Our mark definitions for sole actions and effects pre-date our tab
completion work. This commit adds tab completion actions and
effects to the definition, so they can be sent and received as JSON.
This removes the %http-response special case from gall. In its place,
we implement a subscription regime with the following steps:
- Agent sends %connect to Eyre
- Eyre pokes agent with %handle-http-response, including unique eyre-id
- Agent passes %start-watching to Eyre with eyre-id and unique app-id
- Eyre subscribes to agent on /http-response/app-id
- Agent produces a %http-response-header fact followed by 0 or more
%http-response-data facts and possibly a %http-response-cancel fact
- Agent produces a %kick to close the subscription, which Eyre
interprets as completion of the message.
This works when there is data. There is currently a bug where if the
response has no data in total (as in the case of a naked 404), no
response will be sent.
This also includes lib/http-handler, which implements a convenient
interface for agents that want to respond immediately with all the data.
This lets them avoid carrying extra state to keep track of pending
requests.
This should really have access to your state and the ability to change
it. Perhaps a more minimalist design would be better: just keep track
of the requests, then hand it off to +on-watch when eyre is ready to
receive responses. It's not clear how to pass in the request data in
+on-watch.
For some reason Jael subscriptions aren't starting properly for many
people. Until we can get to the bottom of it, this sets everyone to
start listening directly to the chain.
This augments permission management with invite sending, when setting "positive"
permissions. This matches talk's behavior.
Also implements +full:tr, which renders as ~ship/path, even for local targets.
Allows language-server to commit automatically upon changes. This
is driven by the editor, preventing the autocommit issues seen
with #971. Additionally recalculates syntax issues upon save.
This extends `gol` "backward-inference" typechecking to thread through
cores. Recall that `gol` is used exclusively for receiving more
specific error messages; these changes should have no effect on programs
which already compile successfully.
Before, this would type-fail on the second `|%`.
```
!:
^+ ^?
|%
++ foo *@ud
--
|%
++ foo
?: =(1 1)
2
%foo
--
```
With these changes, it gives a mint-nice at `%foo`. It will also give
you explicit errors if you have the wrong number/names of arms,
including which arms it expects.
This is becoming much more important with static gall, since it's the
first time we've used core subtyping so extensively and in userspace.
This extends `gol` "backward-inference" typechecking to thread through cores. Recall that `gol` is used exclusively for receiving more specific error messages; these changes should have no effect on programs which already compile successfully.
Before, this would type-fail on the second `|%`.
```
!:
^+ ^?
|%
++ foo *@ud
--
|%
++ foo
?: =(1 1)
2
%foo
--
```
With these changes, it gives a mint-nice at `%foo`. It will also give you explicit errors if you have the wrong number/names of arms, including which arms it expects.
This is becoming much more important with static gall, since it's the first time we've used core subtyping so extensively and in userspace.
We were updating our state and then using that when checking if the rift
had incremented. This would never be true, since we'd already set the
new state.
Fixes#1852 again
* origin/invite-app:
chat-hook: upgrade from old state and perform invitatory creation and subscription
invite-hook: crash upon invalid invite received
changed invite peek interface to /:path/:uid
invite: add comments and clean up
chat-js: added invite functionality
chat-hook: added invite functionality
app: added invite app and mark converters to JSON
Signed-off-by: Jared Tobin <jared@tlon.io>
* eth-watcher-2: (21 commits)
eth: move existing chain requests into ethio
eth-watcher: refactor refresh rate to top of file
hook: add pool-group-hook for making invite groups
ethio: add +read-contract for chain state reading
zuse: add delegated-sending address
eth: move eth-watcher's request-rpc into ethio lib
gaze: make compile for latest eth-watcher
drum: start eth-watcher on boot
azimuth-tracker: remove deprecated generator
eth: implement azimuth-tracker using eth-watcher
eth-watcher: ensure logs always sent oldest-first
eth-watcher: allow peers to unconfigured watchdogs
eth-watcher: saner %watch behavior
eth-watcher: implement %clear poke
eth-watcher: store logs in state to implement peer
eth-watcher: move types into /sur file
eth-watcher: properly tag out-peer-data
eth-watcher: single update timer loop
eth-watcher: implement /block peek
eth: turn azimuth-tracker into eth-watcher
...
Signed-off-by: Jared Tobin <jared@tlon.io>
Handle multiple files by keeping a map of text buffers. Also use the
Ford parser so we can parse ford runes. At some point we should load in
libraries when that happens so we have the appropriate types.
This corresponds to hoon-language-server 0.1.1
A simple language server engine, for use with hoonls.py, which presents
the RPC interface expected by editors. Features:
- Syntax error detection
- Rune snippets
- Autocomplete
* philip/tab-complete:
auto: gain and lose types on ?:
auto: handle tab in middle of symbol
auto: support forks
auto: support autocomplete inside wings
auto: fix some crashes on strange wet gates
auto: support multiline tab completion
auto: don't look in context of non-gold cores
easy-print: don't crash if type-check crashes
dojo, drum: change %tab sole-effect to use tanks
dojo, auto: move insert-magic logic to lib/auto
dojo, drum: give tab completion as true output
dojo: add a better function printer
dojo: add tab completion
Signed-off-by: Jared Tobin <jared@tlon.io>
The link used here resolves with a 301 to the proper page for messaging usage, but not actually the 'messaging' section of that page. This commit provides a more direct link to the exact instructions.
This changes the entry-points in lib/auto so that clients never have to
handle magic-spoon. You can specify either a tape of code with a
position index or a preparsed hoon (presumably you ran +insert-magic
before parsing).
This stops slogging the tab completion and intead adds a +sole-effect
for tab completion output. This is morally correct, and it lets dojo
clients show tab completions how they want. For example, web dojo could
implement this as a drop-down box.
Another advantage is that this puts the rendering logic in drum, which
knows the width of the terminal. Thus, we can make sure each match
takes no more than one line by truncating with ellipses. If there's
only one match and it's already fully typed, then we display the whole
type.
It's useful to know what a function takes and produces, so this changes the autocomplete type prettyprinter to emphasize those. This also gives a nice syntax for molds. Examples:
```
-----
add {a/@ b/@} -> @
~zod:dojo> add
-----
term * -> @tas
~zod:dojo> term
-----
sign-transaction {tx/{nonce/@ud gas-price/@ud gas/@ud to/@ux value/@ud data/@ux chain-id/@ux} pk/@} -> @ux
~zod:dojo> sign-transaction🔑ethereum
-----
wind {a/(* -> *) b/(* -> *)} -> * -> ?({$give p/*} {$pass p// q/*} {$slip p/*})
~zod:dojo> wind
```
This is initial support for type-aware tab completion. When you hit tab, it tries to complete the word you're in the middle of using a face or arm in the subject at that point in the code. It also shows all possible matches and their associated types. It's nearly instantaneous. Notes:
- It advances to the longest common prefix, so if you hit tab on `ab` and the only possible results are `abcde` and `abcdz`, then it'll write `abcd` and print both out (with their types).
- If there are fewer than ten matches, it prints the type along with the face. Printing types is too slow to use all the time, but with 10 it's essentially instantaneous.
- The match closest in the subject to you (i.e. smallest axis number) is displayed lowest (closest to your focus).
Examples below, where `<TAB>` represents me hitting tab while my cursor is at that position (the line with the `<TAB>` is not preserved in the actual output).
```
~zod:dojo> eth<TAB>
-----
ethereum #t/<11.qcl {<3.ltb 27.ipf 7.ecf 36.uek 92.bjk 247.ows 51.mvt 126.xjf 41.mac 1.ane $141> <21.yeb 27.ipf 7.ecf 36.uek 92.bjk 247.ows 51.mvt 126.xjf 41.mac 1.ane $141>}>
ethereum-types #t/<3.ltb 27.ipf 7.ecf 36.uek 92.bjk 247.ows 51.mvt 126.xjf 41.mac 1.ane $141>
~zod:dojo> ethereum
~zod:dojo> |= zong=@ud z<TAB>
-----
zing #t/<1.dqs {* <126.xjf 41.mac 1.ane $141>}>
zap #t/<1.iot {tub/{p/{p/@ud q/@ud} q/""} <1.rff {daf/@t <247.ows 51.mvt 126.xjf 41.mac 1.ane $141>}>}>
zuse #t/$309
zong #t/@ud
~zod:dojo> |= zong=@ud zo<TAB>
-----
zong #t/@ud
~zod:dojo> |= zong=@ud zong
~zod:dojo> <TAB>
hoon-version
trel
quip
pole
unit
qual
lone
... about 600 more lines ...
unity
html
zuse
eny
now
our
~zod:dojo>
```
Functionally, this is in a state where I'd be comfortable shipping it. It doesn't interfere with anything if you don't press tab, and it's perfectly OTA-able. I do think its output is a little verbose, but that can be tuned over time as people try it and determine what feels good in practice.
Additional notes:
- There are plenty of similar systems for other languages, but my most direct inspiration is Idris's editor tools. This is implemented for the dojo, but I actually want it in my editor, which is why the meat is all defind in a library. I've only tested on dojo one-liners, so I don't know the performance on large blocks of code.
- The default type printer isn't great for this use case. In particular,
- Cores should not print anything about their context
- The `#t/` should go away
- If it looks like a gate, we should print its return value
- Maybe special handling for molds, but if the above is done, then for example `bone` is `* -> @ud`.
- The worst part about our wing ordering is that it really screws up tab completion. You want to do `point.owner-address` instead of `owner-address.point` because that lets you type `point.ow<TAB>`. I weakly prefer reading it how we do it now, but it's really not great. You could do an (dojo-specific?) alternate syntax of `point;owner-address`; this is a simple transformation.
- Regardless of the above, this should handle the case where we're in the middle of defining a wing; it doesn't right now.
- When a variable is shadowed, we show both of them. We should probably show the shadowed one with a `^`.
- We probably shouldn't print out hundreds of results. Maybe just the closest 50 with ellipses.
- This gets you any face in your subject, regardless of whether its type is reasonable. We could limit that some by copying the `gol` logic in mint, so that if the pseudo-backward-inference engine happens to know what type it should be, you can filter the tab results according to if they nest in that type. This would be "strongly type-aware".
Re-implements the behavior of the previous azimuth-tracker as an app
that pokes and peers eth-watcher. Should have maintained identical
outward semantics to the original.
When configuring a watchdog on a path that already exists, we now
"overwrite" it, meaning we throw away all history and trawl the node
for logs again.
If the only config change is the url, however, we silently modify it,
and simply use it "from this point onward".
This matches the behavior of the original azimuth-tracker.
In order to give an initial response to incoming subscriptions (without
resorting to retrieving that data from chain again) we now store event
log history in state.
Instead of discarding pending-logs entirely after sending out updates,
we add them to the watchdog's history.
Just like pending-logs, we remove from the head during a rewind (though
not before exhausting the pending-logs).
Kicks the update timer on application start, then sets a new timer
whenever it's awoken. This aims to ensure eth-watcher never stops
looking for updates periodically.
No longer overwrite messages' timestamp on-receive, instead keeping whatever
timestamp was set by the sender.
This behavior matches that of the late Hall.
Uses the logic existing in azimuth-tracker to implement a new
eth-watcher, which can look at Ethereum nodes for _any_ events, as
opposed to exclusively a subset of the Azimuth contract's events.
Azimuth-tracker will be reimplemented as a dependent of this in
forthcoming commits.
These were deprecated in favor of azimuth-tracker in #1320.
(Azimuth-tracker, however, isn't a general-purpose Ethereum log watcher
tool. Commits to transform it into a more broadly useful tool are
forthcoming.)
Until now, clients of Jael have had to store the first-seen rift if they
want to reliably detect breaches. Otherwise, they would get a false
positive if they heard an old message about a breach (eg if you kick
azimuth-tracker). Clay and Gall did this correctly, but Ames did not.
Jael already maintains this state, so I added a notification to the
existing subscription that happens whenever it notices a breach (a diff
or full where the new rift is greater than the old one).
Because this is an issue on the live network, I wrote state adapters
for Gall and Clay. The Gall one just removes the rift from our state,
but the Clay one is much more involved because we have to upgrade
instances of the clad monad that are possibly in progress.
Specifically, since more input is possible than before, we must wrap any
in-progress instances of the monad in a function that handles the
potential new input from Jael. This temporarily preservers a copy of
the old kernel, but only until the current commit/merge/update has
completed.
The real solution for Clay is to factor out those IO-heavy instances to
userspace tapp/async/imp/threads, and if an upgrade happens in the
middle, you should simply restart them.
Fixes#1852
* claz-invites-newline:
claz: do invite file reading in +read-invites
claz: ignore empty lines in invites file
Signed-off-by: Jared Tobin <jared@tlon.io>
* publish-fixes:
publish: auto-resubscribe on quit, crash on failed subscription
publish: added permission logic to %serve and import flows
Signed-off-by: Jared Tobin <jared@tlon.io>
fc7901d2 refactored much of +ap-peek, but introduced a bug in the
process. The relevant diff from that commit is as follows:
- =/ =path [ren tyl]
- =/ =vase !>((slag p.u.cug path))
- (ap-slam q.u.cug p.arm vase)
+ =/ index p.u.maybe-arm
+ =/ term q.u.maybe-arm
+ =/ =vase
+ =/ =path [term tyl]
+ =/ raw (slag index path)
+ !> raw
+ (ap-slam term p.arm vase)
Note that [ren tyl] was replaced with [term tyl], where 'term' and 'ren'
are not equal. This commit merely rights that wrong.
* claz-checks:
claz: group state check arms together
claz: factor asserts out of callsites
claz: check pool sizes when inviting
claz: check planet availability for %invites
claz: print proper error messages
Signed-off-by: Jared Tobin <jared@tlon.io>
This test depends on the ames protocol version, and so should probably
be rewritten. It's currently holding up a breach, so it's most
expedient to just disable it for the time being.
09cb5f2 added a %send-point call, which is meant to target the delegated sending
contract. For %invites batches, this was the case. Handling of %single, however,
still sent all calls to the ecliptic contract.
This looks at the call tag to determine the target contract.
Per yosoyubik's commentary in urbit/urbit#1799:
The test is expecting that qeu to not be correct.. when it is.
The test [98 [97 ~ ~] [100 ~ [99 ~ ~]]] is a correct queue if we look at
vertical ordering: (mor 98 97), (mor 98 100) & (mor 100 99) all return
%.y, so vertical ordering is correct.
The previous implementation of +apt:to checked only horizontal ordering
between siblings, in this case that would fail: (mor 97 100) returns
%.n, but that is not how you check correctness of hoon treaps.
The solution is to modify that test with a proper "incorrect" +qeu, for
example: ((soft (qeu)) [97 [98 ~ ~] [100 ~ [99 ~ ~]]]). Vertical
ordering is not correct with any of the children.
* algorithm-tests:
pills: update solid
tests: unit tests for +in (set)
tests: unit tests for +to (queue)
tests: unit tests for +by (map)
tests: unit tests for +differ (diff/merge)
hoon: fix for +uno/uni (#1779) set/map union
hoon: fix for +apt:to (#1778) queue correctness
Signed-off-by: Jared Tobin <jared@tlon.io>
For generating many sendPoint() transactions for the Delegated Sending
contract. Specify what ship to send the invites as, and a path to a file
containing lines of "~ship,~ticket,0xaddress".
Comes with a generator, |claz-invites, for generating such files, given
a star and a range of its children (and an output path).
* odyssey-wip: (31 commits)
chat-cli: Add clarity
drum: Boot with %chat-cli, without %hall & %talk
chat-cli: Cosmetic improvements
chat: Move eval logic out of /lib/chat-json
chat-cli: Properly support deleting local chats
chat-cli: Subscribe to /updates instead of /all
chat: Move eval logic into lib
chat: removed unnecessary cast
chat: removed overly specific pattern match
chat: style fix for a comment
chat: remove poke-noun arms
chat: fixed eval function to disable scry
chat-cli: Simplify message command type & logic
chat-cli: Implement permission management
chat: /primary path provides truncated initial as well as updates
chat: style fixes, removed some redirect bugs from chat
chat-cli: Match store and hook's path handling
chat-cli: Update prompt on-create
chat: changed wire format and quitting subscription properly on ban
chat-cli: Add debug poke for connecting to store
...
Signed-off-by: Jared Tobin <jared@tlon.io>
Since the current implementation of ;leave is silently destroying state
instead of unsubscribing, we disallow running ;leave on local chats and
provide an explicit ;delete instead.
Set security type during ;create. Use ;invite and ;banish to dis/allow
ships from reading and/or writing.
Talks to the group-store to modify permission groups. Scries into
permission-store to check for white- vs blacklist.
Creating a mailbox would refresh the prompt before setting a new
audience, instead of after. This change corrects the behavior.
Also updates glyph binding code and print style.
Renames, refactors, and occasionally rewrites many of the arms used
within the application. Splits +sh into +sh-in and +sh-out, improves
naming for rendering cores, moves arms around for better organization,
and adds descriptions to all arms.
Brings it largely up to parity with Talk, save for features relating to:
- presence & nicknames
- circle management (permissions, sources)
- deprecated message types
In addition to implementing remaining functionality for basic usage
patterns, makes the following changes:
- glyphs per target, not multiple targets
- assume /~ship/path paths are created/used by the chat-hook
Code cleanup pending.
The subtree in pkg/arvo apparently still has a README in it, which had
gotten only slightly out of sync with the overall project README. This
commit updates its 'contributing' section to point at the appropriate
contributing document.
%gall currently prints
[%gall-booting <app> p=<ship> q=<desk>]
whenever it receives a %conf (i.e., when it boots an app). This turns
up in many of the places the old, less-informative '%mo-not-running'
printf did, but it's of similarly little use, and mainly serves to
create redundant line noise. This commit just removes it.
* reclamation:
solid pill
arvo: wires up %trim memory-pressure event stubs
u3: wires up the %trim event in the daemon
u3: adds %trim memory-pressure notification "effect"
u3: tweaks |mass output (and fixes inadvertent truncation)
u3: refactors periodic memory reclamation
u3: factors out u3a_idle() to measure free-lists in a road
u3: adds and uses u3a_full/heap/temp road macros
u3: cleans up whitespace, removes dead code in allocator
* barbus:
hoon: removed unused parser type
hoon: remove extraneous cast
hoon: removed barhax. No hax!
hoon: replace barhax with barbus
hoon: changed barbus to match barhax
hoon: replaced barbus usage with barhax
hoon: add parsing for barhax
hoon: added barhax to replace barbus
hoon: update tall parsing for |$
hoon: replace +* name usage with ++ name |$
hoon: add parsing rules for |$
hoon: add ast for |$
We don't care about the static types in the use-cases where we need to
prevent scry (to prevent accidental data disclosure). We can evaluate
the expression, virtualized and untyped, and then just clam.
Enables .^ in +mule (statically-typed virtualization), by specifying a
scry-handler function that punts the namespace read to a higher
virtualization layer via virtual-nock (mock) 12.
Publish's %serve command makes builds for notes even if the
publish-info file is missing. It now crashes the build if the file is
missing with a one-line ?> asserting that the file is found in the
list of paths associated with the collection.
Updates all Landscape applications to use the
latest version of urbit-ob, from 3.1.1 to 4.1.2.
Removes urbit-ob from applications that don't
use it (Clock, Launch, Weather).
Compiled JS for all the above included in this
commit.
In Publish, users could get in a bad state if they made a post with
valid udon, and subsequently edited to contain invalid udon.
Furthermore, users subscribed to them would get in the same bad state.
This fixes the original bug, and users who are already in the broken
state will be able to run a recovery command: :publish %state-surgery
which will also fix the downstream broken state of their subscribers.
This check required the new type of +type to nest within the old type of
+type, which is wrong. Specifically, this disallowed adding new runes
without a staging procedure (which we didn't successfully complete).
* jt-gall-refactor: (76 commits)
gall: fix issue id in comment
pills: update solid
gall: handle foreign coup success
gall: only print peek bad result if bad
gall: add basic test harness
pills: update solid, brass, ivory
gall: fix obvious nest-failing tisdot
gall: change '-state' to '-core' for +mo and +ap
zuse, gall: deprecate 'club'
zuse, gall, eyre: deprecate 'cush'
zuse, gall, eyre, dojo: deprecate 'cuft'
gall: remove slam-related printfs
gall: remove deprecated 'mak' from 'agents'
gall: use less vertical spacing throughout
gall: add comment re: unpopulated wex
gall: use less vertical separation when wuthepping
gall: fix whitespace
gall: don't define 'move' as a pair
gall: don't give faces to tags
gall: gut some unused stuff
...
Edit post's UI appearance looked quite different
from what creating a new post looked like.
This commit just brings the styling of post
editing UI a bit closer together with new posts.
Read-only chats had a slightly bigger sigil box, looking skewed to
the left. Its copy also had a different line height than the
chat input itself, which was vertically aligned slightly higher.
This commit standardises the sigil box to 32px across both
and brings both to the same, centered vertical alignment
for the chat input and read-only notice.
If your screen wasn't wide enough, the flex rules would destroy the gap
between columns, which destroyed the look of a table altogether.
By removing the 'one-line' class, and moving the margin-left from
the span element to the parent paragraph (for rows that aren't
the header rows), titles wrap onto another line, which enables
a responsive table and firm table margins.
%zuse includes definitions for 'scup' and 'culm', both of which are
superfluous. 'scup' is simply (pair ship desk) and is used only in the
definition of 'culm', a tagged union in which three of the four branches
are commented out (i.e. are unused).
This commit deletes 'scup' and 'culm' and refactors what little code
made use of them.
Additional logic for reducer + array manips
Removing multi-array mutation and comments
Adding comments and working logic.
Fix sigil showing for pending from same aut
Pending messages persist upon circle change
Scaffolding message pending injection
Additional logic for reducer + array manips
Removing multi-array mutation and comments
Adding comments and working logic.
Fix sigil showing for pending from same aut
Pending messages persist upon circle change
Reworking pending boolean logic.
Data structure changed to Map
Checking correct scope of prop, removing dev TODO
Rebase mistake.
