profile
viewpoint
Marcus Weiner mraerino @netlify Cologne, Germany Working on the future of the JAMstack

aergus/collectd-batman-adv 3

a batman-adv plugin for collectd

mraerino/cdn-static 2

A static file CDN hosted on Cloudflare Workers (+KV) and Wasabi

devops-nrw/homepage 1

Our meetup's little homepage. Please improve.

ffddorf/gluon-rfkill-disable 1

Gluon package to disable wifi onoff button

mraerino/awm-postfixadmin-change-password 1

AfterLogic WebMail Lite Password change plugin for systems utilising PostfixAdmin

aergus/u23-projekt-netzwerk-monitoring-slides 0

slides about a network monitoring project for a Freifunk network (in German, cf. collectd-batman-adv)

FactoryCampus/netbox 0

IP address management (IPAM) and data center infrastructure management (DCIM) tool.

Pull request review commentnetlify/netlify-commons

feat(kafka): add new package

+package kafka++import (+	"context"+	"fmt"+	"log/syslog"+	"strings"+	"time"++	kafkalib "github.com/confluentinc/confluent-kafka-go/kafka"+	"github.com/pkg/errors"+	"github.com/sirupsen/logrus"+)++// Config holds all the configuration for this package.+type Config struct {+	Brokers      []string       `json:"brokers"`+	Topic        string         `json:"topic"`+	ProducerConf ProducerConfig `json:"producer" envconfig:"producer"`+	ConsumerConf ConsumerConfig `json:"consumer" envconfig:"consumer"`

these could be renamed since we no longer have methods with the names, right?

smoya

comment created time in 2 days

push eventnetlify/netlify-commons

Marcus Weiner

commit sha 3f245dc79315bb2d700707a94037aca81036b639

Upgrade to Go 1.15 (#149)

view details

push time in 2 days

delete branch netlify/netlify-commons

delete branch : chore/go-1.15

delete time in 2 days

PR merged netlify/netlify-commons

Upgrade to Go 1.15 type: chore

Recently released. I think supporting latest two is the default.

+7 -7

0 comment

2 changed files

mraerino

pr closed time in 2 days

PR opened netlify/netlify-commons

Upgrade to Go 1.15 type: chore

Recently released. I think supporting latest two is the default.

+7 -7

0 comment

2 changed files

pr created time in 2 days

create barnchnetlify/netlify-commons

branch : chore/go-1.15

created branch time in 2 days

delete branch mraerino/deno

delete branch : fix/memory-limit-callback-remove-previous

delete time in 3 days

delete branch mraerino/deno

delete branch : feat/memory-limits

delete time in 3 days

pull request commentdenoland/deno

core: memory limits & callbacks

Now that you say this:

Ah I see you've limited it to support only one callback.. That's ok.

I realized a small thing was missing: https://github.com/denoland/deno/pull/7025

mraerino

comment created time in 3 days

PR opened denoland/deno

Remove previous heap limit callback when registering

Follow-up on #6914

I realized it was missing the removal of a previously registered callback.

+56 -2

0 comment

1 changed file

pr created time in 3 days

create barnchmraerino/deno

branch : fix/memory-limit-callback-remove-previous

created branch time in 3 days

Pull request review commentdenoland/deno

core: memory limits & callbacks

 impl CoreIsolate {     let mut state = state_rc.borrow_mut();     state.op_registry.register(name, op)   }++  /// Registers a callback on the isolate when the memory limits are approached.+  /// Use this to prevent V8 from crashing the process when reaching the limit.+  ///+  /// Calls the closure with the current heap limit and the initial heap limit.+  /// The return value of the closure is set as the new limit.+  pub fn add_near_heap_limit_callback<C>(&mut self, cb: C)+  where+    C: FnMut(usize, usize) -> usize + 'static,+  {+    let boxed_cb = Box::pin(RefCell::new(cb));+    let data = boxed_cb.as_ptr() as *mut c_void;+    self.allocations.near_heap_limit_callback_data =+      Some((boxed_cb, near_heap_limit_callback::<C>));+    self+      .v8_isolate+      .as_mut()+      .unwrap()+      .add_near_heap_limit_callback(near_heap_limit_callback::<C>, data);+  }++  pub fn remove_near_heap_limit_callback(&mut self, heap_limit: usize) {

yeah, this is kinda hard. the point here is: without a callback, V8 crashes with a SIGILL which is impossible to catch in a test. that means it's not possible to assert whether the callback was really not called.

I tested it locally though and the test crashed in the expected way.

mraerino

comment created time in 3 days

Pull request review commentdenoland/deno

core: memory limits & callbacks

 pub enum StartupData<'a> {   None, } +impl StartupData<'_> {+  fn into_options(self) -> (Option<OwnedScript>, Option<Snapshot>) {+    match self {+      Self::Script(script) => (Some(script.into()), None),+      Self::Snapshot(snapshot) => (None, Some(snapshot)),+      Self::None => (None, None),+    }+  }+}+ type JSErrorCreateFn = dyn Fn(JSError) -> ErrBox; +type PinnedRefCell = Pin<Box<RefCell<dyn Any>>>;

good point. removed.

mraerino

comment created time in 3 days

push eventmraerino/deno

Marcus Weiner

commit sha b35347f8112d1a86997f75d35d6e8c2760bdb547

Remove pinning

view details

Marcus Weiner

commit sha 7c8fd8aa7f0bc58281b55eb1d111b79c92ddad75

Test callback remove

view details

push time in 3 days

Pull request review commentdenoland/deno

core: memory limits & callbacks

 pub mod tests {     let mut isolate2 = CoreIsolate::new(startup_data, false);     js_check(isolate2.execute("check.js", "if (a != 3) throw Error('x')"));   }++  #[test]+  fn test_heap_limits() {+    let heap_limits = HeapLimits {+      initial: 0,+      max: 20 * 1024, // 20 kB+    };+    let mut isolate =+      CoreIsolate::with_heap_limits(StartupData::None, heap_limits);+    let cb_handle = isolate.thread_safe_handle();+    isolate.add_near_heap_limit_callback(+      move |current_limit, _initial_limit| {+        cb_handle.terminate_execution();+        current_limit * 2

done

mraerino

comment created time in 3 days

push eventmraerino/deno

Marcus Weiner

commit sha 7556c0b2c02483ded41c2c52f98a0d2d55248853

Assert on callback calls

view details

push time in 3 days

pull request commentdenoland/deno

core: memory limits & callbacks

ok, i can't explain the failures. between the two runs different things failed, so they seem non-deterministic.

my changes only add new interfaces to deno_core that are not being used in cli

mraerino

comment created time in 4 days

push eventmraerino/deno

Marcus Weiner

commit sha 40d48afd50b55b5d234e4a1dd82adb21b516fb88

Rerun CI

view details

push time in 4 days

pull request commentdenoland/deno

core: memory limits & callbacks

did this fail because of a flaky test?

mraerino

comment created time in 4 days

push eventmraerino/deno

Bartek Iwańczuk

commit sha 6fcf06306ed2ea52031a97b918f1e929d7209250

feat(doc): handle imports (#6987) This commit adds additional objects to JSON output of "deno doc" command to facilitate linking between types in different modules.

view details

Marcus Weiner

commit sha 794d693c9a47087cb576836f6fe2057f3f8b88ca

Merge branch 'master' into feat/memory-limits

view details

push time in 4 days

push eventmraerino/deno

Marcus Weiner

commit sha 1ad87bb4b604e32670fa15161a5b24f5943681d1

Add missing link in docs

view details

push time in 4 days

push eventmraerino/deno

Marcus Weiner

commit sha 7f0ffb4def5066ec7586cefa23ff582e5f8c36be

Fix type complexity

view details

push time in 4 days

push eventmraerino/deno

Kitson Kelly

commit sha 076547fbbb52b1df9d6db36dd72353bffa9e4a8b

chore: use matches macro for bool matches (#6904)

view details

Ryan Dahl

commit sha b7942bf0f6f151e172db9b1e08cf4436e8365e8b

Change release build flags to optimize for size (#6907)

view details

Bartek Iwańczuk

commit sha 315efbc0e86d9d848efa4b889aefbc665508598a

fix: downcast from SwcDiagnosticBuffer to OpError (#6909)

view details

Bartek Iwańczuk

commit sha c6917133942c791480cd2aec7297b2a2ee623059

refactor: Use SWC to strip types for "--no-check" flag (#6895)

view details

Luca Casonato

commit sha 1b60840f286bc0203a3bd2900f67557a8ff2c3f6

feat(std/async): add pooledMap utility (#6898)

view details

Nayeem Rahman

commit sha 6afe7bbd395eb34febd25bda48c5cd82f0a4f8b3

fix(std/path): Percent-decode in fromFileUrl() (#6913)

view details

Kid

commit sha 391088c6ff531ac6714e9076eb62f2294793ccdd

docs(std/encoding): Fix TOML docs (#6912)

view details

zfx

commit sha 95597fc6e2ed9975c26eae77ba68ae42ca2c331a

fix(std/tar): directory type bug (#6905)

view details

Divy Srivastava

commit sha 0da4779b178a0c97925c1ef5589d0f24a37f4501

feat(unstable): add "--ignore" flag to deno fmt (#6890)

view details

Nayeem Rahman

commit sha 6e7208bec2911ac0d1729f334fc90bc50b8f9203

fix(cli/rt): Fix file URL to path conversion on Windows (#6920)

view details

Luca Casonato

commit sha 4afb4b6e46de2ed536a3c9828d70d7799b5b6d03

feat: add $STD_VERSION replacement variable in docs (#6922)

view details

Bartek Iwańczuk

commit sha b718e6ff53156a0aae486e570ce7c9cb8a3b822a

upgrade: deno_lint, dprint, swc (#6928) This commit upgrades: deno_lint 0.1.20 dprint-plugin-typescript 0.25.0 swc_ecmascript 0.1.0 SWC is no longer reexported from dprint nor deno_lint.

view details

Ryan Dahl

commit sha 7446230e6b39a0d3ce70b8fffbed3ebc63e8c762

v1.2.2

view details

Bartek Iwańczuk

commit sha d54d4d21bddf0fe80a6f534ec945c655595c5f71

upgrade: swc_ecmascript (#6943)

view details

Luca Casonato

commit sha 9d50c5c1a6fbe6895b6a63dadcba7bacaf9dd578

feat: hash file names in gen cache (#6911)

view details

Ben Stolovitz

commit sha 39693094f14ecba128c8fe9c6a1b8c0aba82fdec

Fix grammatical error in Linter documentation (#6939)

view details

Chayim Refael Friedman

commit sha e9d713c007afd31e22c9edf2b256197ed12336f2

typo (#6937)

view details

Jarrett Helton

commit sha d615ebefe2e306f2877afb40dc603f71263407d6

fix(std/toml): parser error with inline comments (#6942)

view details

Maayan Hanin

commit sha 5fc5e7b54a9fba421dfc473016625a4f592403ed

fix(cli): add support for non-UTF8 source files (#6789) Fixes: #5542

view details

Rauf Islam

commit sha 40ead6cc98ab1d02c8ca0587d24c218e81aa47b7

fix(std/encoding/toml): Add boolean support to stringify (#6941)

view details

push time in 4 days

pull request commentdenoland/rusty_v8

Allow creating snapshot based on other snapshot

cool, i see. thanks for expanding on it 🙌

i'm ok with this not landing right now. good to get the ball rolling though!

mraerino

comment created time in 7 days

Pull request review commentdenoland/rusty_v8

Allow creating snapshot based on other snapshot

 pub struct SnapshotCreator([usize; 1]); impl SnapshotCreator {   /// Create and enter an isolate, and set it up for serialization.   /// The isolate is created from scratch.-  pub fn new(external_references: Option<&'static ExternalReferences>) -> Self {+  pub fn new(+    external_references: Option<&'static ExternalReferences>,+    startup_data: Option<StartupData>,+  ) -> Self {     let mut snapshot_creator: MaybeUninit<Self> = MaybeUninit::uninit();-    let external_references_ptr = if let Some(er) = external_references {-      er.as_ptr()-    } else {-      std::ptr::null()-    };+    let external_references_ptr =+      external_references.map_or_else(std::ptr::null, |er| er.as_ptr());+    let startup_data_ptr = startup_data+      .map_or_else(std::ptr::null_mut, |sd| Box::into_raw(Box::new(sd)));     unsafe {       v8__SnapshotCreator__CONSTRUCT(         &mut snapshot_creator,         external_references_ptr,+        startup_data_ptr,       );       snapshot_creator.assume_init()

wait, how does this have anything to do with my changes? I haven't touched the way the SnapshotCreator is constructed?

mraerino

comment created time in 9 days

pull request commentdenoland/rusty_v8

Allow creating snapshot based on other snapshot

I think in the long term we should use CreateParams for both Isolate and SnapshotCreator. But it's good to follow the C++ api for now.

the bad part is that a snapshot creator will not allow you to set anything except those two params exposed now. since the snapshot creator will initialize the isolate, you cannot set heap limits, for example 😞

mraerino

comment created time in 9 days

pull request commentdenoland/rusty_v8

Allow creating snapshot based on other snapshot

so, i totally missed that a reference would be enough here.

still it needs to be mut according to the C headers. do you think there would be a way of working around that fact?

mraerino

comment created time in 9 days

push eventmraerino/rusty_v8

Marcus Weiner

commit sha 243c33d4c2692797b8b58f6bb1fdb08a2da7b2ae

Use startup data as reference

view details

push time in 9 days

delete branch mraerino/rusty_v8

delete branch : feat/heap-limits

delete time in 10 days

push eventmraerino/deno

Marcus Weiner

commit sha be92a6c87ff4e794ff272d4f2619235751abc0dc

Document callback register method

view details

Marcus Weiner

commit sha 04bb8550198b56340c4cc4cc72d7583494209d65

Test termination on heap limit

view details

push time in 10 days

push eventmraerino/rusty_v8

Marcus Weiner

commit sha 79f9047c491ea44012accaaccf0c090058487368

Fix clippy warnings

view details

push time in 10 days

PR opened denoland/rusty_v8

Allow creating snapshot based on other snapshot

Introduces the ability to initialize the isolate that the SnapshotCreator creates with an optional snapshot blob, just as it's possible on a regular isolate.

This is a small breaking change. If you prefer it to be not breaking, I can introduce a second constructor instead.

+94 -11

0 comment

3 changed files

pr created time in 10 days

create barnchmraerino/rusty_v8

branch : refactor/snapshot-creator

created branch time in 10 days

pull request commentdenoland/deno

WIP: core: memory limits & callbacks

@piscisaureus this needs a release in rusty_v8. were you going to make one soon?

mraerino

comment created time in 10 days

startedvalyala/gozstd

started time in 12 days

startedDataDog/zstd

started time in 12 days

startedklauspost/compress

started time in 12 days

pull request commentdenoland/rusty_v8

Allow managing V8 heap limits

tests added. could you review again?

mraerino

comment created time in 15 days

issue commentdenoland/deno

OOM error when uploading large file

this would allow more control over heap size and preventing OOM: https://github.com/denoland/deno/issues/6916

ry

comment created time in 16 days

issue openeddenoland/deno

Core: Allow fine-grained heap memory control

Problem

Right now, the only way to control the heap memory size in deno_core is via passing --max-heap-size in v8_set_flags.

Several downsides:

  • Applies to all isolates
  • Crashes the process if reached

Possible solution

V8 allows setting the heap size of an isolate via CreateParams: https://chromium.googlesource.com/v8/v8/+/e423f0040333dd3ab410b3c6cab5b71a4f75fa6a/include/v8.h#6409

There is also the NearHeapLimitCallback that allows you to react to memory pressure on a single isolate. It allows increasing the limit or it could terminate the single isolate instead of crashing the whole process.


My goal is to implement this. Just creating the issue for tracking. Work on rusty_v8 is about done: https://github.com/denoland/rusty_v8/pull/427

created time in 16 days

push eventmraerino/rusty_v8

Marcus Weiner

commit sha 28c63791227007ff483d851583102b0e115a6682

Assert on callback count

view details

push time in 16 days

push eventmraerino/rusty_v8

Marcus Weiner

commit sha ee361b982f7aa0366dce6bc02ffdaa8980ee8589

Fix comment position Co-authored-by: Ben Noordhuis <info@bnoordhuis.nl>

view details

push time in 16 days

push eventmraerino/deno

Marcus Weiner

commit sha b340dba23d282964b79632301ad5e19d6d77a244

Document heap limit usage

view details

Marcus Weiner

commit sha 7ea1d7d231df23432572908a3ced5912fd07971e

Expose heap limit callback

view details

push time in 16 days

push eventmraerino/rusty_v8

Marcus Weiner

commit sha fa2f1804cbaa9845874b999d8426b890ab5c183d

Remove debug statement

view details

push time in 16 days

push eventmraerino/rusty_v8

Marcus Weiner

commit sha 1824b1b3fc0dfed427bdb4fd7623777cc8ea6f5f

Refactor test for less flakyness

view details

push time in 16 days

push eventmraerino/rusty_v8

Marcus Weiner

commit sha 91ed338a78bb22f9c1917d7e34d3b508297cbf15

Publish callback type

view details

push time in 16 days

push eventmraerino/rusty_v8

Marcus Weiner

commit sha 844962cf459ddd0c490ec27854522d4d8f4d543e

Improve docs

view details

push time in 16 days

push eventmraerino/rusty_v8

Marcus Weiner

commit sha 0df46ddf887571e0ef1a56d1af1a5458c1479033

Test heap limit and its callback

view details

push time in 16 days

pull request commentdenoland/rusty_v8

Allow managing V8 heap limits

Are you sure you had V8_FROM_SOURCE set when you built it?

somehow missed that concept. using it now, thanks!

mraerino

comment created time in 16 days

Pull request review commentdenoland/rusty_v8

Allow managing V8 heap limits

 impl Isolate {     }   } +  /// Add a callback to invoke in case the heap size is close to the heap limit.+  /// If multiple callbacks are added, only the most recently added callback is+  /// invoked.+  ///+  /// # Safety+  ///+  /// Don't pass a null-pointer for the callback!+  /// Make sure the callback only casts `data` to the type the `data` argument+  /// here.+  pub unsafe fn add_near_heap_limit_callback(

that's the idea

mraerino

comment created time in 16 days

push eventmraerino/rusty_v8

Marcus Weiner

commit sha fbb91ea4d3528107c4da994770014410bf92d7f0

Fix clippy exception

view details

push time in 16 days

push eventmraerino/rusty_v8

Marcus Weiner

commit sha 32568bf274dc01572eae895f9e4e47779b7a8562

Make clippy happy

view details

push time in 16 days

create barnchmraerino/deno

branch : feat/memory-limits

created branch time in 16 days

pull request commentdenoland/rusty_v8

Allow managing V8 heap limits

i'm still getting a linker error when running cargo test -p deno_core in Deno.

I have this change in my core/Cargo.toml:

-rusty_v8 = "0.7.0"
+rusty_v8 = { path = "../../rusty_v8" }

This is how i'm using it: https://github.com/denoland/deno/pull/6914

I did cargo clean in both projects and deno builds fine, but tests fail with the linker problem. Any clues?

mraerino

comment created time in 16 days

PR opened denoland/deno

WIP: Feat/memory limits

<!-- Before submitting a PR, please read https://github.com/denoland/deno/blob/master/docs/contributing.md -->

+58 -10

0 comment

1 changed file

pr created time in 16 days

push eventmraerino/rusty_v8

Marcus Weiner

commit sha 78010d549950e393dd3fe359ed26f73bd855a583

Mark callback setters as safe

view details

Marcus Weiner

commit sha 5aeb0ab891d2af9fbe92a9e633c8275b29b54281

Document heap limit setter

view details

push time in 16 days

Pull request review commentdenoland/rusty_v8

Allow managing V8 heap limits

 impl Isolate {     }   } +  /// Add a callback to invoke in case the heap size is close to the heap limit.+  /// If multiple callbacks are added, only the most recently added callback is+  /// invoked.+  ///+  /// # Safety+  ///+  /// Don't pass a null-pointer for the callback!+  /// Make sure the callback only casts `data` to the type the `data` argument+  /// here.+  pub unsafe fn add_near_heap_limit_callback(

yeah, the caller might have to do some casting but i'm thinking of making a generics-based safe impl in Deno for this

mraerino

comment created time in 16 days

push eventmraerino/rusty_v8

Marcus Weiner

commit sha 48bcffd8373b0761722ffa6a2fd9d6caba419ff3

Allow setting heap limit via CreateParams

view details

push time in 17 days

pull request commentdenoland/rusty_v8

Allow managing V8 heap limits

ok, when trying this out in deno_core i'm getting a linker error...

mraerino

comment created time in 17 days

push eventwdr-data/code4maus

Daniel Diekmeier

commit sha 7fe1f840d5b16f35435c0bc968eaff0952f69cdb

Upload .patternlib from develop

view details

push time in 17 days

push eventmraerino/rusty_v8

Marcus Weiner

commit sha b9c42bae86a564efe335e9c6e360b289fead6dcc

Add missing semicolon

view details

push time in 17 days

Pull request review commentdenoland/rusty_v8

Allow managing V8 heap limits

 bool v8__Isolate__AddMessageListener(v8::Isolate* isolate,   return isolate->AddMessageListener(callback); } +void v8__Isolate__AddNearHeapLimitCallback(v8::Isolate* isolate,+                                           v8::NearHeapLimitCallback callback,+                                           void* data) {+  isolate.AddNearHeapLimitCallback(callback, data);+}++void v8__Isolate__RemoveNearHeapLimitCallback(+    v8::Isolate* isolate, v8::NearHeapLimitCallback callback,+    size_t heap_limit) {+  isolate.RemoveNearHeapLimitCallback(callback, heap_limit);

nope 😄 CI caught it pretty fast

mraerino

comment created time in 17 days

push eventmraerino/rusty_v8

Marcus Weiner

commit sha c6463a7a750d141c2ae62a0f0c98f9479ca8c555

Fix pointer method access

view details

push time in 17 days

PR opened denoland/rusty_v8

Allow managing V8 heap limits

V8 allows setting the max_old_generation_size and max_young_generation_size create parameters through a handy helper: https://chromium.googlesource.com/v8/v8/+/e423f0040333dd3ab410b3c6cab5b71a4f75fa6a/include/v8.h#6409

Also exposes the methods to register a callback when the limit is reached, so a crash of V8 can be avoided.

+93 -0

0 comment

3 changed files

pr created time in 17 days

create barnchmraerino/rusty_v8

branch : feat/heap-limits

created branch time in 17 days

fork mraerino/rusty_v8

V8 javascript bindings for Rust

https://crates.io/crates/rusty_v8

fork in 17 days

push eventwdr-data/code4maus

Daniel Diekmeier

commit sha 0f612e419eaa26f44b39fee9da0c83043965a3f7

Upload .patternlib from develop

view details

push time in 17 days

push eventnetlify/trafficserver

Susan Hinrichs

commit sha c21fd5b7fa2beba47d8999132306266e6eae8514

Remove header set for uninitialized response.

view details

push time in 18 days

startedgithub/glb-director

started time in 21 days

issue commentnetlify/build

Provide Netlify API access to build plugins

Since we have an API which is publicly documented, and provide ways to auth and access it, people will us it in which ever way is most convenient to them and lets the do what they need. So if we limit what they can do in the context of the plugins, then they might revert to doing what I do instead.

The concern here is around third-party plugins which don't clearly tell you if and how they're going to use the token.

It might be a nice idea to expose a full token to local plugins.

erquhart

comment created time in a month

issue commentnetlify/netlify-commons

create a kafka package

I really liked most of the https://github.com/segmentio/kafka-go lib all projects you linked seem to be using it

I think our team will retract that statement. We've been running into nasty issues with the lib: https://github.com/netlify/edge-state/issues/257 https://github.com/netlify/edge-state/issues/243

@smoya is evaluating https://github.com/Shopify/sarama

rybit

comment created time in a month

push eventnetlify/trafficserver

Leif Hedstrom

commit sha 51ae70e1c125063765f2a9465fe748650aa6f141

Validates return values from TSHttpTxnClientAddrGet() Cherry-picked from f9d962b19742146c21c306c9cd804aa54a83301b

view details

push time in a month

startedpaviliondev/discourse-events

started time in a month

delete branch kalkspace/redirects

delete branch : feat/discuss-redirects

delete time in a month

push eventkalkspace/redirects

Marcus Weiner

commit sha 62753fcfb7ebc82d499d9c3de6627f6ffb93d5f1

Update _redirects (#1)

view details

push time in a month

PR merged kalkspace/redirects

Reviewers
Redirect aliases for the forum
+3 -0

0 comment

1 changed file

mraerino

pr closed time in a month

PR opened kalkspace/redirects

Reviewers
Redirect aliases for the forum
+3 -0

0 comment

1 changed file

pr created time in a month

create barnchkalkspace/redirects

branch : feat/discuss-redirects

created branch time in a month

push eventmraerino/netlify-functions-example

Marcus Weiner

commit sha 4bc7cd641d7eb0232b3b7c58e59fac87a5ae343f

Update hello.js

view details

push time in a month

create barnchmraerino/netlify-functions-example

branch : master

created branch time in a month

created repositorymraerino/netlify-functions-example

created time in a month

delete branch kalkspace/website

delete branch : cms/blog/2020-06-30-kalkspace-goes-bigbluebutton

delete time in a month

push eventkalkspace/website

buecherwurmmm

commit sha eb3c2539455f8dfa2216607143bd072f108add83

Update Blog “2020-06-30-kalkspace-goes-bigbluebutton” (#42) * Update Blog “2020-06-30-kalkspace-goes-bigbluebutton” * Update Blog “2020-06-30-kalkspace-goes-bigbluebutton” * Limit size of images in blog posts * Fix overflow of intro heading * Support special characters in blog titles * Auto-format * Break long word in blog post title Co-authored-by: Marcus Weiner <marcus.weiner@gmail.com>

view details

push time in a month

PR merged kalkspace/website

Update Blog “2020-06-30-kalkspace-goes-bigbluebutton” netlify-cms/pending_publish

Automatically generated by Netlify CMS

+25 -9

1 comment

7 changed files

buecherwurmmm

pr closed time in a month

push eventkalkspace/website

Marcus Weiner

commit sha e81487c54bd13142fd7426a24c190c938b6c574d

Limit size of images in blog posts

view details

Marcus Weiner

commit sha 230c5e212b4b1cdb302b1ab10b038ba6e0c2144d

Fix overflow of intro heading

view details

Marcus Weiner

commit sha 2f945df1f2cfd7ea1d8e62e76076e81250220c60

Support special characters in blog titles

view details

Marcus Weiner

commit sha 5dc7c88cfc6ea42e91e197ad87fe1549d406629e

Auto-format

view details

Marcus Weiner

commit sha 1ee868c320d381956d5ee760df3d18a6f8b5097a

Break long word in blog post title

view details

push time in a month

issue commentcapnproto/capnproto-rust

Consider using #![allow(clippy)] at top of generated files

yup, that seems to have done it. thank you! are you going to release this soon?

j-t-d

comment created time in a month

issue commentcapnproto/capnproto-rust

Consider using #![allow(clippy)] at top of generated files

i'm still getting a bunch of warnings, mostly clippy::redundant-field-names

is the plan still to fix individual instances of this instead of ignoring lints in the generated files?

j-t-d

comment created time in a month

startedredsift/redbpf

started time in a month

issue commentsagiegurari/cargo-make

Support workspace makefile extension when running directly from member directory

seems like because of this, the README might be wrong?

When running cargo make for modules which are part of a workspace, you can automatically have the member crates makefile (even if doesn't exist) extend the workspace level makefile.

sagiegurari

comment created time in a month

issue commentnetlify/netlify-commons

create a kafka package

which underlying library should we use?

I really liked most of the https://github.com/segmentio/kafka-go lib all projects you linked seem to be using it

a thing to look out for is that their NewWriter can panic on invalid input: https://github.com/netlify/edge-state/blob/master/kafkaclient/kafka.go#L297-L304

is there a recommended way we should consume/ack messages?

most likely separate calls to Fetch and Commit after it has been applied internally edge-state uses an abstraction that also handles retries and only commits if no error occured: https://github.com/netlify/edge-state/blob/master/kafkaclient/readloop.go


I think a good constructor and a thin wrapper around the library should work well for the start. The segmentio Reader & Writer Implementations are very high-level and easy to use.

rybit

comment created time in a month

issue commentnetlify/gotrue-js

accept pre-hashed passwords

i have not seen this implemented a lot in other services. there are several pitfalls here.

for one, the backend should be able to upgrade the hasing method to provide more security without the frontend being affected by that (breaking change otherwise)

also, the reason we store passwords hashed in a DB is so that a DB leak will not allow the attacker to use the credentials. if we allowed logging in using hashes, leaked DBs could be used for login. I don't think it's considered good practice to do hashing in the browser for login. therefore the proposed method is not necessarily more secure.

The customer stated several reason for wanting this feature:

  • protect against corp firewalls intercepting HTTPS -> that's not something we can defeat against. a lot relies on HTTPS working. users in such a corp setting will have to be aware and mitigate another way
  • leaked logs -> we do audit our code to make sure this doesn't happen. our server logs never contain full payloads
fool

comment created time in a month

pull request commentkalkspace/website

Update Blog “2020-06-30-kalkspace-goes-bigbluebutton”

hier fehlt noch eine CSS änderung, damit das eingefügte bild ordentlich angezeigt wird.

buecherwurmmm

comment created time in a month

push eventkalkspace/website

Marcus Weiner

commit sha 041abf9507621fad5f092d609b0b77bdf633d40d

Enable editorial workflow for the CMS (#39)

view details

push time in a month

PR opened kalkspace/website

Enable editorial workflow for the CMS
+4 -0

0 comment

1 changed file

pr created time in a month

create barnchkalkspace/website

branch : cms-editorial-workflow

created branch time in a month

issue commentrust-analyzer/rust-analyzer

Stack overflow with cyclic dependency error

The dependency is not actually cyclic btw. The compiler is compiling the project just fine.

This can happen when a crate has a dev-depencency on itself. When testing Cargo just builds the crate twice. Once without #[cfg(test)] and one with.

Ok. is there a fix for it or can it just be ignored?

mraerino

comment created time in 2 months

issue commentrust-analyzer/rust-analyzer

Stack overflow with cyclic dependency error

Does the crash happen if you run rust-analyzer analysis-stats . in the project?

Nope.

Output:

[ERROR ra_project_model] cyclic dependency traffic_mesh -> test-helpers
Database loaded, 392 roots, 922.5233ms
Crates in this dir: 16
Total modules found: 96
Total declarations: 851
Total functions: 596
Item Collection: 23.670316s, 0b allocated 0b resident
Total expressions: 40616                                                                        
Expressions of unknown type: 1606 (3%)
Expressions of partially unknown type: 1864 (4%)
Type mismatches: 67
Inference: 32.6445041s, 0b allocated 0b resident
Total: 56.3148737s, 0b allocated 0b resident
mraerino

comment created time in 2 months

issue openedrust-analyzer/rust-analyzer

Stack overflow with cyclic dependency error

I'm now getting a crash on a project where rust-analyzer worked perfectly before.

VSCode output log:

[ERROR ra_project_model] cyclic dependency test-helpers -> traffic_mesh

thread '<unknown>' has overflowed its stack
fatal runtime error: stack overflow
[Info  - 9:56:00 AM] Connection to server got closed. Server will restart.
[Error - 9:56:00 AM] Request textDocument/semanticTokens/range failed.
Error: Connection got disposed.

The dependency is not actually cyclic btw. The compiler is compiling the project just fine.

This is the graph:

test-helpers -- dependency --> traffic_mesh
traffic_mesh -- test-dependency --> test-helpers

I'm running on Linux in Docker (DevContainers).

created time in 2 months

issue commentnetlify/gocommerce

How can I use it ?

Hi! Sadly, we only have what's in the Readme. This project is not really supported by Netlify as a first-tier open-source project.

I'm happy to review a pull request that will improve the Readme. It does not have to be perfect, you could write down how you would run a Go service and we can work out how we'd like to document it.

aKamrani

comment created time in 2 months

more