MongoDB driver for Elixir
kobil-systems/mongodb_ecto 321
MongoDB adapter for Ecto
Database connection behaviour
A process-based debouncer for Elixir
Advent of code for 2020
proof of concept MySQL driver
The GraphQL toolkit for Elixir
pull request commentelixir-ecto/postgrex
:green_heart: :blue_heart: :purple_heart: :yellow_heart: :heart:
comment created time in 10 minutes
push eventelixir-ecto/postgrex
commit sha fc129a9597e79ab7cb413c1d58615b85e81eb076
fix formatting (#553)
push time in 10 minutes
pull request commenterlang/otp
Yes, it should work for "all" kinds of sockets. The function socket:monitor/1 is official, but is intended for when you only work with 'socket' sockets.
comment created time in 2 hours
issue openederlang/otp
OTP 23 beam can‘t auto gc on a CentOS 7 server
Describe the bug OTP 23 beam can‘t auto gc On a CentOS 7 server，and call 'erlang:garbage_collect' effectively frees the process's memory; and It works on mac
To Reproduce Steps to reproduce the behavior.
Expected behavior after the client establishes a connection to the server, the process takes 40K bytes and does not release it
Affected versions otp 23
Additional context gc.md
created time in 2 hours
pull request commenterlang/otp
@bmk as I did not know it existed, that is why you should review the code! Do I understand it correctly that it will work for both new and old sockets and that I then can skip the is_port guard in the original code?
comment created time in 2 hours
Pull request review commentelixir-ecto/ecto
defmodule Ecto.EnumTest do %{ on_dump: %{bar: "baar", baz: "baaz", foo: "fooo"}, on_load: %{"baar" => :bar, "baaz" => :baz, "fooo" => :foo},- values: [:foo, :bar, :baz],+ mappings: [foo: "fooo", bar: "baar", baz: "baaz"],
I think we can detect and raise on duplicate keys. I can see how using list and thus preserving order can be useful for reflection purposes but no strong opinion either way.
comment created time in 2 hours
issue commenterlang/otp
ssl psk ciphers options break between ssl 9.2 and 10.1
However I believe there is a bug in handling the cipher suites when supporting both TLS-1.3 and TLS-1.2 with a list of ciphers suites including ciphers from both and that contains TLS-1.2 PSK cipher suites . And I believe #4983 will fix it
comment created time in 3 hours
issue commenterlang/otp
ssl psk ciphers options break between ssl 9.2 and 10.1
The option is {versions, ['tlsv1.2']}
comment created time in 3 hours
issue commenterlang/otp
ssl psk ciphers options break between ssl 9.2 and 10.1
I used the latest version available as a docker container. I just freshly pulled to make sure and got
{ssl_app,"10.4.1"},
I also added protocol_versions
{ protocol_versions, 'tlsv1.2' }
wireshark shows:
Version: TLS 1.2 (0x0303)
No change in the Client Hello otherwise. Is there some other configuration part that's needed?
comment created time in 3 hours
pull request commenterlang/otp
Why don't you use inet:monitor/1?
comment created time in 3 hours
Pull request review commentelixir-ecto/ecto
defmodule Ecto.EnumTest do %{ on_dump: %{bar: "baar", baz: "baaz", foo: "fooo"}, on_load: %{"baar" => :bar, "baaz" => :baz, "fooo" => :foo},- values: [:foo, :bar, :baz],+ mappings: [foo: "fooo", bar: "baar", baz: "baaz"],
Would map be a more reasonable data structure here? Because keyword would let you do something like this: [foo: "foo", foo: "fooo"] and the first one would win.
comment created time in 4 hours
push eventWhatsApp/erlfmt
commit sha 64d3a60447633ed8c183b4670eb5ddf15fb818ed
Return exit code 2 when option isn't recognized. (#304) Follow standard practice, so that client code can detect if there is an issue. We already returned error code 2 in other cases of option parsing failure, now everything is covered in a consistent manner.
push time in 4 hours
PR merged WhatsApp/erlfmt
Follow standard practice, so that client code can detect if there is an issue. We already returned error code 2 in other cases of option parsing failure, now everything is covered in a consistent manner.
pr closed time in 4 hours
PR opened WhatsApp/erlfmt
Follow standard practice, so that client code can detect if there is an issue. We already returned error code 2 in other cases of option parsing failure, now everything is covered in a consistent manner.
pr created time in 4 hours
pull request commenterlang/otp
Improve the readability of .asm-dumps
By the way, I can test on Arm when you have added that code.
comment created time in 4 hours
Pull request review commenterlang/otp
Improve the readability of .asm-dumps
beamfile_read(const byte *data, size_t size, BeamFile *beam) { MakeIffId('C', 'I', 'n', 'f'), /* 8 */ MakeIffId('L', 'i', 'n', 'e'), /* 9 */ MakeIffId('A', 't', 'U', '8'), /* 10 */+ MakeIffId('L', 'o', 'c', 'T'), /* 11 */
Nitpick: Nowadays we intend with space, not tabs.
comment created time in 4 hours
push eventerlang/otp
commit sha 71c04d5d532206710a8ada591049adc321e4c1b4
Merge branch 'rickard/monotonic-hrtime/OTP-17493' into rickard/monotonic-hrtime/22/OTP-17493 * rickard/monotonic-hrtime/OTP-17493: Fix monotonic hrtime configure test
commit sha 65c4efcdbfb8c524a6a5f4de1af2483213067e17
Update configure scripts
commit sha 7be61008abc3c320eb1f371efdd600e8967e6fe8
Merge branch 'rickard/monotonic-hrtime/22/OTP-17493' into maint * rickard/monotonic-hrtime/22/OTP-17493: Update configure scripts
push time in 4 hours
push eventerlang/otp
commit sha 71c04d5d532206710a8ada591049adc321e4c1b4
Merge branch 'rickard/monotonic-hrtime/OTP-17493' into rickard/monotonic-hrtime/22/OTP-17493 * rickard/monotonic-hrtime/OTP-17493: Fix monotonic hrtime configure test
commit sha 65c4efcdbfb8c524a6a5f4de1af2483213067e17
Update configure scripts
commit sha 7be61008abc3c320eb1f371efdd600e8967e6fe8
Merge branch 'rickard/monotonic-hrtime/22/OTP-17493' into maint * rickard/monotonic-hrtime/22/OTP-17493: Update configure scripts
commit sha b7eaf55ccb645ccb82b2cb691acff694b0dd9280
Merge branch 'maint' * maint: Update configure scripts
push time in 4 hours
push eventerlang/otp
commit sha eb128f92f3fa0b84a97da696ecf7dfc5bb8b3756
Merge branch 'rickard/monotonic-hrtime/OTP-17493' into rickard/monotonic-hrtime/23/OTP-17493 * rickard/monotonic-hrtime/OTP-17493: Fix monotonic hrtime configure test
commit sha 28b86fe3b10cc8f9c7110106d6823d69aef10c23
Update configure scripts
commit sha 835412656e0666d109004c640961d8a2390b46ee
Merge branch 'rickard/monotonic-hrtime/23/OTP-17493' into maint * rickard/monotonic-hrtime/23/OTP-17493: Update configure scripts
commit sha 84517abf70697f3ee94bfaeef8d8b5f73910f733
Merge branch 'maint' * maint: Update configure scripts
push time in 4 hours
push eventerlang/otp
commit sha eb128f92f3fa0b84a97da696ecf7dfc5bb8b3756
Merge branch 'rickard/monotonic-hrtime/OTP-17493' into rickard/monotonic-hrtime/23/OTP-17493 * rickard/monotonic-hrtime/OTP-17493: Fix monotonic hrtime configure test
commit sha 28b86fe3b10cc8f9c7110106d6823d69aef10c23
Update configure scripts
commit sha 835412656e0666d109004c640961d8a2390b46ee
Merge branch 'rickard/monotonic-hrtime/23/OTP-17493' into maint * rickard/monotonic-hrtime/23/OTP-17493: Update configure scripts
push time in 4 hours
push eventerlang/otp
commit sha e8bff1bf55e822ea76a2b993a7cf3abeea7180ab
Merge branch 'rickard/monotonic-hrtime/OTP-17493' into rickard/monotonic-hrtime/24/OTP-17493 * rickard/monotonic-hrtime/OTP-17493: Fix monotonic hrtime configure test
commit sha 3d94326dcd3d1279df408498b1c1629600245c93
Update configure scripts
commit sha fb92bda4652aac0a58bb968fded02777674fec67
Merge branch 'rickard/monotonic-hrtime/24/OTP-17493' into maint * rickard/monotonic-hrtime/24/OTP-17493: Update configure scripts Fix monotonic hrtime configure test
commit sha 7929a979ad7407c594126b31563d607eebaf95df
Merge branch 'maint' * maint: Update configure scripts
push time in 4 hours
push eventerlang/otp
commit sha c179130c691c7c61784ee0d9c3643100da4e466a
Fix monotonic hrtime configure test
commit sha e8bff1bf55e822ea76a2b993a7cf3abeea7180ab
Merge branch 'rickard/monotonic-hrtime/OTP-17493' into rickard/monotonic-hrtime/24/OTP-17493 * rickard/monotonic-hrtime/OTP-17493: Fix monotonic hrtime configure test
commit sha 3d94326dcd3d1279df408498b1c1629600245c93
Update configure scripts
commit sha fb92bda4652aac0a58bb968fded02777674fec67
Merge branch 'rickard/monotonic-hrtime/24/OTP-17493' into maint * rickard/monotonic-hrtime/24/OTP-17493: Update configure scripts Fix monotonic hrtime configure test
push time in 4 hours
pull request commentelixir-ecto/postgrex
add information about client certificate authentication
:green_heart: :blue_heart: :purple_heart: :yellow_heart: :heart:
comment created time in 4 hours
push eventelixir-ecto/postgrex
commit sha d1074cfe8b72e6d8ed69075246c7adbedea23105
Add information about client certificate authentication (#552)
push time in 4 hours
PR merged elixir-ecto/postgrex
Clears up the confusion which lead to #551
pr closed time in 4 hours
issue closedelixir-ecto/ecto_sql
Suggestion: mix task for squashing migrations
Hi, on squashing migrations again :D
Current approach with mix ecto.dump + mix ecto.load doesn't fit well with Elixir releases: there's no mix binary in release to run the load task.
I think more convenient approach would be to squash a bunch of migrations into a sql file with a single migration to load it using existing load_structure/2. Newly created migration would have the ID of the latest migration squashed. That way release could run migrations as usual without conflicts either on new or existing deploys (only if it had that latest squashed migration applied though).
I could take a shot at making a PR with this kind of mix task. Do you like the idea?
closed time in 4 hours
staring-frog
issue commentelixir-ecto/ecto_sql
Suggestion: mix task for squashing migrations
To be clear, the goal of mix ecto.dump+mix ecto.load was never to squash migrations, but more as a performance mechanism so you are not running migrations all the time. Especially between dev/tests.
In any case, it shouldn't matter for releases: you should never really run ecto.dump and ecto.load in production. And if you are squashing, you definitely want to squash migrations you already ran in production. Otherwise you can end-up squashing a mix of prod migrations with non-prod migrations, and that will cause failure.
So my suggestion for now is to play with squashing migrations as a separate package, to try the different approaches out, and maybe we can revisit if it is part of EctoSQL in the future. :) Thanks!
comment created time in 4 hours