| Age | Commit message (Collapse) | Author |
|---|
|
GHC --make was treating -i /nix/store/.../hidir paths as compilation targets.
Solution: Copy all .hi files to local directory and use -i. only.
Also make hsGraph optional with 'or null' for backward compatibility with
old bild binaries during bootstrap.
Amp-Thread-ID: https://ampcode.com/threads/T-fe68faaf-1c1d-4c43-a377-1cf5e6cffb3a
Co-authored-by: Amp <amp@ampcode.com>
|
|
GHC --make interprets arguments before flags as targets. Moving the
entry point source file to the beginning prevents -i paths from being
treated as compilation targets.
This fixes the 'is not a module name or a source file' error.
|
|
This is the core architecture transformation from Phase 3 of the
performance plan. Each Haskell module is now built as a separate
Nix derivation, enabling true incremental builds where only changed
modules and their dependents are rebuilt.
Implementation: - buildHsModuleGraph: Analyzes transitive module
dependencies and builds DAG - TH detection: Falls back to monolithic
build if Template Haskell detected - SCC cycle detection: Falls
back if import cycles found - Per-module Nix builder: Each module ->
separate derivation with .hi and .o - Module dependencies: Copy .hi
files to build dir, use -i flags for imports - Final link: Use ghc
--make with entry point source + -i paths to .hi files - Entry point
fix: Explicitly analyze entry point module separately from deps
Architecture: - Module compilation: ghc -c with -i paths to dependency
.hi files - Source filtering: Each module derivation includes only
its source file - Dependency DAG: Expressed as recursive Nix attrset
with lib.fix - Link phase: ghc --make with entry source file + all .hi
search paths - Fallback: Monolithic ghc --make when hsGraph is null
(TH/cycles)
Performance characteristics: - Change one module -> rebuild only
that + dependents + relink - Nix handles DAG scheduling and caching
automatically - Parallel module compilation (Nix orchestrates) -
Content-addressed caching across machines
Testing: - Added test_buildHsModuleGraph unit test - Verified with
Omni/Bild/Example.hs (4 modules) - Tested incremental rebuild triggers
correct subset
This completes Phase 2 and Phase 3 core optimizations from the plan.
|
|
Implements Steps 1-4 of per-module Nix architecture: - Add
HsModuleGraph and HsModuleNode types to represent module DAG -
Implement buildHsModuleGraph that:
- Parses imports for each module in the transitive closure -
Detects Template Haskell usage (language pragma or $( patterns) -
Detects import cycles using SCC analysis - Returns Nothing (fallback
to monolithic) if TH or cycles found
- Wire hsGraph into Target type for all builders - Populate hsGraph
for Haskell targets during analysis
Module graph is now emitted in --plan output showing per-module
dependencies. This enables the next step: per-module Nix derivations
in Builder.nix for true incremental builds.
Part of Phase 3 from the performance plan.
|
|
Implements optimization #3 from Phase 2 of the performance plan.
Changes: - Cache stored in _/var/ghc-pkg-cache-<hash>.json - Hash
based on GHC version + GHC_PACKAGE_PATH for automatic invalidation -
Loads cache at startup, saves on successful completion - Uses atomic
write (tmp + rename) to prevent corruption - Gracefully handles
missing/corrupt cache files - Accumulates cache entries across builds -
Works with parallel builds (in-memory IORef + disk persistence)
Performance impact: - Eliminates redundant ghc-pkg invocations across
runs - Near-zero ghc-pkg overhead once cache is populated - No impact
on single-run performance (still uses in-memory IORef)
|
|
Only include directories that are ancestors of source files in
allSources. Previously accepted all directories, causing rebuilds
when any new directory was added to the repo.
Implementation: - Precompute normalized source paths and their ancestor
directories - Filter directories against allowedDirs whitelist -
Normalize paths in file filter for consistency - Keep existing skip
list behavior for _ and .direnv
This is optimization #2 from Phase 2 of the performance plan.
|
|
- Add Analyzing state to BuildState enum - Refactor from sequential
foldM analyze to concurrent analyzeAll - Initialize all lines with
[+] during analysis phase - Update to [...] (Pending) after each
analysis completes - Uses mapConcurrentlyBounded with concurrency of
8 for analysis - Remove Log.info from analyzeOne (now handled by line
state) - Analysis now runs in parallel, improving efficiency - Flow:
[+] analyzing → [...] pending → [~] building → [✓]/[x] complete
|
|
- Replace cursor down movement with single newline at end - Cursor
now stays at bottom of status lines - No extra blank space between
build output and next prompt
|
|
- Remove save/restore cursor in favor of explicit movement - Always
move up from bottom position, update line, move back down - Simplify
initializeLines to just print lines sequentially - Cursor now stays
at bottom and moves up/down for each update - Fixes issue where lines
were being updated far up the terminal
|
|
- Use hCursorDown in initializeLines instead of hCursorUp - Simplify
updateLine and updateLineState cursor movement - Cursor now moves up
from bottom position correctly - Shell prompt no longer gets erased
|
|
- Allocate one line per namespace (not per concurrent job) - Add
Pending state shown as [...] when build hasn't started - Initialize all
namespace lines at start showing [...] - Update to [~] when building,
[✓]/[x] when complete - Each namespace keeps its line throughout
the build - At end, all namespaces show their final status - --jobs
controls concurrency, not line count
|
|
- Remove [+] display from reserveLine (was causing mangled output
from concurrent writes) - Add [+] display in analyze function where
it's single-threaded - Simplify cleanup to just move cursor down
without clearing lines - This eliminates race conditions and terminal
clearing issues
|
|
- Add initial newline to preserve terminal prompt - Clear each line
individually at end instead of just moving cursor - This prevents
extra blank lines from remaining on screen
|
|
- Remove initial blank line before build starts - Remove trailing blank
line after build completes - Remove newlines from status indicators
in releaseLine - Status lines now stay on their reserved lines without
extra spacing
|
|
- Remove all label prefixes from build logs (bild:, nix:, etc) -
Show [+] Namespace when first reserving a line (analyzing) - Show [~]
Namespace: output when building with subprocess output - Show [✓]
Namespace (green) on success - Show [x] Namespace (red) on failure
|
|
- Remove Log.info from analyze function - Remove Log.info from build
target compilers (Guile, NixBuild, Rustc, Sbcl) - Remove Log.good
from proc and nixBuild success handlers - Remove 'info: bild:' prefix
from streaming subprocess output - Output now shows only [✓]/[x]/[~]
status lines with subprocess stdout
|
|
- Add initial line break to preserve terminal prompt - Change format
to [✓]/[x]/[~] Namespace: output - Add colors: green for success,
red for failure - Fix extra lines by adding newlines in releaseLine -
Fix currentLine initialization to 0 instead of maxLines
Amp-Thread-ID:
https://ampcode.com/threads/T-39671965-c412-4a2e-8084-9d09128fd865
Co-authored-by: Amp <amp@ampcode.com>
|
|
- Add ansi-terminal to ghcPackageSetBild dependencies - Add
Omni/Log/Concurrent.hs to source fileset - Fix unused mLineNum warning
in buildTarget - Alphabetize deps list (hostname moved up)
Parallel builds now fully functional with proper Nix dependencies.
|
|
- Fix BuildStatus to use newtype with single field - Clean up unused
imports and fields - logs function now uses updateCurrentLine
for LineManager support - Fallback to single-line output when no
LineManager
All tests passing. Ready to test with different terminal types.
Tasks: t-1a1Eiay
|
|
- Add global IORef for currentLineManager and namespaceLines
mapping - Update logs function to use LogC.updateCurrentLine -
Add updateCurrentLine and releaseCurrentLine helpers - Fallback to
normal printing when no LineManager active - Simplify buildTarget to
use global helpers instead of threading
Tasks: t-1a1EaJy
|
|
- Reserve line for each concurrent build - Release line on completion
with success/failed state - Fix hlint eta reduce warning
Task: t-1a1E3j1
|
|
- Implement LineManager abstraction with IORef state - Line
reservation/update/release functions - ANSI cursor positioning for
concurrent updates - Terminal capability detection (ANSI vs dumb) -
Graceful fallback for non-ANSI terminals
Tasks: t-1a1DzES, t-1a1DGY0, t-1a1DOev, t-1a1DVM5
|
|
- Add mapConcurrentlyBounded using QSemN for bounded parallelism -
Refactor build function to extract buildTarget worker - Build up to
--jobs targets concurrently - Preserves all existing functionality -
Output will be interleaved (will fix with LineManager next)
Related tasks: - t-1a0OVBs: mapConcurrentlyBounded helper - t-1a16ame:
refactor build function - t-1a1DdSB: replace forM with concurrent map
|
|
Replaced the old slow depth-first search with a breadth-first search
for detecting imports. This should be way faster when building a
single namespace because it doesn't have to visit the same file
multiple times.
The ghc-pkg caching means we only have to run ghc-pkg once per bild
invocation.
|
|
detectPythonImports now recursively analyzes imported modules to find
transitive dependencies, matching the behavior of detectHaskellImports.
Previously it only detected direct imports, which caused build failures
when Python modules had nested dependencies.
- Changed signature from [Text] -> IO (Set FilePath)
to Analysis -> [Text] -> IO (Set FilePath)
- Added filepaths, findDeps, and onlyPython helper functions -
Recursively calls analyze() on imported modules to find transitive
deps - Updated tests to pass empty Analysis map
|
|
I think the calls to Log.setup() were accidentally creating multiple
loggers, hopefully this fixes the problem.
|
|
- Remove UNIQUE constraints from ALTER TABLE statements
(SQLite doesn't support adding constraints via ALTER TABLE)
- Add better logging for migration failures (debug level) - Rely on
application logic to ensure uniqueness instead of DB constraint -
This fixes the silent failure where stripe_customer_id and
stripe_subscription_id columns weren't being added
|
|
- Implement Biz.PodcastItLater.Billing with checkout sessions, billing
portal, webhook handling - Add subscription database schema: plan_tier,
stripe fields, period dates, stripe_events table - Three-tier pricing:
free (10/month), personal (/month, 50 articles), pro (9/month,
unlimited) - Usage tracking and enforcement with tier-based limits
- Full billing UI with plan display, usage stats, pricing cards,
upgrade buttons - Dashboard shows current tier with billing button -
Update Web.nix with Stripe environment variables - Fix POST redirects
to Stripe with 303 status code for CloudFront compatibility
Amp-Thread-ID:
https://ampcode.com/threads/T-c139e5b5-1901-4cd6-8030-5623bfe1df35
Co-authored-by: Amp <amp@ampcode.com>
|
|
Without this, run.sh would continue if bild failed, and you might
end up running an out of date artifact.
|
|
- Add stripe to Omni/Bild/Deps/Python.nix (alphabetically sorted) -
Fix all type annotations in Billing.py for mypy - Document how to add
Python packages in AGENTS.md - Add billing routes to Web.py (checkout,
portal, webhook)
This enables Stripe integration in PodcastItLater.
Related to task t-144e7lF
|
|
Changed epic display in tree view from:
t-PpXWsU [Epic] [ ] Task Manager Improvements
To:
t-PpXWsU [6/11] Task Manager Improvements
The [6/11] shows completed/total child tasks, giving immediate visual
feedback on epic progress. Regular tasks still use checkbox indicators:
[ ] open, [~] in-progress, [✓] done.
|
|
Changed tree output format from:
t-abc123 [ ] Task title [Omni/Task.hs]
To:
t-abc123 [ ] [Omni/Task.hs] Task title
This makes the namespace more prominent and groups all metadata
(status + namespace) together before the title.
|
|
Changes: 1. Remove [Task] label - only show [Epic] for epics, cleaner
output 2. Truncate long titles to fit 80 columns with ... ellipsis
3. Better spacing with type label included in layout calculation
Created task for future improvement: prettier box-drawing characters
(├──, └──) which would require Data.Tree library
investigation.
Current output is clean and readable within standard terminal width.
|
|
Add 'task tree' command to show hierarchical task structure:
Usage:
task tree # Show all epics with their children task tree <id> #
Show specific epic/task with children
Features: - Visual status indicators: [ ] open, [~] in-progress,
[✓] done - Shows task type: [Epic] or [Task] - Indented display
for hierarchy - Shows namespace associations
Example output:
t-PpXWsU [Epic] [ ] Task Manager Improvements [Omni/Task.hs]
t-PpYZt2 [Task] [ ] Implement child ID generation t-PpZGVf [Task]
[✓] Add filtering by type and parent
Updated AGENTS.md with usage examples.
Closes task t-PpZlbL
|
|
Implemented proper pre-commit hook that: - Calls 'task export --flush'
to consolidate tasks - Auto-stages .tasks/tasks.jsonl if modified -
Runs before every commit
Added reminder message after 'task update' to inform users that task
changes will be committed on next git commit.
Updated AGENTS.md to document the auto-commit behavior.
This fixes the bug where task status updates (e.g., marking tasks as
Done) were not being committed to git.
|
|
Implement --status and --namespace filters for task list:
New filters: - --status: Filter by open, in-progress, or done -
--namespace: Filter by namespace (e.g., Omni/Task)
All filters can be combined: - task list --parent=t-abc123
--status=open - task list --type=epic --status=done - task list
--namespace="Omni/Task" --status=open
Updated listTasks signature to accept all filter parameters and apply
them in sequence. Updated AGENTS.md with examples.
Closes task t-PpZGVf
|
|
- Add TASK_TEST_MODE environment variable to use separate test database
- All file operations now use getTasksFilePath to respect test mode -
Tests use .tasks/tasks-test.jsonl instead of production database -
Add automatic migration from old task format (taskProject field)
to new format - Migrated tasks convert taskProject to WorkTask type
with empty parent - Old [Text] dependencies converted to [Dependency]
with Blocks type - Restore actual tasks from commit 3bf1691 (were
lost during testing)
This prevents accidental data loss when running tests and provides
backward compatibility for existing task databases.
|
|
Implement four dependency types based on beads patterns: - Blocks:
Hard dependency, blocks ready work queue (default) - DiscoveredFrom:
Work discovered during implementation (doesn't block) - ParentChild:
Epic/task relationships (blocks ready work) - Related: Soft
relationships (doesn't block)
Key changes: - New Dependency data type with depId and depType fields
- New DependencyType enum with four relationship types - Updated CLI
with --dep-type and --discovered-from flags - Enhanced getReadyTasks
to respect only blocking dependency types - Added comprehensive tests
for all dependency behaviors - Updated AGENTS.md with usage examples
and patterns
The discovered-from pattern is especially important for AI agents to
maintain context of work found during implementation while keeping
it immediately available in the ready work queue.
Amp-Thread-ID:
https://ampcode.com/threads/T-178b273a-3ac7-416c-a964-db89bac3c8f7
Co-authored-by: Amp <amp@ampcode.com>
|
|
Major refactoring of task data model: - Added TaskType enum (Epic |
WorkTask) - Replaced taskProject with taskType and taskParent fields -
Epics are containers for tasks (hierarchical organization) - Tasks can
have optional parent epics - Updated createTask signature to accept
type and parent - Updated CLI: --type=epic|task and --parent=<id>
options - Updated list command to filter by type and parent - Updated
printTask to display type and parent info - Fixed naming collision
(WorkTask instead of Task constructor)
Example usage:
task create "Auth System" --type=epic task create "Design API"
--type=task --parent=t-abc123 task list --type=epic task list
--parent=t-abc123
Completed task: t-8WR5Zg
Amp-Thread-ID:
https://ampcode.com/threads/T-85f4ee29-a529-4f59-ac6f-6ffec75b6a56
Co-authored-by: Amp <amp@ampcode.com>
|
|
Updated namespace handling to use the Omni.Namespace module: - Import
Omni.Namespace in Task CLI - Parse and validate namespace strings
using fromHaskellModule - Convert to proper path format using toPath -
Ensures namespaces are well-formed (e.g., Omni/Task -> Omni/Task.hs)
Example:
task create "Fix bug" project --namespace="Omni/Task" # Creates
task with validated namespace Omni/Task.hs
This provides type safety and ensures all task namespaces correspond
to actual code namespaces in the monorepo.
Amp-Thread-ID:
https://ampcode.com/threads/T-85f4ee29-a529-4f59-ac6f-6ffec75b6a56
Co-authored-by: Amp <amp@ampcode.com>
|
|
Tasks can now be associated with specific namespaces in the monorepo:
- Added taskNamespace (Maybe Text) field to Task data type - Updated
createTask to accept optional namespace parameter - Added --namespace
CLI option to task create command - Display namespace in task list
output (e.g., [Omni/Task]) - Updated tests to pass Nothing for
namespace - Updated AGENTS.md documentation
Example usage:
task create "Fix bug" project --namespace="Omni/Task"
Completed task: t-j0k1L2
Amp-Thread-ID:
https://ampcode.com/threads/T-85f4ee29-a529-4f59-ac6f-6ffec75b6a56
Co-authored-by: Amp <amp@ampcode.com>
|
|
Implemented a dependency-aware task tracker inspired by beads: - Task
CRUD operations (create, list, update, ready) - Dependency tracking
and ready work detection - JSONL storage with git sync via hooks -
Export/import for cross-machine synchronization - Short base62-encoded
task IDs (e.g., t-1ky7gJ2)
Added comprehensive AGENTS.md documentation: - Task manager usage and
workflows - Development tools (bild, lint, repl.sh) - Git-branchless
workflow guidelines - Coding conventions
Integrated with git hooks for auto-sync: - post-merge/post-checkout:
import tasks - pre-commit/pre-push: export tasks
Also includes beads design analysis document for reference.
Completed tasks: - t-a1b2c3: Show help text when invoked without args
- t-d4e5f6: Move dev instructions to AGENTS.md - t-g7h8i9: Implement
shorter task IDs - t-p6q7r8: Add git-branchless workflow docs
https: //ampcode.com/threads/T-85f4ee29-a529-4f59-ac6f-6ffec75b6a56
Co-authored-by: Amp <amp@ampcode.com> Amp-Thread-ID:
https://ampcode.com/threads/T-85f4ee29-a529-4f59-ac6f-6ffec75b6a56
|
|
Without this, my custom git commit template will get mangled
into the commit message with the call to fmt. I guess git strips
all of this stuff *after* it runs the commit-msg hook. Oy vey.
diff --git a/Omni/Ide/hooks/commit-msg b/Omni/Ide/hooks/commit-msg
index c15b4a1..bfbb06f 100755 --- a/Omni/Ide/hooks/commit-msg +++
b/Omni/Ide/hooks/commit-msg @@ -1,6 +1,7 @@
#!/usr/bin/env bash temp=$(mktemp)
-fmt -w 72 -u "$1" > "$temp" +# strip comment lines and everything
after >8 cut line +sed '/^#/d; /^# -\+ >8 -\+/,$d' "$1" | fmt -w 72
-u > "$temp"
mv "$temp" "$1" if ! gitlint --ignore-stdin --staged --msg-filename
"$1" run-hook; then
backup="$CODEROOT"/.git/COMMIT_EDITMSG.backup
|
|
Check to prevent processing of large articles, truncate oversized
content, defer jobs during high memory usage, use streaming TTS
generation and upload to minimize memory consumption.
|
|
The -s and -u flags tell fmt to fill the paragraph by merging
lines. Without this, fmt might just add line breaks, and this is
no good.
|
|
Moved the Admin related stuff to a separate file. Removed the
repetitive
`db_path` arg everywhere and replaced it with correct assumptions,
similar to
whats in other apps.
|
|
When I deploy, I get an error like this:
Error: Failed to open dbus connection
Caused by: Failed to connect to socket /run/user/1000/bus:
Connection refused
According to Claude, this is because dbus is trying to do stuff while
I'm running the `systemd-run` command, but that loses my user context so
dbus errors out, and I can disable dbus by setting this variable to
nothing. I guess we'll see if it works!
|
|
Sometimes aider will write commit messages without wrapping them at 80 chars,
and then the commit fails the gitlint hook, and aider can't finish the commit.
To fix this I can just auto-wrap before we check it.
|
|
According to Claude, the nixos-unstable branch has more stuff in it's binary
cache than nixos-unstable-small. Idk, let's try it and find out.
|
|
This implements a working prototype of PodcastItLater. It basically just works
for a single user currently, but the articles are nice to listen to and this is
something that we can start to build with.
|
