pardes.git/build.zig.zon, branch mainPardeshttps://git.0x4200.cafe/pardes.git/atom?h=main2026-10-01T14:10:37ZEvery pane pardes spawns gets $PARDES_MOUNT, the session's own directory under 9ns --mntgen ($NINE_MOUNT/pardes/<name>), unset when it is not known2026-10-01T14:10:37ZGabriel Schneidergbrls@0x4200.cafe2026-10-01T14:09:59Zurn:sha1:36a9bfbaba649da3d988608f1cc7834e567df95d
9ns says --mntgen with NINE_MNTGEN=1 (cloud9). Under `9ns --unix` the
mount is the server it dialled, never this session (its socket does not
exist until it starts), so it stays unset there; scripts fall back to
`9p -a "unix!$PARDES_9P"`. Nothing is statted: the name comes from the
session's socket and the environment. selfmount checks both modes.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Release Pardes 0.232026-10-01T12:13:37ZGabriel Schneidergbrls@0x4200.cafe2026-10-01T12:13:37Zurn:sha1:51df4b01f4ea8768b14048841f1c5fdd75128cd3
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Release Pardes 0.222026-10-01T10:16:54ZGabriel Schneidergbrls@0x4200.cafe2026-10-01T10:15:12Zurn:sha1:989f2bae90fde52b5dd36b57c31f535dd011461e
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Every EINVAL or ENOENT refusal a unit test meets must say its errno in words a 9ns mount maps back: pardes re-pins cloud9, whose fs.enameErrno knows acme's address and argument refusals2026-10-01T10:16:43ZGabriel Schneidergbrls@0x4200.cafe2026-10-01T08:54:02Zurn:sha1:94f0c9359dc1b8ccd73bf58812bec9e78fb52073
The tests' request helper asks cloud9's fs.enameErrno, the rules 9ns
turns a refusal's words back into an errno with, for every EINVAL and
ENOENT a request is refused with, and fails the test when a mount
would hear another (EIO, mostly): "this pane has no text to address"
was one. The re-pinned cloud9 maps acme's own refusals, which pardes
keeps in acme's words: no match for regexp, address out of range,
addresses out of order, range past end of body and the rest.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Release Pardes 0.212026-10-01T08:48:11ZGabriel Schneidergbrls@0x4200.cafe2026-10-01T08:48:11Zurn:sha1:c84834523cc53d2a46eea06ae7e26d92df0d022b
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Release Pardes 0.202026-10-01T07:18:10ZGabriel Schneidergbrls@0x4200.cafe2026-10-01T07:18:10Zurn:sha1:3f328a0646268169012c8c0a36840ffaba34823b
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Release Pardes 0.192026-10-01T06:23:55ZGabriel Schneidergbrls@0x4200.cafe2026-10-01T06:23:55Zurn:sha1:a4e17d350a7afdeee4916584c44c4770243c84e4
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
An open past the session's 64 open records reaches a mount's caller as EMFILE, "Too many open files": pardes re-pins cloud9, whose 9ns maps the words2026-10-01T05:57:15ZGabriel Schneidergbrls@0x4200.cafe2026-10-01T05:57:15Zurn:sha1:b7182d34e12130ced98dee7726b3bc87d4584ca9
The refusal's words said too many open files, but 9ns matched none of
its needles and handed the shell Input/output error. The re-pinned
cloud9's 9ns maps them to EMFILE. selfmount opens the root exec for
writing eighty times through the mount and hears EMFILE at the 65th.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Release Pardes 0.182026-10-01T04:51:56ZGabriel Schneidergbrls@0x4200.cafe2026-10-01T04:51:56Zurn:sha1:62ed30e879693ece34c517f238fd6ee1ec3a114b
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Release Pardes 0.172026-10-01T03:59:37ZGabriel Schneidergbrls@0x4200.cafe2026-10-01T03:59:37Zurn:sha1:a909c1728315b6199cb9fca8eb3e9c4bb217ce7b
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>