This PR makes the `github_user_login` field required in the
`LlmTokenClaims`.
We previously added this in
https://github.com/zed-industries/zed/pull/16316 and made it optional
for backwards-compatibility.
It's been more than long enough for all of the previous LLM tokens to
have expired, so we can now make the field required.
Release Notes:
- N/A
This PR reworks our existing billing code in preparation for charging
based on LLM usage.
We aren't yet exercising the new billing-related code outside of
development.
There are some noteworthy changes for our existing LLM usage tracking:
- A new `monthly_usages` table has been added for tracking usage
per-user, per-model, per-month
- The per-month usage measures have been removed, in favor of the
`monthly_usages` table
- All of the per-month metrics in the Clickhouse rows have been changed
from a rolling 30-day window to a calendar month
Release Notes:
- N/A
---------
Co-authored-by: Antonio Scandurra <me@as-cii.com>
Co-authored-by: Richard <richard@zed.dev>
Co-authored-by: Max <max@zed.dev>
This PR drops the `cache_creation_input_tokens_this_month ` and
`cache_read_input_tokens_this_month ` columns from the `usages` table in
the LLM database.
We mistakenly added these in #18834, but these aren't necessary due to
the structure of the `usages` table. We weren't actually using these
columns anywhere.
Release Notes:
- N/A
This PR extends the LLM usage tracking to support tracking usage for
cache writes and reads for Anthropic models.
Release Notes:
- N/A
---------
Co-authored-by: Antonio Scandurra <me@as-cii.com>
Co-authored-by: Antonio <antonio@zed.dev>
This PR reverts the changes to the Clickhouse event rows that were
included in https://github.com/zed-industries/zed/pull/18414.
The changes don't seem to be correct, as they make the row structs
differ from the underlying table schema.
Release Notes:
- N/A
REplace isahc with ureq everywhere gpui is used.
This should allow us to make http requests without libssl; and avoid a
long-tail of panics caused by ishac.
Release Notes:
- (potentially breaking change) updated our http client
---------
Co-authored-by: Mikayla <mikayla@zed.dev>
This PR does not change how things work for settings, but lays the
ground work for the future functionality.
After this change, Zed is prepared to sync more than just
`settings.json` files from local worktree and user config.
* ssh tasks
Part of this work is to streamline the task sync mechanism.
Instead of having an extra set of requests to fetch the task contents
from the server (as remote-via-collab does now and does not cover all
sync cases), we want to reuse the existing mechanism for synchronizing
user and local settings.
* editorconfig
Part of the task is to sync .editorconfig file changes to everyone which
involves sending and storing those configs.
Both ssh (and remove-over-collab) .zed/tasks.json and .editorconfig
files behave similar to .zed/settings.json local files: they belong to a
certain path in a certain worktree; may update over time, changing Zed's
functionality; can be merged hierarchically.
Settings sync follows the same "config file changed -> send to watchers
-> parse and merge locally and on watchers" path that's needed for both
new kinds of files, ergo the messaging layer is extended to send more
types of settings for future watch & parse and merge impls to follow.
Release Notes:
- N/A
I noticed a few places where we were storing `&'static str`s in
`static`s instead of `const`s.
This PR updates them to use `const`.
Release Notes:
- N/A
This restructures the code in `remote` so that it's easier to replace
the current SSH connection with a new one in case of
disconnects/reconnects.
Right now, it successfully reconnects, BUT we're still missing the big
piece on the server-side: keeping the server process alive and
reconnecting to the same process that keeps the project-state.
Release Notes:
- N/A
---------
Co-authored-by: Bennet <bennet@zed.dev>
If specifying a formatter in the settings like this:
"languages": {
"foo": {
"formatter": {
"external": {
"command": "/path/to/foo-formatter"
}
}
}
}
Zed will show an error like this:
Invalid user settings file
data did not match any variant of untagged enum SingleOrVec
This is because the arguments are not optional. The error is hard to
understand, so let's make the arguments actually optional, which makes
the above settings snippet valid.
Release Notes:
- Make external formatter arguments optional
Todo:
* [x] Tooltips for hunk buttons
* [x] Buttons to go to next and previous hunk
* [x] Ellipsis button that opens a context menu with `Revert all`
/cc @iamnbutler @danilo-leal for design 👀
Release Notes:
- Changed the behavior of the git gutter so that diff hunk are expanded
immediately when clicking the gutter, and hunk controls are displayed
above the hunk.
---------
Co-authored-by: Marshall <marshall@zed.dev>
Co-authored-by: Marshall Bowers <elliott.codes@gmail.com>
This PR updates the collab seed script to seed the GitHub users from a
set of static data.
This removes the need to hit the GitHub API to retrieve these users.
Release Notes:
- N/A
This PR fixes the GitHub user retrieval in the database seed script.
The users returned from the [list
users](https://docs.github.com/en/rest/users/users?apiVersion=2022-11-28#list-users)
endpoint don't have a `created_at` timestamp, so we need to fetch them
individually.
I want to rework this further at a later date, this is just a bandaid to
get things working again.
Release Notes:
- N/A
Release Notes:
- (Potentially breaking change) Zed will now use the node installed on
your $PATH (if it is more recent than v18) instead of downloading its
own. You can disable the new behavior with `{"node":
{"disable_path_lookup": true}}` in your settings. We do not yet use
system/project-local node_modules.
---------
Co-authored-by: Mikayla <mikayla@zed.dev>
This pushes the new LanguageServerName type to more places.
As both languages and language servers were identified by Arc<str>, it
was
sometimes hard to tell which was intended.
Release Notes:
- N/A
This PR removes the `replica_id` field from the `MultiBuffer` struct.
We were only ever referencing this field to pass when constructing a
`MultiBuffer`, and never used it outside of that.
Release Notes:
- N/A
This PR makes the `Buffer::apply_ops` method infallible for
`text::Buffer` and `language::Buffer`.
We discovered that `text::Buffer::apply_ops` was only fallible due to
`apply_undo`, which didn't actually need to be fallible.
Release Notes:
- N/A
This PR adds `system_id` to telemetry, which is contained within a new
`global` database (accessible by any release channel of Zed on a single
system). This will help us get a more accurate understanding of user
count, instead of relying on `installationd_id`, which is different per
release channel. This doesn't solve the problem of a user with multiple
machines, but it gets us closer.
Release Notes:
- N/A
This PR adds the ability to display backgrounds for inlay hints within
the editor.
This is controlled by the new `inlay_hints.show_background` setting.
This setting defaults to `false`.
To enable the setting, add the following to your `settings.json`:
```json
{
"inlay_hints": {
"enabled": true,
"show_background": true
}
}
```
When enabled, the inlay hint backgrounds will use the `hint.background`
color from the theme.
| Disabled | Enabled |
|
--------------------------------------------------------------------------------------------------------------------------------------------------
|
--------------------------------------------------------------------------------------------------------------------------------------------------
|
| <img width="1624" alt="Screenshot 2024-09-17 at 4 21 53 PM"
src="https://github.com/user-attachments/assets/5534d09b-1e22-4c6f-9d82-314796ed7d22">
| <img width="1624" alt="Screenshot 2024-09-17 at 4 21 43 PM"
src="https://github.com/user-attachments/assets/6ec58cde-6115-4db4-be95-97c5f2f54b2d">
|
Related issues:
- #12485
- #17392
Release Notes:
- Added an `inlay_hints.show_background` setting to allow displaying
backgrounds for inlay hints in the editor.
- This setting defaults to `false`.
- If enabled, the inlay hint backgrounds will use the `hint.background`
color from the theme.
Fixes https://github.com/zed-industries/zed/issues/5291
Release Notes:
- Fixed a bug where the 'toggle comments' command didn't use the right
comment syntax in JSX and TSX elements.
---------
Co-authored-by: Conrad <conrad@zed.dev>
Co-authored-by: Conrad Irwin <conrad.irwin@gmail.com>
Co-authored-by: Kirill Bulatov <mail4score@gmail.com>
This PR marks the RunPod environment variables as optional in the
Kubernetes template so that we can deploy without them being present.
Collab is already accounting for their absence.
Release Notes:
- N/A
Add `/auto` behind a feature flag that's disabled for now, even for
staff.
We've decided on a different design for context inference, but there are
parts of /auto that will be useful for that, so we want them in the code
base even if they're unused for now.
Release Notes:
- N/A
---------
Co-authored-by: Antonio Scandurra <me@as-cii.com>
Co-authored-by: Marshall Bowers <elliott.codes@gmail.com>
This PR adds a `GET /models` endpoint to the LLM service.
This endpoint returns the models that the authenticated user has access
to.
This is the first step towards populating the models for the hosted
service from the server.
Release Notes:
- N/A
This is a refactor to prepare for adding LSP support in SSH remote
projects.
Release Notes:
- N/A
---------
Co-authored-by: Mikayla <mikayla@zed.dev>
Co-authored-by: Conrad <conrad@zed.dev>
For ssh remoting lsps we'll need to have language server support
factored out of project.
Thus that begins
Release Notes:
- N/A
---------
Co-authored-by: Max Brunsfeld <maxbrunsfeld@gmail.com>
Co-authored-by: Mikayla <mikayla@zed.dev>
This PR makes the `github_user_created_at` field required at ingress
points into collab.
In practice we already have this value passed up, this change just makes
that explicit.
This is a precursor to making it required in the database.
Release Notes:
- N/A
This changes the Zed CLI `zed` to pass along the environment to the Zed
project that it opens (if it opens a new one).
In projects, this CLI environment will now take precedence over any
environment that's acquired by running a login shell in a projects
folder.
The result is that `zed my/folder` now always behaves as if one would
run `zed --foreground` without any previous Zed version running.
Closes#7894Closes#16293
Related issues:
- It fixes the issue described in here:
https://github.com/zed-industries/zed/issues/4977#issuecomment-2305272027
Release Notes:
- Improved the Zed CLI `zed` to pass along the environment as it was on
the CLI to the opened Zed project. That environment is then used when
opening new terminals, spawning tasks, or language servers.
Specifically:
- If Zed was started via `zed my-folder`, a terminal spawned with
`workspace: new terminal` will inherit these environment variables that
existed on the CLI
- Specific language servers that allow looking up the language server
binary in the environments `$PATH` (such as `gopls`, `zls`,
`rust-analyzer` if configured, ...) will look up the language server
binary in the CLI environment too and use that environment when starting
the process.
- Language servers that are _not_ found in the CLI environment (or
configured to not be found in there), will be spawned with the CLI
environment in case that's set. That means users can do something like
`RA_LOG=info zed .` and it will be picked up the rust-analyzer that was
spawned.
Demo/explanation:
https://github.com/user-attachments/assets/455905cc-8b7c-4fc4-b98a-7e027d97cdfa
This PR attaches the `geoip_country_code` that we source from
Cloudflare's `CF-IPCountry` header to the HTTP request spans.
This will allow us to see where traffic is originating geographically.
Release Notes:
- N/A
- Cloudflare provides ISO-3166-1 country code for protectorates. Expand our allowlist to include the territories of countries on the allowlist (US, UK, France, Australia, New Zealand).
- Also include the country_code in the error message when we block.
Co-authored-by: Marshall Bowers <elliott.codes@gmail.com>
This PR fixes an issue where active user counts were being computed
across _all_ measures instead of the per-minute measures.
We now compute them using the tokens per minute measure, as we're
concerned with usage in recent minutes.
Release Notes:
- N/A
Release Notes:
- Fixed the issue related to the project wide search being stuck when
project contains .fifo files
- Might potentially solve the following issue
https://github.com/zed-industries/zed/issues/7360
This PR adds a lifetime spending limit on LLM usage.
Exceeding this limit will prevent further use of the Zed LLM provider.
Currently the cap is $1,000.
Release Notes:
- N/A
This PR makes it so users are upserted by their `github_user_id` instead
of by their `github_login`.
The `github_user_id` is a stable identifier that does not change, while
the `github_login` can change.
In practice we were already using
`get_or_create_user_by_github_account`, which already checks for an
existing user with a `github_user_id` first, so this change doesn't
result in a change in behavior.
This change is primarily for correctness in the event that `create_user`
is called directly, as we want to be upserting by the stable identifier.
Release Notes:
- N/A
This PR makes the `github_user_id` column on the `users` table required
and replaces the index with a unique index.
I have gone through and ensured that all users have a unique
`github_user_id` in the staging and production databases.
Release Notes:
- N/A
This PR adds additional reporting of the active user counts as separate
logs.
We were already reporting these on individual rate limit events/logs,
but it seems like something that would be good to report on independent
of user activity.
Release Notes:
- N/A
This PR fixes an issue where the active user count spanned individual
models.
We now track the active user counts on a per-model basis.
Release Notes:
- N/A
This PR adds traces for when users hit LLM rate limits.
We were already emitting telemetry events for these to Clickhouse, but
it will be handy to have them available in Axiom as well.
Release Notes:
- N/A
This PR adds two Postgrest containers—one for the app database and one
for the LLM database—to the Docker Compose cluster.
Also fixed an issue where `postgres_app.conf` and `postgres_llm.conf`
had been switched.
Release Notes:
- N/A
This PR adds the `is_staff` field to the `upstream rate limit` spans.
Since we use different API keys for staff vs non-staff, it will be
useful to break down the rate limits accordingly.
Release Notes:
- N/A
This PR fixes the writing of LLM rate limit events to Clickhouse.
We had a table in the table name: `llm_rate_limits` instead of
`llm_rate_limit_events`.
I also extracted a helper function to write to Clickhouse so we can use
it anywhere we need to.
Release Notes:
- N/A
This PR reworks how we do checks for model names in the LLM service.
We now normalize the model names using the models defined in the
database.
Release Notes:
- N/A
This PR updates the LLM service to include the GitHub login on its
spans.
We need to pass this information through on the LLM token, so it will
temporarily be `None` until this change is deployed and new tokens have
been issued.
Release Notes:
- N/A
- db deadlock in GetLlmToken for non-staff users
- typo in allowed model name for non-staff users
Release Notes:
- N/A
---------
Co-authored-by: Marshall <marshall@zed.dev>
Co-authored-by: Joseph <joseph@zed.dev>
This PR adds the ability to revoke access tokens for the LLM service.
There is a new `revoked_access_tokens` table that contains the
identifiers (`jti`) of revoked access tokens.
To revoke an access token, insert a record into this table:
```sql
insert into revoked_access_tokens (jti) values ('1e887b9e-37f5-49e8-8feb-3274e5a86b67');
```
We now attach the `jti` as `authn.jti` to the tracing spans so that we
can associate an access token with a given request to the LLM service.
Release Notes:
- N/A
This PR adds feature-flagged access to the LLM service.
We've repurposed the `language-models` feature flag to be used for
providing access to Claude 3.5 Sonnet through the Zed provider.
The remaining RPC endpoints that were previously behind the
`language-models` feature flag are now behind a staff check.
We also put some Zed Pro related messaging behind a feature flag.
Release Notes:
- N/A
---------
Co-authored-by: Max <max@zed.dev>
This PR restricts usage of the LLM service to accounts older than 30
days.
We now store the GitHub user's `created_at` timestamp to check the
GitHub account age. If this is not set—which it won't be for existing
users—then we use the `created_at` timestamp in the Zed database.
Release Notes:
- N/A
---------
Co-authored-by: Max <max@zed.dev>
Now, when an anthropic request is invalid or anthropic's API is down,
we'll expose that to the user instead of just returning a generic 500.
Release Notes:
- N/A
Co-authored-by: Marshall <marshall@zed.dev>
This PR makes it so Zed staff can use a separate Anthropic API key for
the LLM service.
We also added an `is_staff` column to the `usages` table so that we can
exclude staff usage from the "active users" metrics that influence the
rate limits.
Release Notes:
- N/A
---------
Co-authored-by: Max <max@zed.dev>
This PR adds a check to the LLM API token issuance to ensure that we
only issue tokens to users that have accepted the terms of service.
Release Notes:
- N/A
This PR makes it so hitting upstream rate limits from Anthropic result
in an HTTP 429 response instead of an HTTP 500.
To do this we need to surface structured errors out of the `anthropic`
crate.
Release Notes:
- N/A
This adds the requirement for users to accept the terms of service the
first time they send a message with the Cloud provider.
Once this is out and in a nightly, we need to add the check to the
server side too, to authenticate access to the models.
Demo:
https://github.com/user-attachments/assets/0edebf74-8120-4fa2-b801-bb76f04e8a17
Release Notes:
- N/A
This PR makes it so staff members will be exempt from rate limiting by
the LLM service.
This is just a temporary measure until we can tweak the rate-limiting
heuristics.
Staff members are still subject to upstream LLM provider rate limits.
Release Notes:
- N/A
When Anthropic releases a new version of their models, Zed AI users
should always get access to the new version even when using an old
version of zed.
Co-Authored-By: Thorsten <thorsten@zed.dev>
Release Notes:
- N/A
Co-authored-by: Thorsten <thorsten@zed.dev>
This PR adjusts how we display the "mode" collab is running in on the
root endpoint.
It's minor, but it does make things a bit cleaner.
Release Notes:
- N/A
This prevents users from accessing other models, such as OpenAI's GPT-4
or Google's Gemini-Pro.
Staff members can still access all models.
Co-authored-by: Thorsten <thorsten@zed.dev>
Release Notes:
- N/A
---------
Co-authored-by: Thorsten <thorsten@zed.dev>
This will help us as we hit issues with the /workflow and step
resolution. We can override the baked-in prompts and make tweaks, then
import our refinements back into the source tree when we're ready.
Release Notes:
- N/A
This PR removes the unused `ignore_checksum_mismatch` parameter to
`run_database_migrations`.
We were always passing `false`, which meant the behavior didn't need to
be parameterized.
Release Notes:
- N/A
This PR puts the initial infrastructure for the LLM service's database
in place.
The LLM service will be using a separate Postgres database, with its own
set of migrations.
Currently we only connect to the database in development, as we don't
yet have the database setup for the staging/production environments.
Release Notes:
- N/A
This PR updates the LLM service to authorize access to language model
providers based on the requester's country.
We detect the country using Cloudflare's
[`CF-IPCountry`](https://developers.cloudflare.com/fundamentals/reference/http-request-headers/#cf-ipcountry)
header.
The country code is then checked against the list of supported countries
for the given LLM provider. Countries that are not supported will
receive an `HTTP 451: Unavailable For Legal Reasons` response.
Release Notes:
- N/A
This PR introduces a separate backend service for making LLM calls.
It exposes an HTTP interface that can be called by Zed clients. To call
these endpoints, the client must provide a `Bearer` token. These tokens
are issued/refreshed by the collab service over RPC.
We're adding this in a backwards-compatible way. Right now the access
tokens can only be minted for Zed staff, and calling this separate LLM
service is behind the `llm-service` feature flag (which is not
automatically enabled for Zed staff).
Release Notes:
- N/A
---------
Co-authored-by: Marshall <marshall@zed.dev>
Co-authored-by: Marshall Bowers <elliott.codes@gmail.com>
This PR makes it so any Stripe events we receive that occurred over an
hour ago are marked as processed.
We don't want to process an old event long after it occurred and
potentially overwrite more recent updates.
This also makes running collab locally a bit nicer, as we won't be
getting errors for a bunch of older events that will never get processed
successfully.
The period after time after which we consider an event "stale" can be
modified, as needed.
Release Notes:
- N/A
This is just a refactor that we're landing ahead of any functional
changes to make sure we haven't broken anything.
Release Notes:
- N/A
Co-authored-by: Marshall <marshall@zed.dev>
Co-authored-by: Jason <jason@zed.dev>
This PR changes how we report the `geoip_country_code` in the tracing
spans.
I wasn't seeing it come through in the logs, and I think it was because
we didn't declare the field on the initial span.
Release Notes:
- N/A
This PR fixes the `.env.toml` paths, since we inadvertently broke them
in https://github.com/zed-industries/zed/pull/15557.
There's likely a better way we can do this, but I wanted to restore the
previous behavior, for now.
Release Notes:
- N/A
This PR renames the links to the `zed.dev/settings` page to the
`zed.dev/account`.
Some of these spots will likely link out to a marketing page later.
Release Notes:
- N/A
This PR removes the primary/secondary distinction for
`CachedLspAdapter`s.
After #15624 we weren't relying on the `is_primary` field anywhere, so
we can remove it.
Release Notes:
- N/A
This PR increases the frequency at which we poll for Stripe events.
This decreases the amount of time we have to wait in order for changes
in Stripe to be reflected in our system.
We now poll for events every 5 seconds, down from every 5 minutes.
In order to avoid needlessly over-fetching data from Stripe, we put a
cap on the number of pages consisting entirely of already-processed
events that we can see before stopping. This is set to 4, so once we've
seen 4 pages of processed events (400 events in total), we'll stop
fetching subsequent pages.
Release Notes:
- N/A
English is hard.
In US English the forms of "cancel" are as follows:
- `cancel`
- `cancels`
- `canceling`
- `canceled`
- `cancellation`
Note that `cancellation` _does_ use the double `l`, which all the rest
of the forms do not.
Release Notes:
- N/A
This PR updates collab to pick the `zed.dev` URL based on the current
environment rather than always using `https://zed.dev`.
This means that when developing locally we'll use
`http://localhost:3000` to be taken to the locally-running version of
`zed.dev`.
Release Notes:
- N/A
This UI change is behind a `ZedPro` feature flag so that it won't be
visible until we're ready to launch that service.
Release Notes:
- N/A
---------
Co-authored-by: Marshall Bowers <elliott.codes@gmail.com>
Co-authored-by: Marshall <marshall@zed.dev>
This PR updates the rate limits to adapt based on the user's current
plan.
For the free plan rate limits I just took one-tenth of the existing rate
limits (which are now the Pro limits). We can adjust, as needed.
Release Notes:
- N/A
---------
Co-authored-by: Max <max@zed.dev>
This PR adds a new `GET /billing/subscriptions` endpoint to collab for
retrieving the subscriptions to display on the account settings page.
Release Notes:
- N/A
This PR updates the user menu to show the user's current plan.
Also adds a new RPC message to send this information down to the client
when Zed starts.
This is behind a feature flag.
Release Notes:
- N/A
---------
Co-authored-by: Max <max@zed.dev>
This PR reworks how we process Stripe events for reconciliation
purposes.
The previous approach in #15480 turns out to not be workable, on account
of the Stripe event IDs not being strictly in order. This meant that we
couldn't reliably compare two arbitrary event IDs and determine which
one was more recent.
This new approach leans on the guidance that Stripe provides for
webhooks events:
> Webhook endpoints might occasionally receive the same event more than
once. You can guard against duplicated event receipts by logging the
[event IDs](https://docs.stripe.com/api/events/object#event_object-id)
you’ve processed, and then not processing already-logged events.
>
> https://docs.stripe.com/webhooks#handle-duplicate-events
We now record processed Stripe events in the `processed_stripe_events`
table and use this to filter out events that have already been
processed, so we do not process them again.
When retrieving events from the Stripe events API we now buffer the
unprocessed events so that we can sort them by their `created` timestamp
and process them in (roughly) the order they occurred.
Release Notes:
- N/A
We also eliminate the `completion` crate and moved its logic into
`LanguageModelRegistry`.
Release Notes:
- N/A
---------
Co-authored-by: Nathan <nathan@zed.dev>
This PR improves our Stripe event handling by keeping track of the last
event we've seen for each record.
The `billing_customers` and `billing_subscriptions` tables both have a
new `last_stripe_event_id` column. When we apply an event to one of
these records, we store the event ID that was applied.
Then, when we are going through events we can ignore any event that has
an ID that came before the `last_stripe_event_id` (based on the
lexicographical ordering of the IDs).
Release Notes:
- N/A
