profile
viewpoint
Fangdun Cai fundon @viz-rs @vue-bulma @atom-minimap Quanzhou, China https://fundon.me Live & Life <- Postgres <- Rust <- Swift <- JS <- ...

fundon/co-ware 17

Ware inspired, easily create your own middleware layer using generators via co.

fundon/delegate-proxy 8

Super easy delegating method and accessor, using es6 Proxies with less than 15 lines of code

fundon/co-fs-plus 7

co-fs plus, supports fs.walk fs.mkdirp fs.readdir fs.rimraf

fundon/aspect-oriented 6

Aspect Oriented Programming for JavaScript

fundon/awesome-vue 6

A curated list of awesome things related to Vue.js

fundon/Build-Vue-Admin-Panel-Framework 6

Build Vue Admin Panel Framework

fundon/compose-iterator 6

Use iterator for middleware composition

fundon/dotfiles 3

My dotfiles: sh etc.

fundon/apps 2

The funny and cool apps.

Pull request review commentpandas-dev/pandas

PERF: fix regression in creation of resulting index in RollingGroupby

 def _apply(             # Our result will have still kept the column in the result             result = result.drop(columns=column_keys, errors="ignore") -        result_index_data = []-        for key, values in self._groupby.grouper.indices.items():-            for value in values:-                data = [-                    *com.maybe_make_list(key),-                    *com.maybe_make_list(-                        grouped_object_index[value]-                        if grouped_object_index is not None-                        else []-                    ),-                ]-                result_index_data.append(tuple(data))--        result_index = MultiIndex.from_tuples(-            result_index_data, names=result_index_names+        codes = self._groupby.grouper.codes+        levels = self._groupby.grouper.levels++        group_indices = self._groupby.grouper.indices.values()+        if group_indices:+            indexer = np.concatenate(list(self._groupby.grouper.indices.values()))+        else:+            indexer = np.array([], dtype=np.intp)+        codes = [c.take(indexer) for c in codes]++        if grouped_object_index is not None:+            if isinstance(grouped_object_index, MultiIndex):

maybe more clear

idx = grouped_object_index.take(indexer)
if not isinstance(grouped_object_index, MultiIndex):
    idx = MultiIndex.from_arrays(idx)
jorisvandenbossche

comment created time in a few seconds

Pull request review commentpandas-dev/pandas

PERF: fix regression in creation of resulting index in RollingGroupby

 def time_rolling_offset(self, method):         getattr(self.groupby_roll_offset, method)()  +class Groupby2:

how long does this take on prior branch? we want this < 1s

jorisvandenbossche

comment created time in 2 minutes

Pull request review commentpandas-dev/pandas

PERF: fix regression in creation of resulting index in RollingGroupby

 def time_rolling_offset(self, method):         getattr(self.groupby_roll_offset, method)()  +class Groupby2:

can you rename to GroupbyLowNumberOfGroups or similar

jorisvandenbossche

comment created time in 2 minutes

Pull request review commentpandas-dev/pandas

PERF: fix regression in creation of resulting index in RollingGroupby

 def _apply(             # Our result will have still kept the column in the result             result = result.drop(columns=column_keys, errors="ignore") -        result_index_data = []-        for key, values in self._groupby.grouper.indices.items():-            for value in values:-                data = [-                    *com.maybe_make_list(key),-                    *com.maybe_make_list(-                        grouped_object_index[value]-                        if grouped_object_index is not None-                        else []-                    ),-                ]-                result_index_data.append(tuple(data))--        result_index = MultiIndex.from_tuples(-            result_index_data, names=result_index_names+        codes = self._groupby.grouper.codes+        levels = self._groupby.grouper.levels++        group_indices = self._groupby.grouper.indices.values()+        if group_indices:+            indexer = np.concatenate(list(self._groupby.grouper.indices.values()))+        else:+            indexer = np.array([], dtype=np.intp)+        codes = [c.take(indexer) for c in codes]++        if grouped_object_index is not None:

can you add a comment here on what this doing / needed

jorisvandenbossche

comment created time in a minute

Pull request review commentcitusdata/citus

Columnar Store Storage Id

 setup {+    ALTER SEQUENCE cstore.cstore_storageid_seq RESTART WITH 205;

Guess this is to compensate for the id being in the output. Would normalizing work? I forgot if they apply to the isolation tests

pykello

comment created time in a minute

Pull request review commentcitusdata/citus

Columnar Store Storage Id

 CREATE SCHEMA cstore; SET search_path TO cstore; +CREATE SEQUENCE cstore_storageid_seq NO CYCLE;

the name of the sequence reads cstore.cstore_storageid_seq. Would you mind removing the cstore clasifier from the sequence as long as it lives in the cstore schema?

I think this is part of a bigger discussion on naming. Given we want to rename everything to columnar. It is undecided what schema we will put our objects in though.

pykello

comment created time in 7 minutes

Pull request review commentcitusdata/citus

Columnar Store Storage Id

 ROLLBACK TO SAVEPOINT s1; SELECT pg_size_pretty(pg_relation_size('t'));  pg_size_pretty ---------------------------------------------------------------------- 112 kB+ 120 kB (1 row)  COMMIT; -- vacuum should truncate the relation to the usable space VACUUM VERBOSE t; INFO:  statistics for "t":-total file size: 114688, total data size: 10754+storage id: 108

would it make sense to normalize this output away? It is highly dependant on how many columnar tables have been created before entering this suite. This makes later parallelization harder to do while also making it harder to run tests in isolation.

Is there a reason to verify its number here? or anywhere else in tests?

I do feel it is important information to output for debugging production systems etc. Just not for our tests in the VACUUM VERBOSE output

pykello

comment created time in 3 minutes

Pull request review commentcitusdata/citus

Columnar Store Storage Id

 #include "utils/memutils.h" #include "utils/lsyscache.h" #include "utils/rel.h"+#include "utils/relfilenodemap.h"+++/*+ * Content of the first page in main fork, which stores metadata at file+ * level.+ */+typedef struct ColumnarMetapage+{+	/*+	 * Store version of file format used, so we can detect files from+	 * previous versions if we change file format.+	 */+	int versionMajor;+	int versionMinor;

Thanks! I like the idea of starting the page with the version

pykello

comment created time in 18 minutes

Pull request review commentcitusdata/citus

Columnar Store Storage Id

 static void LogRelationStats(Relation rel, int elevel)

function is missing. a comment when it is called. hard to review what we are adding output to. I guess this is VACUUM VERBOSE?

pykello

comment created time in 9 minutes

Pull request review commenttikv/tikv

test: fix test test_ingest_key_manager_delete_file_failed

 fn test_ingest_key_manager_delete_file_failed() {      assert!(!resp.has_error()); -    // regenerate the file add ingest it again, even key manager contains file with the same name,-    // this action should success     fail::remove(deregister_fp); -    // Do upload and ingest again, though key manager contians this file, the ingest action should success.+    let node_id = *cluster.sim.rl().get_node_ids().iter().next().unwrap();+    let save_path = cluster+        .sim+        .rl()+        .importers+        .get(&node_id)+        .unwrap()+        .get_path(&meta);+    // wait until the raw uploaded file is deleted by the async clean up task.+    loop {

add some exit condition to avoid this test running forever.

glorv

comment created time in a few seconds

Pull request review commentcitusdata/citus

Columnar Store Storage Id

 ColumnarOptionsIndexRegclass(void)   /*- * CStoreDataFilesRelationId returns relation id of cstore_data_files.+ * CStoreSkipNodesRelationId returns relation id of cstore_skipnodes.  * TODO: should we cache this similar to citus?  */ static Oid-CStoreDataFilesRelationId(void)+CStoreSkipNodesRelationId(void) {-	return get_relname_relid("cstore_data_files", CStoreNamespaceId());+	return get_relname_relid("cstore_skipnodes", CStoreNamespaceId()); }   /*- * CStoreDataFilesIndexRelationId returns relation id of cstore_data_files_pkey.+ * CStoreSkipNodesIndexRelationId returns relation id of cstore_skipnodes_pkey.  * TODO: should we cache this similar to citus?  */ static Oid-CStoreDataFilesIndexRelationId(void)+CStoreSkipNodesIndexRelationId(void) {-	return get_relname_relid("cstore_data_files_pkey", CStoreNamespaceId());+	return get_relname_relid("cstore_skipnodes_pkey", CStoreNamespaceId()); }   /*- * CStoreSkipNodesRelationId returns relation id of cstore_skipnodes.- * TODO: should we cache this similar to citus?+ * CStoreNamespaceId returns namespace id of the schema we store cstore+ * related tables.  */ static Oid-CStoreSkipNodesRelationId(void)+CStoreNamespaceId(void) {-	return get_relname_relid("cstore_skipnodes", CStoreNamespaceId());+	return get_namespace_oid("cstore", false); }   /*- * CStoreSkipNodesIndexRelationId returns relation id of cstore_skipnodes_pkey.- * TODO: should we cache this similar to citus?+ * ReadMetapage reads metapage for the given relfilenode. It returns+ * false if the relation doesn't have a meta page yet.  */-static Oid-CStoreSkipNodesIndexRelationId(void)+static ColumnarMetapage *+ReadMetapage(RelFileNode relfilenode, bool missingOk) {-	return get_relname_relid("cstore_skipnodes_pkey", CStoreNamespaceId());+	StringInfo metapageBuffer = NULL;+	Oid relationId = RelidByRelfilenode(relfilenode.spcNode,+										relfilenode.relNode);+	if (OidIsValid(relationId))+	{+		Relation relation = relation_open(relationId, NoLock);++		RelationOpenSmgr(relation);+		int nblocks = smgrnblocks(relation->rd_smgr, MAIN_FORKNUM);+		RelationCloseSmgr(relation);++		if (nblocks != 0)+		{+			metapageBuffer = ReadFromSmgr(relation, 0, sizeof(ColumnarMetapage));+		}++		relation_close(relation, NoLock);+	}++	if (metapageBuffer == NULL)+	{+		if (!missingOk)+		{+			elog(ERROR, "columnar metapage was not found");+		}++		return NULL;+	}++	ColumnarMetapage *metapage = palloc0(sizeof(ColumnarMetapage));+	memcpy_s((void *) metapage, sizeof(ColumnarMetapage),+			 metapageBuffer->data, sizeof(ColumnarMetapage));++	return metapage; }   /*- * CStoreNamespaceId returns namespace id of the schema we store cstore- * related tables.+ * InitMetapage initializes metapage for the given relation.  */-static Oid-CStoreNamespaceId(void)+static ColumnarMetapage *+InitMetapage(Relation relation) {-	return get_namespace_oid("cstore", false);+	ColumnarMetapage *metapage = palloc0(sizeof(ColumnarMetapage));+	metapage->storageId = GetNextStorageId();+	metapage->versionMajor = CSTORE_VERSION_MAJOR;+	metapage->versionMinor = CSTORE_VERSION_MINOR;++	/* create the first block */+	Buffer newBuffer = ReadBuffer(relation, P_NEW);+	ReleaseBuffer(newBuffer);++	Assert(sizeof(ColumnarMetapage) <= BLCKSZ - SizeOfPageHeaderData);+	WriteToSmgr(relation, 0, (char *) metapage, sizeof(ColumnarMetapage));++	return metapage;+}+++/*+ * GetNextStorageId returns the next value from the storage id sequence.+ */+static uint64+GetNextStorageId(void)+{+	Oid sequenceId = get_relname_relid("cstore_storageid_seq", CStoreNamespaceId());+	Datum sequenceIdDatum = ObjectIdGetDatum(sequenceId);++	/*+	 * Generate new and unique storage id from sequence.+	 * TODO: should we restrict access to the sequence, which might require+	 * switching security context here?

What can go wrong if this sequence is not protected? I feel we should conclude this TODO before merging.

The only thing I can see going wrong here is that the user decreases the sequence causing the same storage id to be reused. Is there any way we can enforce uniqueness of the storage id when fetching one from the sequence?

When the user decreases the sequence I'd rather throw an error next time we try to create a table saying the storage id was already in use. Then we don't need any restrictions, and it will also guard against a wrap around (which, even on 32 bit, I deem unlikely here, let alone 64 bit we return).

pykello

comment created time in 11 minutes

pull request commentprettier/prettier

Add `Doc[]` as `Doc`

Looks good. Let's revert things in doc-builders and merge.

fisker

comment created time in a few seconds

PR opened sbt/sbt

dotty-library is called scala3-library now

I suspect this fixes https://github.com/lampepfl/dotty/issues/10558 but haven't actually tested it.

+1 -1

0 comment

1 changed file

pr created time in a few seconds

Pull request review commentapple/swift

[WORK IN PROGRESS] Task Groups implementation

 public struct PartialAsyncTask { public struct UnsafeContinuation<T> {   private var context: UnsafeRawPointer -  public func resume(_: __owned T) { }+  /// Return a value into the continuation and make the task schedulable.+  ///+  /// The task will never run synchronously, even if the task does not+  /// need to be resumed on a specific executor.+  ///+  /// This is appropriate when the caller is something "busy", like an event+  /// loop, and doesn't want to be potentially delayed by arbitrary work.+  public func resume(returning _: __owned T) { }

TODO: rebase; we moved those in a previous PR over here, make sure it's all aligned.

ktoso

comment created time in a minute

pull request commentAutomattic/wp-calypso

Update CI image to use bash as the default shell

<!-- icfy-watermark: c52822 --> This PR does not affect the size of JS and CSS bundles shipped to the user's browser.

Generated by performance advisor bot at iscalypsofastyet.com.

scinos

comment created time in a minute

Pull request review commentapple/swift

[WORK IN PROGRESS] Task Groups implementation

 void IRGenSILFunction::visitFunctionRefBaseInst(FunctionRefBaseInst *i) {       fn, NotForDefinition, false /*isDynamicallyReplaceableImplementation*/,       isa<PreviousDynamicFunctionRefInst>(i));   llvm::Value *value;+//  auto isSpecialAsyncWithoutCtxtSize =+//      fn->isAsync() && fn->getName().equals("swift_task_future_wait");   auto isSpecialAsyncWithoutCtxtSize =-      fn->isAsync() && fn->getName().equals("swift_task_future_wait");+      fn->isAsync() && (+          fn->getName().equals("swift_task_future_wait") ||+          fn->getName().equals("swift_task_group_poll") // TODO: do we need this?

TODO: not actively used today; but it may be nicer to call directly into swift_task_group_poll so we don't pollute the swift_task_future_wait impl with the special case for if it's a group child.

ktoso

comment created time in a minute

issue openedprettier/prettier

.editorconfig file with default Prettier settings

Hi,

it's not an issue - just a question. Is any where in GitHub or elsewhere .editorconfig file that would consist of all default prettier config?

Thanks!

created time in 2 minutes

Pull request review commentelastic/elasticsearch

Record timestamp field range in index metadata

+/*+ * Licensed to Elasticsearch under one or more contributor+ * license agreements. See the NOTICE file distributed with+ * this work for additional information regarding copyright+ * ownership. Elasticsearch 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.elasticsearch.index.shard;++import org.elasticsearch.common.Nullable;+import org.elasticsearch.common.io.stream.StreamInput;+import org.elasticsearch.common.io.stream.StreamOutput;+import org.elasticsearch.common.io.stream.Writeable;+import org.elasticsearch.common.xcontent.ToXContentFragment;+import org.elasticsearch.common.xcontent.XContentBuilder;+import org.elasticsearch.common.xcontent.XContentParser;++import java.io.IOException;+import java.util.ArrayList;+import java.util.Arrays;+import java.util.List;+import java.util.Objects;+import java.util.stream.IntStream;++import static org.elasticsearch.index.shard.ShardLongFieldRange.LONG_FIELD_RANGE_VERSION_INTRODUCED;++/**+ * Class representing an (inclusive) range of {@code long} values in a field in an index which may comprise multiple shards. This+ * information is accumulated shard-by-shard, and we keep track of which shards are represented in this value. Only once all shards are+ * represented should this information be considered accurate for the index.+ */+public class IndexLongFieldRange implements Writeable, ToXContentFragment {++    /**+     * Sentinel value indicating that no information is currently available, for instance because the index has just been created.+     */+    public static final IndexLongFieldRange UNKNOWN = new IndexLongFieldRange(new int[0], Long.MAX_VALUE, Long.MIN_VALUE);++    /**+     * Sentinel value indicating an empty range, for instance because the field is missing or has no values in any shard.+     */+    public static final IndexLongFieldRange EMPTY = new IndexLongFieldRange(null, Long.MAX_VALUE, Long.MIN_VALUE);++    /**+     * Sentinel value indicating the actual range may change in the future.+     */+    public static final IndexLongFieldRange MUTABLE = new IndexLongFieldRange(null, Long.MIN_VALUE, Long.MAX_VALUE);

I'm leaving that logic up to you :smile: I definitely think they're both complete, but if you want to implement some of the coordinator-node-side logic as more methods on IndexLongFieldRange then that's fine by me.

The EMPTY case sounds to me to be suitable for rewriting on the coordinator: if we require docs with a timestamp in a certain range, and the index contains no docs with a timestamp, then should we not skip this index?

DaveCTurner

comment created time in 2 minutes

push eventethereum/go-ethereum

Martin Holst Swende

commit sha aba0c234c29c72860c369ec97553716a3fad11cd

cmd/geth: make tests run quicker + use less memory and disk (#21919)

view details

push time in 2 minutes

push eventAutomattic/wp-calypso

Rosalind W

commit sha 3d2d1ba6b29168f721269f63e1d676f377e99bed

Adding support for spacer blocks to web notification rendering (#46019)

view details

Filippo Di Trapani

commit sha bd0a180e2f485ac5ef9124ccd29de5edf13be15d

Update/site level links (#46028) * Update renewal plan link * Add link to notice Co-authored-by: fditrapani <filippo.di.trapani@a8c.com>

view details

Filippo Di Trapani

commit sha 8e8c34d3cca84a52f41a905c3020af08fb2f5f1c

Purchases: Hide plan controls for non-owners at site and account-level (#45979) * Hide plan controls to non owners * Add notice messaging for non-owners * Cleanup unecessary code * Cleanup unecessary code * Use `purchase.userId` instead of `isCurrentUserCurrentPlanOwner` * Remove unused prop * Hide Jetpack upgrade button * Cleanup redundent code * Fix test * Fix formatting Co-authored-by: fditrapani <filippo.di.trapani@a8c.com> Co-authored-by: Michael Cain <michael@michaeldcain.com>

view details

Caleb Bauermeister

commit sha 59303d1b8dedc2acc7147ee44714f73981033956

Jetpack Cloud: Update SSH credentials in place flow behind feature flag (#46024)

view details

Sergio Cinos

commit sha 96747065421e19f9e132744d16cce2a3cc30c587

Use checkstyle to detect eslint errors and warnings (#46044)

view details

Glen Davies

commit sha b3d3062e61115fe35a55f577c8eaa9d86cfaa254

Update Newspack to 1.13.1 - adds alignment options to post carousel block (#46023) * Update Newspack to 1.13.1 - adds alignment options to post carousel block\ * Update yarn lock Co-authored-by: Glen Davies <glen.davies@a8c.com>

view details

Payton Swick

commit sha 48c216ecd93546c5e23a6d94c0b819817909dbcd

Make sure WPOrderReviewList never renders undefined (#46057)

view details

Allison Levine

commit sha c5e9b2539007a6b78fb6bf842096bd3436a0414d

Revert Convert to blocks button override code (#45552) Now that Core has the "Convert to blocks" button as part of the classic block (as of Gutenberg 8.6), we can remove this override code from wpcom-block-editor. Reverts #43902.

view details

Jarda Snajdr

commit sha 268ef9584c20675e8869ea502a3826f7263fd40d

Remove the isSectionIsomorphic selector (#46036) * Boot: import setRoute action without aliasing * Give consistent names to page.js middlewares that setup locale and section Avoids name conflicts with Redux action names and allows imports without aliasing. * setSectionMiddleware: save the section object to context * Server: read the section.isomorphic flag from context.section rather than from Redux Server routing should read all information from routing `context` and shouldn't depend on Redux. The Redux store should be used only by the React UI when rendering. Also updates the server routing test. It's nice that it no longer needs to mock the Redux store -- the routing code doesn't use it. * Remove the unused isSectionIsomorphic selector

view details

Louis Laugesen

commit sha ac744543cbc19a654224e1637fc3e5502599d08b

Update/remove coming soon option (#46034) * Try to hide Coming Soon v1 option if coming-soon-v2 flag is set * types * Add tests

view details

Jarda Snajdr

commit sha 4ab7e6f742e062feb1a616f21556aaa2844b676d

Media: fix runtime TypeError when navigating to page with null selected site (#46080)

view details

Dave Smith

commit sha 2903e64306ab95b51a32ed2378bf7b3745fee23d

Avoid multiple selected menu items (#46050) * Properly determine selected item * Add class to indicate menu is selected * Add dedicated classes for parent and child menu items * Match styles with WPAdmin * Auto expand section if selected or child is selected. * Only auto expand section for selected item once

view details

Sergio Cinos

commit sha 2f93bde7157e98a322c6d9fec20fc5b3a331560c

Delete disabled tests (#46064)

view details

Harris Papazolgou

commit sha 07ef7540d35815071379fae9b13d503177cffd2f

Use masterbar height var instead of explicit values (#46035) * Use masterbar height var instead of explicit values * Adds comment with source file * Allows border-radius style so that it is in par with wp-admin Co-authored-by: cpap <papazoglou.charalampos@gmail.com>

view details

Ramon

commit sha 6ea502644f291a558ce7a65b72d9195f9b64ff81

We're using `coming-soon-v2` in other commits. This just standardizes it for Gutenboarding. (#46076)

view details

Marin Atanasov

commit sha b9f08ca0941056e6a7ee7de8c2562fa7a1be08b8

i18n: Add form encoding to GlotPress postRequest method (#46088)

view details

Marin Atanasov

commit sha 7e38671e7f8fef2d8896cd33f0a9bed6c389efd0

Components: Use FormTextarea for TranslatableTextarea (#46092)

view details

Harris Papazolgou

commit sha 5768829f5cabafddb64a5a529ba70974112bfac7

Nav Unification: Me section routing (#46048) * Clicking me should: - focus on the sidebar when already focused to me sections - focus to me content - fixed bug when clicking my sites after me wasn't working - also removed unneeded flag in read url * Fix comments * Make conditional more predictive Co-authored-by: cpap <papazoglou.charalampos@gmail.com>

view details

Michael P. Pfeiffer

commit sha 62aff13313a96b6cdf73bd4b2c6d80cbda5e44a4

Nav unification: add flyout menu to sidebar (#46046) * Nav unification: add flyout menu to sidebar * Remove bold font weight on hovered item * Show flyout only on screen sizes >782px * Remove var added in merge conflict * Selected state overrides hover state * Ensure blue background only appears in selected state * Adjust hover text color in reader

view details

Filippo Di Trapani

commit sha bc8a35b2cd35fa9548622f68c0ade3f814f291ca

Purchases: update upcoming renewal link (#46068) * Purchases: update upcoming renewal link * Fix prop names Co-authored-by: fditrapani <filippo.di.trapani@a8c.com>

view details

push time in 2 minutes

PR merged ethereum/go-ethereum

cmd/geth: make tests run quicker + use less memory and disk

This PR fixes a lot of the tests in cmd/geth so they run faster and consume less memory, which should improve the appveyor/travis speeds. The tests run about twice as fast on my laptop.

  • Adds nousb
  • Lower the cache to 64
  • Use syncmode=full, avoid allocating fast-sync bloom
  • Use ropsten instead of mainnet. No alloc-portion makes writing the genesis quicker
  • Combine the three welcome-attachers to use one instance
  • Make les server use lightkdf during import
+71 -89

0 comment

5 changed files

holiman

pr closed time in 2 minutes

pull request commentelastic/elasticsearch

Create system indices on-demand

I've split out the infrastructure parts of this PR into #65604.

pugnascotia

comment created time in 2 minutes

push eventRocketChat/Rocket.Chat

Diego Sampaio

commit sha 12c8f1359d04337780620c6ffd367c3a57b8c86c

Bump version to 3.9.0-rc.0

view details

Diego Sampaio

commit sha 7ee00d1d25f6839cbb95b53c9a01b1952f32e2c2

Merge remote-tracking branch 'origin/develop' into release-candidate

view details

Diego Sampaio

commit sha 920892181f984529ddda0cd01685f2ce5e4c8b72

Bump version to 3.9.0-rc.1

view details

Diego Sampaio

commit sha c68082df878934e126f3477ceb7c5a07706955d1

Merge remote-tracking branch 'origin/develop' into release-candidate

view details

Diego Sampaio

commit sha eab56ee9eb588856b7d3f1c7044add66ab37ea46

Bump version to 3.9.0-rc.2

view details

Diego Sampaio

commit sha 99ff86c2448875fe60651401d98a5015f07a03ed

Merge remote-tracking branch 'origin/develop' into release-candidate

view details

Diego Sampaio

commit sha 443dd87bcf76eead3f40f2f0af3bf92ab0b63143

Bump version to 3.9.0-rc.3

view details

Diego Sampaio

commit sha 188ccdeee1b2fd11a1ddaa363397d855930a5b23

Merge remote-tracking branch 'origin/develop' into release-candidate

view details

Diego Sampaio

commit sha 69cdd69f54dc6278192ab046d52bbbfcc3f447ab

Bump version to 3.9.0-rc.4

view details

Diego Sampaio

commit sha a89532d9ef89c0bebdf1ad8a5a3803c18b97af5a

Merge remote-tracking branch 'origin/develop' into release-candidate

view details

Diego Sampaio

commit sha 79522800e82ef6fb60ca511a6cb24b5644b3017a

Bump version to 3.9.0-rc.5

view details

Diego Sampaio

commit sha e4cfc14bd8df460a45bdee04e0143cd1214b07b7

Bump version to 3.9.0

view details

Diego Sampaio

commit sha d05e26f4fc0d770bfdaed7e8b81a55dbc85dc1d3

Merge pull request #19706 from RocketChat/release-3.9.0 Release 3.9.0

view details

Diego Sampaio

commit sha ac6535f7685a9ee22064c60aac65331b692c3e27

Bump version to 3.10.0-develop

view details

Guilherme Gazzo

commit sha 4b3d9a62479a9cd196ff4117b2a06b64dad4a06d

Merge pull request #19720 from RocketChat/develop-sync Merge master into develop & Set version to 3.10.0-develop

view details

Guilherme Gazzo

commit sha abb80f0da18238db2bd5165c44d9d021affaca68

[FIX] UIKit Modal not scrolling (#19690)

view details

Douglas Fabris

commit sha 3ea8e7835c4e228dda150c71fea998a08e24d88a

[IMPROVE] Rewrite contextualbar OTR panel (#19674)

view details

Martin Schoeler

commit sha 9a224afa17263a0c6cc326e1f1d3297e2da1a14a

Merge branch 'develop' into CustomScroll

view details

push time in 2 minutes

issue commentPrismJS/prism

Incomplete support for CSP

Thank you for doing all this research @edukisto!

I think that Prism should support all keywords and directives that it into a WD or RC of CSP. We can't assume that people only use Prism to highlight the latest version of CSP, so we do have to support some older versions as well. To answer your questions:

  1. CSP3, yes. Trusted Types, no. TT is only an Editor's draft right now and "is provided for discussion only and may change at any moment."
  2. It's not part of the CSP spec, so no.
  3. Yes, and...
  4. yes. The UI Security and the Visibility API seems to be fairly official, so let's support it.
  5. No, people might want to highlight code containing these tokens, so let's keep them.
  6. Yes, please.
edukisto

comment created time in 2 minutes

issue commentapache/incubator-mxnet

No matching distribution found for mxnet==1.7.0.post1

I think the problem was that the windows pip package for mxnet has become too large to publish on PyPI.

StevenJokess

comment created time in 2 minutes

Pull request review commentapple/swift

[WORK IN PROGRESS] Task Groups implementation

 struct SomeStruct {  func acceptAutoclosureNonAsync(_: @autoclosure () -> Int) async { } func acceptAutoclosureAsync(_: @autoclosure () async -> Int) async { }+func acceptAutoclosureAsyncThrows(_: @autoclosure () async throws -> Int) async { }+func acceptAutoclosureAsyncThrowsRethrows(_: @autoclosure () async throws -> Int) async rethrows { }+func acceptEscapingAsyncThrowsRethrows(_: @escaping (inout String) async throws -> Int) async rethrows { }+func acceptEscapingAsyncThrowsRethrows<R>(_: (inout String) async throws -> Int) async rethrows -> R { }

TODO: remove from this PR

ktoso

comment created time in 3 minutes

pull request commentnodejs/node

net: refactor to use more primordials

Benchmark CI: https://ci.nodejs.org/view/Node.js%20benchmark/job/benchmark-node-micro-benchmarks/734/

aduh95

comment created time in 3 minutes

issue commentlovell/sharp

cc: error: arm64: No such file or directory / compilation of both libvips and sharp from source

Have the libjpeg header files been installed? https://packages.debian.org/buster/libjpeg62-turbo-dev

konghouyin

comment created time in 3 minutes

pull request commentelastic/elasticsearch

Add infrastructure for managing system indices

Pinging @elastic/es-core-infra (Team:Core/Infra)

pugnascotia

comment created time in 3 minutes

Pull request review commentapple/swift

[WORK IN PROGRESS] Task Groups implementation

 func test_taskGroup_add() async throws -> Int {  func test_taskGroup_addHandles() async throws -> Int {   await try Task.withGroup(resultType: Int.self) { group in-    let one = await group.addWithHandle {+    let one = await group.add {

addWithHandle is gone, it is not necessary, we simply return the handle always and it is discardable

ktoso

comment created time in 3 minutes

more