This is an automated email from the ASF dual-hosted git repository.
hello-stephen pushed a commit to branch master
in repository https://gitbox.apache.org/repos/asf/doris-website.git
The following commit(s) were added to refs/heads/master by this push:
new f812228db9f add weekly report 2026-08-17 (#4088)
f812228db9f is described below
commit f812228db9f92a7c86faa67f6331858ef8bf6582
Author: Mingyu Chen (Rayner) <[email protected]>
AuthorDate: Wed Aug 26 14:09:45 2026 +0800
add weekly report 2026-08-17 (#4088)
Add community weekly report for Aug 17 – Aug 23, 2026 (Week 34).
🤖 Generated with [Claude Code](https://claude.com/claude-code)
https://claude.ai/code/session_01UhYif4nBH1XEhrJJaG2Xtd
Co-authored-by: morningman <[email protected]>
Co-authored-by: Claude Fable 5 <[email protected]>
---
src/pages/community-report/_reports/2026-08-17.mdx | 385 +++++++++++++++++++++
1 file changed, 385 insertions(+)
diff --git a/src/pages/community-report/_reports/2026-08-17.mdx
b/src/pages/community-report/_reports/2026-08-17.mdx
new file mode 100644
index 00000000000..f7f8e1c2c78
--- /dev/null
+++ b/src/pages/community-report/_reports/2026-08-17.mdx
@@ -0,0 +1,385 @@
+{/* Auto-generated by community-radar `report export-mdx` — do not edit by
hand.
+ `report` is English-only structured data laid out by doris-website. */}
+
+export const label = "Aug 17 – Aug 23, 2026";
+export const week = "Week 34, 2026";
+export const stats = [
+ {
+ "num": "97",
+ "label": "Merged PRs"
+ },
+ {
+ "num": "15",
+ "label": "New issues"
+ },
+ {
+ "num": "49",
+ "label": "Contributors"
+ }
+];
+
+export const report = {
+ "summary": {
+ "lead": "Doris landed 97 PRs this week with 49 contributors driving 170+
changes across seven focus areas, with kernel engineering and the real-time
data warehouse still leading the mix while compute-storage separation, the
multi-modal lakehouse, and agent observability continue to accelerate in
parallel. The capability gravity is clearly shifting toward an
object-storage-native cloud-native engine with unified lakehouse read/write
paths, with Variant V2, enterprise security/governan [...]
+ "highlights": [
+ {
+ "title": "An object-storage-native engine takes shape",
+ "narrative": "Async cache writes, page prefetch, and I/O coalescing
keep cold scans off the query thread; the SNII inverted index reads object
storage directly without local rebuild; BE and Recycler now share one object
client with consistent credentials, errors, and metrics; a backend-selection
extension framework lands. Cold-start latency, operational consistency, and",
+ "prs": [
+ {
+ "num": 65658,
+ "title": "[feature](file cache) Add asynchronous file cache
writes",
+ "url": "https://github.com/apache/doris/pull/65658";
+ },
+ {
+ "num": 66520,
+ "title": "[feature](file cache) Add page prefetch and I/O
coalescing for cold queries",
+ "url": "https://github.com/apache/doris/pull/66520";
+ },
+ {
+ "num": 66052,
+ "title": "[feature](inverted index) Add SNII inverted index
storage format",
+ "url": "https://github.com/apache/doris/pull/66052";
+ },
+ {
+ "num": 66350,
+ "title": "[refactor](storage) Unify BE and Recycler object
clients",
+ "url": "https://github.com/apache/doris/pull/66350";
+ },
+ {
+ "num": 65173,
+ "title": "[feature](coordinator) Add backend selection extension
framework",
+ "url": "https://github.com/apache/doris/pull/65173";
+ }
+ ]
+ },
+ {
+ "title": "Lakehouse write and read paths reach parity",
+ "narrative": "Paimon 2.0 transactional writes close the external-table
write gap; Iceberg HDFS handles open lazily to cut NameNode RPCs and propagate
delete-file stats for scan pruning; Lance vector search moves to multi-BE
distributed execution while preserving global Top-K. Read and write paths
across Paimon, Iceberg, and Lance are becoming first-class.",
+ "prs": [
+ {
+ "num": 66790,
+ "title": "[feature](paimon) Add Paimon 2.0 transactional writes",
+ "url": "https://github.com/apache/doris/pull/66790";
+ },
+ {
+ "num": 66773,
+ "title": "[feat](iceberg) HDFS lazy open + iceberg delete file
file_size propagation",
+ "url": "https://github.com/apache/doris/pull/66773";
+ },
+ {
+ "num": 66848,
+ "title": "[Feature](lance) Support distributed vector search",
+ "url": "https://github.com/apache/doris/issues/66848";
+ }
+ ]
+ },
+ {
+ "title": "Variant V2 becomes the default shredded path",
+ "narrative": "Variant V2 ingestion removes redundant metadata decoding
and type inference across children; native shredded execution preserves typed
sub-columns on the BE without forced collapse; a default-on switch routes
planning and storage to V2 automatically. Out-of-the-box performance and type
fidelity for JSON and semi-structured workloads advance together.",
+ "prs": [
+ {
+ "num": 66709,
+ "title": "[improvement](variant) Optimize Variant V2 ingestion and
STRING casts",
+ "url": "https://github.com/apache/doris/pull/66709";
+ },
+ {
+ "num": 66705,
+ "title": "[feature](be) Add native shredded Variant V2 execution",
+ "url": "https://github.com/apache/doris/pull/66705";
+ },
+ {
+ "num": 66858,
+ "title": "[feature](variant) Enable Variant V2 by default",
+ "url": "https://github.com/apache/doris/pull/66858";
+ }
+ ]
+ },
+ {
+ "title": "Enterprise security and governance gain extension points",
+ "narrative": "TDE KMS identifiers now cover Aliyun, Ranger, GCP, and
Azure with portable endpoint and auth config; an authorization plugin SPI moves
Ranger out of fe-core, making the policy engine replaceable without touching
kernel. Plug-in surfaces for transparent encryption and row-level access
control are opened up.",
+ "prs": [
+ {
+ "num": 64561,
+ "title": "[feat](tde) Add KMS provider metadata",
+ "url": "https://github.com/apache/doris/pull/64561";
+ },
+ {
+ "num": 66770,
+ "title": "[feat](authorization) introduce an authorization plugin
SPI and move the Ranger sources out of fe-core",
+ "url": "https://github.com/apache/doris/pull/66770";
+ }
+ ]
+ },
+ {
+ "title": "Open-source SPIs open up downstream extensibility",
+ "narrative": "A cloud snapshot handler SPI with Env override lets
enterprise implementations load via ServiceLoader; the snapshot manager factory
becomes a source-overridable contract; an all-in-one 4.1 Docker image offers a
single entry point for integration testing. The API gap between downstream
customizations and upstream kernel is being closed.",
+ "prs": [
+ {
+ "num": 66916,
+ "title": "[refactor](snapshot) Add cloud snapshot handler SPI and
Env override",
+ "url": "https://github.com/apache/doris/pull/66916";
+ },
+ {
+ "num": 66923,
+ "title": "[refactor](snapshot) Allow overriding snapshot manager
factory source",
+ "url": "https://github.com/apache/doris/pull/66923";
+ },
+ {
+ "num": 66983,
+ "title": "[feat](docker) add an all-in-one image for 4.1
integration testing",
+ "url": "https://github.com/apache/doris/pull/66983";
+ }
+ ]
+ }
+ ],
+ "numbers": {
+ "mergedPrs": 97,
+ "newIssues": 15,
+ "contributors": 49
+ }
+ },
+ "repos": [
+ {
+ "repo": "apache/doris",
+ "scenarios": [
+ {
+ "name": "Multi-modal Lakehouse",
+ "mergedNarrative": "Multi-modal lakehouse work this week polishes
external source correctness and execution. External row-count caches invalidate
with metadata, Parquet and Iceberg Variant fixes forward-port to master, JDBC
catalogs reject PostgreSQL precision changes, Iceberg COUNT(*) derives from
live manifests, Paimon pushes LIMIT into split planning, and lance third-party
patches land—removing correctness and",
+ "merged": [
+ {
+ "num": 64160,
+ "title": "[fix](meta cache) Invalidate external row count cache
after metadata changes",
+ "url": "https://github.com/apache/doris/pull/64160";
+ },
+ {
+ "num": 66413,
+ "title": "[fix](variant) Forward-port Parquet and external
Variant fixes to master",
+ "url": "https://github.com/apache/doris/pull/66413";
+ },
+ {
+ "num": 66677,
+ "title": "[fix](jdbc) reject a PostgreSQL datetime value the
cached precision cannot hold",
+ "url": "https://github.com/apache/doris/pull/66677";
+ },
+ {
+ "num": 66750,
+ "title": "[fix](fe) Preserve Paimon error message
capitalization",
+ "url": "https://github.com/apache/doris/pull/66750";
+ },
+ {
+ "num": 66778,
+ "title": "[fix](iceberg) Derive count pushdown from live
manifests",
+ "url": "https://github.com/apache/doris/pull/66778";
+ }
+ ],
+ "inProgressNarrative": "Multi-modal lakehouse work strengthens
external catalogs alongside Parquet and Iceberg reads. Newly opened pieces
backfill Parquet columns absent from a file's physical schema after schema
evolution, while in-flight fixes curb redundant remote table enumeration on
bypass-mode Iceberg REST, retain Azure vended credentials, convert Iceberg
mapping mismatches from BE crashes into",
+ "inProgress": []
+ },
+ {
+ "name": "Real-time Data Warehouse",
+ "mergedNarrative": "Real-time DWH work this week sharpens both
planner and write path. MTMV partition lineage analysis sheds FE CPU, write
backpressure reflects the real pipeline more accurately, BucketHashAggregate
fuses shuffle and aggregation at translation, multiple correctness regressions
in TopN, scan replica selection, and fragment lifecycle are fixed, and
FD-redundant group-by keys",
+ "merged": [
+ {
+ "num": 63899,
+ "title": "[improvement](mtmv) Optimize MTMV partition lineage
check",
+ "url": "https://github.com/apache/doris/pull/63899";
+ },
+ {
+ "num": 66847,
+ "title": "[perf](load) Optimize write performance with improved
backpressure control",
+ "url": "https://github.com/apache/doris/pull/66847";
+ }
+ ],
+ "inProgressNarrative": "Real-time data warehouse work keeps
sharpening the planner, execution, and ecosystem parity. Highlights include
vectorized nullable comparison predicates, reuse of identical CTE-referencing
scalar subqueries, MV scan stats calibrated against actual row counts,
Trino-compatible bounding-box accessors and a third argument for
regexp_extract_all, TIMESTAMPTZ-preserving microsecond interval arithmetic, and
short-circuit point queries retaining",
+ "inProgress": [
+ {
+ "num": 66050,
+ "title": "[Feature](func) Support the third argument for
regexp_extract_all and regexp_extract_all_array",
+ "url": "https://github.com/apache/doris/pull/66050";
+ },
+ {
+ "num": 66688,
+ "title": "[opt](expr) vectorize nullable comparison predicates",
+ "url": "https://github.com/apache/doris/pull/66688";
+ },
+ {
+ "num": 66781,
+ "title": "[opt](nereids) Reuse identical uncorrelated scalar
subqueries referencing the same CTE",
+ "url": "https://github.com/apache/doris/pull/66781";
+ },
+ {
+ "num": 66814,
+ "title": "[feature](geo) Add st_xmax/st_xmin/st_ymax/st_ymin
functions",
+ "url": "https://github.com/apache/doris/pull/66814";
+ },
+ {
+ "num": 66853,
+ "title": "[feature](nereids) Calibrate mv scan stats by its
actual row count",
+ "url": "https://github.com/apache/doris/pull/66853";,
+ "isNew": true
+ }
+ ]
+ },
+ {
+ "name": "Compute-Storage Separation & Cloud-Native",
+ "mergedNarrative": "Cloud-native work this week systematically
corrects Meta Service, recycler, and commit-path details: rowset recycling
marking, parallel cumulative point advancement, checkpoint triggering,
group-commit routing, V2 delete bitmap blob handling, instance tombstone
retention, and Broker Load transaction cleanup. Together with config validation
and third-party build upgrades, multi-version multi-component coordination on
cloud becomes",
+ "merged": [
+ {
+ "num": 65550,
+ "title": "[fix](cloud) Only mark prepare rowsets before recycle",
+ "url": "https://github.com/apache/doris/pull/65550";
+ },
+ {
+ "num": 65915,
+ "title": "[fix](cloud) Correct cumulative point advancement when
enable parallel cumulative compaction",
+ "url": "https://github.com/apache/doris/pull/65915";
+ },
+ {
+ "num": 66154,
+ "title": "[fix](cloud) Trigger periodic cloud checkpoints
through edit log rolls",
+ "url": "https://github.com/apache/doris/pull/66154";
+ },
+ {
+ "num": 66585,
+ "title": "[fix](group commit) Fix group commit routing for
virtual compute groups",
+ "url": "https://github.com/apache/doris/pull/66585";
+ },
+ {
+ "num": 66697,
+ "title": "[feature](compute group) Support show compute groups
in non cloud mode",
+ "url": "https://github.com/apache/doris/pull/66697";
+ }
+ ],
+ "inProgressNarrative": "Compute-storage separation work this week
broadens object-store coverage and hardens BE for the cloud. New pieces unlock
S3 Express One Zone and keyless GCS vaults over GKE Workload Identity, while
pre-rowset delete-bitmap cleanup now uses point removes to ease FDB pressure.
Additional fixes address ARM64 correctness, K8s BE registration failure",
+ "inProgress": []
+ },
+ {
+ "name": "Agent Observability",
+ "mergedNarrative": "Agent observability work this week tightens
Variant and inverted-index search correctness. A typed-path Bloom Filter
regression, expanded Variant V1/V2 segment benchmarks, and fixes for deleted
documents and extra predicates in score TopN give AI/agent query paths on
semi-structured and full-text data more reliable behavior and reproducible
measurement.",
+ "merged": [
+ {
+ "num": 64573,
+ "title": "[test](variant) Add typed path bloom filter
regression",
+ "url": "https://github.com/apache/doris/pull/64573";
+ },
+ {
+ "num": 65821,
+ "title": "[fix](score) disable search topn with extra
predicates",
+ "url": "https://github.com/apache/doris/pull/65821";
+ },
+ {
+ "num": 66668,
+ "title": "[test](be) Extend Variant V1/V2 segment benchmarks",
+ "url": "https://github.com/apache/doris/pull/66668";
+ },
+ {
+ "num": 66979,
+ "title": "[fix](inverted index) Handle deleted documents in
score TopN",
+ "url": "https://github.com/apache/doris/pull/66979";
+ }
+ ],
+ "inProgressNarrative": "Agent observability work is hardening
Routine Load JSON semantics and inverted-index internals. One stream lets JSON
loads auto-complete unspecified columns against the base table schema for
clearer ingestion, while another stream removes a misused global null bitmap in
sub-readers and untangles SNII scoring and cache keying across analyzer
generations so",
+ "inProgress": [
+ {
+ "num": 64071,
+ "title": "[feature](routine-load) support fill_missing_columns
for json format to auto-complete unspecified columns",
+ "url": "https://github.com/apache/doris/pull/64071";
+ }
+ ]
+ },
+ {
+ "name": "Ecosystem Integration",
+ "mergedNarrative": "Ecosystem integration work fills gaps with
Spark/Hive, ADBC, and Arrow Flight SQL. The new stack table function matches
Spark/Hive output semantics, ADBC now materializes ListView variants correctly,
and Flight SQL exposes DATETIME as timezone-naive timestamps. Doris now
exchanges data across engines and clients with more faithful semantics.",
+ "merged": [
+ {
+ "num": 66734,
+ "title": "[feature](function) Support Spark/Hive-compatible
stack table function",
+ "url": "https://github.com/apache/doris/pull/66734";
+ },
+ {
+ "num": 66819,
+ "title": "[fix](adbc) Materialize Arrow list view arrays",
+ "url": "https://github.com/apache/doris/pull/66819";
+ },
+ {
+ "num": 66873,
+ "title": "[fix](exec) Return DATETIME as naive Arrow timestamp
(#10492)",
+ "url": "https://github.com/apache/doris/pull/66873";
+ }
+ ]
+ },
+ {
+ "name": "Security & Governance",
+ "mergedNarrative": "Security and governance work hardens log
redaction. Plaintext credentials in HTTP auth and Stream Load diagnostic paths
are masked, IV/CIPHER columns in information_schema.encryption_keys and
tls_private_key_password in backend_configuration are likewise masked, and a
new ldap_allow_empty_pass setting blocks LDAP login with empty
passwords—substantially reducing sensitive data exposure under default
privileges.",
+ "merged": [
+ {
+ "num": 61440,
+ "title": "[enhance](auth) introduction of configuration property
to prohibit login with empty LDAP password",
+ "url": "https://github.com/apache/doris/pull/61440";
+ },
+ {
+ "num": 66618,
+ "title": "[fix](auth) Mask credentials in authentication and
Stream Load logs",
+ "url": "https://github.com/apache/doris/pull/66618";
+ },
+ {
+ "num": 66834,
+ "title": "[fix](security) Mask sensitive fields in encryption
keys schema table",
+ "url": "https://github.com/apache/doris/pull/66834";
+ },
+ {
+ "num": 66836,
+ "title": "[fix](security) Mask private key password in backend
configuration",
+ "url": "https://github.com/apache/doris/pull/66836";
+ }
+ ],
+ "inProgressNarrative": "Security and governance work closes a
credential-exposure gap in config exports: after the BE side began masking the
TLS private-key password, the same FE config key remained unmasked and is now
redacted in Config.java so ADMIN SHOW FRONTEND CONFIG no longer returns the
plaintext secret.",
+ "inProgress": [
+ {
+ "num": 67007,
+ "title": "[fix](fe) Mask FE TLS private key and key store
passwords in config dumps",
+ "url": "https://github.com/apache/doris/pull/67007";,
+ "isNew": true
+ }
+ ]
+ },
+ {
+ "name": "Kernel & Engineering",
+ "mergedNarrative": "Kernel engineering this week advances
execution-stack refactors and build hygiene in parallel. ReadSchema unifies
column contracts across storage readers, row-binlog derivation moves into the
block-transform chain, and BE cold builds shrink via PCH slimming,
olap_common.h decoupling, and dwarf-5 line-tables debug info, with
DORIS_DEV_DEBUG_INFO locking in compile-time guardrails at configure time.",
+ "merged": [
+ {
+ "num": 66472,
+ "title": "[refactor](scan) Unify read schemas and column
ordinals across storage readers",
+ "url": "https://github.com/apache/doris/pull/66472";
+ },
+ {
+ "num": 66883,
+ "title": "[refactor](storage) derive row binlog in the transform
chain; delete RowBinlogSegmentWriter",
+ "url": "https://github.com/apache/doris/pull/66883";
+ },
+ {
+ "num": 66901,
+ "title": "[opt](build) Enforce BE compile-time hygiene at
configure time",
+ "url": "https://github.com/apache/doris/pull/66901";
+ },
+ {
+ "num": 66960,
+ "title": "[opt](build) Add DORIS_DEV_DEBUG_INFO=line-tables and
default the UT / macOS compile-check builds to it",
+ "url": "https://github.com/apache/doris/pull/66960";
+ }
+ ],
+ "inProgressNarrative": "Kernel and engineering work tightens
encodings, parsing, constant folding, and external-format safety. Segment V3
floats revert to plain encoding, backup table references survive replay, and
Parquet metadata allocation is bounded against malformed inputs. TIMESTAMPTZ
interval arithmetic, SHOW paging limits, parser dispatch by first token, and
left-factored primaryExpression grammars are also",
+ "inProgress": [
+ {
+ "num": 66670,
+ "title": "[feature](http) Add structured type metadata to table
schema API",
+ "url": "https://github.com/apache/doris/pull/66670";
+ }
+ ]
+ }
+ ],
+ "demand": []
+ }
+ ]
+};
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
