Merge pull request #3491 from volodya-lombrozo/3490_add_refs

feat(#3490): Attempt To Speed Up The `add-refs.xsl`
objectionary · Nov 15, 2024 · a0a4afe · a0a4afe
2 parents e4427a8 + 3230154
commit a0a4afe
Show file tree

Hide file tree

Showing 3 changed files with 86 additions and 49 deletions.
diff --git a/README.md b/README.md
@@ -35,30 +35,30 @@ All of them have something **we don't tolerate**:
 
 * types ([why?](https://www.yegor256.com/2020/11/10/typing-without-types.html))
 * static/class methods or attributes
-([why?](http://www.yegor256.com/2014/05/05/oop-alternative-to-utility-classes.html))
+  ([why?](http://www.yegor256.com/2014/05/05/oop-alternative-to-utility-classes.html))
 * classes ([why?](http://www.yegor256.com/2016/09/20/oop-without-classes.html))
 * implementation inheritance
-([why?](http://www.yegor256.com/2016/09/13/inheritance-is-procedural.html))
+  ([why?](http://www.yegor256.com/2016/09/13/inheritance-is-procedural.html))
 * mutability
-([why?](http://www.yegor256.com/2014/06/09/objects-should-be-immutable.html)
-and
-[why not?](https://www.yegor256.com/2016/09/07/gradients-of-immutability.html))
+  ([why?](http://www.yegor256.com/2014/06/09/objects-should-be-immutable.html)
+  and
+  [why not?](https://www.yegor256.com/2016/09/07/gradients-of-immutability.html))
 * NULL ([why?](http://www.yegor256.com/2014/05/13/why-null-is-bad.html))
 * global scope
-([why?](https://www.yegor256.com/2018/07/03/global-variables.html))
+  ([why?](https://www.yegor256.com/2018/07/03/global-variables.html))
 * type casting
-([why?](http://www.yegor256.com/2015/04/02/class-casting-is-anti-pattern.html))
+  ([why?](http://www.yegor256.com/2015/04/02/class-casting-is-anti-pattern.html))
 * reflection
-([why?](https://www.yegor256.com/2022/06/05/reflection-means-hidden-coupling.html))
+  ([why?](https://www.yegor256.com/2022/06/05/reflection-means-hidden-coupling.html))
 * scalar types and data primitives
 * annotations
-([why?](http://www.yegor256.com/2016/04/12/java-annotations-are-evil.html))
+  ([why?](http://www.yegor256.com/2016/04/12/java-annotations-are-evil.html))
 * operators
 * traits and mixins
-([why?](https://www.yegor256.com/2017/03/07/traits-and-mixins.html))
+  ([why?](https://www.yegor256.com/2017/03/07/traits-and-mixins.html))
 * flow control statements (`for`, `while`, `if`, etc)
 * [syntactic sugar](https://en.wikipedia.org/wiki/Syntactic_sugar)
-([why?](https://github.com/objectionary/eo/issues/51))
+  ([why?](https://github.com/objectionary/eo/issues/51))
 
 ## Quick Start
 
@@ -242,7 +242,7 @@ Read about integration with Maven,
 
 ## Benchmark
 
-This is how many milliseconds were spend on different
+This is how many milliseconds were spent on different
 XSL stylesheets during the execution of `mvn install` of
 the `eo-runtime` module:
 
@@ -275,6 +275,21 @@ We show only the first 16 most expensive XSL stylesheets.
 
 <!-- benchmark_end -->
 
+You can run this benchmark locally with the following commands.
+First, to generate the `measures.csv` file:
+
+```shell
+mvn clean install --errors --batch-mode -Deo.xslMeasuresFile=measures.csv
+```
+
+Then, to generate the report:
+
+```shell
+awk -F ',' '{ a[$1]+=$2; s+=$2; } END { for (k in a) \
+ printf("%s.xsl\t%d\t%0.2f%%\n", k, a[k], 100 * a[k]/s)}' \
+ eo-runtime/measures.csv | sort -g -k 2 | tail -r | column -t | head "-16"
+```
+
 ## How to Contribute
 
 Fork repository, make changes, then send us
@@ -297,4 +312,5 @@ to enhance the performance of EO components:
 [![YourKit](https://www.yourkit.com/images/yklogo.png)](https://www.yourkit.com)
 
 [cargo]: https://doc.rust-lang.org/cargo/getting-started/installation.html
+
 [benchmark-gha]: https://github.com/objectionary/eo/actions/runs/11770647245
diff --git a/eo-parser/src/main/resources/org/eolang/parser/add-refs.xsl b/eo-parser/src/main/resources/org/eolang/parser/add-refs.xsl
@@ -22,7 +22,7 @@ LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
 OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
 SOFTWARE.
 -->
-<xsl:stylesheet xmlns:xsl="http://www.w3.org/1999/XSL/Transform" id="add-refs" version="2.0">
+<xsl:stylesheet xmlns:xsl="http://www.w3.org/1999/XSL/Transform" xmlns:eo="https://www.eolang.org" xmlns:xs="http://www.w3.org/2001/XMLSchema" id="add-refs" version="2.0">
   <!--
   Here we go through all objects and find what their @base
   are referring to. If we find the object they refer to,
@@ -38,6 +38,7 @@ SOFTWARE.
   defined in this particular file.
   -->
   <xsl:output encoding="UTF-8" method="xml"/>
+  <xsl:key name="o-by-name" match="o[@name]" use="@name"/>
   <xsl:template match="o[not(@base='bytes' and /program/metas/meta[head='package' and tail='org.eolang'] and /program/objects/o[@name='bytes'])]">
     <xsl:apply-templates select="." mode="not-bytes"/>
   </xsl:template>
@@ -49,44 +50,41 @@ SOFTWARE.
   </xsl:template>
   <xsl:template match="o[@base!='$' and @base!='^']" mode="no-dots">
     <xsl:variable name="current" select="."/>
+    <xsl:variable name="source" select="eo:closest($current, key('o-by-name', @base))"/>
     <xsl:copy>
-      <xsl:variable name="parent" select="ancestor::*[o[@name=$current/@base]][1]"/>
-      <xsl:if test="$parent">
-        <xsl:variable name="source" select="$parent/o[@name=$current/@base]"/>
-        <xsl:if test="$parent">
-          <xsl:if test="count($source)!=1">
-            <xsl:message terminate="yes">
-              <xsl:text>Duplicate names inside "</xsl:text>
-              <xsl:value-of select="@name"/>
-              <xsl:text>", the base is "</xsl:text>
-              <xsl:value-of select="@base"/>
-              <xsl:text>" at the line #</xsl:text>
+      <xsl:if test="$source">
+        <xsl:if test="count($source)!=1">
+          <xsl:message terminate="yes">
+            <xsl:text>Duplicate names inside "</xsl:text>
+            <xsl:value-of select="@name"/>
+            <xsl:text>", the base is "</xsl:text>
+            <xsl:value-of select="@base"/>
+            <xsl:text>" at the line #</xsl:text>
+            <xsl:value-of select="@line"/>
+            <xsl:text> pointing to </xsl:text>
+            <xsl:for-each select="$source">
+              <xsl:if test="position()&gt;1">
+                <xsl:text>, </xsl:text>
+              </xsl:if>
+              <xsl:text>&lt;</xsl:text>
+              <xsl:value-of select="name(.)"/>
+              <xsl:text>/&gt;</xsl:text>
+              <xsl:text> at line #</xsl:text>
               <xsl:value-of select="@line"/>
-              <xsl:text> pointing to </xsl:text>
-              <xsl:for-each select="$source">
-                <xsl:if test="position()&gt;1">
-                  <xsl:text>, </xsl:text>
-                </xsl:if>
-                <xsl:text>&lt;</xsl:text>
-                <xsl:value-of select="name(.)"/>
-                <xsl:text>/&gt;</xsl:text>
-                <xsl:text> at line #</xsl:text>
-                <xsl:value-of select="@line"/>
-              </xsl:for-each>
-              <xsl:text>; it's internal bug</xsl:text>
-            </xsl:message>
-          </xsl:if>
-          <xsl:if test="not($source/@line)">
-            <xsl:message terminate="yes">
-              <xsl:text>Attribute @line is absent at "</xsl:text>
-              <xsl:value-of select="$source/@name"/>
-              <xsl:text>"</xsl:text>
-            </xsl:message>
-          </xsl:if>
-          <xsl:attribute name="ref">
-            <xsl:value-of select="$source/@line"/>
-          </xsl:attribute>
+            </xsl:for-each>
+            <xsl:text>; it's internal bug</xsl:text>
+          </xsl:message>
         </xsl:if>
+        <xsl:if test="not($source/@line)">
+          <xsl:message terminate="yes">
+            <xsl:text>Attribute @line is absent at "</xsl:text>
+            <xsl:value-of select="$source/@name"/>
+            <xsl:text>"</xsl:text>
+          </xsl:message>
+        </xsl:if>
+        <xsl:attribute name="ref">
+          <xsl:value-of select="$source/@line"/>
+        </xsl:attribute>
       </xsl:if>
       <xsl:apply-templates select="node()|@*"/>
     </xsl:copy>
@@ -96,4 +94,27 @@ SOFTWARE.
       <xsl:apply-templates select="node()|@*"/>
     </xsl:copy>
   </xsl:template>
+  <xsl:function name="eo:closest" as="node()?">
+    <xsl:param name="current-node" as="node()"/>
+    <xsl:param name="nodes" as="node()*"/>
+    <xsl:variable name="intersecting-nodes" select="$nodes[eo:has-intersecting-route($current-node, .)]"/>
+    <xsl:for-each select="$intersecting-nodes">
+      <xsl:sort select="string-length(eo:get-route(.))" order="descending" data-type="number"/>
+      <xsl:if test="position() = 1">
+        <xsl:copy-of select="."/>
+      </xsl:if>
+    </xsl:for-each>
+  </xsl:function>
+  <xsl:function name="eo:has-intersecting-route" as="xs:boolean">
+    <xsl:param name="node1" as="node()"/>
+    <xsl:param name="node2" as="node()"/>
+    <xsl:variable name="route1" select="eo:get-route($node1)"/>
+    <xsl:variable name="route2" select="eo:get-route($node2)"/>
+    <xsl:sequence select="starts-with($route1, $route2)"/>
+  </xsl:function>
+  <xsl:function name="eo:get-route" as="xs:string">
+    <xsl:param name="node" as="node()"/>
+    <xsl:variable name="ancestors" select="$node/ancestor::*"/>
+    <xsl:sequence select="string-join(for $ancestor in $ancestors return generate-id($ancestor), '/')"/>
+  </xsl:function>
 </xsl:stylesheet>
diff --git a/src/test/groovy/check-xsl-id.groovy b/src/test/groovy/check-xsl-id.groovy
@@ -34,7 +34,7 @@ project = new File('.')
 project.traverse(
   type         : FileType.FILES,
   preDir       : { if (it.name == 'target') return FileVisitResult.SKIP_SUBTREE },
-  nameFilter   : ~/.*\.xsl/
+  nameFilter   : ~/.*\.xsl|xs3p.xsl/
 ) {
   it ->
     String id = new XmlSlurper().parse(it).@id