udapted docs

.gitignore

@@ -1,3 +1,5 @@
 venv
 instance
 __pycache__
+found.pi
+

README.md

@@ -5,6 +5,7 @@ Mathy is a playground for experimenting with streamed arithmetic on very large i
 ## What's Here
 - `mathstream/` – Streaming big-int helpers (`StreamNumber`, arithmetic primitives, garbage collection utilities). See `mathstream/README.md` for an in-depth guide.
 - `collatz_ui/` + `collatz.py` – A curses dashboard that visualizes Collatz walks. The UI pushes work onto a background thread so the main loop can keep the interface responsive while mathstream evaluates each step.
+- `pi_finder/` + `find_my.py` – Nilakantha-based π explorer with both CLI output and a curses dashboard powered by mathstream primitives.
 - `stupid.py` – Experimental Collatz-based encoder/decoder that maps ASCII message chunks to Collatz seeds and replays the sequence to recover the text.
 - `seed_start.py` – Utility for seeding `start.txt` with huge streamed additions sourced from randomness or deterministic sequences.
 - `tests/` – Sample digit files used by the regression script in `test.py`.
@@ -19,13 +20,15 @@ pip install -e .
 From there you can explore the components:
 - Run the regression checks: `python test.py`
 - Launch the Collatz dashboard: `python collatz.py`
+- Stream π digits to the terminal: `python find_my.py --ui`
+- Print a quick π approximation: `python find_my.py --digits 200`
 - Try the Collatz encoder prototype: `python stupid.py`
 - Seed the running total file: `python seed_start.py --seed 100 --mode seq`
 
 All streamed results land in `instance/log/`. If you need a clean slate, call `mathstream.clear_logs()` or delete that directory between runs.
 
 ## Experiments and Next Steps
-- **Streaming worker** – The Collatz UI's worker thread (`collatz_ui/app.py`) demonstrates how to keep long-running streamed computations off the main event loop.
+- **Streaming workers** – The Collatz and π dashboards (`collatz_ui/app.py`, `pi_finder/ui/app.py`) showcase background threads feeding long-running computations into a curses UI.
 - **Collatz encoding** – `stupid.py` currently encodes a message by turning ASCII blocks into integers and treating the first block as the Collatz seed. Extending this to use the full sequence (or to target specific steps) is an open avenue for exploration.
 - **More tests** – `test.py` covers the primary mathstream operations. Add pytest cases or property tests under `tests/` as new behaviours land.
 

WORK.md

@@ -2,7 +2,7 @@
 
 Figured I’d write down how the whole thing is wired, more like a brain dump than polished docs. If you’re diving in to fix something, this should save you a bunch of spelunking.
 
-> New top-level `README.md` now mirrors the high-level pitch, cross-links experiments (`collatz_ui`, `stupid.py`, `seed_start.py`), and lists quick-start commands (`pip install -e .`, `python test.py`, etc.). Treat this file as the deep dive into the mathstream core.
+> New top-level `README.md` now mirrors the high-level pitch, cross-links experiments (`collatz_ui`, `pi_finder`, `find_my.py`, `stupid.py`, `seed_start.py`), and lists quick-start commands (`pip install -e .`, `python test.py`, etc.). Treat this file as the deep dive into the mathstream core.
 
 ## Directory map (so you know where to poke)
 

mathstream/README.md

@@ -57,6 +57,7 @@ Each arithmetic call writes its result back into `instance/log` (configurable vi
 - **Sign-aware** – Addition, subtraction, multiplication, division (`//` behavior), modulo, and exponentiation (non-negative exponents) all respect operand sign. Division/modulo follow Python’s floor-division rules.
 - **Utilities** – `clear_logs()` wipes prior staged results so you can start fresh.
 - **Manual freeing** – Call `stream.free()` (or `free_stream(stream)`) once you are done with a staged number to release its reference immediately. Logger metadata keeps per-path reference counts so the final free removes the backing file on the spot.
+- **GC toggle** – Need total control over when files disappear? Flip `mathstream.set_manual_free_only(True)` so automatic finalizers stop unlinking staged files; they will persist until you call `free()` (or `collect_garbage`). Use `mathstream.manual_free_only_enabled()` to inspect the current setting.
 - **Parity helpers** – `is_even` and `is_odd` inspect the streamed digits without materializing the integer.
 - **Garbage collection** – `collect_garbage(score_threshold)` computes a score from file age, access count, and reference count (tracked in `instance/mathstream_logs.sqlite`, freshly truncated each run). Files whose score meets or exceeds the threshold are deleted, letting you tune how aggressively to reclaim space. Both staged results and literal caches participate. Use `tracked_files()` or `active_streams()` to inspect current state.