lemke_handClaude Sonnet 5
12d70469cd
There were two full copies of the interactive startup script: eco/
startup_inline.py (the package copy, run via `eco_cli.py`'s `run -m eco.
startup_inline`, used by both the eco console-script and scripts/eco-dev)
and a second copy at the repo root, execed by startup_inline_new.py -- the
back-compat shim the real /sf/bernina/bin/{eco,eco_new,eco-in,eco_from}
wrappers hardcode by absolute path (root-owned, not editable here).
The two had already drifted: the root copy was silently missing the
kernel_registry.install_shell_logger() call the package copy had gained,
so every session started the normal way at a console (the actual daily
entry point) was never logged at all, despite eco.logs' viewer assuming
every console session has a log. Repointed the shim to exec eco/
startup_inline.py directly instead of the sibling root copy, and deleted
the now-fully-redundant root copy -- one real script left to maintain, so
this class of drift can't recur.
Verified live: ran a real ipython session through startup_inline_new.py
exactly as the /sf/bernina/bin wrappers do (`run startup_inline_new.py -l
-s bernina`), confirmed a fresh ~/.eco/kernel_logs/*.jsonl appeared with
the probe command actually logged.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
33 lines
1.5 KiB
Python
33 lines
1.5 KiB
Python
#!/usr/bin/env python
|
|
# Back-compat alias for eco/startup_inline.py, the one real startup script
|
|
# (this file's own content used to differ -- see git history -- and
|
|
# eco_cli.py, desktop_app.py and docs/installation.md already assumed
|
|
# startup_inline.py's namespace-based semantics, not this file's old
|
|
# `{scope}.init()` call, which no longer exists on eco.bernina). Kept only
|
|
# because several /sf/bernina/bin wrapper scripts (eco, eco_new, eco-in,
|
|
# eco_from) hardcode this filename by absolute path and aren't editable
|
|
# here (root-owned).
|
|
#
|
|
# Used to exec a second copy of the script that lived at this same
|
|
# directory level (a root-level startup_inline.py, deleted now) instead of
|
|
# eco/startup_inline.py -- the two had drifted apart (confirmed 2026-09-12:
|
|
# the root copy was silently missing the kernel_registry.install_shell_
|
|
# logger() call the package copy had gained, meaning every session started
|
|
# through these /sf/bernina/bin wrappers was never actually logged despite
|
|
# eco.logs' viewer assuming it would be). Pointing this shim straight at
|
|
# eco/startup_inline.py leaves exactly one real copy to maintain, so this
|
|
# class of drift can't recur.
|
|
#
|
|
# exec'd (not run_path'd) so this executes in the caller's own namespace,
|
|
# exactly as if eco/startup_inline.py had been run directly -- matching how
|
|
# IPython's `run <path>` normally populates the interactive namespace.
|
|
from pathlib import Path
|
|
|
|
exec(
|
|
compile(
|
|
(Path(__file__).parent / "eco" / "startup_inline.py").read_text(),
|
|
"eco/startup_inline.py",
|
|
"exec",
|
|
)
|
|
)
|