[Bio] / Sprout / AttributesDBD.xml Repository:
ViewVC logotype

Diff of /Sprout/AttributesDBD.xml

Parent Directory Parent Directory | Revision Log Revision Log | View Patch Patch

revision 1.6, Thu Apr 12 06:00:02 2007 UTC revision 1.10, Tue Aug 12 06:06:02 2008 UTC
# Line 1  Line 1 
 <?xml version='1.0' standalone='yes'?>  
1  <Database>  <Database>
2    <Title>Attributes Database</Title>    <Title>Attributes Database</Title>
3      <Notes>The Attributes Database allows applications to upload and retrieve
4    custom data for NMPDR data objects. It uses the ERDB database system (which
5    is also used by the [[SproutDatabase]] and the [[SaplingDatabase]]) to
6    store the attributes. Attributes differ from conventional data in that they
7    are unstructured: new attributes can be added without modifying the database
8    design. A single attribute database services the entire NMPDR complex.
9    %BR%%BR%
10    Attributes are organized by _attribute key_. Attribute values are
11    assigned to _objects_. In the real world, objects have types and IDs;
12    however, to the attribute database only the ID matters. In general, it
13    is possible to determine the object type from the format of the ID, but
14    it doesn't matter to the database.
15    %BR%%BR%
16    The actual attribute values are stored as a relationship between the attribute
17    keys and the objects. There can be multiple values for a single key/object pair.
18    
19    </Notes>
20    <Entities>    <Entities>
21      <Entity name="AttributeGroup" keyType="name-string">      <Entity name="AttributeGroup" keyType="name-string">
22        <Notes>A [i]group[/i] is a named set of attribute keys. An attribute key can belong to        <DisplayInfo theme="seed" col="3" row="1" caption="Attribute Group" />
23          <Notes>A group is a named set of attribute keys. An attribute key can belong to
24        many groups or none at all.</Notes>        many groups or none at all.</Notes>
25          <Asides>Groups provide a useful mechanism for classifying certain types of attribute
26          keys. An attribute key's group membership is updated using a table of checkboxes
27          in the attribute web interface. The interface also allows new groups to be created.
28          There is a group for attributes that are only used for %FIG{ProteinEncodingGroups}%
29          and another for attributes that require NMPDR support. Groups can also be used
30          to specify attribute key white lists for individual applications.</Asides>
31      </Entity>      </Entity>
32      <Entity name="AttributeKey" keyType="name-string">      <Entity name="AttributeKey" keyType="name-string">
33        <Notes>An [i]attribute key[/i] describes named values attached to a particular type of object.        <DisplayInfo theme="seed" col="3" row="3" caption="Attribute Key" />
34          <Notes>An attribute key describes named values attached to a particular type of object.
35        Each instance of this entity contains the attribute's description and its name. An attribute        Each instance of this entity contains the attribute's description and its name. An attribute
36        key name must consist of letters, numbers, periods, underscores, and hyphens.</Notes>        key name must consist of letters, numbers, periods, underscores, and/or hyphens.</Notes>
37        <Fields>        <Fields>
38          <Field name="description" type="text">          <Field name="description" type="text">
39            <Notes>Text describing the attribute. This should explain what the attribute means and            <Notes>Text describing the attribute. This should explain what the attribute means and
40            what kinds of values are appropriate.</Notes>            what kinds of values are appropriate.</Notes>
41          </Field>          </Field>
42          <Field name="data-type" type="name-string">          <Field name="relationship-name" type="name-string">
43            <Notes>ERDB data type for the attribute. This serves as documentation only, but            <Notes>Name of the relationship containing the attribute data. This relationship
44            that may change at some future point.</Notes>            must look exactly like HasValueFor. The only difference is the table name.</Notes>
45              <Asides>In order to improve performance of batch updates, some attributes
46              are stored in private relationship tables. When retrieving attributes, we need
47              to check each attribute key's private table for the data. The tables have
48              the same structure as HasValueFor. They are separated solely for performance
49              reasons.</Asides>
50          </Field>          </Field>
51        </Fields>        </Fields>
52      </Entity>      </Entity>
53      <Entity name="TargetObject" keyType="string">      <Entity name="TargetObject" keyType="string">
54        <Notes>An [i]object[/i] represents a keyed thing to which an attribute can be assigned.        <DisplayInfo theme="neutral" col="3" row="5" caption="Target Object" />
55          <Notes>An object represents a keyed thing to which an attribute can be assigned.
56        In order to improve performance, this table has no data in it. It is simply required by        In order to improve performance, this table has no data in it. It is simply required by
57        the semantics of the database.</Notes>        the semantics of the database.</Notes>
58      </Entity>      </Entity>
59    </Entities>    </Entities>
60    <Relationships>    <Relationships>
61        <Relationship name="IsEvidencedBy" to="TargetObject" from="AttributeKey" arity="MM" converse="HasEvidence">
62          <DisplayInfo theme="seed" col="3.5" row="4" fixed="1" caption="Is\nEvidenced\nBy" />
63          <Notes>This relationship connects a feature (target object) to its evidence
64          attributes.</Notes>
65          <Asides>This table exists to separate evidence codes from other attributes for
66          performance reasons. Evidence codes are built from scratch several times a week.
67          Erasing an entire table is faster than erasing individual rows, so putting the
68          evidence codes in their own private table provides significant benefits.</Asides>
69          <Fields>
70            <Field name="subkey" type="string">
71                <Notes>Some attribute keys are split into two parts. The real key is controlled and
72                must correspond to a key in the AttributeKey table. The second part of the key
73                resides in this field. When the attribute value is presented to the user, the two
74                parts of the key are joined together around the splitter value.</Notes>
75            </Field>
76            <Field name="value" type="text">
77              <Notes>This is the actual attribute value. Note that a single key/object pair may in fact
78              have many values, and therefore many entries in this table.</Notes>
79            </Field>
80          </Fields>
81          <FromIndex>
82            <Notes>This index sorts all the values for a single attribute key by the subkey value.</Notes>
83            <IndexFields>
84                <IndexField name="subkey" order="ascending" />
85            </IndexFields>
86          </FromIndex>
87          <ToIndex>
88            <Notes>This index sorts all the values for an object by attribute key and the subkey.</Notes>
89            <IndexFields>
90                <IndexField name="from-link" order="ascending" />
91                <IndexField name="subkey" order="ascending" />
92            </IndexFields>
93          </ToIndex>
94          <Indexes>
95            <Index>
96                <Notes>This index allows searching for an attribute value by its prefix.</Notes>
97                <IndexFields>
98                    <IndexField name="value" order="ascending" />
99                </IndexFields>
100            </Index>
101            <Index>
102                <Notes>This index allows searching for values within keys.</Notes>
103                <IndexFields>
104                    <IndexField name="from-link" order="ascending" />
105                    <IndexField name="value" order="ascending" />
106                </IndexFields>
107            </Index>
108            <Index>
109                <Notes>This index sorts all the values for a single attribute key by the target object ID.</Notes>
110                <IndexFields>
111                    <IndexField name="from-link" order="ascending" />
112                    <IndexField name="to-link" order="ascending" />
113                </IndexFields>
114            </Index>
115          </Indexes>
116        </Relationship>
117      <Relationship name="IsInGroup" from="AttributeKey" to="AttributeGroup" arity="MM">      <Relationship name="IsInGroup" from="AttributeKey" to="AttributeGroup" arity="MM">
118          <DisplayInfo theme="seed" caption="Is\nGroup\nOf" />
119        <Notes>This relationship connects an attribute key to the groups of which it is a member.</Notes>        <Notes>This relationship connects an attribute key to the groups of which it is a member.</Notes>
120      </Relationship>      </Relationship>
121      <Relationship name="HasValueFor" from="AttributeKey" to="TargetObject" arity="MM">      <Relationship name="HasValueFor" from="AttributeKey" to="TargetObject" arity="MM" converse="HasValueOf">
122          <DisplayInfo theme="seed" col="2.5" row="4" fixed="1" caption="Has\nValue\nFor" />
123        <Notes>This relationship connects an attribute key to its objects. It contains the        <Notes>This relationship connects an attribute key to its objects. It contains the
124        attribute values, and is therefore the primary data repository for the database.</Notes>        attribute values, and is therefore the primary data repository for the database.</Notes>
125        <Fields>        <Fields>
# Line 52  Line 140 
140              <IndexField name="subkey" order="ascending" />              <IndexField name="subkey" order="ascending" />
141          </IndexFields>          </IndexFields>
142        </FromIndex>        </FromIndex>
143          <ToIndex>
144            <Notes>This index sorts all the values for an object by attribute key and the subkey.</Notes>
145            <IndexFields>
146                <IndexField name="from-link" order="ascending" />
147                <IndexField name="subkey" order="ascending" />
148            </IndexFields>
149          </ToIndex>
150        <Indexes>        <Indexes>
151          <Index>          <Index>
152              <Notes>This index allows searching for an attribute value by its prefix.</Notes>              <Notes>This index allows searching for an attribute value by its prefix.</Notes>
# Line 66  Line 161 
161                  <IndexField name="value" order="ascending" />                  <IndexField name="value" order="ascending" />
162              </IndexFields>              </IndexFields>
163          </Index>          </Index>
164            <Index>
165                <Notes>This index sorts all the values for a single attribute key by the target object ID.</Notes>
166                <IndexFields>
167                    <IndexField name="from-link" order="ascending" />
168                    <IndexField name="to-link" order="ascending" />
169                </IndexFields>
170            </Index>
171        </Indexes>        </Indexes>
172      </Relationship>      </Relationship>
173    </Relationships>    </Relationships>
174      <Shapes>
175      </Shapes>
176  </Database>  </Database>
Legend:
Removed from v.1.6  
changed lines
  Added in v.1.10
MCS Webmaster
 ViewVC Help
Powered by ViewVC 1.0.3