Apache/lucene
1
0
Fork 0
mirror of https://github.com/apache/lucene.git synced 2025-03-06 08:19:23 +00:00
Code Issues Packages Projects Releases Wiki Activity

Rename prefix to prefixLength, suffix to suffixLength. (#13629)

Browse Source
This commit is contained in:
zhouhui 2024-08-02 23:12:34 +08:00 committed by GitHub
parent 1ce1156de4
commit cb8bc75750
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
3 changed files with 49 additions and 46 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

28
lucene/core/src/java/org/apache/lucene/codecs/lucene90/blocktree/SegmentTermsEnum.java
View File

@ -277,10 +277,10 @@ final class SegmentTermsEnum extends BaseTermsEnum {
// System.out.println(" skip rewind!");
// }
}
assert length == f.prefix;
assert length == f.prefixLength;
} else {
f.nextEnt = -1;
f.prefix = length;
f.prefixLength = length;
f.state.termBlockOrd = 0;
f.fpOrig = f.fp = fp;
f.lastSubFP = -1;
@ -488,7 +488,7 @@ final class SegmentTermsEnum extends BaseTermsEnum {
// toHex(targetLabel));
// }
validIndexPrefix = currentFrame.prefix;
validIndexPrefix = currentFrame.prefixLength;
// validIndexPrefix = targetUpto;
currentFrame.scanToFloorFrame(target);
@ -550,7 +550,7 @@ final class SegmentTermsEnum extends BaseTermsEnum {
}
// validIndexPrefix = targetUpto;
validIndexPrefix = currentFrame.prefix;
validIndexPrefix = currentFrame.prefixLength;
currentFrame.scanToFloorFrame(target);
@ -772,7 +772,7 @@ final class SegmentTermsEnum extends BaseTermsEnum {
// targetLabel);
// }
validIndexPrefix = currentFrame.prefix;
validIndexPrefix = currentFrame.prefixLength;
// validIndexPrefix = targetUpto;
currentFrame.scanToFloorFrame(target);
@ -830,7 +830,7 @@ final class SegmentTermsEnum extends BaseTermsEnum {
}
// validIndexPrefix = targetUpto;
validIndexPrefix = currentFrame.prefix;
validIndexPrefix = currentFrame.prefixLength;
currentFrame.scanToFloorFrame(target);
@ -868,7 +868,7 @@ final class SegmentTermsEnum extends BaseTermsEnum {
while (true) {
SegmentTermsEnumFrame f = getFrame(ord);
assert f != null;
final BytesRef prefix = new BytesRef(term.get().bytes, 0, f.prefix);
final BytesRef prefix = new BytesRef(term.get().bytes, 0, f.prefixLength);
if (f.nextEnt == -1) {
out.println(
" frame "
@ -879,7 +879,7 @@ final class SegmentTermsEnum extends BaseTermsEnum {
+ f.fp
+ (f.isFloor ? (" (fpOrig=" + f.fpOrig + ")") : "")
+ " prefixLen="
+ f.prefix
+ f.prefixLength
+ " prefix="
+ prefix
+ (f.nextEnt == -1 ? "" : (" (of " + f.entCount + ")"))
@ -907,7 +907,7 @@ final class SegmentTermsEnum extends BaseTermsEnum {
+ f.fp
+ (f.isFloor ? (" (fpOrig=" + f.fpOrig + ")") : "")
+ " prefixLen="
+ f.prefix
+ f.prefixLength
+ " prefix="
+ prefix
+ " nextEnt="
@ -932,12 +932,14 @@ final class SegmentTermsEnum extends BaseTermsEnum {
}
if (fr.index != null) {
assert !isSeekFrame || f.arc != null : "isSeekFrame=" + isSeekFrame + " f.arc=" + f.arc;
if (f.prefix > 0 && isSeekFrame && f.arc.label() != (term.byteAt(f.prefix - 1) & 0xFF)) {
if (f.prefixLength > 0
&& isSeekFrame
&& f.arc.label() != (term.byteAt(f.prefixLength - 1) & 0xFF)) {
out.println(
" broken seek state: arc.label="
+ (char) f.arc.label()
+ " vs term byte="
+ (char) (term.byteAt(f.prefix - 1) & 0xFF));
+ (char) (term.byteAt(f.prefixLength - 1) & 0xFF));
throw new RuntimeException("seek state is broken");
}
BytesRef output = Util.get(fr.index, prefix);
@ -965,7 +967,7 @@ final class SegmentTermsEnum extends BaseTermsEnum {
if (f == currentFrame) {
break;
}
if (f.prefix == validIndexPrefix) {
if (f.prefixLength == validIndexPrefix) {
isSeekFrame = false;
}
ord++;
@ -1046,7 +1048,7 @@ final class SegmentTermsEnum extends BaseTermsEnum {
// Note that the seek state (last seek) has been
// invalidated beyond this depth
validIndexPrefix = Math.min(validIndexPrefix, currentFrame.prefix);
validIndexPrefix = Math.min(validIndexPrefix, currentFrame.prefixLength);
// if (DEBUG) {
// System.out.println(" reset validIndexPrefix=" + validIndexPrefix);
// }

61
lucene/core/src/java/org/apache/lucene/codecs/lucene90/blocktree/SegmentTermsEnumFrame.java
View File

@ -59,7 +59,7 @@ final class SegmentTermsEnumFrame {
final ByteArrayDataInput floorDataReader = new ByteArrayDataInput();
// Length of prefix shared by all terms in this block
int prefix;
int prefixLength;
// Number of entries (term or sub-block) in this block
int entCount;
@ -318,11 +318,11 @@ final class SegmentTermsEnumFrame {
assert nextEnt != -1 && nextEnt < entCount
: "nextEnt=" + nextEnt + " entCount=" + entCount + " fp=" + fp;
nextEnt++;
suffix = suffixLengthsReader.readVInt();
suffixLength = suffixLengthsReader.readVInt();
startBytePos = suffixesReader.getPosition();
ste.term.setLength(prefix + suffix);
ste.term.setLength(prefixLength + suffixLength);
ste.term.grow(ste.term.length());
suffixesReader.readBytes(ste.term.bytes(), prefix, suffix);
suffixesReader.readBytes(ste.term.bytes(), prefixLength, suffixLength);
ste.termExists = true;
}
@ -346,11 +346,11 @@ final class SegmentTermsEnumFrame {
: "nextEnt=" + nextEnt + " entCount=" + entCount + " fp=" + fp;
nextEnt++;
final int code = suffixLengthsReader.readVInt();
suffix = code >>> 1;
suffixLength = code >>> 1;
startBytePos = suffixesReader.getPosition();
ste.term.setLength(prefix + suffix);
ste.term.setLength(prefixLength + suffixLength);
ste.term.grow(ste.term.length());
suffixesReader.readBytes(ste.term.bytes(), prefix, suffix);
suffixesReader.readBytes(ste.term.bytes(), prefixLength, suffixLength);
if ((code & 1) == 0) {
// A normal term
ste.termExists = true;
@ -375,7 +375,7 @@ final class SegmentTermsEnumFrame {
// floor blocks we "typically" get
public void scanToFloorFrame(BytesRef target) {
if (!isFloor || target.length <= prefix) {
if (!isFloor || target.length <= prefixLength) {
// if (DEBUG) {
// System.out.println(" scanToFloorFrame skip: isFloor=" + isFloor + " target.length=" +
// target.length + " vs prefix=" + prefix);
@ -383,7 +383,7 @@ final class SegmentTermsEnumFrame {
return;
}
final int targetLabel = target.bytes[target.offset + prefix] & 0xFF;
final int targetLabel = target.bytes[target.offset + prefixLength] & 0xFF;
// if (DEBUG) {
// System.out.println(" scanToFloorFrame fpOrig=" + fpOrig + " targetLabel=" +
@ -497,7 +497,7 @@ final class SegmentTermsEnumFrame {
// Used only by assert
private boolean prefixMatches(BytesRef target) {
for (int bytePos = 0; bytePos < prefix; bytePos++) {
for (int bytePos = 0; bytePos < prefixLength; bytePos++) {
if (target.bytes[target.offset + bytePos] != ste.term.byteAt(bytePos)) {
return false;
}
@ -553,7 +553,7 @@ final class SegmentTermsEnumFrame {
}
private int startBytePos;
private int suffix;
private int suffixLength;
private long subCode;
CompressionAlgorithm compressionAlg = CompressionAlgorithm.NO_COMPRESSION;
@ -584,7 +584,7 @@ final class SegmentTermsEnumFrame {
do {
nextEnt++;
suffix = suffixLengthsReader.readVInt();
suffixLength = suffixLengthsReader.readVInt();
// if (DEBUG) {
// BytesRef suffixBytesRef = new BytesRef();
@ -596,16 +596,16 @@ final class SegmentTermsEnumFrame {
// }
startBytePos = suffixesReader.getPosition();
suffixesReader.skipBytes(suffix);
suffixesReader.skipBytes(suffixLength);
// Loop over bytes in the suffix, comparing to the target
final int cmp =
Arrays.compareUnsigned(
suffixBytes,
startBytePos,
startBytePos + suffix,
startBytePos + suffixLength,
target.bytes,
target.offset + prefix,
target.offset + prefixLength,
target.offset + target.length);
if (cmp < 0) {
@ -674,7 +674,7 @@ final class SegmentTermsEnumFrame {
assert prefixMatches(target);
suffix = suffixLengthsReader.readVInt();
suffixLength = suffixLengthsReader.readVInt();
// TODO early terminate when target length unequals suffix + prefix.
// But we need to keep the same status with scanToTermLeaf.
int start = nextEnt;
@ -684,16 +684,16 @@ final class SegmentTermsEnumFrame {
while (start <= end) {
int mid = (start + end) >>> 1;
nextEnt = mid + 1;
startBytePos = mid * suffix;
startBytePos = mid * suffixLength;
// Binary search bytes in the suffix, comparing to the target.
cmp =
Arrays.compareUnsigned(
suffixBytes,
startBytePos,
startBytePos + suffix,
startBytePos + suffixLength,
target.bytes,
target.offset + prefix,
target.offset + prefixLength,
target.offset + target.length);
if (cmp < 0) {
start = mid + 1;
@ -701,7 +701,7 @@ final class SegmentTermsEnumFrame {
end = mid - 1;
} else {
// Exact match!
suffixesReader.setPosition(startBytePos + suffix);
suffixesReader.setPosition(startBytePos + suffixLength);
fillTerm();
// if (DEBUG) System.out.println(" found!");
return SeekStatus.FOUND;
@ -724,14 +724,14 @@ final class SegmentTermsEnumFrame {
// If binary search ended at the less term, and greater term exists.
// We need to advance to the greater term.
if (cmp < 0) {
startBytePos += suffix;
startBytePos += suffixLength;
nextEnt++;
}
suffixesReader.setPosition(startBytePos + suffix);
suffixesReader.setPosition(startBytePos + suffixLength);
fillTerm();
} else {
seekStatus = SeekStatus.END;
suffixesReader.setPosition(startBytePos + suffix);
suffixesReader.setPosition(startBytePos + suffixLength);
if (exactOnly) {
fillTerm();
}
@ -769,7 +769,7 @@ final class SegmentTermsEnumFrame {
nextEnt++;
final int code = suffixLengthsReader.readVInt();
suffix = code >>> 1;
suffixLength = code >>> 1;
// if (DEBUG) {
// BytesRef suffixBytesRef = new BytesRef();
@ -782,7 +782,7 @@ final class SegmentTermsEnumFrame {
// }
startBytePos = suffixesReader.getPosition();
suffixesReader.skipBytes(suffix);
suffixesReader.skipBytes(suffixLength);
ste.termExists = (code & 1) == 0;
if (ste.termExists) {
state.termBlockOrd++;
@ -796,9 +796,9 @@ final class SegmentTermsEnumFrame {
Arrays.compareUnsigned(
suffixBytes,
startBytePos,
startBytePos + suffix,
startBytePos + suffixLength,
target.bytes,
target.offset + prefix,
target.offset + prefixLength,
target.offset + target.length);
if (cmp < 0) {
@ -819,7 +819,8 @@ final class SegmentTermsEnumFrame {
// us to position to the next term after
// the target, so we must recurse into the
// sub-frame(s):
ste.currentFrame = ste.pushFrame(null, ste.currentFrame.lastSubFP, prefix + suffix);
ste.currentFrame =
ste.pushFrame(null, ste.currentFrame.lastSubFP, prefixLength + suffixLength);
ste.currentFrame.loadBlock();
while (ste.currentFrame.next()) {
ste.currentFrame = ste.pushFrame(null, ste.currentFrame.lastSubFP, ste.term.length());
@ -864,9 +865,9 @@ final class SegmentTermsEnumFrame {
}
private void fillTerm() {
final int termLength = prefix + suffix;
final int termLength = prefixLength + suffixLength;
ste.term.setLength(termLength);
ste.term.grow(termLength);
System.arraycopy(suffixBytes, startBytePos, ste.term.bytes(), prefix, suffix);
System.arraycopy(suffixBytes, startBytePos, ste.term.bytes(), prefixLength, suffixLength);
}
}

6
lucene/core/src/java/org/apache/lucene/codecs/lucene90/blocktree/Stats.java
View File

@ -116,10 +116,10 @@ public class Stats {
nonFloorBlockCount++;
}
if (blockCountByPrefixLen.length <= frame.prefix) {
blockCountByPrefixLen = ArrayUtil.grow(blockCountByPrefixLen, 1 + frame.prefix);
if (blockCountByPrefixLen.length <= frame.prefixLength) {
blockCountByPrefixLen = ArrayUtil.grow(blockCountByPrefixLen, 1 + frame.prefixLength);
}
blockCountByPrefixLen[frame.prefix]++;
blockCountByPrefixLen[frame.prefixLength]++;
startBlockCount++;
totalBlockSuffixBytes += frame.totalSuffixBytes;
totalUncompressedBlockSuffixBytes += frame.suffixesReader.length();