Skip to content

GH-3263: Add DictionaryPage.decode to allow dictionary reuse in the ColumnReaderBase ctor #3264

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
wgtmac merged 7 commits into from
Aug 15, 2025
Merged

GH-3263: Add DictionaryPage.decode to allow dictionary reuse in the ColumnReaderBase ctor #3264

Changes from 1 commit
Commits
Show all changes
7 commits
Select commit Hold shift + click to select a range
7ac4953
Add c-tor to ColumnReaderBase to allow dictionary reuse
pyckle Aug 6, 2025
73f68c7
add getDictionary function to PageReader
pyckle Aug 7, 2025
a9f07e1
run spotless
pyckle Aug 7, 2025
a220df6
add converter logic mistakenly removed
pyckle Aug 7, 2025
e6930f9
move getDictionary to DictionaryPage
pyckle Aug 10, 2025
a69918a
improve naming
Aug 11, 2025
69b45e2
test that getting a dict from decode rather than decoding bytes works
Aug 14, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
35 changes: 22 additions & 13 deletions parquet-column/src/main/java/org/apache/parquet/column/impl/ColumnReaderBase.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -445,30 +445,39 @@ void writeValue() {
* @param converter a converter that materializes the values in this column in the current record
* @param writerVersion writer version string from the Parquet file being read
*/
ColumnReaderBase(
protected ColumnReaderBase(
ColumnDescriptor path, PageReader pageReader, PrimitiveConverter converter, ParsedVersion writerVersion) {
this(path, readDict(path, pageReader), pageReader, converter, writerVersion);
}

protected ColumnReaderBase(
ColumnDescriptor path, Dictionary dictionary, PageReader dataPageReader, PrimitiveConverter converter, ParsedVersion writerVersion) {
this.path = Objects.requireNonNull(path, "path cannot be null");
this.pageReader = Objects.requireNonNull(pageReader, "pageReader cannot be null");
this.pageReader = Objects.requireNonNull(dataPageReader, "pageReader cannot be null");
this.converter = Objects.requireNonNull(converter, "converter cannot be null");
this.writerVersion = writerVersion;
this.maxDefinitionLevel = path.getMaxDefinitionLevel();
DictionaryPage dictionaryPage = pageReader.readDictionaryPage();
this.dictionary = dictionary;
if (dictionary != null && converter.hasDictionarySupport()) {
converter.setDictionary(dictionary);
}
this.totalValueCount = pageReader.getTotalValueCount();
if (totalValueCount <= 0) {
throw new ParquetDecodingException("totalValueCount '" + totalValueCount + "' <= 0");
}
}

private static Dictionary readDict(ColumnDescriptor path, PageReader pageReader) {
Dictionary dictionary = null;
DictionaryPage dictionaryPage = Objects.requireNonNull(pageReader,"Page reader cannot be null").readDictionaryPage();
if (dictionaryPage != null) {
try {
this.dictionary = dictionaryPage.getEncoding().initDictionary(path, dictionaryPage);
if (converter.hasDictionarySupport()) {
converter.setDictionary(dictionary);
}
dictionary = dictionaryPage.getEncoding().initDictionary(path, dictionaryPage);
} catch (IOException e) {
throw new ParquetDecodingException("could not decode the dictionary for " + path, e);
}
} else {
this.dictionary = null;
}
this.totalValueCount = pageReader.getTotalValueCount();
if (totalValueCount <= 0) {
throw new ParquetDecodingException("totalValueCount '" + totalValueCount + "' <= 0");
}
return dictionary;
}

boolean isFullyConsumed() {
Expand Down