {
2141 :     # For a relationship we count the to-link because it's usually more
2142 :     # numerous. Note we're automatically converting to the SQL form
2143 :     # of the field name (to_link vs. to-link).
2144 :     $countedField = "to_link";
2145 :     }
2146 : parrello 1.45 # Create the SQL command suffix to get the desired records.
2147 :     my ($suffix, $mappedNameListRef, $mappedNameHashRef) = $self->_SetupSQL($objectNames,
2148 :     $filter);
2149 :     # Prefix it with text telling it we want a record count.
2150 :     my $firstObject = $mappedNameListRef->[0];
2151 : parrello 1.47 my $command = "SELECT COUNT($firstObject.$countedField) $suffix";
2152 : parrello 1.45 # Prepare and execute the command.
2153 :     my $sth = $self->_GetStatementHandle($command, $params);
2154 :     # Get the count value.
2155 :     ($retVal) = $sth->fetchrow_array();
2156 :     # Check for a problem.
2157 :     if (! defined($retVal)) {
2158 :     if ($sth->err) {
2159 :     # Here we had an SQL error.
2160 :     Confess("Error retrieving row count: " . $sth->errstr());
2161 :     } else {
2162 :     # Here we have no result.
2163 :     Confess("No result attempting to retrieve row count.");
2164 :     }
2165 :     }
2166 :     # Return the result.
2167 :     return $retVal;
2168 :     }
2169 :    
2170 : parrello 1.1 =head3 ComputeObjectSentence
2171 :    
2172 : parrello 1.18 C<< my $sentence = $erdb->ComputeObjectSentence($objectName); >>
2173 : parrello 1.1
2174 :     Check an object name, and if it is a relationship convert it to a relationship sentence.
2175 :    
2176 :     =over 4
2177 :    
2178 :     =item objectName
2179 :    
2180 :     Name of the entity or relationship.
2181 :    
2182 :     =item RETURN
2183 :    
2184 :     Returns a string containing the entity name or a relationship sentence.
2185 :    
2186 :     =back
2187 :    
2188 :     =cut
2189 :    
2190 :     sub ComputeObjectSentence {
2191 : parrello 1.10 # Get the parameters.
2192 :     my ($self, $objectName) = @_;
2193 :     # Set the default return value.
2194 :     my $retVal = $objectName;
2195 :     # Look for the object as a relationship.
2196 :     my $relTable = $self->{_metaData}->{Relationships};
2197 :     if (exists $relTable->{$objectName}) {
2198 :     # Get the relationship sentence.
2199 :     $retVal = _ComputeRelationshipSentence($objectName, $relTable->{$objectName});
2200 :     }
2201 :     # Return the result.
2202 :     return $retVal;
2203 : parrello 1.1 }
2204 :    
2205 :     =head3 DumpRelations
2206 :    
2207 : parrello 1.18 C<< $erdb->DumpRelations($outputDirectory); >>
2208 : parrello 1.1
2209 :     Write the contents of all the relations to tab-delimited files in the specified directory.
2210 :     Each file will have the same name as the relation dumped, with an extension of DTX.
2211 :    
2212 :     =over 4
2213 :    
2214 :     =item outputDirectory
2215 :    
2216 :     Name of the directory into which the relation files should be dumped.
2217 :    
2218 :     =back
2219 :    
2220 :     =cut
2221 :    
2222 :     sub DumpRelations {
2223 : parrello 1.10 # Get the parameters.
2224 :     my ($self, $outputDirectory) = @_;
2225 :     # Now we need to run through all the relations. First, we loop through the entities.
2226 :     my $metaData = $self->{_metaData};
2227 :     my $entities = $metaData->{Entities};
2228 :     for my $entityName (keys %{$entities}) {
2229 :     my $entityStructure = $entities->{$entityName};
2230 :     # Get the entity's relations.
2231 :     my $relationList = $entityStructure->{Relations};
2232 :     # Loop through the relations, dumping them.
2233 :     for my $relationName (keys %{$relationList}) {
2234 :     my $relation = $relationList->{$relationName};
2235 :     $self->_DumpRelation($outputDirectory, $relationName, $relation);
2236 :     }
2237 :     }
2238 :     # Next, we loop through the relationships.
2239 :     my $relationships = $metaData->{Relationships};
2240 :     for my $relationshipName (keys %{$relationships}) {
2241 :     my $relationshipStructure = $relationships->{$relationshipName};
2242 :     # Dump this relationship's relation.
2243 :     $self->_DumpRelation($outputDirectory, $relationshipName, $relationshipStructure->{Relations}->{$relationshipName});
2244 :     }
2245 : parrello 1.1 }
2246 :    
2247 : parrello 1.51 =head3 InsertValue
2248 :    
2249 :     C<< $erdb->InsertValue($entityID, $fieldName, $value); >>
2250 :    
2251 :     This method will insert a new value into the database. The value must be one
2252 :     associated with a secondary relation, since primary values cannot be inserted:
2253 :     they occur exactly once. Secondary values, on the other hand, can be missing
2254 :     or multiply-occurring.
2255 :    
2256 :     =over 4
2257 :    
2258 :     =item entityID
2259 :    
2260 :     ID of the object that is to receive the new value.
2261 :    
2262 :     =item fieldName
2263 :    
2264 :     Field name for the new value-- this includes the entity name, since
2265 :     field names are of the format I<objectName>C<(>I<fieldName>C<)>.
2266 :    
2267 :     =item value
2268 :    
2269 :     New value to be put in the field.
2270 :    
2271 :     =back
2272 :    
2273 :     =cut
2274 :    
2275 :     sub InsertValue {
2276 :     # Get the parameters.
2277 :     my ($self, $entityID, $fieldName, $value) = @_;
2278 :     # Parse the entity name and the real field name.
2279 :     if ($fieldName =~ /^([^(]+)\(([^)]+)\)/) {
2280 :     my $entityName = $1;
2281 :     my $fieldTitle = $2;
2282 :     # Get its descriptor.
2283 :     if (!$self->IsEntity($entityName)) {
2284 :     Confess("$entityName is not a valid entity.");
2285 :     } else {
2286 :     my $entityData = $self->{_metaData}->{Entities}->{$entityName};
2287 :     # Find the relation containing this field.
2288 :     my $fieldHash = $entityData->{Fields};
2289 : parrello 1.52 if (! exists $fieldHash->{$fieldTitle}) {
2290 : parrello 1.51 Confess("$fieldTitle not found in $entityName.");
2291 :     } else {
2292 :     my $relation = $fieldHash->{$fieldTitle}->{relation};
2293 :     if ($relation eq $entityName) {
2294 :     Confess("Cannot do InsertValue on primary field $fieldTitle of $entityName.");
2295 :     } else {
2296 :     # Now we can create an INSERT statement.
2297 :     my $dbh = $self->{_dbh};
2298 :     my $fixedName = _FixName($fieldTitle);
2299 : parrello 1.53 my $statement = "INSERT INTO $relation (id, $fixedName) VALUES(?, ?)";
2300 : parrello 1.51 # Execute the command.
2301 :     $dbh->SQL($statement, 0, $entityID, $value);
2302 :     }
2303 :     }
2304 :     }
2305 :     } else {
2306 :     Confess("$fieldName is not a valid field name.");
2307 :     }
2308 :     }
2309 :    
2310 : parrello 1.1 =head3 InsertObject
2311 :    
2312 : parrello 1.78 C<< $erdb->InsertObject($objectType, \%fieldHash); >>
2313 : parrello 1.1
2314 :     Insert an object into the database. The object is defined by a type name and then a hash
2315 :     of field names to values. Field values in the primary relation are represented by scalars.
2316 :     (Note that for relationships, the primary relation is the B<only> relation.)
2317 :     Field values for the other relations comprising the entity are always list references. For
2318 :     example, the following line inserts an inactive PEG feature named C<fig|188.1.peg.1> with aliases
2319 :     C<ZP_00210270.1> and C<gi|46206278>.
2320 :    
2321 : parrello 1.18 C<< $erdb->InsertObject('Feature', { id => 'fig|188.1.peg.1', active => 0, feature-type => 'peg', alias => ['ZP_00210270.1', 'gi|46206278']}); >>
2322 : parrello 1.1
2323 :     The next statement inserts a C<HasProperty> relationship between feature C<fig|158879.1.peg.1> and
2324 :     property C<4> with an evidence URL of C<http://seedu.uchicago.edu/query.cgi?article_id=142>.
2325 :    
2326 : parrello 1.57 C<< $erdb->InsertObject('HasProperty', { 'from-link' => 'fig|158879.1.peg.1', 'to-link' => 4, evidence => 'http://seedu.uchicago.edu/query.cgi?article_id=142'}); >>
2327 : parrello 1.1
2328 :     =over 4
2329 :    
2330 :     =item newObjectType
2331 :    
2332 :     Type name of the object to insert.
2333 :    
2334 :     =item fieldHash
2335 :    
2336 :     Hash of field names to values.
2337 :    
2338 :     =back
2339 :    
2340 :     =cut
2341 :    
2342 :     sub InsertObject {
2343 : parrello 1.10 # Get the parameters.
2344 :     my ($self, $newObjectType, $fieldHash) = @_;
2345 :     # Denote that so far we appear successful.
2346 :     my $retVal = 1;
2347 :     # Get the database handle.
2348 :     my $dbh = $self->{_dbh};
2349 :     # Get the relation list.
2350 :     my $relationTable = $self->_GetRelationTable($newObjectType);
2351 :     # Loop through the relations. We'll build insert statements for each one. If a relation is
2352 :     # secondary, we may end up generating multiple insert statements. If an error occurs, we
2353 :     # stop the loop.
2354 :     my @relationList = keys %{$relationTable};
2355 :     for (my $i = 0; $retVal && $i <= $#relationList; $i++) {
2356 :     my $relationName = $relationList[$i];
2357 :     my $relationDefinition = $relationTable->{$relationName};
2358 :     # Get the relation's fields. For each field we will collect a value in the corresponding
2359 :     # position of the @valueList array. If one of the fields is missing, we will add it to the
2360 :     # @missing list.
2361 :     my @fieldList = @{$relationDefinition->{Fields}};
2362 :     my @fieldNameList = ();
2363 :     my @valueList = ();
2364 :     my @missing = ();
2365 :     my $recordCount = 1;
2366 :     for my $fieldDescriptor (@fieldList) {
2367 :     # Get the field name and save it. Note we need to fix it up so the hyphens
2368 :     # are converted to underscores.
2369 :     my $fieldName = $fieldDescriptor->{name};
2370 :     push @fieldNameList, _FixName($fieldName);
2371 :     # Look for the named field in the incoming structure. Note that we are looking
2372 :     # for the real field name, not the fixed-up one!
2373 :     if (exists $fieldHash->{$fieldName}) {
2374 :     # Here we found the field. Stash it in the value list.
2375 :     my $value = $fieldHash->{$fieldName};
2376 :     push @valueList, $value;
2377 :     # If the value is a list, we may need to increment the record count.
2378 :     if (ref $value eq "ARRAY") {
2379 :     my $thisCount = @{$value};
2380 :     if ($recordCount == 1) {
2381 :     # Here we have our first list, so we save its count.
2382 :     $recordCount = $thisCount;
2383 :     } elsif ($recordCount != $thisCount) {
2384 :     # Here we have a second list, so its length has to match the
2385 :     # previous lists.
2386 :     Trace("Field $value in new $newObjectType object has an invalid list length $thisCount. Expected $recordCount.") if T(0);
2387 :     $retVal = 0;
2388 :     }
2389 :     }
2390 :     } else {
2391 :     # Here the field is not present. Flag it as missing.
2392 :     push @missing, $fieldName;
2393 :     }
2394 :     }
2395 :     # If we are the primary relation, add the new-record flag.
2396 :     if ($relationName eq $newObjectType) {
2397 :     push @valueList, 1;
2398 :     push @fieldNameList, "new_record";
2399 :     }
2400 :     # Only proceed if there are no missing fields.
2401 :     if (@missing > 0) {
2402 :     Trace("Relation $relationName for $newObjectType skipped due to missing fields: " .
2403 :     join(' ', @missing)) if T(1);
2404 :     } else {
2405 :     # Build the INSERT statement.
2406 :     my $statement = "INSERT INTO $relationName (" . join (', ', @fieldNameList) .
2407 :     ") VALUES (";
2408 :     # Create a marker list of the proper size and put it in the statement.
2409 :     my @markers = ();
2410 :     while (@markers < @fieldNameList) { push @markers, '?'; }
2411 :     $statement .= join(', ', @markers) . ")";
2412 :     # We have the insert statement, so prepare it.
2413 :     my $sth = $dbh->prepare_command($statement);
2414 :     Trace("Insert statement prepared: $statement") if T(3);
2415 :     # Now we loop through the values. If a value is scalar, we use it unmodified. If it's
2416 :     # a list, we use the current element. The values are stored in the @parameterList array.
2417 :     my $done = 0;
2418 :     for (my $i = 0; $i < $recordCount; $i++) {
2419 :     # Clear the parameter list array.
2420 :     my @parameterList = ();
2421 :     # Loop through the values.
2422 :     for my $value (@valueList) {
2423 :     # Check to see if this is a scalar value.
2424 :     if (ref $value eq "ARRAY") {
2425 :     # Here we have a list value. Pull the current entry.
2426 :     push @parameterList, $value->[$i];
2427 :     } else {
2428 :     # Here we have a scalar value. Use it unmodified.
2429 :     push @parameterList, $value;
2430 :     }
2431 :     }
2432 :     # Execute the INSERT statement with the specified parameter list.
2433 :     $retVal = $sth->execute(@parameterList);
2434 :     if (!$retVal) {
2435 :     my $errorString = $sth->errstr();
2436 : parrello 1.78 Confess("Error inserting into $relationName: $errorString");
2437 : parrello 1.85 } else {
2438 :     Trace("Insert successful using $parameterList[0].") if T(3);
2439 : parrello 1.10 }
2440 :     }
2441 :     }
2442 :     }
2443 : parrello 1.78 # Return a 1 for backward compatability.
2444 :     return 1;
2445 :     }
2446 :    
2447 :     =head3 UpdateEntity
2448 :    
2449 :     C<< $erdb->UpdateEntity($entityName, $id, \%fields); >>
2450 :    
2451 :     Update the values of an entity. This is an unprotected update, so it should only be
2452 :     done if the database resides on a database server.
2453 :    
2454 :     =over 4
2455 :    
2456 :     =item entityName
2457 :    
2458 :     Name of the entity to update. (This is the entity type.)
2459 :    
2460 :     =item id
2461 :    
2462 :     ID of the entity to update. If no entity exists with this ID, an error will be thrown.
2463 :    
2464 :     =item fields
2465 :    
2466 :     Reference to a hash mapping field names to their new values. All of the fields named
2467 :     must be in the entity's primary relation, and they cannot any of them be the ID field.
2468 :    
2469 :     =back
2470 :    
2471 :     =cut
2472 :    
2473 :     sub UpdateEntity {
2474 :     # Get the parameters.
2475 :     my ($self, $entityName, $id, $fields) = @_;
2476 :     # Get a list of the field names being updated.
2477 :     my @fieldList = keys %{$fields};
2478 :     # Verify that the fields exist.
2479 :     my $checker = $self->GetFieldTable($entityName);
2480 :     for my $field (@fieldList) {
2481 :     if ($field eq 'id') {
2482 :     Confess("Cannot update the ID field for entity $entityName.");
2483 :     } elsif ($checker->{$field}->{relation} ne $entityName) {
2484 :     Confess("Cannot find $field in primary relation of $entityName.");
2485 :     }
2486 :     }
2487 :     # Build the SQL statement.
2488 :     my @sets = ();
2489 :     my @valueList = ();
2490 :     for my $field (@fieldList) {
2491 :     push @sets, _FixName($field) . " = ?";
2492 :     push @valueList, $fields->{$field};
2493 :     }
2494 :     my $command = "UPDATE $entityName SET " . join(", ", @sets) . " WHERE id = ?";
2495 :     # Add the ID to the list of binding values.
2496 :     push @valueList, $id;
2497 :     # Call SQL to do the work.
2498 :     my $rows = $self->{_dbh}->SQL($command, 0, @valueList);
2499 :     # Check for errors.
2500 :     if ($rows == 0) {
2501 :     Confess("Entity $id of type $entityName not found.");
2502 :     }
2503 : parrello 1.1 }
2504 :    
2505 :     =head3 LoadTable
2506 :    
2507 : parrello 1.78 C<< my $results = $erdb->LoadTable($fileName, $relationName, $truncateFlag); >>
2508 : parrello 1.1
2509 : parrello 1.9 Load data from a tab-delimited file into a specified table, optionally re-creating the table
2510 :     first.
2511 : parrello 1.1
2512 :     =over 4
2513 :    
2514 :     =item fileName
2515 :    
2516 :     Name of the file from which the table data should be loaded.
2517 :    
2518 :     =item relationName
2519 :    
2520 :     Name of the relation to be loaded. This is the same as the table name.
2521 :    
2522 :     =item truncateFlag
2523 :    
2524 :     TRUE if the table should be dropped and re-created, else FALSE
2525 :    
2526 :     =item RETURN
2527 :    
2528 : parrello 1.28 Returns a statistical object containing a list of the error messages.
2529 : parrello 1.1
2530 :     =back
2531 :    
2532 :     =cut
2533 :     sub LoadTable {
2534 : parrello 1.10 # Get the parameters.
2535 :     my ($self, $fileName, $relationName, $truncateFlag) = @_;
2536 :     # Create the statistical return object.
2537 :     my $retVal = _GetLoadStats();
2538 :     # Trace the fact of the load.
2539 :     Trace("Loading table $relationName from $fileName") if T(2);
2540 :     # Get the database handle.
2541 :     my $dbh = $self->{_dbh};
2542 : parrello 1.22 # Get the input file size.
2543 :     my $fileSize = -s $fileName;
2544 : parrello 1.10 # Get the relation data.
2545 :     my $relation = $self->_FindRelation($relationName);
2546 :     # Check the truncation flag.
2547 :     if ($truncateFlag) {
2548 :     Trace("Creating table $relationName") if T(2);
2549 : parrello 1.19 # Compute the row count estimate. We take the size of the load file,
2550 :     # divide it by the estimated row size, and then multiply by 1.5 to
2551 :     # leave extra room. We postulate a minimum row count of 1000 to
2552 :     # prevent problems with incoming empty load files.
2553 :     my $rowSize = $self->EstimateRowSize($relationName);
2554 :     my $estimate = FIG::max($fileSize * 1.5 / $rowSize, 1000);
2555 : parrello 1.10 # Re-create the table without its index.
2556 : parrello 1.19 $self->CreateTable($relationName, 0, $estimate);
2557 : parrello 1.10 # If this is a pre-index DBMS, create the index here.
2558 :     if ($dbh->{_preIndex}) {
2559 :     eval {
2560 :     $self->CreateIndex($relationName);
2561 :     };
2562 :     if ($@) {
2563 :     $retVal->AddMessage($@);
2564 :     }
2565 :     }
2566 :     }
2567 : parrello 1.3 # Load the table.
2568 : parrello 1.10 my $rv;
2569 :     eval {
2570 : parrello 1.20 $rv = $dbh->load_table(file => $fileName, tbl => $relationName);
2571 : parrello 1.10 };
2572 :     if (!defined $rv) {
2573 : parrello 1.3 $retVal->AddMessage($@) if ($@);
2574 : parrello 1.73 $retVal->AddMessage("Table load failed for $relationName using $fileName: " . $dbh->error_message);
2575 : parrello 1.10 Trace("Table load failed for $relationName.") if T(1);
2576 :     } else {
2577 : parrello 1.22 # Here we successfully loaded the table.
2578 :     $retVal->Add("tables");
2579 :     my $size = -s $fileName;
2580 :     Trace("$size bytes loaded into $relationName.") if T(2);
2581 : parrello 1.10 # If we're rebuilding, we need to create the table indexes.
2582 : parrello 1.70 if ($truncateFlag) {
2583 :     # Indexes are created here for PostGres. For PostGres, indexes are
2584 :     # best built at the end. For MySQL, the reverse is true.
2585 :     if (! $dbh->{_preIndex}) {
2586 :     eval {
2587 :     $self->CreateIndex($relationName);
2588 :     };
2589 :     if ($@) {
2590 :     $retVal->AddMessage($@);
2591 :     }
2592 :     }
2593 :     # The full-text index (if any) is always built last, even for MySQL.
2594 :     # First we need to see if this table has a full-text index. Only
2595 :     # primary relations are allowed that privilege.
2596 : parrello 1.81 Trace("Checking for full-text index on $relationName.") if T(2);
2597 : parrello 1.70 if ($self->_IsPrimary($relationName)) {
2598 : parrello 1.82 $self->CreateSearchIndex($relationName);
2599 : parrello 1.10 }
2600 :     }
2601 :     }
2602 : parrello 1.20 # Analyze the table to improve performance.
2603 : parrello 1.61 Trace("Analyzing and compacting $relationName.") if T(3);
2604 : olson 1.16 $dbh->vacuum_it($relationName);
2605 : parrello 1.61 Trace("$relationName load completed.") if T(3);
2606 : parrello 1.10 # Return the statistics.
2607 :     return $retVal;
2608 : parrello 1.1 }
2609 :    
2610 : parrello 1.82 =head3 CreateSearchIndex
2611 :    
2612 :     C<< $erdb->CreateSearchIndex($objectName); >>
2613 :    
2614 :     Check for a full-text search index on the specified entity or relationship object, and
2615 :     if one is required, rebuild it.
2616 :    
2617 :     =over 4
2618 :    
2619 :     =item objectName
2620 :    
2621 :     Name of the entity or relationship to be indexed.
2622 :    
2623 :     =back
2624 :    
2625 :     =cut
2626 :    
2627 :     sub CreateSearchIndex {
2628 :     # Get the parameters.
2629 :     my ($self, $objectName) = @_;
2630 :     # Get the relation's entity/relationship structure.
2631 :     my $structure = $self->_GetStructure($objectName);
2632 :     # Get the database handle.
2633 :     my $dbh = $self->{_dbh};
2634 :     Trace("Checking for search fields in $objectName.") if T(3);
2635 :     # Check for a searchable fields list.
2636 :     if (exists $structure->{searchFields}) {
2637 :     # Here we know that we need to create a full-text search index.
2638 :     # Get an SQL-formatted field name list.
2639 :     my $fields = join(", ", _FixNames(@{$structure->{searchFields}}));
2640 :     # Create the index. If it already exists, it will be dropped.
2641 :     $dbh->create_index(tbl => $objectName, idx => "search_idx",
2642 :     flds => $fields, kind => 'fulltext');
2643 :     Trace("Index created for $fields in $objectName.") if T(2);
2644 :     }
2645 :     }
2646 :    
2647 : parrello 1.73 =head3 DropRelation
2648 : parrello 1.1
2649 : parrello 1.73 C<< $erdb->DropRelation($relationName); >>
2650 : parrello 1.1
2651 : parrello 1.73 Physically drop a relation from the database.
2652 : parrello 1.1
2653 :     =over 4
2654 :    
2655 : parrello 1.73 =item relationName
2656 : parrello 1.1
2657 : parrello 1.73 Name of the relation to drop. If it does not exist, this method will have
2658 :     no effect.
2659 : parrello 1.1
2660 :     =back
2661 :    
2662 :     =cut
2663 :    
2664 : parrello 1.73 sub DropRelation {
2665 : parrello 1.10 # Get the parameters.
2666 : parrello 1.73 my ($self, $relationName) = @_;
2667 :     # Get the database handle.
2668 :     my $dbh = $self->{_dbh};
2669 :     # Drop the relation. The method used here has no effect if the relation
2670 :     # does not exist.
2671 : parrello 1.74 Trace("Invoking DB Kernel to drop $relationName.") if T(3);
2672 :     $dbh->drop_table(tbl => $relationName);
2673 : parrello 1.1 }
2674 :    
2675 : parrello 1.78 =head3 MatchSqlPattern
2676 :    
2677 :     C<< my $matched = ERDB::MatchSqlPattern($value, $pattern); >>
2678 :    
2679 :     Determine whether or not a specified value matches an SQL pattern. An SQL
2680 :     pattern has two wild card characters: C<%> that matches multiple characters,
2681 :     and C<_> that matches a single character. These can be escaped using a
2682 :     backslash (C<\>). We pull this off by converting the SQL pattern to a
2683 :     PERL regular expression. As per SQL rules, the match is case-insensitive.
2684 :    
2685 :     =over 4
2686 :    
2687 :     =item value
2688 :    
2689 :     Value to be matched against the pattern. Note that an undefined or empty
2690 :     value will not match anything.
2691 :    
2692 :     =item pattern
2693 :    
2694 :     SQL pattern against which to match the value. An undefined or empty pattern will
2695 :     match everything.
2696 :    
2697 :     =item RETURN
2698 :    
2699 :     Returns TRUE if the value and pattern match, else FALSE.
2700 :    
2701 :     =back
2702 :    
2703 :     =cut
2704 :    
2705 :     sub MatchSqlPattern {
2706 :     # Get the parameters.
2707 :     my ($value, $pattern) = @_;
2708 :     # Declare the return variable.
2709 :     my $retVal;
2710 :     # Insure we have a pattern.
2711 :     if (! defined($pattern) || $pattern eq "") {
2712 :     $retVal = 1;
2713 :     } else {
2714 :     # Break the pattern into pieces around the wildcard characters. Because we
2715 :     # use parentheses in the split function's delimiter expression, we'll get
2716 :     # list elements for the delimiters as well as the rest of the string.
2717 :     my @pieces = split /([_%]|\\[_%])/, $pattern;
2718 :     # Check some fast special cases.
2719 :     if ($pattern eq '%') {
2720 :     # A null pattern matches everything.
2721 :     $retVal = 1;
2722 :     } elsif (@pieces == 1) {
2723 :     # No wildcards, so we have a literal comparison. Note we're case-insensitive.
2724 :     $retVal = (lc($value) eq lc($pattern));
2725 :     } elsif (@pieces == 2 && $pieces[1] eq '%') {
2726 :     # A wildcard at the end, so we have a substring match. This is also case-insensitive.
2727 :     $retVal = (lc(substr($value, 0, length($pieces[0]))) eq lc($pieces[0]));
2728 :     } else {
2729 :     # Okay, we have to do it the hard way. Convert each piece to a PERL pattern.
2730 :     my $realPattern = "";
2731 :     for my $piece (@pieces) {
2732 :     # Determine the type of piece.
2733 :     if ($piece eq "") {
2734 :     # Empty pieces are ignored.
2735 :     } elsif ($piece eq "%") {
2736 :     # Here we have a multi-character wildcard. Note that it can match
2737 :     # zero or more characters.
2738 :     $realPattern .= ".*"
2739 :     } elsif ($piece eq "_") {
2740 :     # Here we have a single-character wildcard.
2741 :     $realPattern .= ".";
2742 :     } elsif ($piece eq "\\%" || $piece eq "\\_") {
2743 :     # This is an escape sequence (which is a rare thing, actually).
2744 :     $realPattern .= substr($piece, 1, 1);
2745 :     } else {
2746 :     # Here we have raw text.
2747 :     $realPattern .= quotemeta($piece);
2748 :     }
2749 :     }
2750 :     # Do the match.
2751 :     $retVal = ($value =~ /^$realPattern$/i ? 1 : 0);
2752 :     }
2753 :     }
2754 :     # Return the result.
2755 :     return $retVal;
2756 :     }
2757 :    
2758 : parrello 1.6 =head3 GetEntity
2759 :    
2760 : parrello 1.18 C<< my $entityObject = $erdb->GetEntity($entityType, $ID); >>
2761 : parrello 1.6
2762 :     Return an object describing the entity instance with a specified ID.
2763 :    
2764 :     =over 4
2765 :    
2766 :     =item entityType
2767 :    
2768 :     Entity type name.
2769 :    
2770 :     =item ID
2771 :    
2772 :     ID of the desired entity.
2773 :    
2774 :     =item RETURN
2775 :    
2776 :     Returns a B<DBObject> representing the desired entity instance, or an undefined value if no
2777 :     instance is found with the specified key.
2778 :    
2779 :     =back
2780 :    
2781 :     =cut
2782 :    
2783 :     sub GetEntity {
2784 : parrello 1.10 # Get the parameters.
2785 :     my ($self, $entityType, $ID) = @_;
2786 :     # Create a query.
2787 : parrello 1.45 my $query = $self->Get([$entityType], "$entityType(id) = ?", [$ID]);
2788 : parrello 1.10 # Get the first (and only) object.
2789 :     my $retVal = $query->Fetch();
2790 :     # Return the result.
2791 :     return $retVal;
2792 : parrello 1.6 }
2793 :    
2794 : parrello 1.69 =head3 GetChoices
2795 :    
2796 :     C<< my @values = $erdb->GetChoices($entityName, $fieldName); >>
2797 :    
2798 :     Return a list of all the values for the specified field that are represented in the
2799 :     specified entity.
2800 :    
2801 :     Note that if the field is not indexed, then this will be a very slow operation.
2802 :    
2803 :     =over 4
2804 :    
2805 :     =item entityName
2806 :    
2807 :     Name of an entity in the database.
2808 :    
2809 :     =item fieldName
2810 :    
2811 :     Name of a field belonging to the entity. This is a raw field name without
2812 :     the standard parenthesized notation used in most calls.
2813 :    
2814 :     =item RETURN
2815 :    
2816 :     Returns a list of the distinct values for the specified field in the database.
2817 :    
2818 :     =back
2819 :    
2820 :     =cut
2821 :    
2822 :     sub GetChoices {
2823 :     # Get the parameters.
2824 :     my ($self, $entityName, $fieldName) = @_;
2825 :     # Declare the return variable.
2826 :     my @retVal;
2827 :     # Get the entity data structure.
2828 :     my $entityData = $self->_GetStructure($entityName);
2829 :     # Get the field.
2830 :     my $fieldHash = $entityData->{Fields};
2831 :     if (! exists $fieldHash->{$fieldName}) {
2832 :     Confess("$fieldName not found in $entityName.");
2833 :     } else {
2834 :     # Get the name of the relation containing the field.
2835 :     my $relation = $fieldHash->{$fieldName}->{relation};
2836 :     # Fix up the field name.
2837 :     my $realName = _FixName($fieldName);
2838 :     # Get the database handle.
2839 :     my $dbh = $self->{_dbh};
2840 :     # Query the database.
2841 :     my $results = $dbh->SQL("SELECT DISTINCT $realName FROM $relation");
2842 :     # Clean the results. They are stored as a list of lists, and we just want the one list.
2843 :     @retVal = sort map { $_->[0] } @{$results};
2844 :     }
2845 :     # Return the result.
2846 :     return @retVal;
2847 :     }
2848 :    
2849 : parrello 1.6 =head3 GetEntityValues
2850 :    
2851 : parrello 1.18 C<< my @values = $erdb->GetEntityValues($entityType, $ID, \@fields); >>
2852 : parrello 1.6
2853 : parrello 1.68 Return a list of values from a specified entity instance. If the entity instance
2854 :     does not exist, an empty list is returned.
2855 : parrello 1.6
2856 :     =over 4
2857 :    
2858 :     =item entityType
2859 :    
2860 :     Entity type name.
2861 :    
2862 :     =item ID
2863 :    
2864 :     ID of the desired entity.
2865 :    
2866 :     =item fields
2867 :    
2868 :     List of field names, each of the form I<objectName>C<(>I<fieldName>C<)>.
2869 :    
2870 :     =item RETURN
2871 :    
2872 :     Returns a flattened list of the values of the specified fields for the specified entity.
2873 :    
2874 :     =back
2875 :    
2876 :     =cut
2877 :    
2878 :     sub GetEntityValues {
2879 : parrello 1.10 # Get the parameters.
2880 :     my ($self, $entityType, $ID, $fields) = @_;
2881 :     # Get the specified entity.
2882 :     my $entity = $self->GetEntity($entityType, $ID);
2883 :     # Declare the return list.
2884 :     my @retVal = ();
2885 :     # If we found the entity, push the values into the return list.
2886 :     if ($entity) {
2887 :     push @retVal, $entity->Values($fields);
2888 :     }
2889 :     # Return the result.
2890 :     return @retVal;
2891 : parrello 1.6 }
2892 : parrello 1.1
2893 : parrello 1.7 =head3 GetAll
2894 :    
2895 : parrello 1.18 C<< my @list = $erdb->GetAll(\@objectNames, $filterClause, \@parameters, \@fields, $count); >>
2896 : parrello 1.7
2897 :     Return a list of values taken from the objects returned by a query. The first three
2898 :     parameters correspond to the parameters of the L</Get> method. The final parameter is
2899 :     a list of the fields desired from each record found by the query. The field name
2900 :     syntax is the standard syntax used for fields in the B<ERDB> system--
2901 :     B<I<objectName>(I<fieldName>)>-- where I<objectName> is the name of the relevant entity
2902 :     or relationship and I<fieldName> is the name of the field.
2903 :    
2904 :     The list returned will be a list of lists. Each element of the list will contain
2905 :     the values returned for the fields specified in the fourth parameter. If one of the
2906 :     fields specified returns multiple values, they are flattened in with the rest. For
2907 :     example, the following call will return a list of the features in a particular
2908 :     spreadsheet cell, and each feature will be represented by a list containing the
2909 :     feature ID followed by all of its aliases.
2910 :    
2911 : parrello 1.78 C<< @query = $erdb->Get(['ContainsFeature', 'Feature'], "ContainsFeature(from-link) = ?", [$ssCellID], ['Feature(id)', 'Feature(alias)']); >>
2912 : parrello 1.7
2913 :     =over 4
2914 :    
2915 :     =item objectNames
2916 :    
2917 :     List containing the names of the entity and relationship objects to be retrieved.
2918 :    
2919 :     =item filterClause
2920 :    
2921 :     WHERE/ORDER BY clause (without the WHERE) to be used to filter and sort the query. The WHERE clause can
2922 :     be parameterized with parameter markers (C<?>). Each field used must be specified in the standard form
2923 :     B<I<objectName>(I<fieldName>)>. Any parameters specified in the filter clause should be added to the
2924 :     parameter list as additional parameters. The fields in a filter clause can come from primary
2925 :     entity relations, relationship relations, or secondary entity relations; however, all of the
2926 :     entities and relationships involved must be included in the list of object names.
2927 :    
2928 :     =item parameterList
2929 :    
2930 :     List of the parameters to be substituted in for the parameters marks in the filter clause.
2931 :    
2932 :     =item fields
2933 :    
2934 :     List of the fields to be returned in each element of the list returned.
2935 :    
2936 :     =item count
2937 :    
2938 :     Maximum number of records to return. If omitted or 0, all available records will be returned.
2939 :    
2940 :     =item RETURN
2941 :    
2942 :     Returns a list of list references. Each element of the return list contains the values for the
2943 :     fields specified in the B<fields> parameter.
2944 :    
2945 :     =back
2946 :    
2947 :     =cut
2948 :     #: Return Type @@;
2949 :     sub GetAll {
2950 : parrello 1.10 # Get the parameters.
2951 :     my ($self, $objectNames, $filterClause, $parameterList, $fields, $count) = @_;
2952 :     # Translate the parameters from a list reference to a list. If the parameter
2953 :     # list is a scalar we convert it into a singleton list.
2954 :     my @parmList = ();
2955 :     if (ref $parameterList eq "ARRAY") {
2956 : parrello 1.45 Trace("GetAll parm list is an array.") if T(4);
2957 : parrello 1.10 @parmList = @{$parameterList};
2958 :     } else {
2959 : parrello 1.45 Trace("GetAll parm list is a scalar: $parameterList.") if T(4);
2960 : parrello 1.10 push @parmList, $parameterList;
2961 :     }
2962 : parrello 1.30 # Insure the counter has a value.
2963 :     if (!defined $count) {
2964 :     $count = 0;
2965 :     }
2966 :     # Add the row limit to the filter clause.
2967 :     if ($count > 0) {
2968 :     $filterClause .= " LIMIT $count";
2969 :     }
2970 : parrello 1.10 # Create the query.
2971 : parrello 1.45 my $query = $self->Get($objectNames, $filterClause, \@parmList);
2972 : parrello 1.10 # Set up a counter of the number of records read.
2973 :     my $fetched = 0;
2974 :     # Loop through the records returned, extracting the fields. Note that if the
2975 :     # counter is non-zero, we stop when the number of records read hits the count.
2976 :     my @retVal = ();
2977 :     while (($count == 0 || $fetched < $count) && (my $row = $query->Fetch())) {
2978 :     my @rowData = $row->Values($fields);
2979 :     push @retVal, \@rowData;
2980 :     $fetched++;
2981 :     }
2982 : parrello 1.54 Trace("$fetched rows returned in GetAll.") if T(SQL => 4);
2983 : parrello 1.10 # Return the resulting list.
2984 :     return @retVal;
2985 : parrello 1.7 }
2986 :    
2987 : parrello 1.55 =head3 Exists
2988 :    
2989 :     C<< my $found = $sprout->Exists($entityName, $entityID); >>
2990 :    
2991 :     Return TRUE if an entity exists, else FALSE.
2992 :    
2993 :     =over 4
2994 :    
2995 :     =item entityName
2996 :    
2997 :     Name of the entity type (e.g. C<Feature>) relevant to the existence check.
2998 :    
2999 :     =item entityID
3000 :    
3001 :     ID of the entity instance whose existence is to be checked.
3002 :    
3003 :     =item RETURN
3004 :    
3005 :     Returns TRUE if the entity instance exists, else FALSE.
3006 :    
3007 :     =back
3008 :    
3009 :     =cut
3010 :     #: Return Type $;
3011 :     sub Exists {
3012 :     # Get the parameters.
3013 :     my ($self, $entityName, $entityID) = @_;
3014 :     # Check for the entity instance.
3015 :     Trace("Checking existence of $entityName with ID=$entityID.") if T(4);
3016 :     my $testInstance = $self->GetEntity($entityName, $entityID);
3017 :     # Return an existence indicator.
3018 :     my $retVal = ($testInstance ? 1 : 0);
3019 :     return $retVal;
3020 :     }
3021 :    
3022 : parrello 1.18 =head3 EstimateRowSize
3023 :    
3024 :     C<< my $rowSize = $erdb->EstimateRowSize($relName); >>
3025 :    
3026 :     Estimate the row size of the specified relation. The estimated row size is computed by adding
3027 :     up the average length for each data type.
3028 :    
3029 :     =over 4
3030 :    
3031 :     =item relName
3032 :    
3033 :     Name of the relation whose estimated row size is desired.
3034 :    
3035 :     =item RETURN
3036 :    
3037 :     Returns an estimate of the row size for the specified relation.
3038 :    
3039 :     =back
3040 :    
3041 :     =cut
3042 :     #: Return Type $;
3043 :     sub EstimateRowSize {
3044 :     # Get the parameters.
3045 :     my ($self, $relName) = @_;
3046 :     # Declare the return variable.
3047 :     my $retVal = 0;
3048 :     # Find the relation descriptor.
3049 :     my $relation = $self->_FindRelation($relName);
3050 :     # Get the list of fields.
3051 :     for my $fieldData (@{$relation->{Fields}}) {
3052 :     # Get the field type and add its length.
3053 :     my $fieldLen = $TypeTable{$fieldData->{type}}->{avgLen};
3054 :     $retVal += $fieldLen;
3055 :     }
3056 :     # Return the result.
3057 :     return $retVal;
3058 :     }
3059 :    
3060 : parrello 1.38 =head3 GetFieldTable
3061 :    
3062 :     C<< my $fieldHash = $self->GetFieldTable($objectnName); >>
3063 :    
3064 :     Get the field structure for a specified entity or relationship.
3065 :    
3066 :     =over 4
3067 :    
3068 :     =item objectName
3069 :    
3070 :     Name of the desired entity or relationship.
3071 :    
3072 :     =item RETURN
3073 :    
3074 :     The table containing the field descriptors for the specified object.
3075 :    
3076 :     =back
3077 :    
3078 :     =cut
3079 :    
3080 :     sub GetFieldTable {
3081 :     # Get the parameters.
3082 :     my ($self, $objectName) = @_;
3083 :     # Get the descriptor from the metadata.
3084 :     my $objectData = $self->_GetStructure($objectName);
3085 :     # Return the object's field table.
3086 :     return $objectData->{Fields};
3087 :     }
3088 :    
3089 : parrello 1.73 =head3 SplitKeywords
3090 :    
3091 :     C<< my @keywords = ERDB::SplitKeywords($keywordString); >>
3092 :    
3093 :     This method returns a list of the positive keywords in the specified
3094 :     keyword string. All of the operators will have been stripped off,
3095 :     and if the keyword is preceded by a minus operator (C<->), it will
3096 :     not be in the list returned. The idea here is to get a list of the
3097 :     keywords the user wants to see. The list will be processed to remove
3098 :     duplicates.
3099 :    
3100 :     It is possible to create a string that confuses this method. For example
3101 :    
3102 :     frog toad -frog
3103 :    
3104 :     would return both C<frog> and C<toad>. If this is a problem we can deal
3105 :     with it later.
3106 :    
3107 :     =over 4
3108 :    
3109 :     =item keywordString
3110 :    
3111 :     The keyword string to be parsed.
3112 :    
3113 :     =item RETURN
3114 :    
3115 :     Returns a list of the words in the keyword string the user wants to
3116 :     see.
3117 :    
3118 :     =back
3119 :    
3120 :     =cut
3121 :    
3122 :     sub SplitKeywords {
3123 :     # Get the parameters.
3124 :     my ($keywordString) = @_;
3125 :     # Make a safety copy of the string. (This helps during debugging.)
3126 :     my $workString = $keywordString;
3127 :     # Convert operators we don't care about to spaces.
3128 :     $workString =~ tr/+"()<>/ /;
3129 :     # Split the rest of the string along space boundaries. Note that we
3130 :     # eliminate any words that are zero length or begin with a minus sign.
3131 :     my @wordList = grep { $_ && substr($_, 0, 1) ne "-" } split /\s+/, $workString;
3132 :     # Use a hash to remove duplicates.
3133 :     my %words = map { $_ => 1 } @wordList;
3134 :     # Return the result.
3135 :     return sort keys %words;
3136 :     }
3137 :    
3138 :     =head3 ValidateFieldName
3139 :    
3140 :     C<< my $okFlag = ERDB::ValidateFieldName($fieldName); >>
3141 :    
3142 :     Return TRUE if the specified field name is valid, else FALSE. Valid field names must
3143 :     be hyphenated words subject to certain restrictions.
3144 :    
3145 :     =over 4
3146 :    
3147 :     =item fieldName
3148 :    
3149 :     Field name to be validated.
3150 :    
3151 :     =item RETURN
3152 :    
3153 :     Returns TRUE if the field name is valid, else FALSE.
3154 :    
3155 :     =back
3156 :    
3157 :     =cut
3158 :    
3159 :     sub ValidateFieldName {
3160 :     # Get the parameters.
3161 :     my ($fieldName) = @_;
3162 :     # Declare the return variable. The field name is valid until we hear
3163 :     # differently.
3164 :     my $retVal = 1;
3165 : parrello 1.78 # Compute the maximum name length.
3166 :     my $maxLen = $TypeTable{'name-string'}->{maxLen};
3167 : parrello 1.73 # Look for bad stuff in the name.
3168 :     if ($fieldName =~ /--/) {
3169 :     # Here we have a doubled minus sign.
3170 :     Trace("Field name $fieldName has a doubled hyphen.") if T(1);
3171 :     $retVal = 0;
3172 :     } elsif ($fieldName !~ /^[A-Za-z]/) {
3173 :     # Here the field name is missing the initial letter.
3174 :     Trace("Field name $fieldName does not begin with a letter.") if T(1);
3175 :     $retVal = 0;
3176 : parrello 1.78 } elsif (length($fieldName) > $maxLen) {
3177 :     # Here the field name is too long.
3178 :     Trace("Maximum field name length is $maxLen. Field name must be truncated to " . substr($fieldName,0, $maxLen) . ".");
3179 : parrello 1.73 } else {
3180 : parrello 1.75 # Strip out the minus signs. Everything remaining must be a letter,
3181 :     # underscore, or digit.
3182 : parrello 1.73 my $strippedName = $fieldName;
3183 :     $strippedName =~ s/-//g;
3184 : parrello 1.75 if ($strippedName !~ /^(\w|\d)+$/) {
3185 : parrello 1.73 Trace("Field name $fieldName contains illegal characters.") if T(1);
3186 :     $retVal = 0;
3187 :     }
3188 :     }
3189 :     # Return the result.
3190 :     return $retVal;
3191 :     }
3192 :    
3193 :     =head3 ReadMetaXML
3194 :    
3195 :     C<< my $rawMetaData = ERDB::ReadDBD($fileName); >>
3196 :    
3197 :     This method reads a raw database definition XML file and returns it.
3198 :     Normally, the metadata used by the ERDB system has been processed and
3199 :     modified to make it easier to load and retrieve the data; however,
3200 :     this method can be used to get the data in its raw form.
3201 :    
3202 :     =over 4
3203 :    
3204 :     =item fileName
3205 :    
3206 :     Name of the XML file to read.
3207 :    
3208 :     =item RETURN
3209 :    
3210 :     Returns a hash reference containing the raw XML data from the specified file.
3211 :    
3212 :     =back
3213 :    
3214 :     =cut
3215 :    
3216 :     sub ReadMetaXML {
3217 :     # Get the parameters.
3218 :     my ($fileName) = @_;
3219 :     # Read the XML.
3220 :     my $retVal = XML::Simple::XMLin($fileName, %XmlOptions, %XmlInOpts);
3221 :     Trace("XML metadata loaded from file $fileName.") if T(1);
3222 :     # Return the result.
3223 :     return $retVal;
3224 :     }
3225 :    
3226 :     =head3 GetEntityFieldHash
3227 :    
3228 :     C<< my $fieldHashRef = ERDB::GetEntityFieldHash($structure, $entityName); >>
3229 :    
3230 :     Get the field hash of the named entity in the specified raw XML structure.
3231 :     The field hash may not exist, in which case we need to create it.
3232 :    
3233 :     =over 4
3234 :    
3235 :     =item structure
3236 :    
3237 :     Raw XML structure defininng the database. This is not the run-time XML used by
3238 :     an ERDB object, since that has all sorts of optimizations built-in.
3239 :    
3240 :     =item entityName
3241 :    
3242 :     Name of the entity whose field structure is desired.
3243 :    
3244 :     =item RETURN
3245 :    
3246 :     Returns the field hash used to define the entity's fields.
3247 :    
3248 :     =back
3249 :    
3250 :     =cut
3251 :    
3252 :     sub GetEntityFieldHash {
3253 :     # Get the parameters.
3254 :     my ($structure, $entityName) = @_;
3255 :     # Get the entity structure.
3256 :     my $entityData = $structure->{Entities}->{$entityName};
3257 :     # Look for a field structure.
3258 :     my $retVal = $entityData->{Fields};
3259 :     # If it doesn't exist, create it.
3260 :     if (! defined($retVal)) {
3261 :     $entityData->{Fields} = {};
3262 :     $retVal = $entityData->{Fields};
3263 :     }
3264 :     # Return the result.
3265 :     return $retVal;
3266 :     }
3267 :    
3268 :     =head3 WriteMetaXML
3269 :    
3270 :     C<< ERDB::WriteMetaXML($structure, $fileName); >>
3271 :    
3272 :     Write the metadata XML to a file. This method is the reverse of L</ReadMetaXML>, and is
3273 :     used to update the database definition. It must be used with care, however, since it
3274 :     will only work on a raw structure, not on the processed structure created by an ERDB
3275 :     constructor.
3276 :    
3277 :     =over 4
3278 :    
3279 :     =item structure
3280 :    
3281 :     XML structure to be written to the file.
3282 :    
3283 :     =item fileName
3284 :    
3285 :     Name of the output file to which the updated XML should be stored.
3286 :    
3287 :     =back
3288 :    
3289 :     =cut
3290 :    
3291 :     sub WriteMetaXML {
3292 :     # Get the parameters.
3293 :     my ($structure, $fileName) = @_;
3294 :     # Compute the output.
3295 :     my $fileString = XML::Simple::XMLout($structure, %XmlOptions, %XmlOutOpts);
3296 :     # Write it to the file.
3297 :     my $xmlOut = Open(undef, ">$fileName");
3298 :     print $xmlOut $fileString;
3299 :     }
3300 :    
3301 :    
3302 :     =head3 HTMLNote
3303 :    
3304 :     Convert a note or comment to HTML by replacing some bulletin-board codes with HTML. The codes
3305 :     supported are C<[b]> for B<bold>, C<[i]> for I<italics>, and C<[p]> for a new paragraph.
3306 :     Except for C<[p]>, all the codes are closed by slash-codes. So, for
3307 :     example, C<[b]Feature[/b]> displays the string C<Feature> in boldface.
3308 :    
3309 :     C<< my $realHtml = ERDB::HTMLNote($dataString); >>
3310 :    
3311 :     =over 4
3312 :    
3313 :     =item dataString
3314 :    
3315 :     String to convert to HTML.
3316 :    
3317 :     =item RETURN
3318 :    
3319 :     An HTML string derived from the input string.
3320 :    
3321 :     =back
3322 :    
3323 :     =cut
3324 :    
3325 :     sub HTMLNote {
3326 :     # Get the parameter.
3327 :     my ($dataString) = @_;
3328 :     # HTML-escape the text.
3329 :     my $retVal = CGI::escapeHTML($dataString);
3330 :     # Substitute the bulletin board codes.
3331 :     $retVal =~ s!\[(/?[bi])\]!<$1>!g;
3332 :     $retVal =~ s!\[p\]!</p><p>!g;
3333 : parrello 1.83 $retVal =~ s!\[link\s+([^\]]+)\]!<a href="$1">!g;
3334 :     $retVal =~ s!\[/link\]!</a>!g;
3335 : parrello 1.73 # Return the result.
3336 :     return $retVal;
3337 :     }
3338 :    
3339 :    
3340 : parrello 1.48 =head2 Data Mining Methods
3341 :    
3342 : parrello 1.38 =head3 GetUsefulCrossValues
3343 :    
3344 :     C<< my @attrNames = $sprout->GetUsefulCrossValues($sourceEntity, $relationship); >>
3345 :    
3346 :     Return a list of the useful attributes that would be returned by a B<Cross> call
3347 :     from an entity of the source entity type through the specified relationship. This
3348 :     means it will return the fields of the target entity type and the intersection data
3349 :     fields in the relationship. Only primary table fields are returned. In other words,
3350 :     the field names returned will be for fields where there is always one and only one
3351 :     value.
3352 :    
3353 :     =over 4
3354 :    
3355 :     =item sourceEntity
3356 :    
3357 :     Name of the entity from which the relationship crossing will start.
3358 :    
3359 :     =item relationship
3360 :    
3361 :     Name of the relationship being crossed.
3362 :    
3363 :     =item RETURN
3364 :    
3365 :     Returns a list of field names in Sprout field format (I<objectName>C<(>I<fieldName>C<)>.
3366 :    
3367 :     =back
3368 :    
3369 :     =cut
3370 :     #: Return Type @;
3371 :     sub GetUsefulCrossValues {
3372 :     # Get the parameters.
3373 :     my ($self, $sourceEntity, $relationship) = @_;
3374 :     # Declare the return variable.
3375 :     my @retVal = ();
3376 :     # Determine the target entity for the relationship. This is whichever entity is not
3377 :     # the source entity. So, if the source entity is the FROM, we'll get the name of
3378 :     # the TO, and vice versa.
3379 :     my $relStructure = $self->_GetStructure($relationship);
3380 :     my $targetEntityType = ($relStructure->{from} eq $sourceEntity ? "to" : "from");
3381 :     my $targetEntity = $relStructure->{$targetEntityType};
3382 :     # Get the field table for the entity.
3383 :     my $entityFields = $self->GetFieldTable($targetEntity);
3384 :     # The field table is a hash. The hash key is the field name. The hash value is a structure.
3385 :     # For the entity fields, the key aspect of the target structure is that the {relation} value
3386 :     # must match the entity name.
3387 :     my @fieldList = map { "$targetEntity($_)" } grep { $entityFields->{$_}->{relation} eq $targetEntity }
3388 :     keys %{$entityFields};
3389 :     # Push the fields found onto the return variable.
3390 :     push @retVal, sort @fieldList;
3391 :     # Get the field table for the relationship.
3392 :     my $relationshipFields = $self->GetFieldTable($relationship);
3393 :     # Here we have a different rule. We want all the fields other than "from-link" and "to-link".
3394 :     # This may end up being an empty set.
3395 :     my @fieldList2 = map { "$relationship($_)" } grep { $_ ne "from-link" && $_ ne "to-link" }
3396 :     keys %{$relationshipFields};
3397 :     # Push these onto the return list.
3398 :     push @retVal, sort @fieldList2;
3399 :     # Return the result.
3400 :     return @retVal;
3401 :     }
3402 :    
3403 : parrello 1.48 =head3 FindColumn
3404 :    
3405 :     C<< my $colIndex = ERDB::FindColumn($headerLine, $columnIdentifier); >>
3406 :    
3407 :     Return the location a desired column in a data mining header line. The data
3408 :     mining header line is a tab-separated list of column names. The column
3409 :     identifier is either the numerical index of a column or the actual column
3410 :     name.
3411 :    
3412 :     =over 4
3413 :    
3414 :     =item headerLine
3415 :    
3416 :     The header line from a data mining command, which consists of a tab-separated
3417 :     list of column names.
3418 :    
3419 :     =item columnIdentifier
3420 :    
3421 :     Either the ordinal number of the desired column (1-based), or the name of the
3422 :     desired column.
3423 :    
3424 :     =item RETURN
3425 :    
3426 :     Returns the array index (0-based) of the desired column.
3427 :    
3428 :     =back
3429 :    
3430 :     =cut
3431 :    
3432 :     sub FindColumn {
3433 :     # Get the parameters.
3434 :     my ($headerLine, $columnIdentifier) = @_;
3435 :     # Declare the return variable.
3436 :     my $retVal;
3437 :     # Split the header line into column names.
3438 :     my @headers = ParseColumns($headerLine);
3439 :     # Determine whether we have a number or a name.
3440 :     if ($columnIdentifier =~ /^\d+$/) {
3441 :     # Here we have a number. Subtract 1 and validate the result.
3442 :     $retVal = $columnIdentifier - 1;
3443 :     if ($retVal < 0 || $retVal > $#headers) {
3444 :     Confess("Invalid column identifer \"$columnIdentifier\": value out of range.");
3445 :     }
3446 :     } else {
3447 :     # Here we have a name. We need to find it in the list.
3448 :     for (my $i = 0; $i <= $#headers && ! defined($retVal); $i++) {
3449 :     if ($headers[$i] eq $columnIdentifier) {
3450 :     $retVal = $i;
3451 :     }
3452 :     }
3453 :     if (! defined($retVal)) {
3454 :     Confess("Invalid column identifier \"$columnIdentifier\": value not found.");
3455 :     }
3456 :     }
3457 :     # Return the result.
3458 :     return $retVal;
3459 :     }
3460 :    
3461 :     =head3 ParseColumns
3462 :    
3463 : parrello 1.50 C<< my @columns = ERDB::ParseColumns($line); >>
3464 : parrello 1.48
3465 :     Convert the specified data line to a list of columns.
3466 :    
3467 :     =over 4
3468 :    
3469 :     =item line
3470 :    
3471 :     A data mining input, consisting of a tab-separated list of columns terminated by a
3472 :     new-line.
3473 :    
3474 :     =item RETURN
3475 :    
3476 :     Returns a list consisting of the column values.
3477 :    
3478 :     =back
3479 :    
3480 :     =cut
3481 :    
3482 :     sub ParseColumns {
3483 :     # Get the parameters.
3484 : parrello 1.50 my ($line) = @_;
3485 : parrello 1.48 # Chop off the line-end.
3486 :     chomp $line;
3487 :     # Split it into a list.
3488 :     my @retVal = split(/\t/, $line);
3489 :     # Return the result.
3490 :     return @retVal;
3491 :     }
3492 :    
3493 : parrello 1.71 =head2 Virtual Methods
3494 :    
3495 :     =head3 CleanKeywords
3496 :    
3497 :     C<< my $cleanedString = $erdb->CleanKeywords($searchExpression); >>
3498 :    
3499 :     Clean up a search expression or keyword list. This is a virtual method that may
3500 :     be overridden by the subclass. The base-class method removes extra spaces
3501 :     and converts everything to lower case.
3502 :    
3503 :     =over 4
3504 :    
3505 :     =item searchExpression
3506 :    
3507 :     Search expression or keyword list to clean. Note that a search expression may
3508 :     contain boolean operators which need to be preserved. This includes leading
3509 :     minus signs.
3510 :    
3511 :     =item RETURN
3512 :    
3513 :     Cleaned expression or keyword list.
3514 :    
3515 :     =back
3516 :    
3517 :     =cut
3518 :    
3519 :     sub CleanKeywords {
3520 :     # Get the parameters.
3521 :     my ($self, $searchExpression) = @_;
3522 :     # Lower-case the expression and copy it into the return variable. Note that we insure we
3523 :     # don't accidentally end up with an undefined value.
3524 :     my $retVal = lc($searchExpression || "");
3525 :     # Remove extra spaces.
3526 :     $retVal =~ s/\s+/ /g;
3527 :     $retVal =~ s/(^\s+)|(\s+$)//g;
3528 :     # Return the result.
3529 :     return $retVal;
3530 :     }
3531 :    
3532 : parrello 1.73 =head3 GetSourceObject
3533 :    
3534 :     C<< my $source = $erdb->GetSourceObject($entityName); >>
3535 :    
3536 :     Return the object to be used in loading special attributes of the specified entity. The
3537 :     algorithm for loading special attributes is stored in the C<DataGen> elements of the
3538 :     XML
3539 :    
3540 : parrello 1.1 =head2 Internal Utility Methods
3541 :    
3542 : parrello 1.70 =head3 _RelationMap
3543 :    
3544 :     C<< my @relationMap = _RelationMap($mappedNameHashRef, $mappedNameListRef); >>
3545 :    
3546 :     Create the relation map for an SQL query. The relation map is used by B<DBObject>
3547 :     to determine how to interpret the results of the query.
3548 :    
3549 :     =over 4
3550 :    
3551 :     =item mappedNameHashRef
3552 :    
3553 :     Reference to a hash that maps modified object names to real object names.
3554 :    
3555 :     =item mappedNameListRef
3556 :    
3557 :     Reference to a list of modified object names in the order they appear in the
3558 :     SELECT list.
3559 :    
3560 :     =item RETURN
3561 :    
3562 :     Returns a list of 2-tuples. Each tuple consists of an object name as used in the
3563 :     query followed by the actual name of that object. This enables the B<DBObject> to
3564 :     determine the order of the tables in the query and which object name belongs to each
3565 :     mapped object name. Most of the time these two values are the same; however, if a
3566 :     relation occurs twice in the query, the relation name in the field list and WHERE
3567 :     clause will use a mapped name (generally the actual relation name with a numeric
3568 :     suffix) that does not match the actual relation name.
3569 :    
3570 :     =back
3571 :    
3572 :     =cut
3573 :    
3574 :     sub _RelationMap {
3575 :     # Get the parameters.
3576 :     my ($mappedNameHashRef, $mappedNameListRef) = @_;
3577 :     # Declare the return variable.
3578 :     my @retVal = ();
3579 :     # Build the map.
3580 :     for my $mappedName (@{$mappedNameListRef}) {
3581 :     push @retVal, [$mappedName, $mappedNameHashRef->{$mappedName}];
3582 :     }
3583 :     # Return it.
3584 :     return @retVal;
3585 :     }
3586 :    
3587 :    
3588 :     =head3 _SetupSQL
3589 : parrello 1.45
3590 :     Process a list of object names and a filter clause so that they can be used to
3591 :     build an SQL statement. This method takes in a reference to a list of object names
3592 :     and a filter clause. It will return a corrected filter clause, a list of mapped
3593 :     names and the mapped name hash.
3594 :    
3595 :     This is an instance method.
3596 :    
3597 :     =over 4
3598 :    
3599 :     =item objectNames
3600 :    
3601 :     Reference to a list of the object names to be included in the query.
3602 :    
3603 :     =item filterClause
3604 :    
3605 :     A string containing the WHERE clause for the query (without the C<WHERE>) and also
3606 :     optionally the C<ORDER BY> and C<LIMIT> clauses.
3607 :    
3608 : parrello 1.70 =item matchClause
3609 :    
3610 :     An optional full-text search clause. If specified, it will be inserted at the
3611 :     front of the WHERE clause. It should already be SQL-formatted; that is, the
3612 :     field names should be in the form I<table>C<.>I<fieldName>.
3613 :    
3614 : parrello 1.45 =item RETURN
3615 :    
3616 :     Returns a three-element list. The first element is the SQL statement suffix, beginning
3617 :     with the FROM clause. The second element is a reference to a list of the names to be
3618 :     used in retrieving the fields. The third element is a hash mapping the names to the
3619 :     objects they represent.
3620 :    
3621 :     =back
3622 :    
3623 :     =cut
3624 :    
3625 :     sub _SetupSQL {
3626 : parrello 1.70 my ($self, $objectNames, $filterClause, $matchClause) = @_;
3627 : parrello 1.45 # Adjust the list of object names to account for multiple occurrences of the
3628 :     # same object. We start with a hash table keyed on object name that will
3629 :     # return the object suffix. The first time an object is encountered it will
3630 :     # not be found in the hash. The next time the hash will map the object name
3631 :     # to 2, then 3, and so forth.
3632 :     my %objectHash = ();
3633 :     # This list will contain the object names as they are to appear in the
3634 :     # FROM list.
3635 :     my @fromList = ();
3636 :     # This list contains the suffixed object name for each object. It is exactly
3637 :     # parallel to the list in the $objectNames parameter.
3638 :     my @mappedNameList = ();
3639 :     # Finally, this hash translates from a mapped name to its original object name.
3640 :     my %mappedNameHash = ();
3641 :     # Now we create the lists. Note that for every single name we push something into
3642 :     # @fromList and @mappedNameList. This insures that those two arrays are exactly
3643 :     # parallel to $objectNames.
3644 :     for my $objectName (@{$objectNames}) {
3645 :     # Get the next suffix for this object.
3646 :     my $suffix = $objectHash{$objectName};
3647 :     if (! $suffix) {
3648 :     # Here we are seeing the object for the first time. The object name
3649 :     # is used as is.
3650 :     push @mappedNameList, $objectName;
3651 :     push @fromList, $objectName;
3652 :     $mappedNameHash{$objectName} = $objectName;
3653 :     # Denote the next suffix will be 2.
3654 :     $objectHash{$objectName} = 2;
3655 :     } else {
3656 :     # Here we've seen the object before. We construct a new name using
3657 :     # the suffix from the hash and update the hash.
3658 :     my $mappedName = "$objectName$suffix";
3659 :     $objectHash{$objectName} = $suffix + 1;
3660 :     # The FROM list has the object name followed by the mapped name. This
3661 :     # tells SQL it's still the same table, but we're using a different name
3662 :     # for it to avoid confusion.
3663 :     push @fromList, "$objectName $mappedName";
3664 :     # The mapped-name list contains the real mapped name.
3665 :     push @mappedNameList, $mappedName;
3666 :     # Finally, enable us to get back from the mapped name to the object name.
3667 :     $mappedNameHash{$mappedName} = $objectName;
3668 :     }
3669 :     }
3670 :     # Begin the SELECT suffix. It starts with
3671 :     #
3672 :     # FROM name1, name2, ... nameN
3673 :     #
3674 :     my $suffix = "FROM " . join(', ', @fromList);
3675 : parrello 1.70 # Now for the WHERE. First, we need a place for the filter string.
3676 :     my $filterString = "";
3677 :     # We will also keep a list of conditions to add to the WHERE clause in order to link
3678 :     # entities and relationships as well as primary relations to secondary ones.
3679 :     my @joinWhere = ();
3680 : parrello 1.45 # Check for a filter clause.
3681 :     if ($filterClause) {
3682 :     # Here we have one, so we convert its field names and add it to the query. First,
3683 :     # We create a copy of the filter string we can work with.
3684 : parrello 1.70 $filterString = $filterClause;
3685 : parrello 1.45 # Next, we sort the object names by length. This helps protect us from finding
3686 :     # object names inside other object names when we're doing our search and replace.
3687 :     my @sortedNames = sort { length($b) - length($a) } @mappedNameList;
3688 :     # The final preparatory step is to create a hash table of relation names. The
3689 :     # table begins with the relation names already in the SELECT command. We may
3690 :     # need to add relations later if there is filtering on a field in a secondary
3691 :     # relation. The secondary relations are the ones that contain multiply-
3692 :     # occurring or optional fields.
3693 :     my %fromNames = map { $_ => 1 } @sortedNames;
3694 :     # We are ready to begin. We loop through the object names, replacing each
3695 :     # object name's field references by the corresponding SQL field reference.
3696 :     # Along the way, if we find a secondary relation, we will need to add it
3697 :     # to the FROM clause.
3698 :     for my $mappedName (@sortedNames) {
3699 :     # Get the length of the object name plus 2. This is the value we add to the
3700 :     # size of the field name to determine the size of the field reference as a
3701 :     # whole.
3702 :     my $nameLength = 2 + length $mappedName;
3703 :     # Get the real object name for this mapped name.
3704 :     my $objectName = $mappedNameHash{$mappedName};
3705 :     Trace("Processing $mappedName for object $objectName.") if T(4);
3706 :     # Get the object's field list.
3707 :     my $fieldList = $self->GetFieldTable($objectName);
3708 :     # Find the field references for this object.
3709 :     while ($filterString =~ m/$mappedName\(([^)]*)\)/g) {
3710 :     # At this point, $1 contains the field name, and the current position
3711 :     # is set immediately after the final parenthesis. We pull out the name of
3712 :     # the field and the position and length of the field reference as a whole.
3713 :     my $fieldName = $1;
3714 :     my $len = $nameLength + length $fieldName;
3715 :     my $pos = pos($filterString) - $len;
3716 :     # Insure the field exists.
3717 :     if (!exists $fieldList->{$fieldName}) {
3718 :     Confess("Field $fieldName not found for object $objectName.");
3719 :     } else {
3720 :     Trace("Processing $fieldName at position $pos.") if T(4);
3721 :     # Get the field's relation.
3722 :     my $relationName = $fieldList->{$fieldName}->{relation};
3723 :     # Now we have a secondary relation. We need to insure it matches the
3724 :     # mapped name of the primary relation. First we peel off the suffix
3725 :     # from the mapped name.
3726 :     my $mappingSuffix = substr $mappedName, length($objectName);
3727 :     # Put the mapping suffix onto the relation name to get the
3728 :     # mapped relation name.
3729 :     my $mappedRelationName = "$relationName$mappingSuffix";
3730 :     # Insure the relation is in the FROM clause.
3731 :     if (!exists $fromNames{$mappedRelationName}) {
3732 :     # Add the relation to the FROM clause.
3733 :     if ($mappedRelationName eq $relationName) {
3734 :     # The name is un-mapped, so we add it without
3735 :     # any frills.
3736 :     $suffix .= ", $relationName";
3737 :     push @joinWhere, "$objectName.id = $relationName.id";
3738 :     } else {
3739 :     # Here we have a mapping situation.
3740 :     $suffix .= ", $relationName $mappedRelationName";
3741 :     push @joinWhere, "$mappedRelationName.id = $mappedName.id";
3742 :     }
3743 :     # Denote we have this relation available for future fields.
3744 :     $fromNames{$mappedRelationName} = 1;
3745 :     }
3746 :     # Form an SQL field reference from the relation name and the field name.
3747 :     my $sqlReference = "$mappedRelationName." . _FixName($fieldName);
3748 :     # Put it into the filter string in place of the old value.
3749 :     substr($filterString, $pos, $len) = $sqlReference;
3750 :     # Reposition the search.
3751 :     pos $filterString = $pos + length $sqlReference;
3752 :     }
3753 :     }
3754 :     }
3755 : parrello 1.70 }
3756 :     # The next step is to join the objects together. We only need to do this if there
3757 :     # is more than one object in the object list. We start with the first object and
3758 :     # run through the objects after it. Note also that we make a safety copy of the
3759 :     # list before running through it, because we shift off the first object before
3760 :     # processing the rest.
3761 :     my @mappedObjectList = @mappedNameList;
3762 :     my $lastMappedObject = shift @mappedObjectList;
3763 :     # Get the join table.
3764 :     my $joinTable = $self->{_metaData}->{Joins};
3765 :     # Loop through the object list.
3766 :     for my $thisMappedObject (@mappedObjectList) {
3767 :     # Look for a join using the real object names.
3768 :     my $lastObject = $mappedNameHash{$lastMappedObject};
3769 :     my $thisObject = $mappedNameHash{$thisMappedObject};
3770 :     my $joinKey = "$lastObject/$thisObject";
3771 :     if (!exists $joinTable->{$joinKey}) {
3772 :     # Here there's no join, so we throw an error.
3773 :     Confess("No join exists to connect from $lastMappedObject to $thisMappedObject.");
3774 :     } else {
3775 :     # Get the join clause.
3776 :     my $unMappedJoin = $joinTable->{$joinKey};
3777 :     # Fix the names.
3778 :     $unMappedJoin =~ s/$lastObject/$lastMappedObject/;
3779 :     $unMappedJoin =~ s/$thisObject/$thisMappedObject/;
3780 :     push @joinWhere, $unMappedJoin;
3781 :     # Save this object as the last object for the next iteration.
3782 :     $lastMappedObject = $thisMappedObject;
3783 :     }
3784 :     }
3785 :     # Now we need to handle the whole ORDER BY / LIMIT thing. The important part
3786 :     # here is we want the filter clause to be empty if there's no WHERE filter.
3787 :     # We'll put the ORDER BY / LIMIT clauses in the following variable.
3788 :     my $orderClause = "";
3789 :     # This is only necessary if we have a filter string in which the ORDER BY
3790 :     # and LIMIT clauses can live.
3791 :     if ($filterString) {
3792 : parrello 1.45 # Locate the ORDER BY or LIMIT verbs (if any). We use a non-greedy
3793 :     # operator so that we find the first occurrence of either verb.
3794 :     if ($filterString =~ m/^(.*?)\s*(ORDER BY|LIMIT)/g) {
3795 :     # Here we have an ORDER BY or LIMIT verb. Split it off of the filter string.
3796 :     my $pos = pos $filterString;
3797 :     $orderClause = $2 . substr($filterString, $pos);
3798 :     $filterString = $1;
3799 :     }
3800 : parrello 1.70 }
3801 :     # All the things that are supposed to be in the WHERE clause of the
3802 :     # SELECT command need to be put into @joinWhere so we can string them
3803 :     # together. We begin with the match clause. This is important,
3804 :     # because the match clause's parameter mark must precede any parameter
3805 :     # marks in the filter string.
3806 :     if ($matchClause) {
3807 :     push @joinWhere, $matchClause;
3808 :     }
3809 :     # Add the filter string. We put it in parentheses to avoid operator
3810 :     # precedence problems with the match clause or the joins.
3811 :     if ($filterString) {
3812 :     Trace("Filter string is \"$filterString\".") if T(4);
3813 :     push @joinWhere, "($filterString)";
3814 :     }
3815 :     # String it all together into a big filter clause.
3816 :     if (@joinWhere) {
3817 :     $suffix .= " WHERE " . join(' AND ', @joinWhere);
3818 :     }
3819 :     # Add the sort or limit clause (if any).
3820 :     if ($orderClause) {
3821 :     $suffix .= " $orderClause";
3822 : parrello 1.45 }
3823 :     # Return the suffix, the mapped name list, and the mapped name hash.
3824 :     return ($suffix, \@mappedNameList, \%mappedNameHash);
3825 :     }
3826 :    
3827 : parrello 1.70 =head3 _GetStatementHandle
3828 : parrello 1.45
3829 :     This method will prepare and execute an SQL query, returning the statement handle.
3830 :     The main reason for doing this here is so that everybody who does SQL queries gets
3831 :     the benefit of tracing.
3832 :    
3833 :     This is an instance method.
3834 :    
3835 :     =over 4
3836 :    
3837 :     =item command
3838 :    
3839 :     Command to prepare and execute.
3840 :    
3841 :     =item params
3842 :    
3843 :     Reference to a list of the values to be substituted in for the parameter marks.
3844 :    
3845 :     =item RETURN
3846 :    
3847 :     Returns a prepared and executed statement handle from which the caller can extract
3848 :     results.
3849 :    
3850 :     =back
3851 :    
3852 :     =cut
3853 :    
3854 :     sub _GetStatementHandle {
3855 :     # Get the parameters.
3856 :     my ($self, $command, $params) = @_;
3857 :     # Trace the query.
3858 :     Trace("SQL query: $command") if T(SQL => 3);
3859 :     Trace("PARMS: '" . (join "', '", @{$params}) . "'") if (T(SQL => 4) && (@{$params} > 0));
3860 :     # Get the database handle.
3861 :     my $dbh = $self->{_dbh};
3862 :     # Prepare the command.
3863 :     my $sth = $dbh->prepare_command($command);
3864 :     # Execute it with the parameters bound in.
3865 : parrello 1.75 $sth->execute(@{$params}) || Confess("SELECT error: " . $sth->errstr());
3866 : parrello 1.45 # Return the statement handle.
3867 :     return $sth;
3868 :     }
3869 :    
3870 : parrello 1.70 =head3 _GetLoadStats
3871 : parrello 1.1
3872 :     Return a blank statistics object for use by the load methods.
3873 :    
3874 :     This is a static method.
3875 :    
3876 :     =cut
3877 :    
3878 : parrello 1.29 sub _GetLoadStats{
3879 : parrello 1.28 return Stats->new();
3880 : parrello 1.1 }
3881 :    
3882 : parrello 1.70 =head3 _DumpRelation
3883 : parrello 1.1
3884 : parrello 1.73 Dump the specified relation to the specified output file in tab-delimited format.
3885 : parrello 1.1
3886 :     This is an instance method.
3887 :    
3888 :     =over 4
3889 :    
3890 :     =item outputDirectory
3891 :    
3892 :     Directory to contain the output file.
3893 :    
3894 :     =item relationName
3895 :    
3896 :     Name of the relation to dump.
3897 :    
3898 :     =item relation
3899 :    
3900 :     Structure describing the relation to be dumped.
3901 :    
3902 :     =back
3903 :    
3904 :     =cut
3905 :    
3906 :     sub _DumpRelation {
3907 : parrello 1.10 # Get the parameters.
3908 :     my ($self, $outputDirectory, $relationName, $relation) = @_;
3909 :     # Open the output file.
3910 :     my $fileName = "$outputDirectory/$relationName.dtx";
3911 :     open(DTXOUT, ">$fileName") || Confess("Could not open dump file $fileName: $!");
3912 :     # Create a query for the specified relation.
3913 :     my $dbh = $self->{_dbh};
3914 :     my $query = $dbh->prepare_command("SELECT * FROM $relationName");
3915 :     # Execute the query.
3916 :     $query->execute() || Confess("SELECT error dumping $relationName.");
3917 :     # Loop through the results.
3918 :     while (my @row = $query->fetchrow) {
3919 :     # Escape any tabs or new-lines in the row text.
3920 :     for my $field (@row) {
3921 :     $field =~ s/\n/\\n/g;
3922 :     $field =~ s/\t/\\t/g;
3923 :     }
3924 :     # Tab-join the row and write it to the output file.
3925 :     my $rowText = join("\t", @row);
3926 :     print DTXOUT "$rowText\n";
3927 :     }
3928 :     # Close the output file.
3929 :     close DTXOUT;
3930 : parrello 1.1 }
3931 :    
3932 : parrello 1.70 =head3 _GetStructure
3933 : parrello 1.1
3934 :     Get the data structure for a specified entity or relationship.
3935 :    
3936 :     This is an instance method.
3937 :    
3938 :     =over 4
3939 :    
3940 :     =item objectName
3941 :    
3942 :     Name of the desired entity or relationship.
3943 :    
3944 :     =item RETURN
3945 :    
3946 :     The descriptor for the specified object.
3947 :    
3948 :     =back
3949 :    
3950 :     =cut
3951 :    
3952 :     sub _GetStructure {
3953 : parrello 1.10 # Get the parameters.
3954 :     my ($self, $objectName) = @_;
3955 :     # Get the metadata structure.
3956 :     my $metadata = $self->{_metaData};
3957 :     # Declare the variable to receive the descriptor.
3958 :     my $retVal;
3959 :     # Get the descriptor from the metadata.
3960 :     if (exists $metadata->{Entities}->{$objectName}) {
3961 :     $retVal = $metadata->{Entities}->{$objectName};
3962 :     } elsif (exists $metadata->{Relationships}->{$objectName}) {
3963 :     $retVal = $metadata->{Relationships}->{$objectName};
3964 :     } else {
3965 :     Confess("Object $objectName not found in database.");
3966 :     }
3967 :     # Return the descriptor.
3968 :     return $retVal;
3969 : parrello 1.1 }
3970 :    
3971 : parrello 1.70
3972 :    
3973 :     =head3 _GetRelationTable
3974 : parrello 1.1
3975 :     Get the list of relations for a specified entity or relationship.
3976 :    
3977 :     This is an instance method.
3978 :    
3979 :     =over 4
3980 :    
3981 :     =item objectName
3982 :    
3983 :     Name of the desired entity or relationship.
3984 :    
3985 :     =item RETURN
3986 :    
3987 :     A table containing the relations for the specified object.
3988 :    
3989 :     =back
3990 :    
3991 :     =cut
3992 :    
3993 :     sub _GetRelationTable {
3994 : parrello 1.10 # Get the parameters.
3995 :     my ($self, $objectName) = @_;
3996 :     # Get the descriptor from the metadata.
3997 :     my $objectData = $self->_GetStructure($objectName);
3998 :     # Return the object's relation list.
3999 :     return $objectData->{Relations};
4000 : parrello 1.1 }
4001 :    
4002 : parrello 1.70 =head3 _ValidateFieldNames
4003 : parrello 1.1
4004 :     Determine whether or not the field names are valid. A description of the problems with the names
4005 :     will be written to the standard error output. If there is an error, this method will abort. This is
4006 :     a static method.
4007 :    
4008 :     =over 4
4009 :    
4010 :     =item metadata
4011 :    
4012 :     Metadata structure loaded from the XML data definition.
4013 :    
4014 :     =back
4015 :    
4016 :     =cut
4017 :    
4018 :     sub _ValidateFieldNames {
4019 : parrello 1.10 # Get the object.
4020 :     my ($metadata) = @_;
4021 :     # Declare the return value. We assume success.
4022 :     my $retVal = 1;
4023 :     # Loop through the sections of the database definition.
4024 :     for my $section ('Entities', 'Relationships') {
4025 :     # Loop through the objects in this section.
4026 :     for my $object (values %{$metadata->{$section}}) {
4027 :     # Loop through the object's fields.
4028 :     for my $fieldName (keys %{$object->{Fields}}) {
4029 : parrello 1.73 # If this field name is invalid, set the return value to zero
4030 :     # so we know we encountered an error.
4031 :     if (! ValidateFieldName($fieldName)) {
4032 : parrello 1.10 $retVal = 0;
4033 :     }
4034 :     }
4035 :     }
4036 :     }
4037 :     # If an error was found, fail.
4038 :     if ($retVal == 0) {
4039 :     Confess("Errors found in field names.");
4040 :     }
4041 : parrello 1.1 }
4042 :    
4043 : parrello 1.70 =head3 _LoadRelation
4044 : parrello 1.1
4045 :     Load a relation from the data in a tab-delimited disk file. The load will only take place if a disk
4046 :     file with the same name as the relation exists in the specified directory.
4047 :    
4048 :     This is an instance method.
4049 :    
4050 :     =over 4
4051 :    
4052 :     =item dbh
4053 :    
4054 :     DBKernel object for accessing the database.
4055 :    
4056 :     =item directoryName
4057 :    
4058 :     Name of the directory containing the tab-delimited data files.
4059 :    
4060 :     =item relationName
4061 :    
4062 :     Name of the relation to load.
4063 :    
4064 :     =item rebuild
4065 :    
4066 :     TRUE if the table should be dropped and re-created before loading.
4067 :    
4068 :     =item RETURN
4069 :    
4070 :     Returns a statistical object describing the number of records read and a list of error messages.
4071 :    
4072 :     =back
4073 :    
4074 :     =cut
4075 :    
4076 :     sub _LoadRelation {
4077 : parrello 1.10 # Get the parameters.
4078 :     my ($self, $directoryName, $relationName, $rebuild) = @_;
4079 :     # Create the file name.
4080 :     my $fileName = "$directoryName/$relationName";
4081 :     # If the file doesn't exist, try adding the .dtx suffix.
4082 :     if (! -e $fileName) {
4083 :     $fileName .= ".dtx";
4084 :     if (! -e $fileName) {
4085 :     $fileName = "";
4086 :     }
4087 :     }
4088 :     # Create the return object.
4089 :     my $retVal = _GetLoadStats();
4090 :     # If a file exists to load the table, its name will be in $fileName. Otherwise, $fileName will
4091 :     # be a null string.
4092 :     if ($fileName ne "") {
4093 :     # Load the relation from the file.
4094 :     $retVal = $self->LoadTable($fileName, $relationName, $rebuild);
4095 :     } elsif ($rebuild) {
4096 :     # Here we are rebuilding, but no file exists, so we just re-create the table.
4097 :     $self->CreateTable($relationName, 1);
4098 :     }
4099 :     # Return the statistics from the load.
4100 :     return $retVal;
4101 : parrello 1.1 }
4102 :    
4103 : parrello 1.73
4104 : parrello 1.70 =head3 _LoadMetaData
4105 : parrello 1.1
4106 : parrello 1.86 C<< my $metadata = ERDB::_LoadMetaData($filename); >>
4107 :    
4108 : parrello 1.1 This method loads the data describing this database from an XML file into a metadata structure.
4109 :     The resulting structure is a set of nested hash tables containing all the information needed to
4110 :     load or use the database. The schema for the XML file is F<ERDatabase.xml>.
4111 :    
4112 :     This is a static method.
4113 :    
4114 :     =over 4
4115 :    
4116 :     =item filename
4117 :    
4118 :     Name of the file containing the database definition.
4119 :    
4120 :     =item RETURN
4121 :    
4122 :     Returns a structure describing the database.
4123 :    
4124 :     =back
4125 :    
4126 :     =cut
4127 :    
4128 :     sub _LoadMetaData {
4129 : parrello 1.10 # Get the parameters.
4130 :     my ($filename) = @_;
4131 : parrello 1.75 Trace("Reading DBD from $filename.") if T(2);
4132 : parrello 1.10 # Slurp the XML file into a variable. Extensive use of options is used to insure we
4133 :     # get the exact structure we want.
4134 : parrello 1.73 my $metadata = ReadMetaXML($filename);
4135 : parrello 1.10 # Before we go any farther, we need to validate the field and object names. If an error is found,
4136 :     # the method below will fail.
4137 :     _ValidateFieldNames($metadata);
4138 :     # Next we need to create a hash table for finding relations. The entities and relationships are
4139 :     # implemented as one or more database relations.
4140 :     my %masterRelationTable = ();
4141 :     # Loop through the entities.
4142 :     my $entityList = $metadata->{Entities};
4143 :     for my $entityName (keys %{$entityList}) {
4144 :     my $entityStructure = $entityList->{$entityName};
4145 :     #
4146 : parrello 1.12 # The first step is to create all the entity's default values. For C<Field> elements,
4147 : parrello 1.10 # the relation name must be added where it is not specified. For relationships,
4148 :     # the B<from-link> and B<to-link> fields must be inserted, and for entities an B<id>
4149 :     # field must be added to each relation. Finally, each field will have a C<PrettySort> attribute
4150 :     # added that can be used to pull the implicit fields to the top when displaying the field
4151 :     # documentation. The PrettySort values are 1-based and indicate in which pass through a
4152 :     # relation's data the field should be displayed-- 1 for the first pass, 2 for the second,
4153 :     # and so on.
4154 :     #
4155 :     # Fix up this entity.
4156 :     _FixupFields($entityStructure, $entityName, 2, 3);
4157 :     # Add the ID field.
4158 :     _AddField($entityStructure, 'id', { type => $entityStructure->{keyType},
4159 :     relation => $entityName,
4160 :     Notes => { content => "Unique identifier for this \[b\]$entityName\[/b\]." },
4161 :     PrettySort => 1});
4162 :     #
4163 :     # The current field list enables us to quickly find the relation containing a particular field.
4164 :     # We also need a list that tells us the fields in each relation. We do this by creating a
4165 :     # Relations structure in the entity structure and collating the fields into it based on their
4166 :     # C<relation> property. There is one tricky bit, which is that every relation has to have the
4167 :     # C<id> field in it. Note also that the field list is put into a C<Fields> member of the
4168 :     # relation's structure so that it looks more like the entity and relationship structures.
4169 :     #
4170 :     # First we need to create the relations list.
4171 :     my $relationTable = { };
4172 :     # Loop through the fields. We use a list of field names to prevent a problem with
4173 :     # the hash table cursor losing its place during the loop.
4174 :     my $fieldList = $entityStructure->{Fields};
4175 :     my @fieldNames = keys %{$fieldList};
4176 :     for my $fieldName (@fieldNames) {
4177 :     my $fieldData = $fieldList->{$fieldName};
4178 :     # Get the current field's relation name.
4179 :     my $relationName = $fieldData->{relation};
4180 :     # Insure the relation exists.
4181 :     if (!exists $relationTable->{$relationName}) {
4182 :     $relationTable->{$relationName} = { Fields => { } };
4183 :     }
4184 :     # Add the field to the relation's field structure.
4185 :     $relationTable->{$relationName}->{Fields}->{$fieldName} = $fieldData;
4186 :     }
4187 :     # Now that we've organized all our fields by relation name we need to do some serious
4188 :     # housekeeping. We must add the C<id> field to every relation and convert each relation
4189 :     # to a list of fields. First, we need the ID field itself.
4190 :     my $idField = $fieldList->{id};
4191 :     # Loop through the relations.
4192 :     for my $relationName (keys %{$relationTable}) {
4193 :     my $relation = $relationTable->{$relationName};
4194 :     # Get the relation's field list.
4195 :     my $relationFieldList = $relation->{Fields};
4196 :     # Add the ID field to it. If the field's already there, it will not make any
4197 :     # difference.
4198 :     $relationFieldList->{id} = $idField;
4199 :     # Convert the field set from a hash into a list using the pretty-sort number.
4200 :     $relation->{Fields} = _ReOrderRelationTable($relationFieldList);
4201 :     # Add the relation to the master table.
4202 :     $masterRelationTable{$relationName} = $relation;
4203 :     }
4204 :     # The indexes come next. The primary relation will have a unique-keyed index based on the ID field.
4205 :     # The other relations must have at least one index that begins with the ID field. In addition, the
4206 :     # metadata may require alternate indexes. We do those alternate indexes first. To begin, we need to
4207 :     # get the entity's field list and index list.
4208 :     my $indexList = $entityStructure->{Indexes};
4209 :     # Loop through the indexes.
4210 :     for my $indexData (@{$indexList}) {
4211 :     # We need to find this index's fields. All of them should belong to the same relation.
4212 :     # The ID field is an exception, since it's in all relations.
4213 :     my $relationName = '0';
4214 :     for my $fieldDescriptor (@{$indexData->{IndexFields}}) {
4215 :     # Get this field's name.
4216 :     my $fieldName = $fieldDescriptor->{name};
4217 :     # Only proceed if it is NOT the ID field.
4218 :     if ($fieldName ne 'id') {
4219 :     # Find the relation containing the current index field.
4220 :     my $thisName = $fieldList->{$fieldName}->{relation};
4221 :     if ($relationName eq '0') {
4222 :     # Here we're looking at the first field, so we save its relation name.
4223 :     $relationName = $thisName;
4224 :     } elsif ($relationName ne $thisName) {
4225 :     # Here we have a field mismatch.
4226 :     Confess("Mixed index: field $fieldName does not belong to relation $relationName.");
4227 :     }
4228 :     }
4229 :     }
4230 :     # Now $relationName is the name of the relation that contains this index. Add the index structure
4231 :     # to the relation.
4232 :     push @{$relationTable->{$relationName}->{Indexes}}, $indexData;
4233 :     }
4234 :     # Now each index has been put in a relation. We need to add the primary index for the primary
4235 :     # relation.
4236 :     push @{$relationTable->{$entityName}->{Indexes}},
4237 :     { IndexFields => [ {name => 'id', order => 'ascending'} ], Unique => 'true',
4238 :     Notes => { content => "Primary index for $entityName." }
4239 :     };
4240 :     # The next step is to insure that each relation has at least one index that begins with the ID field.
4241 :     # After that, we convert each relation's index list to an index table. We first need to loop through
4242 :     # the relations.
4243 :     for my $relationName (keys %{$relationTable}) {
4244 :     my $relation = $relationTable->{$relationName};
4245 :     # Get the relation's index list.
4246 :     my $indexList = $relation->{Indexes};
4247 :     # Insure this relation has an ID index.
4248 :     my $found = 0;
4249 :     for my $index (@{$indexList}) {
4250 :     if ($index->{IndexFields}->[0]->{name} eq "id") {
4251 :     $found = 1;
4252 :     }
4253 :     }
4254 :     if ($found == 0) {
4255 :     push @{$indexList}, { IndexFields => [ {name => 'id', order => 'ascending'} ] };
4256 :     }
4257 :     # Now we need to convert the relation's index list to an index table. We begin by creating
4258 :     # an empty table in the relation structure.
4259 :     $relation->{Indexes} = { };
4260 :     # Loop through the indexes.
4261 :     my $count = 0;
4262 :     for my $index (@{$indexList}) {
4263 :     # Add this index to the index table.
4264 : parrello 1.75 _AddIndex("idx$count", $relation, $index);
4265 : parrello 1.10 # Increment the counter so that the next index has a different name.
4266 :     $count++;
4267 :     }
4268 :     }
4269 :     # Finally, we add the relation structure to the entity.
4270 :     $entityStructure->{Relations} = $relationTable;
4271 :     }
4272 :     # Loop through the relationships. Relationships actually turn out to be much simpler than entities.
4273 :     # For one thing, there is only a single constituent relation.
4274 :     my $relationshipList = $metadata->{Relationships};
4275 :     for my $relationshipName (keys %{$relationshipList}) {
4276 :     my $relationshipStructure = $relationshipList->{$relationshipName};
4277 :     # Fix up this relationship.
4278 :     _FixupFields($relationshipStructure, $relationshipName, 2, 3);
4279 :     # Format a description for the FROM field.
4280 :     my $fromEntity = $relationshipStructure->{from};
4281 : parrello 1.83 my $fromComment = "[b]id[/b] of the source [b][link #$fromEntity]$fromEntity\[/link][/b].";
4282 : parrello 1.10 # Get the FROM entity's key type.
4283 :     my $fromType = $entityList->{$fromEntity}->{keyType};
4284 :     # Add the FROM field.
4285 :     _AddField($relationshipStructure, 'from-link', { type => $fromType,
4286 :     relation => $relationshipName,
4287 :     Notes => { content => $fromComment },
4288 :     PrettySort => 1});
4289 :     # Format a description for the TO field.
4290 :     my $toEntity = $relationshipStructure->{to};
4291 : parrello 1.83 my $toComment = "[b]id[/b] of the target [b][link #$toEntity]$toEntity\[/link][/b].";
4292 : parrello 1.10 # Get the TO entity's key type.
4293 :     my $toType = $entityList->{$toEntity}->{keyType};
4294 :     # Add the TO field.
4295 :     _AddField($relationshipStructure, 'to-link', { type=> $toType,
4296 :     relation => $relationshipName,
4297 :     Notes => { content => $toComment },
4298 :     PrettySort => 1});
4299 :     # Create an index-free relation from the fields.
4300 :     my $thisRelation = { Fields => _ReOrderRelationTable($relationshipStructure->{Fields}),
4301 :     Indexes => { } };
4302 :     $relationshipStructure->{Relations} = { $relationshipName => $thisRelation };
4303 :     # Create the FROM and TO indexes.
4304 :     _CreateRelationshipIndex("From", $relationshipName, $relationshipStructure);
4305 :     _CreateRelationshipIndex("To", $relationshipName, $relationshipStructure);
4306 :     # Add the relation to the master table.
4307 :     $masterRelationTable{$relationshipName} = $thisRelation;
4308 :     }
4309 :     # Now store the master relation table in the metadata structure.
4310 :     $metadata->{RelationTable} = \%masterRelationTable;
4311 :     # Our final task is to create the join table. The join table is a hash that describes all
4312 :     # the join clauses for traveling through the relationships. The join clause is an equality
4313 :     # condition that can be put into a WHERE clause in order to join two objects. Two relationships
4314 :     # can be joined if they share an entity in common; and an entity can be joined to a relationship
4315 :     # if the entity is at either end of the relationship.
4316 :     my %joinTable = ();
4317 :     # Loop through the entities.
4318 :     for my $entityName (keys %{$entityList}) {
4319 :