Merge pull request #47 from CBIIT/sts-related-updates

Sts related updates

cypher/mdb-indexes.4.4.cypher

@@ -12,8 +12,8 @@ create fulltext index termValueDefn for (t:term) on each [t.value, t.origin_defi
 
 create fulltext index tagKeyValue for (g:tag) on each [g.key, g.value];
 
-create fulltext index nodeHandle on (n:node) on each [n.handle];
-create fulltext index edgeHandle on (n:relationship) on each [n.handle];
-create fulltext index propHandle on (n:property) on each [n.handle];
+create fulltext index nodeHandle for (n:node) on each [n.handle];
+create fulltext index edgeHandle for (n:relationship) on each [n.handle];
+create fulltext index propHandle for (n:property) on each [n.handle];
 
 

docs/_modules/bento_meta/mdb/mdb.html

@@ -323,8 +323,8 @@ <h1>Source code for bento_meta.mdb.mdb</h1><div class="highlight"><pre>
         <span class="n">qry</span> <span class="o">=</span> <span class="p">(</span>
             <span class="s2">&quot;match (t:term {nanoid:$nanoid}) &quot;</span>
             <span class="s2">&quot;where not exists(t._to) &quot;</span>
-            <span class="s2">&quot;with t &quot;</span>
-            <span class="s2">&quot;optional match (t)-[:has_origin]-&gt;(o:origin) &quot;</span>
+            <span class="s2">&quot;with t, t.origin_name as origin_name &quot;</span>
+            <span class="s2">&quot;optional match (o:origin {name: origin_name}) &quot;</span>
             <span class="s2">&quot;where not exists(o._to) &quot;</span>
             <span class="s2">&quot;return t as term, o as origin &quot;</span>
             <span class="p">)</span>

docs/_sources/mdb-conventions.rst.txt

@@ -81,7 +81,7 @@ _______________________________________________
 
 For integration with the STS, and for performance, the Neo4j instance
 of an MDB requires certain indexes to be established. These are
-specified in `these Cypher statements <githubref>`_. The primary
+specified in `these Cypher statements <https://github.com/CBIIT/bento-mdb/blob/main/changelogs/mdb_setup_changelog.xml>`_. The primary
 requirement is that fulltext, Lucene-based indexes should be created
 on entity ``description`` and Term ``origin_definition`` properties to
 enable "search box" like queries over the the entire graph. Regular
@@ -120,7 +120,7 @@ on `Cypher <https://neo4j.com/docs/cypher-manual/current/>`_.)
 
 * *Concept*: For any `(c:concept)` and `(k:concept)` where `(n)--(c)` and `(n)--(k)` return exactly the same graph nodes `[n1, n2, ...]`, one of `c` or `k` should be removed from the database.
 
-* *Tag*: For two tag nodes, all of whose properties except nanoid are identical in key and value, and which are linked to exactly the same graph nodes, one must be removed.
+* *Tag*: Any node should have *at most one* Tag node with a given pair of values for `key` and `value` properties. Tags should attach to at most one node; that is, the `()-[:has_tag]->(:tag)` relationship is one-to-many, not many-to-many. Practically, this means that to use Tags to group a set of different nodes with a key-value pair, a separate Tag with that key-value pair should be created and linked to each nodes. 
 
 * *Nanoid*: Each unique unversioned graph node as defined above must have a unique nanoid.
 

docs/mdb-conventions.html

@@ -164,7 +164,7 @@ <h2>MDB entity properties required for consistency<a class="headerlink" href="#m
 <h2>MDB indexes needed for additional functionality<a class="headerlink" href="#mdb-indexes-needed-for-additional-functionality" title="Permalink to this heading"></a></h2>
 <p>For integration with the STS, and for performance, the Neo4j instance
 of an MDB requires certain indexes to be established. These are
-specified in <a class="reference external" href="githubref">these Cypher statements</a>. The primary
+specified in <a class="reference external" href="https://github.com/CBIIT/bento-mdb/blob/main/changelogs/mdb_setup_changelog.xml">these Cypher statements</a>. The primary
 requirement is that fulltext, Lucene-based indexes should be created
 on entity <code class="docutils literal notranslate"><span class="pre">description</span></code> and Term <code class="docutils literal notranslate"><span class="pre">origin_definition</span></code> properties to
 enable “search box” like queries over the the entire graph. Regular
@@ -200,7 +200,7 @@ <h3>Uniqueness and Existence of Entities<a class="headerlink" href="#uniqueness-
 <li><p><em>Value Set</em>: For <cite>(p:property)</cite> with <cite>p.value_domain == “value_set”</cite>, then one and only one value_set <cite>v</cite> with <cite>(p)-[:has_value_set]-&gt;(v:value_set)</cite> must exist.</p></li>
 <li><p><em>Term</em>: For a term <cite>(t:term)</cite>, only one graph node with <cite>[t.origin_name, t.origin_id, t.origin_version]</cite> may exist, <em>even in a versioned MDB</em>.</p></li>
 <li><p><em>Concept</em>: For any <cite>(c:concept)</cite> and <cite>(k:concept)</cite> where <cite>(n)–(c)</cite> and <cite>(n)–(k)</cite> return exactly the same graph nodes <cite>[n1, n2, …]</cite>, one of <cite>c</cite> or <cite>k</cite> should be removed from the database.</p></li>
-<li><p><em>Tag</em>: For two tag nodes, all of whose properties except nanoid are identical in key and value, and which are linked to exactly the same graph nodes, one must be removed.</p></li>
+<li><p><em>Tag</em>: Any node should have <em>at most one</em> Tag node with a given pair of values for <cite>key</cite> and <cite>value</cite> properties. Tags should attach to at most one node; that is, the <cite>()-[:has_tag]-&gt;(:tag)</cite> relationship is one-to-many, not many-to-many. Practically, this means that to use Tags to group a set of different nodes with a key-value pair, a separate Tag with that key-value pair should be created and linked to each nodes.</p></li>
 <li><p><em>Nanoid</em>: Each unique unversioned graph node as defined above must have a unique nanoid.</p>
 <ul>
 <li><p>The nanoid <em>shall not change</em> when a graph node is updated, provided above conditions remain satisfied upon update.</p></li>

python/docs/mdb-conventions.rst

@@ -81,7 +81,7 @@ _______________________________________________
 
 For integration with the STS, and for performance, the Neo4j instance
 of an MDB requires certain indexes to be established. These are
-specified in `these Cypher statements <githubref>`_. The primary
+specified in `these Cypher statements <https://github.com/CBIIT/bento-mdb/blob/main/changelogs/mdb_setup_changelog.xml>`_. The primary
 requirement is that fulltext, Lucene-based indexes should be created
 on entity ``description`` and Term ``origin_definition`` properties to
 enable "search box" like queries over the the entire graph. Regular
@@ -120,7 +120,7 @@ on `Cypher <https://neo4j.com/docs/cypher-manual/current/>`_.)
 
 * *Concept*: For any `(c:concept)` and `(k:concept)` where `(n)--(c)` and `(n)--(k)` return exactly the same graph nodes `[n1, n2, ...]`, one of `c` or `k` should be removed from the database.
 
-* *Tag*: For two tag nodes, all of whose properties except nanoid are identical in key and value, and which are linked to exactly the same graph nodes, one must be removed.
+* *Tag*: Any node should have *at most one* Tag node with a given pair of values for `key` and `value` properties. Tags should attach to at most one node; that is, the `()-[:has_tag]->(:tag)` relationship is one-to-many, not many-to-many. Practically, this means that to use Tags to group a set of different nodes with a key-value pair, a separate Tag with that key-value pair should be created and linked to each nodes. 
 
 * *Nanoid*: Each unique unversioned graph node as defined above must have a unique nanoid.
 

python/pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "bento-meta"
-version = "0.2.2"
+version = "0.2.3"
 description = "Python drivers for Bento Metamodel Database"
 authors = [
   { name="Mark A. Jensen", email = "mark.jensen@nih.gov"},
@@ -20,7 +20,7 @@ classifiers = [
 
 [tool.poetry]
 name = "bento-meta"
-version = "0.2.2"
+version = "0.2.3"
 description = "Python drivers for Bento Metamodel Database"
 authors = [
     "Mark A. Jensen <mark.jensen@nih.gov>",

python/src/bento_meta/mdb/mdb.py

@@ -244,8 +244,8 @@ def get_term_by_id(self, nanoid):
         qry = (
             "match (t:term {nanoid:$nanoid}) "
             "where not exists(t._to) "
-            "with t "
-            "optional match (t)-[:has_origin]->(o:origin) "
+            "with t, t.origin_name as origin_name "
+            "optional match (o:origin {name: origin_name}) "
             "where not exists(o._to) "
             "return t as term, o as origin "
             )

python/tox.ini

@@ -7,5 +7,6 @@ deps = pytest
      docker-compose
      pytest-docker
 commands =
-      pytest
+      pytest -m "not slow"
+