Fixes #37102 - webpack 5 #9834

MariaAga · 2023-09-14T15:43:54Z

Details and discussion here: https://community.theforeman.org/t/resurrection-of-the-client-side-infrastructure-upgrade-effort/32340

bundler.d/assets.rb

package.json

ehelms · 2023-09-18T19:06:13Z

package.json

-    "node-sass": "^4.5.0",
-    "optimize-css-assets-webpack-plugin": "^3.2.0",
+    "mini-css-extract-plugin": "2.7.5",
+    "node-sass": "6.0.0",


Is this an update we can do independent of the webpack change?

no - sass-loader only supports node-sass v6 from sass-loader v11.1.0
and sass-loader v11.0.0 only supports wepack 5

ehelms · 2023-09-18T19:06:21Z

package.json

    "cross-env": "^5.2.0",
-    "css-loader": "^0.23.1",
+    "css-loader": "4.3.0",


Is this an update we can do independent of the webpack change?

css-loader 4 requires webpack 4, so we can't do it separetly

ehelms · 2023-09-18T19:07:11Z

I've tried to pick out a few spots where my goal is to identify any changes we can make independently so that we are able to do the small update round trip with packaging and then reduce the scope of changes required at one time for this PR.

developer_docs/getting-started.asciidoc

app/assets/javascripts/late_load.js

ekohl · 2023-09-19T09:23:54Z

app/helpers/reactjs_helper.rb

-    global_css_tags(global_plugins_list).join.html_safe
+  def js_tags_for(requested_plugins)
+    requested_plugins.map do |plugin|
+      javascript_tag("window.tfm.tools.loadPluginModule('/webpack/#{plugin.to_s.tr('-', '_')}','#{plugin.to_s.tr('-', '_')}','./index');".html_safe)


In global_js_tags you use plugin[:id] and here just plugin. That should be consistent.

ekohl · 2023-09-19T09:25:36Z

app/helpers/reactjs_helper.rb

-  def webpacked_plugins_with_global_js
-    global_js_tags(global_plugins_list).join.html_safe
+  def other_webpack_plugin(plugin_name, file)
+    javascript_tag("window.tfm.tools.loadPluginModule('/webpack/#{plugin_name.to_s.tr('-', '_')}','#{plugin_name.to_s.tr('-', '_')}','./#{file}_index');".html_safe)


Given the repetition of this conversion I think we should have a common way for it. I'm not exactly sure where that should live, but the Foreman plugin code feels like a natural place. Perhaps we already have such a thing.

ekohl · 2023-09-19T09:27:19Z

app/helpers/reactjs_helper.rb

-      end
+  def get_webpack_foreman_vendor_js
+    if ENV['RAILS_ENV'] == 'production'
+      javascript_include_tag("/webpack/#{File.basename(Dir.glob('public/webpack/foreman-vendor*production*js')[0])}")


Please avoid globbing for these methods. If there's some left over file it may end up using the wrong file. It also creates additional IO for every request at runtime. Previously we had some JSON file which had the mapping. Isn't that something that can still be done?

Previously we had to create the JSON manifest files to make rails helpers to work correctly (it's a hack built upon the webpack-rails gem). Now that we don't use the gem and don't reference the files from Ruby code, I don't see a good reason to keep that special configuration in webpack to generate the manifest file.
This concept inspired heavily b jsbundling-rails gem, which is the default option for rails7. The idea behind it is to reduce knowledge of the JS building process and serve the files as pure static files.

In our specific case, there always should be only one vendor file in the folder, but we don't know the hash ID for it. Would it be better to assert single file here, and prevent the app from loading in case more than one file exists?

Globbing can be unpredictable. sprockets also has a random component in its manifest filename and they at least have a warning if multiple files are found:

https://github.com/rails/sprockets/blob/5b040f3bec503ded8a98c0c911c7a77dd1f5bf1b/lib/sprockets/manifest_utils.rb#L40-L43

And you still haven't addressed my concern that this isn't stored in the application program, but loaded every time. I don't insist on a manifest, but your current approach causes additional IO and can make upgrades less predictable (because there can be a point where the new files are deployed, but the application not yet restarted).

I'm not sure what the best place is, but an application initializer would be my first thought.

And you still haven't addressed my concern that this isn't stored in the application program, but loaded every time. I don't insist on a manifest, but your current approach causes additional IO and can make upgrades less predictable (because there can be a point where the new files are deployed, but the application not yet restarted).

What about having the filename cached once?
Something along the lines of:

def self.manifest_file logger.warn("too many foreman-vendor files") if Dir.glob('public/webpack/foreman-vendor*production*js').length > 1 @manifest_file ||= "/webpack/#{File.basename(Dir.glob('public/webpack/foreman-vendor*production*js')[0])}" end def get_webpack_foreman_vendor_js javascript_include_tag(self.manifest_file) end

This would make the upgrade predictable (until the next restart) and avoid extra IO.

I would be worried about multiple workers. When Puma forks of a new worker, would it determine the manifest file again? In production mode I would expect it to be part of an initializer so it happens before forking.

And a small nit that I would use:

def self.manifest_file @manifest_file ||= begin files = Dir.glob('public/webpack/foreman-vendor*production*js') logger.warn("too many foreman-vendor files") if files.length > 1 "/webpack/#{File.basename(files.first)}" end end

Because your logger statement would still glob on every request.

I was thinking of doing:

def read_webpack_manifest root = File.expand_path(File.dirname(__FILE__) + "/../..") file = File.read(root + '/public/webpack/manifest.json') JSON.parse(file) end def get_webpack_foreman_vendor_js data = read_webpack_manifest foreman_vendor_js = data['assetsByChunkName']['foreman-vendor'].find { |value| value.end_with?('.js') } javascript_include_tag("/webpack/#{foreman_vendor_js}") end

@MariaAga The whole idea is to get rid of the manifest, but if we decide to leave the manifest we shall use it of course.

app/helpers/reactjs_helper.rb

ekohl · 2023-09-19T09:29:41Z

app/helpers/reactjs_helper.rb

+    if ENV['RAILS_ENV'] == 'production'
+      javascript_include_tag("/webpack/#{File.basename(Dir.glob('public/webpack/foreman-vendor*production*js')[0])}")
+    else
+      javascript_include_tag("/webpack/#{File.basename(Dir.glob('public/webpack/foreman-vendor*development*js')[0])}")


This implies in testing we also use the development version. Is that correct?

The filename depends on NODE_ENV should I change RAILS_ENV to NODE_ENV?
can NODE_ENV be something that is not production or development?

Not sure about NODE_ENV. Like with #9834 (comment), can't we write some JSON file at a predictable location with the filename? I suspect that determining this at runtime is likely to cause problems at some point.

As far as tests go we don't need webpack file to be loaded anyway (we don't do integration tests using RAILS_ENV=test).

As I have mentioned before: since we need a single file there anyway, would it be OK to assert and prevent loading the app in case more than one file exists? Having the manifest file requires knowledge transfer between the webpack and rails parts of the code, and I try to simplify both webpack configuration and rails loading process.

In a long term, I think this part also should be part of the foreman federated module Maria already mentioned in the thread. This should remove the need to know the exact filename entirely.

Shouldn't it be:

Suggested change

javascript_include_tag("/webpack/#{File.basename(Dir.glob('public/webpack/foreman-vendor*development*js')[0])}")

javascript_include_tag("/webpack/#{File.basename(Dir.glob('public/webpack/foreman-vendor*#{ENV.fetch('NODE_ENV', 'production')}*js')[0])}")

ehelms · 2023-10-11T20:30:47Z

Should I be able to npm install at this point? I am seeing the following:

$ npm install
npm ERR! code ERESOLVE
npm ERR! ERESOLVE unable to resolve dependency tree
npm ERR! 
npm ERR! While resolving: [email protected]
npm ERR! Found: [email protected]
npm ERR! node_modules/webpack
npm ERR!   dev webpack@"5.75.0" from the root project
npm ERR! 
npm ERR! Could not resolve dependency:
npm ERR! peer webpack@"^4.0.0" from [email protected]
npm ERR! node_modules/optimize-css-assets-webpack-plugin
npm ERR!   dev optimize-css-assets-webpack-plugin@"^4.0.0" from the root project
npm ERR! 
npm ERR! Fix the upstream dependency conflict, or retry
npm ERR! this command with --force, or --legacy-peer-deps
npm ERR! to accept an incorrect (and potentially broken) dependency resolution.

adamruzicka · 2023-10-12T12:20:12Z

@ehelms Seems to work for me. Is your katello patched as well? Did you wipe node_modules before npm install? What version of node do you have?

ehelms · 2023-10-12T13:17:32Z

@ehelms Seems to work for me. Is your katello patched as well? Did you wipe node_modules before npm install? What version of node do you have?

No Katello involved in my test. I tried this with 18 and 16 as that is what I had local. Perhaps I made a wrong assumption that this would already enable doing it with newer Nodes. I can re-test with 12.

adamruzicka · 2023-10-12T16:22:48Z

I have node 14, but I can't tell if that has any impact

MariaAga · 2023-10-12T16:25:33Z

Node 14 is recommended and tested. Making node 16+ available will be possible after this PR but requires some changes that should be done it a new PR.
Also I recommend deleting package-lock.json and node_modules when testing this PR.

ehelms · 2023-10-12T18:28:27Z

Node 14 is recommended and tested. Making node 16+ available will be possible after this PR but requires some changes that should be done it a new PR.

Looks like this also currently requires Node 14?

npm ERR! code ENOTSUP
npm ERR! notsup Unsupported engine for [email protected]: wanted: {"node":">= 14.15.0"} (current: {"node":"12.20.1","npm":"6.14.10"})
npm ERR! notsup Not compatible with your version of node/npm: [email protected]
npm ERR! notsup Not compatible with your version of node/npm: [email protected]
npm ERR! notsup Required: {"node":">= 14.15.0"}
npm ERR! notsup Actual:   {"npm":"6.14.10","node":"12.20.1"}

If yes, and there is no way to support Node 12, we will have to consider that transition in our build environment. It would be nice to be able to only do one change at a time. We need to understand if that is possible or impossible.

ShimShtein · 2023-10-17T07:10:12Z

npm ERR! notsup Unsupported engine for [email protected]: wanted: {"node":">= 14.15.0"} (current: {"node":"12.20.1","npm":"6.14.10"})

@MariaAga , can we use [email protected] ? I see they dropped the support for node 12 in >=4.0.0: webpack-contrib/css-minimizer-webpack-plugin@88a5674

@ehelms , can you try locally if you can work with lower version of the plugin?

ehelms · 2023-10-17T18:18:51Z

npm ERR! notsup Unsupported engine for [email protected]: wanted: {"node":">= 14.15.0"} (current: {"node":"12.20.1","npm":"6.14.10"})

@MariaAga , can we use [email protected] ? I see they dropped the support for node 12 in >=4.0.0: webpack-contrib/css-minimizer-webpack-plugin@88a5674

@ehelms , can you try locally if you can work with lower version of the plugin?

Looks like it's a no go, but due to webpack itself:

pm ERR! notsup Unsupported engine for [email protected]: wanted: {"node":">=14.15.0"} (current: {"node":"12.20.1","npm":"6.14.10"})
npm ERR! notsup Not compatible with your version of node/npm: [email protected]
npm ERR! notsup Not compatible with your version of node/npm: [email protected]
npm ERR! notsup Required: {"node":">=14.15.0"}
npm ERR! notsup Actual:   {"npm":"6.14.10","node":"12.20.1"}

ehelms · 2023-10-18T00:59:30Z

Based on the nodejs 14 requirement, I was curious to, from a build perspective, look at if our packages would build against 14 and then if Foreman would. I haven't tested the results, but everything built (and thanks to Copr) I have a repo with only those nodejs and Foreman package in it:

https://copr.fedorainfracloud.org/coprs/ehelms/foreman-nightly-staging/

ehelms · 2023-10-18T18:41:52Z

Based on the nodejs 14 requirement, I was curious to, from a build perspective, look at if our packages would build against 14 and then if Foreman would. I haven't tested the results, but everything built (and thanks to Copr) I have a repo with only those nodejs and Foreman package in it:

https://copr.fedorainfracloud.org/coprs/ehelms/foreman-nightly-staging/

I took the Foreman build here for a test drive and poked around at the UI -- I found no issues. This leads me to believe we can do an update to Node 14 in our build environment with our current code base prior to this change and reduce the friction.

Looking back at history, [1] was the reason we had ended up in a chicken-egg problem with prior attempts because Node 14 in the RPM environment came with an older NPM. However, I believe that we can update to Node 14 as is cleanly and that [1] is not as relevant to get to Webpack 5 (and can be a nice to have later on).

@MariaAga @ShimShtein can you confirm?

[1] theforeman/foreman-js#404

ehelms · 2023-10-20T03:21:34Z

theforeman/tool_belt#569 -- this will swap build environment to Node 14.

ehelms · 2023-10-26T16:24:08Z

Can you rebase this? I've done a round of updates to Node 14.

app/helpers/layout_helper.rb

MariaAga · 2023-11-02T12:42:29Z

Thanks everyone for your input, Updated some loading order logic so Jquery should work as expected now. The PR still needs some cleaning up + info comments + test fixes

Ron-Lavi

Nice work @MariaAga !! thanks for working hard on getting this in.

Something that we should pay attention to regarding dev environment compilation,
I made a change in a string and after 8 seconds the webpack compiled,
later on regular refresh the string didn't change, only after hard refresh it did.

after that I changed the string again and even when webpack got compiled successfully the page with the string remained the same.

is there any caching mechanism we are using?

Thanks

app/assets/javascripts/application.js

Ron-Lavi · 2023-12-26T19:14:47Z

webpack/assets/javascripts/react_app/common/AwaitedMount.js

+import componentRegistry from '../components/componentRegistry';
+
+// Mounts a component after all plugins have been imported to make sure that all plugins are available to the component
+export const AwaitedMount = ({ component, data, flattenData }) => {


there's a warning on every page about Component name already taken:
from the stack trace it's originated from this file.

Was this compiled with foreman-js 12.2.2? If so, that's the real culprit, not this line.
12.2.3 should fix everything.

should we drop the version pin to v12.2.0 now? https://github.com/theforeman/foreman/pull/9834/files#diff-7ae45ad102eab3b6d7e7896acd08c427a9b25b346470d7bc6507b6481575d519R24 ?

going to try with v12.2.3

12.2.3 should be fine. only .1 and .2 are broken.

I'll drop the pin, good catch

yeah looks like this is fixed in the latest commit here and using v12.2.3

apologies, there was an issue that plugins weren't compiling well using bundle exec rake webpack:compile
it worked with npx webpack --config config/webpack.config.js

and now the issue is visible again

Ron-Lavi · 2023-12-26T19:32:19Z

app/views/layouts/base.html.erb

@@ -33,18 +30,20 @@
    </script>
    <%= javascript_include_tag "locale/#{FastGettext.locale}/app" %>
    <%= locale_js_tags %>
-
    <%= yield(:head) %>


just before the end of the tags we are getting now many <style> tags which I think comes from this yield:

as of now, I don't see an impact on existing pages.

Ron-Lavi

@MariaAga Can you check if clicking on the kebab toggle in the host details page fails for you as well?

I got some error regarding onKebabToggle being undefined, I wonder if there is some issue with the React Context provider.

verified and it works on the develop branch.

MariaAga · 2024-01-02T13:50:21Z

Also saw this issue now, that I will fix

ekohl · 2024-01-02T14:04:08Z

I played used this while working on theforeman/foreman_monitoring#94 and as a developer there it appeared to work as I expected.

Ron-Lavi · 2024-01-03T13:06:31Z

@MariaAga Can you check if clicking on the kebab toggle in the host details page fails for you as well?

I got some error regarding onKebabToggle being undefined, I wonder if there is some issue with the React Context provider.

verified and it works on the develop branch.

Update: Works for me on the latest commit with @theforeman/vendor v12.2.3

MariaAga · 2024-01-03T13:20:57Z

I got some error regarding onKebabToggle being undefined, I wonder if there is some issue with the React Context provider.
This is still happening with bootdisk plugin I think. I have a solution for it, will push when I'll resolve the rest of the issues (400 style tags, and deprecation text in the UI)

Ron-Lavi · 2024-01-03T19:35:48Z

I got some error regarding onKebabToggle being undefined, I wonder if there is some issue with the React Context provider. This is still happening with bootdisk plugin I think. I have a solution for it, will push when I'll resolve the rest of the issues (400 style tags, and deprecation text in the UI)

apologies, there was an issue that plugins weren't compiling well using bundle exec rake webpack:compile
it worked with npx webpack --config config/webpack.config.js

and now the issue with clicking the host details kebab is happening again, my plugins are:

MariaAga · 2024-01-10T09:54:11Z

bundle exec rake webpack:compile is defaulting to NODE_ENV=production I'll remove/edit the suggestion to use it.
Instead of adding context files (like Templates/CardItem/CardTemplate/index.js and HostDetails/ActionsBar/index.js) to ModuleFederationPlugin shared field, I have wrapped the contexts with forceSingleton which resolved the issues in host details page.
Changed style-loader to use singletonStyleTag - https://webpack.js.org/loaders/style-loader/#singletonstyletag
Added an id=plugin so we can track which plugin css rules come from as this option disables source maps.

Procfile

ekohl · 2024-01-11T15:05:15Z

app/helpers/layout_helper.rb

+  def javascript_include_tag(*params, **kwargs)
+    # Workaround for overriding javascript load with webpack_asset_paths, should be removed when webpack_asset_paths is removed
+    if kwargs[:source] == "webpack_asset_paths"
+      kwargs[:webpacked]


There's no super call here, so it does nothing. Is that intentional, or did you mean to unconditionally call super(...)?

we dont want to call super cause it returns a script tag,
and the webpacked key contains for example: webpacked=>"<script>\n//<![CDATA[\nwindow.tfm.tools.loadPluginModule('/webpack/foreman_remote_execution','foreman_remote_execution','./index');\n//]]>\n</script>"
which we declare in: webpacked: webpacked_plugins_js_for(plugin_name.to_sym)

app/helpers/reactjs_helper.rb

package.json

ekohl · 2024-01-19T17:39:27Z

@MariaAga and all reviewers: I'm proposing we set a date to merge this and deal with any fallout after. We should announce this date on Discourse and other places so people are aware of it. I'm proposing Monday 29 or Tuesday 30. That gives us a full week to iterate on it and prepare, while also giving us a week before FOSDEM + cfgmgmtcamp. Looking at the 3.10 schedule we'd have about 2 weeks until stabilization week to deal with any fallout, or roll back if it's really too bad. How does that sound?

evgeni · 2024-01-19T17:48:00Z

.packit.yaml

@@ -18,7 +18,7 @@ downstream_package_name: foreman

 actions:
  post-upstream-clone:
-    - "wget https://raw.githubusercontent.com/theforeman/foreman-packaging/rpm/develop/packages/foreman/foreman/foreman.spec -O foreman.spec"


we should not forget to drop the [HACK] use evgeni's temporary repo with webpack5 builds commit before merging (we can also do so now, the only downside would be failing packit builds, but that's IMHO fine)

evgeni · 2024-01-19T17:51:18Z

I'm proposing Monday 29 or Tuesday 30

Can we aim for Friday the 26th instead? That would allow us to come in fresh on Monday and see what the pipelines and everyone has to say about the result.

ekohl · 2024-01-19T19:04:01Z

I'm not a fan of merging something potentially breaking on Friday, but I can see arguments for letting pipelines run. I can't speak for everyone, but my Monday is also pretty full with meetings which why I also said Tuesday. Friday would be OK with me. If people get stuck with things on the weekend, they can submit PRs and we can merge them while waiting on pipelines during meetings.

config/webpack.config.js

ekohl

Let's break some pipelines and development environments by merging this :)

As a reminder, https://community.theforeman.org/t/webpack-5-merge-this-friday-2024-01-26/36581 recommends removing your node_modules directory entirely and run npm install fresh.

theforeman-bot added Waiting on contributor Not yet reviewed labels Sep 14, 2023

github-actions bot added UI Legacy JS PRs making changes in the legacy Javascript stack Docs labels Sep 14, 2023

ehelms reviewed Sep 18, 2023

View reviewed changes

bundler.d/assets.rb Outdated Show resolved Hide resolved

ehelms reviewed Sep 18, 2023

View reviewed changes

package.json Outdated Show resolved Hide resolved

ehelms reviewed Sep 18, 2023

View reviewed changes

MariaAga force-pushed the webpack5-module-federations branch from 01eca52 to f19a2a4 Compare September 19, 2023 09:11

theforeman-bot added Waiting on contributor and removed Waiting on contributor labels Sep 19, 2023

ekohl reviewed Sep 19, 2023

View reviewed changes

ianballou mentioned this pull request Sep 25, 2023

webpack5 Katello/katello#10735

Merged

ekohl reviewed Oct 31, 2023

View reviewed changes

app/helpers/layout_helper.rb Outdated Show resolved Hide resolved

MariaAga force-pushed the webpack5-module-federations branch from f19a2a4 to ce1dda6 Compare November 2, 2023 12:41

Ron-Lavi reviewed Dec 26, 2023

View reviewed changes

evgeni force-pushed the webpack5-module-federations branch from cba583b to a7c624a Compare January 3, 2024 12:28

MariaAga force-pushed the webpack5-module-federations branch from a7c624a to 0e88bf5 Compare January 10, 2024 09:54

ekohl reviewed Jan 11, 2024

View reviewed changes

Procfile Outdated Show resolved Hide resolved

ekohl reviewed Jan 11, 2024

View reviewed changes

MariaAga commented Jan 18, 2024

View reviewed changes

app/helpers/reactjs_helper.rb Show resolved Hide resolved

MariaAga force-pushed the webpack5-module-federations branch from 2698590 to e8f9435 Compare January 19, 2024 15:05

evgeni reviewed Jan 19, 2024

View reviewed changes

package.json Outdated Show resolved Hide resolved

MariaAga force-pushed the webpack5-module-federations branch from e8f9435 to e06460f Compare January 19, 2024 15:12

evgeni reviewed Jan 19, 2024

View reviewed changes

MariaAga commented Jan 25, 2024

View reviewed changes

config/webpack.config.js Show resolved Hide resolved

MariaAga force-pushed the webpack5-module-federations branch from e06460f to ecf7586 Compare January 25, 2024 12:15

Fixes #37102 - webpack 5

fa778e0

MariaAga force-pushed the webpack5-module-federations branch from 9333d6c to fa778e0 Compare January 25, 2024 13:21

pr-processor bot removed the Waiting on contributor label Jan 25, 2024

ekohl changed the title ~~webpack 5~~ Fixes #37102 - webpack 5 Jan 26, 2024

ekohl approved these changes Jan 26, 2024

View reviewed changes

pr-processor bot removed the Not yet reviewed label Jan 26, 2024

ekohl merged commit 89c6104 into theforeman:develop Jan 26, 2024
22 of 23 checks passed

ekohl mentioned this pull request Jan 30, 2024

Refs #37102 - detect plugins webpack by looking for remoteEntry.js #10025

Merged

	javascript_include_tag("/webpack/#{File.basename(Dir.glob('public/webpack/foreman-vendordevelopmentjs')[0])}")
	javascript_include_tag("/webpack/#{File.basename(Dir.glob('public/webpack/foreman-vendor#{ENV.fetch('NODE_ENV', 'production')}js')[0])}")

Fixes #37102 - webpack 5 #9834

Fixes #37102 - webpack 5 #9834

Conversation

MariaAga commented Sep 14, 2023 • edited Loading

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

ehelms commented Sep 18, 2023

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

ehelms commented Oct 11, 2023

adamruzicka commented Oct 12, 2023

ehelms commented Oct 12, 2023

adamruzicka commented Oct 12, 2023

MariaAga commented Oct 12, 2023 • edited Loading

ehelms commented Oct 12, 2023

ShimShtein commented Oct 17, 2023

ehelms commented Oct 17, 2023

ehelms commented Oct 18, 2023

ehelms commented Oct 18, 2023

ehelms commented Oct 20, 2023

ehelms commented Oct 26, 2023

MariaAga commented Nov 2, 2023 • edited Loading

Ron-Lavi left a comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Ron-Lavi Jan 3, 2024 • edited Loading

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Ron-Lavi left a comment

Choose a reason for hiding this comment

MariaAga commented Jan 2, 2024

ekohl commented Jan 2, 2024

Ron-Lavi commented Jan 3, 2024

MariaAga commented Jan 3, 2024

Ron-Lavi commented Jan 3, 2024

MariaAga commented Jan 10, 2024

Choose a reason for hiding this comment

Choose a reason for hiding this comment

ekohl commented Jan 19, 2024

Choose a reason for hiding this comment

evgeni commented Jan 19, 2024

ekohl commented Jan 19, 2024

ekohl left a comment

Choose a reason for hiding this comment

MariaAga commented Sep 14, 2023 •

edited

Loading

MariaAga commented Oct 12, 2023 •

edited

Loading

MariaAga commented Nov 2, 2023 •

edited

Loading

Ron-Lavi Jan 3, 2024 •

edited

Loading