Gabriel39 commented on code in PR #66227:
URL: https://github.com/apache/doris/pull/66227#discussion_r4128781008
##########
fe/fe-core/src/main/java/org/apache/doris/datasource/paimon/source/PaimonScanNode.java:
##########
@@ -411,10 +727,330 @@ private void setPaimonParams(TFileRangeDesc rangeDesc,
PaimonSplit paimonSplit)
String fileFormat = getFileFormat(paimonSplit.getPathString());
if (split != null) {
+ // use jni reader / paimon-cpp reader / paimon-rust reader
rangeDesc.setFormatType(TFileFormatType.FORMAT_JNI);
- // A logical DataSplit may span multiple files, so keep it intact
for the JNI reader.
- fileDesc.setReaderType(TPaimonReaderType.PAIMON_JNI);
- fileDesc.setPaimonSplit(PaimonUtil.encodeObjectToString(split));
+ // paimon-cpp and paimon-rust both consume Paimon native binary
serialization,
+ // which only supports DataSplit. Any other split type falls back
to JNI.
+ boolean nativeSplit = split instanceof DataSplit;
+ // Fallback-read splits stay on JNI: FallbackDataSplit extends
+ // DataSplit, so the instanceof above passes, but its serializer
+ // appends an isFallback byte after the ordinary split that the
+ // pinned rust decoder rejects outright ("trailing bytes after
+ // DataSplit" — it requires full-buffer consumption), and even a
+ // permissive decode would still lack the second table identity
+ // needed to honor the fallback-side discriminator. Both sides of a
+ // FallbackReadFileStoreTable wrap their splits, so the table
+ // wrapper is gated as a whole (any split from it routes to JNI)
+ // until the rust ABI represents both sides; the FallbackSplit
+ // interface also catches a wrapper split regardless of how the
+ // table was resolved here.
+ boolean fallbackRead = split instanceof
FallbackReadFileStoreTable.FallbackSplit
+ || processedTable instanceof FallbackReadFileStoreTable;
+ // Serialize the same effective table that planning and the JNI
reader use.
+ // Relation options such as t@options('read.batch-size'='1') are
applied by
+ // getProcessedTable() (doInitialize caches it in processedTable),
and the
+ // rust reader derives its read batch size from the schema options
— the raw
+ // cached table would silently drop the override. Copies,
delegates and
+ // fallback wrappers of getProcessedTable() are still
FileStoreTable, so the
+ // instanceof gate keeps its semantics.
+ Table paimonTable = processedTable;
+ FileStoreTable paimonFileStoreTable =
+ paimonTable instanceof FileStoreTable ? (FileStoreTable)
paimonTable : null;
+ // query-auth.enabled tables stay on JNI: when catalog
authorization
+ // succeeds with no row filter or column mask, Paimon still leaves
an
+ // ordinary DataSplit (restricted results use QueryAuthSplit and
are
+ // already handled by the nativeSplit gate above), so this table
shape
+ // passes the compound gate — but the shipped schema keeps
+ // query-auth.enabled=true and the pinned rust ReadBuilder rejects
+ // every such table (its CoreOptions::ensure_read_authorized fails
+ // closed because the client cannot enforce the row filter / column
+ // masking), turning a valid authorized scan into a BE-open
failure.
+ // Until the authorization result can be transported and enforced
by
+ // the rust ABI, these tables route to JNI.
+ boolean queryAuthTable = false;
+ // REST-token tables stay on JNI: doInitialize snapshots
+ // RESTTokenFileIO.validToken().token() into the backend storage
+ // properties, discarding expireAtMillis and the REST refresh
+ // context, so the shipped credentials look static — but the
+ // pinned rust table reuses one option map with no refresh
+ // callback, while paimon 1.4.2's JNI RESTTokenFileIO checks
+ // expiry before each file operation and obtains a replacement
+ // token. A queued or long scan that crosses the token TTL would
+ // start on rust and later fail authentication. Gate until the
+ // rust ABI can refresh and atomically update credentials.
+ boolean restTokenTable = false;
+ // Partial-update / aggregation tables with deletion vectors only
pass
+ // the rust reader in the fully materialized shape: the pinned rust
+ // read_pk rejects merge-engine=partial-update/aggregation with
+ // deletion-vectors.merge-on-read=true outright, and otherwise
requires
+ // every split to be compacted and known free of retract rows
+ // (DataSplit::is_fully_materialized_pk_dv). Their ordinary
DataSplits
+ // sail through the compound gate above, so without this check a
valid
+ // Java/JNI scan reaches BE and the rust open fails. Deduplicate
stays
+ // rust-eligible: its read_pk routes uncompacted splits to the KV
+ // reader, which applies the attached per-file DVs.
merge-on-read=true
+ // is a table option, so the whole table routes to JNI;
+ // non-materialized splits are gated per split below.
+ boolean puAggDeletionVectors = false;
+ boolean dvMergeOnRead = false;
+ boolean deduplicateIgnoreDelete = false;
+ boolean rustUnsupportedMergeOption = false;
+ if (paimonFileStoreTable != null) {
+ // A renewable REST token reached the shipped properties as a
+ // plain value; only the table's FileIO type reveals it
expires.
+ // Null-safe: a table handle whose FileIO is not resolved stays
+ // rust-eligible, mirroring the CoreOptions null-safety below.
+ restTokenTable = paimonFileStoreTable.fileIO() instanceof
RESTTokenFileIO;
+ CoreOptions resolvedCoreOptions =
paimonFileStoreTable.coreOptions();
+ // Null-safe: a table handle whose CoreOptions is not resolved
+ // (e.g. some wrapper shapes) stays rust-eligible rather than
+ // failing the scan here — the rust open itself rejects such a
+ // table if the option is really set.
+ if (resolvedCoreOptions != null) {
+ queryAuthTable = resolvedCoreOptions.queryAuthEnabled();
+ CoreOptions.MergeEngine mergeEngine =
resolvedCoreOptions.mergeEngine();
+ if (resolvedCoreOptions.deletionVectorsEnabled()
+ && (mergeEngine ==
CoreOptions.MergeEngine.PARTIAL_UPDATE
+ || mergeEngine ==
CoreOptions.MergeEngine.AGGREGATE)) {
+ puAggDeletionVectors = true;
+ // The merge-engine and deletion-vectors.enabled checks
+ // above resolve through the Java CoreOptions
accessors,
+ // which the table builds from this same schema options
+ // map — the one the BE rust reader deserializes from
+ // the shipped schema JSON — so they cannot diverge
from
+ // what BE sees. merge-on-read has no Java accessor in
+ // paimon 1.4, so it is read raw from the map, with the
+ // rust parsing semantics (any case-insensitive "true"
+ // is on, default false).
+ TableSchema dvSchema = paimonFileStoreTable.schema();
+ Map<String, String> dvOptions = dvSchema == null ?
null : dvSchema.options();
+ String mergeOnRead = dvOptions == null
+ ? null :
dvOptions.get(DELETION_VECTORS_MERGE_ON_READ);
+ dvMergeOnRead = "true".equalsIgnoreCase(mergeOnRead);
+ }
+ // deduplicate.ignore-delete=true tables stay on JNI:
+ // Java's DeduplicateMergeFunction skips retract records
+ // when the option is set — including old, uncompacted
+ // files that still contain them — but the pinned rust
+ // read_pk does not pass table options into its
+ // deduplicate merge: it picks the latest row and omits
+ // the key when that row is DELETE/UPDATE_BEFORE. An
+ // uncompacted insert followed by a delete therefore
+ // returns the insert through JNI but silently disappears
+ // through rust. Gate the option until the rust merge
+ // implements it.
+ if (mergeEngine == CoreOptions.MergeEngine.DEDUPLICATE
+ && resolvedCoreOptions.ignoreDelete()) {
+ deduplicateIgnoreDelete = true;
+ }
+ // Non-DV merge options the pinned rust read rejects: Java
+ // supports partial-update.remove-record-on-delete /
+ // aggregation.remove-record-on-delete and the wider
+ // per-field retract matrix, but the rust
+ // PartialUpdateConfig / AggregationConfig validations
+ // return Unsupported for them — and the DV-derived gates
+ // above only cover deletion-vector tables, so an ordinary
+ // non-DV DataSplit with one of these options would pass
the
+ // compound gate and fail during the rust merge
+ // construction. Mirror the exact rust key matrix
(presence,
+ // not values) against the same schema options map BE
+ // deserializes.
+ if (mergeEngine == CoreOptions.MergeEngine.PARTIAL_UPDATE
+ || mergeEngine ==
CoreOptions.MergeEngine.AGGREGATE) {
+ TableSchema mergeSchema =
paimonFileStoreTable.schema();
+ Map<String, String> mergeOptions =
+ mergeSchema == null ? null :
mergeSchema.options();
+ rustUnsupportedMergeOption = mergeOptions != null
+ && hasRustUnsupportedMergeOption(mergeOptions,
mergeEngine);
+ }
+ }
+ }
+ // paimon-rust additionally requires (a) FileScannerV2: the V1
FileScanner
+ // explicitly rejects PAIMON_RUST, so with enable_file_scanner_v2
disabled
+ // the split falls back to JNI instead of encoding a rust request
that the
+ // selected scanner cannot consume, and (b) a FileStoreTable: BE
opens the
+ // table via paimon_table_from_schema_json, which needs the
resolved
+ // TableSchema that only FileStoreTable exposes via schema(). If
the table
+ // is not a FileStoreTable (e.g. a sys table backed by DataSplit),
we cannot
+ // ship a schema JSON, so fall back to CPP / JNI rather than
sending an
+ // incomplete PAIMON_RUST request that BE would reject.
+ //
+ // The paimon-rust S3 bridge maps static credentials, anonymous
+ // access (AWS_CREDENTIALS_PROVIDER_TYPE=ANONYMOUS -> s3.anonymous)
+ // and assume-role (AWS_ROLE_ARN / AWS_EXTERNAL_ID ->
+ // s3.assumed.role.*), but the remaining credential-provider modes
+ // are ambient JVM provider chains (ENV, SYSTEM_PROPERTIES,
+ // WEB_IDENTITY, CONTAINER, INSTANCE_PROFILE) with no paimon-rust
+ // equivalent — rust would silently sign with whatever the ambient
+ // chain resolves to. Gate those modes away from the rust reader
+ // here so the configured provider is honored via the JNI path.
+ boolean providerModeTranslatable = true;
+ String providerType = backendStorageProperties == null
+ ? null :
backendStorageProperties.get("AWS_CREDENTIALS_PROVIDER_TYPE");
+ String mode = providerType == null ? "DEFAULT" :
providerType.trim().toUpperCase(Locale.ROOT);
+ if (mode.isEmpty()) {
+ mode = "DEFAULT";
+ }
+ String location = source.getTableLocation();
+ if (location != null && (location.startsWith("s3://") ||
location.startsWith("s3a://"))) {
+ // Java DEFAULT may resolve JVM properties or anonymous
credentials;
+ // Rust's ambient chain is different. Only explicit keys or
explicit
+ // anonymous access can cross this boundary without changing
identity.
+ String accessKey = backendStorageProperties == null
+ ? null :
backendStorageProperties.get("AWS_ACCESS_KEY");
+ String secretKey = backendStorageProperties == null
+ ? null :
backendStorageProperties.get("AWS_SECRET_KEY");
+ boolean staticKeys = accessKey != null &&
!accessKey.trim().isEmpty()
+ && secretKey != null && !secretKey.trim().isEmpty();
+ providerModeTranslatable = mode.equals("ANONYMOUS") ||
(mode.equals("DEFAULT") && staticKeys);
+ } else if (providerType != null) {
+ providerModeTranslatable = mode.equals("DEFAULT") ||
mode.equals("ANONYMOUS");
+ // OSS has no anonymous FileIO mode in the pinned Rust
dependency.
+ if (mode.equals("ANONYMOUS") && location != null &&
location.startsWith("oss://")) {
+ providerModeTranslatable = false;
+ }
+ }
+ // Incremental scans (binlog / changelog / delta / diff) must stay
+ // on the JNI path: this wire format carries only an ordinary
+ // DataSplit and the rust reader invokes TableRead::to_arrow, but
+ // paimon 1.4 marks incremental splits as streaming (which the
+ // pinned rust deserializer rejects), diff requires a separate
+ // IncrementalPlan instead of an ordinary plan, and ordinary
+ // primary-key reads can merge versions rather than return the
+ // changes — until the C ABI transports the mode and plan, the
+ // rust reader cannot express any of these.
+ TableScanParams incrementalParams = getScanParams();
+ boolean isIncremental = incrementalParams != null &&
incrementalParams.incrementalRead();
+ // ORC TIMESTAMP_WITH_LOCAL_TIME_ZONE schemas stay on JNI: the
pinned
+ // paimon-rust ORC decoder materializes LTZ instants shifted by the
+ // writer timezone (an upstream crate limitation), so a logical ORC
+ // DataSplit that selects rust (e.g. with force_jni_scanner=true or
+ // when raw conversion is unavailable) returns a different instant
+ // than JNI — applying the session timezone in BE cannot repair an
+ // epoch already shifted during decode. Two bypasses are covered:
+ // (a) the format must come from EVERY member file — paimon allows
+ // per-level file.format, so one DataSplit can mix Parquet and ORC
+ // files and the split path's suffix (the first file) would hide
+ // the ORC members; (b) the LTZ search must recurse into nested
+ // types — an LTZ under MAP/ARRAY/ROW reaches the same shifted ORC
+ // decode through the container's field materialization. Parquet
+ // files with any LTZ, and ORC without any recursive LTZ, stay
+ // rust-eligible. nativeSplit only guards the cast — non-DataSplit
+ // splits already route to JNI.
+ boolean orcLtzSchema = paimonFileStoreTable != null
+ && nativeSplit
+ && splitHasOrcFile((DataSplit) split)
+ && paimonFileStoreTable.schema().fields().stream()
+ .anyMatch(field ->
containsTimestampLtz(field.type()));
+ // data-file.external-paths splits stay on JNI (see
+ // splitHasExternalFiles): the rust table's single FileIO cannot
+ // serve an external file's backend. nativeSplit only guards the
+ // cast — non-DataSplit splits already route to JNI.
+ boolean externalFileSplit = nativeSplit &&
splitHasExternalFiles((DataSplit) split);
+ // Projected VARIANT columns stay on JNI: the rust leaf feeds its
+ // Arrow arrays to the slot serdes, and DataTypeVariantV2SerDe::
+ // read_column_from_arrow unconditionally returns
+ // NOT_IMPLEMENTED_ERROR — a nested Variant (ARRAY / MAP / STRUCT
+ // containing one) reaches the same decoder through the container
+ // serdes. desc carries only the slots this query projects, so a
+ // table whose VARIANT column is not projected still scans on
+ // rust. Gate until the rust leaf has a Variant Arrow decoder.
+ boolean projectedVariant = desc.getSlots().stream()
+ .anyMatch(slot ->
PaimonUtil.containsVariant(slot.getType()));
+ // Scheme capability gate: the pinned paimon-rust storage
+ // dispatcher (io/storage.rs) selects the FileIO parser from the
+ // table location's URI scheme, and libpaimon_c.a compiles in
+ // separate COS, OBS, GCS and Azdls parsers besides the OSS and S3
+ // ones. Doris normalizes every object store's credentials into
+ // the AWS_* / use_path_style aliases (see the *Properties storage
+ // classes), which the BE rust bridge translates only into the
+ // fs.oss.* and s3.* key families — a cosn:// / obs:// / gs:// /
+ // abfs:// warehouse would reach its scheme's parser without the
+ // key family it reads (fs.cosn.userinfo.*, fs.obs.*, gcs.*,
+ // azure.*) and fail the open instead of using JNI. Only the
+ // schemes whose property translation is implemented and
+ // open-tested (s3 / s3a / oss, via RUST_VERIFIED_LOCATION_SCHEMES)
+ // plus the credential-free hdfs and local-filesystem parsers stay
+ // rust-eligible; every other scheme falls back to JNI. A null
+ // location also routes to JNI: the rust path needs the
+ // paimon_table that only a real location can provide (BE rejects
+ // a split without it).
+ boolean schemeCapabilityVerified =
isRustVerifiedLocationScheme(source.getTableLocation());
+ // An hdfs:// location is scheme-verified only together with the
credential-free
+ // backend shape: the backend storage properties that ship to BE
also carry an
+ // HDFS catalog's authentication (kerberos principal / keytab,
proxy user, HA
+ // nameservice config), none of which the pinned rust HDFS parser
reads — the
+ // scan would open as the BE process's ambient identity instead of
the
+ // catalog's configured one and fail the access JNI honors. See
+ // isRustVerifiedHdfsBackend.
+ boolean hdfsBackendVerified =
!isHdfsLocationScheme(source.getTableLocation())
+ || isRustVerifiedHdfsBackend(backendStorageProperties);
+ // With merge-on-read=true the whole table already routes to JNI
(dvMergeOnRead);
+ // for the remaining partial-update/aggregation DV tables, a split
that is
+ // not fully materialized (uncompacted level-0 data, or
retractions not
+ // known to be applied — even a split with no deletion file
attached yet)
+ // fails the rust is_fully_materialized_pk_dv guard, so it falls
back per
+ // split instead of turning into a BE-open failure. nativeSplit and
+ // !fallbackRead only guard the cast — those splits already route
to JNI.
+ boolean splitDvNotMaterialized = puAggDeletionVectors &&
!dvMergeOnRead
+ && nativeSplit && !fallbackRead
+ && !isFullyMaterializedPkDvSplit((DataSplit) split);
+ boolean canUseRust = sessionVariable.isEnablePaimonRustReader()
+ && sessionVariable.enableFileScannerV2 && nativeSplit &&
!fallbackRead
+ && !isIncremental && providerModeTranslatable &&
!queryAuthTable
+ && !restTokenTable
+ && !dvMergeOnRead && !splitDvNotMaterialized
+ && !orcLtzSchema && !projectedVariant && !externalFileSplit
+ && !deduplicateIgnoreDelete && !rustUnsupportedMergeOption
+ && schemeCapabilityVerified && hdfsBackendVerified
+ && paimonFileStoreTable != null;
+ if (canUseRust) {
+ if (rustReaderCapabilities == null) {
+ rustReaderCapabilities = new
PaimonRustReaderCapabilities(paimonFileStoreTable, desc);
+ }
+ canUseRust = rustReaderCapabilities.canRead((DataSplit) split);
+ }
+ if (canUseRust) {
+ fileDesc.setReaderType(TPaimonReaderType.PAIMON_RUST);
Review Comment:
Fixed with an optional BE heartbeat capability, defaulting to false when
absent. FE emits PAIMON_RUST only when every eligible backend advertises
support; otherwise it uses the JNI enum and Java split encoding. Tests cover
empty/old/new/mixed backend pools, heartbeat replay/image persistence, and
clearing support after downgrade. The changed BE heartbeat translation unit
compiles.
##########
fe/fe-core/src/main/java/org/apache/doris/datasource/paimon/source/PaimonRustReaderCapabilities.java:
##########
@@ -0,0 +1,257 @@
+// Licensed to the Apache Software Foundation (ASF) under one
+// or more contributor license agreements. See the NOTICE file
+// distributed with this work for additional information
+// regarding copyright ownership. The ASF licenses this file
+// to you under the Apache License, Version 2.0 (the
+// "License"); you may not use this file except in compliance
+// with the License. You may obtain a copy of the License at
+//
+// http://www.apache.org/licenses/LICENSE-2.0
+//
+// Unless required by applicable law or agreed to in writing,
+// software distributed under the License is distributed on an
+// "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
+// KIND, either express or implied. See the License for the
+// specific language governing permissions and limitations
+// under the License.
+
+package org.apache.doris.datasource.paimon.source;
+
+import org.apache.doris.analysis.SlotDescriptor;
+import org.apache.doris.analysis.TupleDescriptor;
+
+import com.google.common.collect.ImmutableSet;
+import org.apache.paimon.CoreOptions;
+import org.apache.paimon.io.DataFileMeta;
+import org.apache.paimon.schema.TableSchema;
+import org.apache.paimon.table.FileStoreTable;
+import org.apache.paimon.table.source.DataSplit;
+import org.apache.paimon.types.ArrayType;
+import org.apache.paimon.types.DataField;
+import org.apache.paimon.types.DataType;
+import org.apache.paimon.types.DataTypeRoot;
+import org.apache.paimon.types.DecimalType;
+import org.apache.paimon.types.MapType;
+import org.apache.paimon.types.RowType;
+import org.apache.paimon.types.VarCharType;
+
+import java.util.Arrays;
+import java.util.HashMap;
+import java.util.HashSet;
+import java.util.List;
+import java.util.Map;
+import java.util.Set;
+import java.util.concurrent.ConcurrentHashMap;
+
+/** Compatibility checks for the pinned paimon-rust reader, beyond storage
capabilities. */
+final class PaimonRustReaderCapabilities {
+ private static final Set<String> SUPPORTED_AGGREGATE_NAMES =
ImmutableSet.of(
+ "sum", "product", "min", "max", "last_value", "first_value",
"last_non_null_value",
+ "first_non_null_value", "first_not_null_value", "bool_and",
"bool_or", "listagg");
+ private final FileStoreTable table;
+ private final TableSchema schema;
+ private final boolean tableCompatible;
+ private final Map<Long, Boolean> compatibleFileSchemas = new
ConcurrentHashMap<>();
+
+ PaimonRustReaderCapabilities(FileStoreTable table, TupleDescriptor tuple) {
+ this.table = table;
+ this.schema = table.schema();
+ this.tableCompatible = schema != null && hasFullNestedProjection(tuple)
+ && hasCompatibleAggregates(schema);
+ }
+
+ boolean canRead(DataSplit split) {
+ if (!tableCompatible) {
+ return false;
+ }
+ // The pinned merge reader retains losing input batches until an
output batch fills.
+ // Neither zero deletes nor a small read.batch-size bounds this across
multiple files.
+ if (!schema.primaryKeys().isEmpty() && split.dataFiles().size() > 1) {
+ return false;
+ }
+ for (DataFileMeta file : split.dataFiles()) {
+ // Sort-merge retains consumed batches until it emits enough rows.
Retracts can
+ // produce an unbounded zero-output prefix; aggregation also
rejects retracts.
+ // Unknown counts must stay on JNI, including old files without
this statistic.
+ if (!schema.primaryKeys().isEmpty() &&
file.deleteRowCount().orElse(-1L) != 0L) {
+ return false;
+ }
+ if (file.schemaId() != schema.id() &&
!compatibleFileSchemas.computeIfAbsent(
+ file.schemaId(), this::hasCompatibleFileSchema)) {
+ return false;
+ }
+ }
+ return true;
+ }
+
+ private static boolean hasFullNestedProjection(TupleDescriptor tuple) {
+ for (SlotDescriptor slot : tuple.getSlots()) {
+ // The ABI projects only root names, while Arrow struct SerDes
bind by ordinal.
+ // A pruned slot must use JNI's recursive read type, even inside
arrays or maps.
+ if (slot.getType().isComplexType() && slot.getColumn() != null
+ && !slot.getType().equals(slot.getColumn().getType())) {
+ return false;
+ }
+ }
+ return true;
+ }
+
+ private boolean hasCompatibleFileSchema(long id) {
+ try {
+ TableSchema fileSchema = table.schemaManager().schema(id);
+ // Match historical types by ID: renames and added fields do not
require value casts.
+ return fileSchema != null &&
!hasIncompatibleEvolution(fileSchema.fields(), schema.fields());
+ } catch (RuntimeException e) {
+ // Failure to establish compatibility must not opt a historical
file into Rust.
+ return false;
+ }
+ }
+
+ private static boolean hasIncompatibleEvolution(List<DataField> oldFields,
List<DataField> newFields) {
+ Map<Integer, DataType> oldTypes = new HashMap<>();
+ for (DataField field : oldFields) {
+ oldTypes.put(field.id(), field.type());
+ }
+ for (DataField field : newFields) {
+ DataType oldType = oldTypes.get(field.id());
+ if (oldType != null && hasIncompatibleEvolution(oldType,
field.type())) {
+ return true;
+ }
+ }
+ return false;
+ }
+
+ private static boolean hasIncompatibleEvolution(DataType oldType, DataType
newType) {
+ // Java rounds the decimal string of a floating value; Arrow scales
the binary value.
+ // Even an in-range value such as DOUBLE 1.005 can therefore round
differently.
+ if (newType instanceof DecimalType && (oldType.getTypeRoot() ==
DataTypeRoot.FLOAT
+ || oldType.getTypeRoot() == DataTypeRoot.DOUBLE)) {
+ return true;
+ }
+ int oldWidth = integerWidth(oldType.getTypeRoot());
+ DataTypeRoot oldRoot = oldType.getTypeRoot();
+ DataTypeRoot newRoot = newType.getTypeRoot();
+ boolean targetTimestamp = newRoot ==
DataTypeRoot.TIMESTAMP_WITHOUT_TIME_ZONE
+ || newRoot == DataTypeRoot.TIMESTAMP_WITH_LOCAL_TIME_ZONE;
+ // Java accepts epoch-day/millisecond strings and interprets numeric
timestamps as seconds.
+ // Arrow instead parses calendar strings or reuses numeric timestamp
ticks, including in
+ // nested values. Keep these historical casts on JNI until the pinned
reader matches Java.
+ boolean sourceString = oldRoot == DataTypeRoot.CHAR || oldRoot ==
DataTypeRoot.VARCHAR;
+ if ((sourceString && (newRoot == DataTypeRoot.DATE || targetTimestamp))
+ || (oldWidth > 0 && targetTimestamp)) {
+ return true;
+ }
+ int newWidth = integerWidth(newRoot);
+ if (newWidth > 0) {
+ // Floating-point and decimal sources also differ under narrowing;
only integer
+ // identity/widening casts have the same range and value semantics
in both readers.
+ return oldWidth == 0 || oldWidth > newWidth;
+ }
+ if (oldType instanceof RowType && newType instanceof RowType) {
Review Comment:
Fixed by failing closed on unproven constructed-type casts, including ROW to
STRING. Added a persisted ROW fixture verifying Java output and recursive
routing coverage for ROW/ARRAY/MAP conversions to text.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
