Skip to content

Commit

Permalink
8260581: IGV: enhance node search
Browse files Browse the repository at this point in the history
Allow users to search by node id or name by default, show partial matches when
searching for a specific property, show 'All N matching nodes' entry only if
relevant, and rank results by level of matching.

Co-authored-by: Christian Hagedorn <chagedorn@openjdk.org>
Reviewed-by: chagedorn, vlivanov, xliu
  • Loading branch information
2 people authored and Vladimir Ivanov committed Feb 3, 2021
1 parent 9037615 commit ae2c5f0
Show file tree
Hide file tree
Showing 3 changed files with 95 additions and 24 deletions.
Original file line number Diff line number Diff line change
@@ -1,5 +1,5 @@
/*
* Copyright (c) 2008, 2015, Oracle and/or its affiliates. All rights reserved.
* Copyright (c) 2008, 2021, Oracle and/or its affiliates. All rights reserved.
* DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER.
*
* This code is free software; you can redistribute it and/or modify it
Expand Down Expand Up @@ -274,6 +274,13 @@ void removeOutputSlot(OutputSlot s) {
public String[] getLines() {
if (lines == null) {
updateLines();
// Set the "label" property of each input node, so that by default
// search is done on the node label (without line breaks). See also
// class NodeQuickSearch in the View module.
for (InputNode n : getSource().getSourceNodes()) {
String label = resolveString(diagram.getNodeText(), n.getProperties());
n.getProperties().setProperty("label", label);
}
}
return lines;
}
Expand Down
Original file line number Diff line number Diff line change
@@ -1,2 +1,2 @@
javac.source=1.7
javac.source=1.8
javac.compilerargs=-Xlint -Xlint:-serial
Original file line number Diff line number Diff line change
@@ -1,5 +1,5 @@
/*
* Copyright (c) 2008, 2016, Oracle and/or its affiliates. All rights reserved.
* Copyright (c) 2008, 2021, Oracle and/or its affiliates. All rights reserved.
* DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER.
*
* This code is free software; you can redistribute it and/or modify it
Expand Down Expand Up @@ -29,6 +29,7 @@
import com.sun.hotspot.igv.data.Properties.RegexpPropertyMatcher;
import com.sun.hotspot.igv.data.services.InputGraphProvider;
import com.sun.hotspot.igv.util.LookupHistory;
import java.util.Collections;
import java.util.HashSet;
import java.util.List;
import java.util.Set;
Expand All @@ -46,7 +47,7 @@
*/
public class NodeQuickSearch implements SearchProvider {

private static final String DEFAULT_PROPERTY = "name";
private static final String DEFAULT_PROPERTY = "label";

/**
* Method is called by infrastructure when search operation was requested.
Expand All @@ -66,18 +67,17 @@ public void evaluate(SearchRequest request, SearchResponse response) {
final String[] parts = query.split("=", 2);

String name;
String rawValue;
String value;

if (parts.length == 1) {
name = DEFAULT_PROPERTY;
value = ".*" + Pattern.quote(parts[0]) + ".*";
rawValue = parts[0];
value = ".*" + Pattern.quote(rawValue) + ".*";
} else {
name = parts[0];
value = parts[1];
}

if (value.isEmpty()) {
value = ".*";
rawValue = parts[1];
value = (rawValue.isEmpty() ? "" : Pattern.quote(rawValue)) + ".*";
}

final InputGraphProvider p = LookupHistory.getLast(InputGraphProvider.class);
Expand Down Expand Up @@ -109,25 +109,40 @@ public void evaluate(SearchRequest request, SearchResponse response) {
if (matches != null) {
final Set<InputNode> set = new HashSet<>(matches);
final InputGraph theGraph = p.getGraph() != matchGraph ? matchGraph : null;
response.addResult(new Runnable() {
@Override
public void run() {
final EditorTopComponent comp = EditorTopComponent.getActive();
if (comp != null) {
if (theGraph != null) {
comp.getDiagramModel().selectGraph(theGraph);
// Show "All N matching nodes" entry only if 1) there are
// multiple matches and 2) the query does not only contain
// digits (it is rare to select all nodes whose id contains a
// certain subsequence of digits).
if (matches.size() > 1 && !rawValue.matches("\\d+")) {
if (!response.addResult(new Runnable() {
@Override
public void run() {
final EditorTopComponent comp = EditorTopComponent.getActive();
if (comp != null) {
if (theGraph != null) {
comp.getDiagramModel().selectGraph(theGraph);
}
comp.setSelectedNodes(set);
comp.requestActive();
}
comp.setSelectedNodes(set);
comp.requestActive();
}
},
"All " + matches.size() + " matching nodes (" + name + "=" + value + ")" + (theGraph != null ? " in " + theGraph.getName() : "")
)) {
return;
}
},
"All " + matches.size() + " matching nodes (" + name + "=" + value + ")" + (theGraph != null ? " in " + theGraph.getName() : "")
);
}

// Rank the matches.
Collections.sort(matches,
(InputNode a, InputNode b) ->
compareByRankThenNumVal(rawValue,
a.getProperties().get(name),
b.getProperties().get(name)));

// Single matches
for (final InputNode n : matches) {
response.addResult(new Runnable() {
if (!response.addResult(new Runnable() {
@Override
public void run() {
final EditorTopComponent comp = EditorTopComponent.getActive();
Expand All @@ -143,7 +158,9 @@ public void run() {
}
},
n.getProperties().get(name) + " (" + n.getId() + " " + n.getProperties().get("name") + ")" + (theGraph != null ? " in " + theGraph.getName() : "")
);
)) {
return;
}
}
}
} else {
Expand Down Expand Up @@ -173,4 +190,51 @@ public void run() {
}
return null;
}

/**
* Compare two matches for a given query, first by rank (see rankMatch()
* below) and then by numeric value, if applicable.
*/
private int compareByRankThenNumVal(String qry, String prop1, String prop2) {
int key1 = rankMatch(qry, prop1);
int key2 = rankMatch(qry, prop2);
if (key1 == key2) {
// If the matches have the same rank, compare the numeric values of
// their first words, if applicable.
try {
key1 = Integer.parseInt(prop1.split("\\W+")[0]);
key2 = Integer.parseInt(prop2.split("\\W+")[0]);
} catch (Exception e) {
// Not applicable, return equality value.
return 0;
}
}
return Integer.compare(key1, key2);
}

/**
* Rank a match by splitting the property into words. Full matches of a word
* rank highest, followed by partial matches at the word start, followed by
* the rest of matches in increasing size of the partially matched word, for
* example:
*
* rank("5", "5 AddI") = 1 (full match of first word)
* rank("5", "554 MulI") = 2 (start match of first word)
* rank("5", "25 AddL") = 3 (middle match of first word with excess 1)
* rank("5", "253 AddL") = 4 (middle match of first word with excess 2)
*/
private int rankMatch(String qry, String prop) {
String query = qry.toLowerCase();
String property = prop.toLowerCase();
for (String component : property.split("\\W+")) {
if (component.equals(query)) {
return 1;
} else if (component.startsWith(query)) {
return 2;
} else if (component.contains(query)) {
return component.length() - query.length() + 2;
}
}
return Integer.MAX_VALUE;
}
}

1 comment on commit ae2c5f0

@openjdk-notifier
Copy link

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Please sign in to comment.