- I upgraded the actor system (Control.Concurrent.Actor) to support
typed per-actor methods with type-level names, parameter lists and
return values
- For extensible sum types, I decided to use the 'vary' package, while
extensible product still uses 'HList' (other options are to get both
from 'extensible' (Sum, Prod) or from 'vinyl' (Rec, CoRed))
- Since 'vary' is new, it uses GHC2021 and thus requires GHC 9
- So I decided to make the leap into upgrading to an LTS with GHC 9
- Vervis was at LTS 18, at first I tried 19 and 20 but there were
dependency version problems, in particular I couldn't solve them for
the 'darcs' package even when I switched to latest LTS which is 22
- So I decided it's time to do the long-waiting task of switching to
calling-darcs-process instead of using darcs as a library
All of this together creates a huge commit, but 3 big things are now
solved :)
- No factories inserted to DB yet
- Admin user list specified in settings
- Admins can create a Factory via C2S
- Creating other actors via C2S the old way still works as well
- Factory S2S handler implementation still blank
The major changes:
- No more hit and hit-* packages
- No more diagram and SVG font stuff
- Dependency on darcs is not yet removed
- No more persistent-graph i.e. recursive SQL queries
- Some hit-network stuff still in use, now part of Vervis itself
- Git operations use the git command-line program, a convenient API is provided
in Data.Git.Local
- No more patch parsing and no diff view, patches are treated as a piece of
text and displayed as-is (although I could add syntax-highlighting as a nice
touch for the HTML view)
Vervis currently supports only direct grants. The new process supports
delegation chains as well. This commit just implements the new process
as a new function, without yet using it anywhere. The next commits will
plug it into Deck actor handlers.
Limitations:
- Verification of proofs not implemeneted yet, just generation.
Verification probably coming in the next commit.
- Only jcs-eddsa-2022 is supported. Can add more cryptosuites once
they're updated for the requirements of the VC Data Integrity spec.
- Bug: The proofs aren't stored in the DB versions of outgoing activities, i.e.
HTTP GETing an activity won't include the proof. Probably not urgent
to fix. Ideally, change the whole PersistJSONObject/Envelope/etc.
thing to allow to serialize the activity exactly once.
This patch makes Vervis temporarily unusable, because all actors' inbox POST
handlers use the new system, but the actual federation handler code hasn't been
ported. The next patches will port all the S2S activities supported so far, as
well as C2S.
New iteration of the ActivityPub delivery implementation and interface.
Advantages over previous interface:
* When sending a ByteString body, the sender is explicitly passed as a
parameter instead of JSON-parsing it out of the ByteString
* Clear 3 operations provided: Send, Resend and Forward
* Support for per-actor keys
* Actor-type-specific functions (e.g. deliverRemoteDB_D) removed
* Only the most high-level API is exposed to Activity handler code, making
handler code more concise and clear
Also added in this patch:
* Foundation for per-actor key support
* 1 key per actor allowed in DB
* Disabled C2S and S2S handlers now un-exported for clarity
* Audience and capability parsing automatically done for all C2S handlers
* Audience and activity composition automatically done for Vervis.Client
builder functions
Caveats:
* Actor documents still don't link to their per-actor keys; that should be the
last piece to complete per-actor key support
* No moderation and anti-spam tools yet
* Delivery API doesn't yet have good integration of persistence layer, e.g.
activity is separately encoded into bytestring for DB and for HTTP; this will
be improved in the next iteration
* Periodic delivery now done in 3 separate steps, running sequentially; it
simplifies the code, but may be changed for efficiency/robustness in the next
iterations
* Periodic delivery collects per-actor keys in a
1-DB-transaction-for-each-delivery fashion, rather than grabbing them in the
big Esqueleto query (or keeping the signed output in the DB; this isn't done
currently to allow for smooth actor key renewal)
* No support yet in the API for delivery where the actor key has already been
fetched, rather than doing a DB transaction to grab it; such support would be
just an optimization, so it's low-priority, but will be added in later
iterations
Both Git and Darcs are supported
- Darcs implementation applies right on the bare repo, I haven't tested to make
sure it works right (federated MR demo is going to be only for Git)
- Git implementation clones to temporary repo, runs `git am` on it to apply,
then pushes to the real bare repo (because `git am` doesn't work on bare
repos; I haven't tested yet to see how it handles conflicts; cloning and
pushing should be efficient since the refs are just hardlinked rather than
copied)
For now it's implemented only for Git:
If tracker is a local loom, and a (local or remote) origin repo is specified,
but no patches are provided, then generate them ourselves!
* Clone the (local) target repo
* Add the (local or remote) origin repo as a git remote
* Make sure target branch is an ancestor of the origin branch
* Generate patches for the commits that origin adds on top of target
* Insert them into our DB
What's missing:
- Match patch VCS, origin repo VCS and target repo VCS
- Hook into postPersonInboxR
- If only origin is provided, generate patches / otherwise somehow remember the
commits proposed
Person inbox handler:
- Invite: Parse and insert to inbox
- Grant: Parse and insert to inbox
Repo/Deck/Loom inbox handler:
- Invite: Parse and remember as Collab record in DB for later
- Accept: Send a Grant (and remember it in DB)
Along with inviteC and acceptC, the Invite->Accept->Grant flow is now fully
federated, yay!
What's missing is UI for actually using it. Coming soon.
Giving access now starts with an Invite activity, followed by Accept from the
Invite's recipient. Finally, the resource sends a Grant, which is the actual
OCap.
Only Vervis.Hook (for the hook programs) and Vervis.Application (for the web
app program itself) are exposed now. I suppose more will be needed for tests.
But now there's a clear distinction what's used by other components and what
isn't.
The steps are:
- Parse activity ID and match with the authenticated sender
- For local activity (we got via forwarding), find in DB
- For remote activity, cache in DB
- Insert activity to recipient's inbox
What's not there yet is the actual logic of handling specific activities.
This is such a huge patch, it's probably impossible to tell what it does by
looking at the code. One thing is clear: It changes *everything* :P so here's
an overview:
- There are now 5 types of actors, each having its own top-level route
- So projects, repos, etc. are no longer "under" sharers
- Actor routes are now based on their KeyHashid, there are no "idents" anymore,
i.e. URLs look random and don't contain user or repo names
- No sharers anymore; people and groups are distinct entities not sharing a
common namespace or anything like that
- Project has been renamed to Deck and it simply means a ticket tracker; repos
are no longer "under" projects
- In addition to Person, Group, Repo and Deck, there's a new actor type Loom,
which is a patch tracker; i.e. Repo actors don't manage MRs anymore
- All C2S and S2S is temporarily disabled, because huge changes to the whole
code are required and I'll do them gradually in the next patches
- Since form-based actions are implemented using C2S, they're disabled as well,
so Vervis is now essentially read-only
- Some views have been temporarily removed, e.g. repo history and commit view
- A huge set of DB migrations has been added to adapt the DB to these changes;
I haven't tested them yet on a read DB so there may be errors there; I'll fix
them in the next patches if I find any (probably going to test on the main
instance where Vervis itself is hosted...)
- Some modules got tech upgrades, e.g. LocalActor became a higher-kinded type
and a similar pattern is probably relevant for several other types
- There's an 'Actor' entity in the DB schema now, and all 5 actor types use it
for common things like inbox and outbox
- Although inbox and outbox are used only by Actor, so essentially could be
removed, I haven't removed them; that's because I wonder if at some point
users can have a tree of inboxes much like in email; I don't have an excuse
for Outbox, but anyway, leaving them as is for now
- Workflows, roles and collaborators are partially removed/unused until I
figure out a sane federated way to provide these features
- Since repo routes don't contain a "sharer" anymore, SSH URIs are now simpler,
they already look like user@host/repo regardless of who "controls" that repo
