private string[] InitTasksPackages(Config config)
{
// LUCENENET specific - changing the logic a bit
// to add all referenced assemblies by default.
// The alt.tasks.packages parameter still exists, but
// it is only necessary for assemblies that are not
// referenced by the host assembly.
ISet <string> result = new JCG.HashSet <string>();
string alts = config.Get("alt.tasks.packages", null);
string dfltPkg = typeof(PerfTask).Assembly.GetName().Name;
IEnumerable <string> referencedAssemblies = AssemblyUtils.GetReferencedAssemblies().Select(a => a.GetName().Name);
result.Add(dfltPkg);
if (alts == null)
{
result.UnionWith(referencedAssemblies);
return(result.ToArray());
}
foreach (string alt in alts.Split(',').TrimEnd())
{
result.Add(alt);
}
result.UnionWith(referencedAssemblies);
return(result.ToArray());
}
// TODO: Remove warning after API has been finalized
/// <summary>
/// WARNING: The List is not necessarily in order of the the positions </summary>
/// <returns> Collection of <see cref="T:byte[]"/> payloads </returns>
/// <exception cref="System.IO.IOException"> if there is a low-level I/O error </exception>
public override ICollection <byte[]> GetPayload()
{
var matchPayload = new JCG.HashSet <byte[]>();
for (var cell = first; cell != null; cell = cell.next)
{
if (cell.IsPayloadAvailable)
{
matchPayload.UnionWith(cell.GetPayload());
}
}
return(matchPayload);
}
public virtual void Inform(IResourceLoader loader)
{
IList <string> files = SplitFileNames(stopTypesFiles);
if (files.Count > 0)
{
stopTypes = new JCG.HashSet <string>();
foreach (string file in files)
{
IList <string> typesLines = GetLines(loader, file.Trim());
stopTypes.UnionWith(typesLines);
}
}
}
/// <summary>
/// Returns all files in use by this segment. </summary>
public virtual ICollection <string> GetFiles()
{
// Start from the wrapped info's files:
ISet <string> files = new JCG.HashSet <string>(Info.GetFiles());
// TODO we could rely on TrackingDir.getCreatedFiles() (like we do for
// updates) and then maybe even be able to remove LiveDocsFormat.files().
// Must separately add any live docs files:
Info.Codec.LiveDocsFormat.Files(this, files);
// Must separately add any field updates files
foreach (ISet <string> updateFiles in genUpdatesFiles.Values)
{
files.UnionWith(updateFiles);
}
return(files);
}
internal virtual void SealFlushedSegment(FlushedSegment flushedSegment)
{
if (Debugging.AssertsEnabled)
{
Debugging.Assert(flushedSegment != null);
}
SegmentCommitInfo newSegment = flushedSegment.segmentInfo;
IndexWriter.SetDiagnostics(newSegment.Info, IndexWriter.SOURCE_FLUSH);
IOContext context = new IOContext(new FlushInfo(newSegment.Info.DocCount, newSegment.GetSizeInBytes()));
bool success = false;
try
{
if (indexWriterConfig.UseCompoundFile)
{
filesToDelete.UnionWith(IndexWriter.CreateCompoundFile(infoStream, directory, CheckAbort.NONE, newSegment.Info, context));
newSegment.Info.UseCompoundFile = true;
}
// Have codec write SegmentInfo. Must do this after
// creating CFS so that 1) .si isn't slurped into CFS,
// and 2) .si reflects useCompoundFile=true change
// above:
codec.SegmentInfoFormat.SegmentInfoWriter.Write(directory, newSegment.Info, flushedSegment.fieldInfos, context);
// TODO: ideally we would freeze newSegment here!!
// because any changes after writing the .si will be
// lost...
// Must write deleted docs after the CFS so we don't
// slurp the del file into CFS:
if (flushedSegment.liveDocs != null)
{
int delCount = flushedSegment.delCount;
if (Debugging.AssertsEnabled)
{
Debugging.Assert(delCount > 0);
}
if (infoStream.IsEnabled("DWPT"))
{
infoStream.Message("DWPT", "flush: write " + delCount + " deletes gen=" + flushedSegment.segmentInfo.DelGen);
}
// TODO: we should prune the segment if it's 100%
// deleted... but merge will also catch it.
// TODO: in the NRT case it'd be better to hand
// this del vector over to the
// shortly-to-be-opened SegmentReader and let it
// carry the changes; there's no reason to use
// filesystem as intermediary here.
SegmentCommitInfo info = flushedSegment.segmentInfo;
Codec codec = info.Info.Codec;
codec.LiveDocsFormat.WriteLiveDocs(flushedSegment.liveDocs, directory, info, delCount, context);
newSegment.DelCount = delCount;
newSegment.AdvanceDelGen();
}
success = true;
}
finally
{
if (!success)
{
if (infoStream.IsEnabled("DWPT"))
{
infoStream.Message("DWPT", "hit exception creating compound file for newly flushed segment " + newSegment.Info.Name);
}
}
}
}
/// <summary>
/// The <see cref="SubSpans"/> are ordered in the same doc, so there is a possible match.
/// Compute the slop while making the match as short as possible by advancing
/// all <see cref="SubSpans"/> except the last one in reverse order.
/// </summary>
private bool ShrinkToAfterShortestMatch()
{
matchStart = subSpans[subSpans.Length - 1].Start;
matchEnd = subSpans[subSpans.Length - 1].End;
var possibleMatchPayloads = new JCG.HashSet <byte[]>();
if (subSpans[subSpans.Length - 1].IsPayloadAvailable)
{
possibleMatchPayloads.UnionWith(subSpans[subSpans.Length - 1].GetPayload());
}
IList <byte[]> possiblePayload = null;
int matchSlop = 0;
int lastStart = matchStart;
int lastEnd = matchEnd;
for (int i = subSpans.Length - 2; i >= 0; i--)
{
Spans prevSpans = subSpans[i];
if (collectPayloads && prevSpans.IsPayloadAvailable)
{
possiblePayload = new List <byte[]>(prevSpans.GetPayload()); // LUCENENET specific - using copy constructor instead of AddRange()
}
int prevStart = prevSpans.Start;
int prevEnd = prevSpans.End;
while (true) // Advance prevSpans until after (lastStart, lastEnd)
{
if (!prevSpans.Next())
{
inSameDoc = false;
more = false;
break; // Check remaining subSpans for final match.
}
else if (matchDoc != prevSpans.Doc)
{
inSameDoc = false; // The last subSpans is not advanced here.
break; // Check remaining subSpans for last match in this document.
}
else
{
int ppStart = prevSpans.Start;
int ppEnd = prevSpans.End; // Cannot avoid invoking .end()
if (!DocSpansOrdered(ppStart, ppEnd, lastStart, lastEnd))
{
break; // Check remaining subSpans.
} // prevSpans still before (lastStart, lastEnd)
else
{
prevStart = ppStart;
prevEnd = ppEnd;
if (collectPayloads && prevSpans.IsPayloadAvailable)
{
possiblePayload = new List <byte[]>(prevSpans.GetPayload()); // LUCENENET specific - using copy constructor instead of AddRange()
}
}
}
}
if (collectPayloads && possiblePayload != null)
{
possibleMatchPayloads.UnionWith(possiblePayload);
}
Debug.Assert(prevStart <= matchStart);
if (matchStart > prevEnd) // Only non overlapping spans add to slop.
{
matchSlop += (matchStart - prevEnd);
}
/* Do not break on (matchSlop > allowedSlop) here to make sure
* that subSpans[0] is advanced after the match, if any.
*/
matchStart = prevStart;
lastStart = prevStart;
lastEnd = prevEnd;
}
bool match = matchSlop <= allowedSlop;
if (collectPayloads && match && possibleMatchPayloads.Count > 0)
{
matchPayload.AddRange(possibleMatchPayloads);
}
return(match); // ordered and allowed slop
}
public virtual void TestIndexing()
{
DirectoryInfo tmpDir = CreateTempDir("TestNeverDelete");
BaseDirectoryWrapper d = NewFSDirectory(tmpDir);
// We want to "see" files removed if Lucene removed
// them. this is still worth running on Windows since
// some files the IR opens and closes.
if (d is MockDirectoryWrapper)
{
((MockDirectoryWrapper)d).NoDeleteOpenFile = false;
}
RandomIndexWriter w = new RandomIndexWriter(Random, d, NewIndexWriterConfig(TEST_VERSION_CURRENT, new MockAnalyzer(Random)).SetIndexDeletionPolicy(NoDeletionPolicy.INSTANCE));
w.IndexWriter.Config.SetMaxBufferedDocs(TestUtil.NextInt32(Random, 5, 30));
w.Commit();
ThreadJob[] indexThreads = new ThreadJob[Random.Next(4)];
long stopTime = (J2N.Time.NanoTime() / J2N.Time.MillisecondsPerNanosecond) + AtLeast(1000); // LUCENENET: Use NanoTime() rather than CurrentTimeMilliseconds() for more accurate/reliable results
for (int x = 0; x < indexThreads.Length; x++)
{
indexThreads[x] = new ThreadAnonymousClass(w, stopTime, NewStringField, NewTextField);
indexThreads[x].Name = "Thread " + x;
indexThreads[x].Start();
}
ISet <string> allFiles = new JCG.HashSet <string>();
DirectoryReader r = DirectoryReader.Open(d);
while (J2N.Time.NanoTime() / J2N.Time.MillisecondsPerNanosecond < stopTime) // LUCENENET: Use NanoTime() rather than CurrentTimeMilliseconds() for more accurate/reliable results
{
IndexCommit ic = r.IndexCommit;
if (Verbose)
{
Console.WriteLine("TEST: check files: " + ic.FileNames);
}
allFiles.UnionWith(ic.FileNames);
// Make sure no old files were removed
foreach (string fileName in allFiles)
{
Assert.IsTrue(SlowFileExists(d, fileName), "file " + fileName + " does not exist");
}
DirectoryReader r2 = DirectoryReader.OpenIfChanged(r);
if (r2 != null)
{
r.Dispose();
r = r2;
}
Thread.Sleep(1);
}
r.Dispose();
foreach (ThreadJob t in indexThreads)
{
t.Join();
}
w.Dispose();
d.Dispose();
System.IO.Directory.Delete(tmpDir.FullName, true);
}
//-----------------------------------------------------------------------------
//
// calcChainedFollowPos. Modify the previously calculated followPos sets
// to implement rule chaining. NOT described by Aho
//
//-----------------------------------------------------------------------------
internal virtual void CalcChainedFollowPos(RBBINode tree)
{
IList <RBBINode> endMarkerNodes = new JCG.List <RBBINode>();
IList <RBBINode> leafNodes = new JCG.List <RBBINode>();
// get a list of all endmarker nodes.
tree.FindNodes(endMarkerNodes, RBBINode.endMark);
// get a list all leaf nodes
tree.FindNodes(leafNodes, RBBINode.leafChar);
// Collect all leaf nodes that can start matches for rules
// with inbound chaining enabled, which is the union of the
// firstPosition sets from each of the rule root nodes.
IList <RBBINode> ruleRootNodes = new JCG.List <RBBINode>();
AddRuleRootNodes(ruleRootNodes, tree);
ISet <RBBINode> matchStartNodes = new JCG.HashSet <RBBINode>();
foreach (RBBINode node in ruleRootNodes)
{
if (node.fChainIn)
{
matchStartNodes.UnionWith(node.fFirstPosSet);
}
}
// Iterate over all leaf nodes,
//
foreach (RBBINode tNode in leafNodes)
{
RBBINode endNode = null;
// Identify leaf nodes that correspond to overall rule match positions.
// These include an endMarkerNode in their followPos sets.
foreach (RBBINode endMarkerNode in endMarkerNodes)
{
if (tNode.fFollowPos.Contains(endMarkerNode))
{
endNode = tNode;
break;
}
}
if (endNode == null)
{
// node wasn't an end node. Try again with the next.
continue;
}
// We've got a node that can end a match.
// Line Break Specific hack: If this node's val correspond to the $CM char class,
// don't chain from it.
// TODO: Add rule syntax for this behavior, get specifics out of here and
// into the rule file.
if (fRB.fLBCMNoChain)
{
int c = this.fRB.fSetBuilder.GetFirstChar(endNode.fVal);
if (c != -1)
{
// c == -1 occurs with sets containing only the {eof} marker string.
int cLBProp = UChar.GetIntPropertyValue(c, UProperty.Line_Break);
if (cLBProp == LineBreak.CombiningMark)
{
continue;
}
}
}
// Now iterate over the nodes that can start a match, looking for ones
// with the same char class as our ending node.
foreach (RBBINode startNode in matchStartNodes)
{
if (startNode.fType != RBBINode.leafChar)
{
continue;
}
if (endNode.fVal == startNode.fVal)
{
// The end val (character class) of one possible match is the
// same as the start of another.
// Add all nodes from the followPos of the start node to the
// followPos set of the end node, which will have the effect of
// letting matches transition from a match state at endNode
// to the second char of a match starting with startNode.
endNode.fFollowPos.UnionWith(startNode.fFollowPos);
}
}
}
}
/// <summary>
/// Encodes an input string into an output phonetic representation, given a set of possible origin languages.
/// </summary>
/// <param name="input">String to phoneticise; a string with dashes or spaces separating each word.</param>
/// <param name="languageSet"></param>
/// <returns>A phonetic representation of the input; a string containing '-'-separated phonetic representations of the input.</returns>
public virtual string Encode(string input, LanguageSet languageSet)
{
IDictionary <string, IList <Rule> > rules = Rule.GetInstanceMap(this.nameType, RuleType.RULES, languageSet);
// rules common across many (all) languages
IDictionary <string, IList <Rule> > finalRules1 = Rule.GetInstanceMap(this.nameType, this.ruleType, "common");
// rules that apply to a specific language that may be ambiguous or wrong if applied to other languages
IDictionary <string, IList <Rule> > finalRules2 = Rule.GetInstanceMap(this.nameType, this.ruleType, languageSet);
// tidy the input
// lower case is a locale-dependent operation
input = input.ToLowerInvariant().Replace('-', ' ').Trim();
if (this.nameType == NameType.GENERIC)
{
if (input.Length >= 2 && input.Substring(0, 2 - 0).Equals("d'", StringComparison.Ordinal))
{ // check for d'
string remainder = input.Substring(2);
string combined = "d" + remainder;
return("(" + Encode(remainder) + ")-(" + Encode(combined) + ")");
}
foreach (string l in NAME_PREFIXES[this.nameType])
{
// handle generic prefixes
if (input.StartsWith(l + " ", StringComparison.Ordinal))
{
// check for any prefix in the words list
string remainder = input.Substring(l.Length + 1); // input without the prefix
string combined = l + remainder; // input with prefix without space
return("(" + Encode(remainder) + ")-(" + Encode(combined) + ")");
}
}
}
IList <string> words = WHITESPACE.Split(input).TrimEnd();
ISet <string> words2 = new JCG.HashSet <string>();
// special-case handling of word prefixes based upon the name type
switch (this.nameType)
{
case NameType.SEPHARDIC:
foreach (string aWord in words)
{
string[] parts = aWord.Split('\'').TrimEnd();
string lastPart = parts[parts.Length - 1];
words2.Add(lastPart);
}
words2.ExceptWith(NAME_PREFIXES[this.nameType]);
break;
case NameType.ASHKENAZI:
words2.UnionWith(words);
words2.ExceptWith(NAME_PREFIXES[this.nameType]);
break;
case NameType.GENERIC:
words2.UnionWith(words);
break;
default:
throw new InvalidOperationException("Unreachable case: " + this.nameType);
}
if (this.concat)
{
// concat mode enabled
input = Join(words2, " ");
}
else if (words2.Count == 1)
{
// not a multi-word name
//input = words.iterator().next();
input = words[0];
}
else
{
// encode each word in a multi-word name separately (normally used for approx matches)
StringBuilder result = new StringBuilder();
foreach (string word in words2)
{
result.Append("-").Append(Encode(word));
}
// return the result without the leading "-"
return(result.ToString(1, result.Length - 1));
}
PhonemeBuilder phonemeBuilder = PhonemeBuilder.Empty(languageSet);
// loop over each char in the input - we will handle the increment manually
for (int i = 0; i < input.Length;)
{
RulesApplication rulesApplication =
new RulesApplication(rules, input, phonemeBuilder, i, maxPhonemes).Invoke();
i = rulesApplication.I;
phonemeBuilder = rulesApplication.PhonemeBuilder;
}
// Apply the general rules
phonemeBuilder = ApplyFinalRules(phonemeBuilder, finalRules1);
// Apply the language-specific rules
phonemeBuilder = ApplyFinalRules(phonemeBuilder, finalRules2);
return(phonemeBuilder.MakeString());
}
