Tree:
72d6815492
master
v0.1a
${ noResults }
168 Commits (72d6815492640aae48d3ebc658a50c51ed9c470a)
| Author | SHA1 | Message | Date |
|---|---|---|---|
 Julian Noble
|
72d6815492 |
make.tcl overview: interim single-column layout (cell-wrap revisit noted)
User decision: -choicecolumns back to 1 on both the rich overview and the lean dispatch definition - the long one-line summaries make the two-column layout too wide without table cell wrapping. Revisit -choicecolumns 2 when cell wrapping is implemented; the '## FORM' header-line dropping in the overview cells is accepted. Max render width drops ~250 -> 168 columns. help/colour suites 16/16. punkshell 0.41.2. Assisted-by: harness=claude; primary-model=claude-opus-5[1m]; api-location=anthropic.com |
1 week ago |
|
Julian Noble
|
157715fadd |
make.tcl: rich top-level help overview (i info style)
User-directed: bare 'make.tcl help' / 'make.tcl' / '-help' now render a per-subcommand overview in the repl 'i info' style - each cell carries the subcommand's one-line summary plus its own auto-generated bracket-notation synopsis line(s) (per-form lines for the multi-form tool/buildsuite), grouped by the existing SUBGROUPS in two columns. Stock punk::args machinery: rich -choicelabels built at lazy resolve time by ::punkboot::argdoc::overview_labels (punk::args::synopsis -noheader per subcommand id, '## FORM' headers dropped for compactness) + -choicecolumns 2, hung on the separate (script)::punkboot.overview id so the label cost (a few ms) is paid only when the top-level help renders - the lean (script)::punkboot definition remains the dispatch surface (now also -choicecolumns 2 for its unknown-subcommand error render) and is deliberately excluded from the capability probe. Plain PUNKBOOT_PLAIN fallback help unchanged. maketclhelp toplevel test updated to pin the overview (tool/buildsuite form synopsis lines, ESC-free); help/colour/bakelist suites 25/25. punkshell 0.41.1. Assisted-by: harness=claude; primary-model=claude-opus-5[1m]; api-location=anthropic.com |
1 week ago |
|
Julian Noble
|
13dc78191b |
make.tcl: restricted kitname choices + single memoized mapvfs read
User-directed pair completing the declaration-authoritative arc: 1. bake/bakelist kitnames are -choicerestricted 1: the configured kit names + @group selectors (computed into the declaration when the mapping parses at definition time) now GATE at dispatch - an unknown kit is a punk::args choice error before any build (the no-build guarantee moves to the gate), unambiguous prefixes resolve to canonical names, and the dry-run help mirrors the verdict ('help bake punk91 -confirm 0' is now a pointed choice error naming '-confirm' rather than a value-swallow). The handlers' own validation remains the backstop where declarations cannot gate: PUNKBOOT_PLAIN degraded mode, and a definition-time mapping-read failure (the choices clause is omitted entirely - no gate). 2. All kit-mapping consumers share ONE memoized parse per invocation: new punkboot::lib::mapvfs_model front door (locate+parse, keyed by resolved mapfile/rtbase/sourcefolder/target - a PUNK_MAPVFS_CONFIG override keys its own entry) consumed by the definition-time choices (now passing the real bin/runtime rtbase - mapvfs_parse is pure, so the models are interchangeable) and the five handler sites (selective bake, main bake, bakelist, vfslibs %platform% derivation, check smoke listing). Success is cached; a throwing parse deliberately is not, so each caller diagnoses fresh. maketclbakelist.test: unknown-name pins updated to the dispatch choice error + PUNKBOOT_PLAIN backstop variants (runner gains envoverrides); maketclhelp.test: bake dry-run case now pins the choice-gate rejection and 'help bake punk91' acceptance. Full punkexe subtree 114 tests, 0 fail. punkshell 0.41.0. Assisted-by: harness=claude; primary-model=claude-opus-5[1m]; api-location=anthropic.com |
1 week ago |
|
Julian Noble
|
120ca92b49 |
make.tcl help: pure dry-run + received-args report (G-143 follow-up)
User-directed: the bare-action carve-out is removed - 'make.tcl help
buildsuite build' now falls through to the punk::args diagnosis like any
other incomplete line (exit 1). Empirically the error table carries the
form's argument rows alongside the all-forms synopsis and per-form
reasons, so the diagnosis IS the form documentation and the carve-out
bought nothing but an exit code. Accepted lines gain a one-line
received-args report after the usage table, built from the parse
result's 'received' key (defaulted opts omitted):
'dry-run: line accepted (form build) - action = build | -test = 0 |
toolname = punkzip'. This makes punk::args' value-swallow visible
('kitname = punk91 -confirm 0' for the out-of-order line) - an interim
make.tcl-side clue until punk::args grows an annotated success render.
bake/bakelist kitname -choicerestricted 0 confirmed deliberate
(discoverability-only choices; handler validation authoritative;
define-time mapvfs parse best-effort). maketclhelp.test pins updated;
help/tool/colour suites 20/20. punkshell 0.40.3.
Assisted-by: harness=claude; primary-model=claude-opus-5[1m]; api-location=anthropic.com
|
1 week ago |
|
Julian Noble
|
0604d23729 |
make.tcl help dry-runs the supplied command line through the declaration (G-143 follow-up)
User-directed second refinement: 'make.tcl help <subcommand> ?arg ...?'
(and '<subcommand> ?arg ...? -help') now parses the words through the
subcommand's punk::args declaration exactly as dispatch would (form
auto-selection, choice prefixes). An accepted line renders the matched
action's single-form usage ('help tool build -test 0 punkzip', 'help
tool bu'); a rejected line (unknown action, option-first line, unknown
flag in option position) emits the same punk::args noformmatch
diagnosis dispatch gives - all-forms synopsis plus per-form reasons -
on stderr, exit 1: the error table is itself usage documentation, so
help verdicts mirror dispatch verdicts. Carve-out: a bare action
word/prefix renders its form directly, since forms with required
values ('buildsuite build' needs a suitename) would fail a strict
dry-run. Replaces 0.40.1's leading-word-only selection and option-first
whole-usage fallback. Empirical punk::args findings recorded (archived
goal file + probes): flag-like words at/after the first value position
parse as values; -regexprepass/-regexprefail are honoured during form
matching (a '-regexprefail {^-}' value disambiguates flag-led lines -
the punk::auto_exec::hash latent selection ambiguity). maketclhelp.test
pins updated; help/tool/colour suites 20/20. punkshell 0.40.2.
Assisted-by: harness=claude; primary-model=claude-opus-5[1m]; api-location=anthropic.com
|
1 week ago |
|
Julian Noble
|
77485434ef |
make.tcl help takes the subcommand's own command line (G-143 refinement)
User-directed reshape of the help-depth interface: 'make.tcl help
<subcommand> ?arg ...?' (and '<subcommand> ?arg ...? -help') now accepts
the subcommand's command line as typed and renders the usage most
specific to it - 'make.tcl help tool build -test 0 punkzip' shows the
tool build form via its leading action word (the same literal-leader
word punk::args form auto-selection keys on at dispatch). Words after
the action are tolerated and ignored; an option-first line falls back to
the whole subcommand's usage; single-form subjects (e.g 'help bake
punk91 -confirm 0') ignore the words entirely; an unknown action word on
a multi-form subject stays a pointed exit-1 usage error. The 0.40.0
numeric form-index acceptance ('help tool 2') is withdrawn as
unintuitive. Probe recorded: the repl 'i' (punk::ns::cmdhelp) tolerates
trailing argument words but does not form-narrow flat multi-form
commands - make.tcl help now exceeds it there. maketclhelp.test pins
updated (12 tests, all green + tool/colour suites); punkshell 0.40.1.
Assisted-by: harness=claude; primary-model=claude-opus-5[1m]; api-location=anthropic.com
|
1 week ago |
|
Julian Noble
|
6711302b5c |
G-143 achieved: make.tcl multi-form tool/buildsuite definitions + per-action help depth
tool/buildsuite/help are multi-form punk::args definitions (one @form per action, literal single-choice action leaders + choicelabels): 'help tool' / 'help buildsuite' render one synopsis line per action; 'make.tcl help <subcommand> <action>' - equivalently '<subcommand> <action> -help', or a 0-based form index ('help tool 2') - renders that action's single-form usage; unknown actions and out-of-range indexes are pointed punk::args errors (exit 1). tool dispatch parses through the definition (unknown actions/flags: exit-1 usage errors). User-approved deviation: the declared positional model puts options before tool names ('tool build -test 0 <name> ...'); docs/agent guidance updated to match and a flag-shaped tool name earns a stderr hint; the PUNKBOOT_PLAIN degraded scan keeps the historic flag-anywhere parse. buildsuite driver-arg passthrough unchanged. G-144 consumer follow-through: all 17 per-subcommand @form -synopsis overrides retired (automatic @cmd -name bracket-notation synopses; top-level make.tcl override kept deliberately). Pinned by new punkexe/maketclhelp.test (12 tests); full punkexe subtree green (114 tests, 0 fail). Goal flipped to achieved and archived; punkshell 0.40.0. Assisted-by: harness=claude; primary-model=claude-opus-5[1m]; api-location=anthropic.com |
1 week ago |
|
Julian Noble
|
543dd1df40 |
G-144 achieved: punk::args auto-synopsis @cmd -name fallback for (script) ids
Auto-generated synopses for (script)-prefixed constructed definition ids now lead with the declared @cmd -name (the invocation name - e.g 'make.tcl modules' for (script)::punkboot::modules) instead of the raw id, across punk::args::synopsis full/summary renders and the usage/arg_error synopsis sections that funnel through them (punk::args 0.12.7 -> 0.13.0). The fallback is gated on the (script) prefix only (user-confirmed narrowing): (autodef) ids are excluded - their id-minus-tag IS the invocation (arg_error's existing tag strip surfaces it) while their @cmd -name is a display label ('destroy', 'Object: ::x'); (shared)/(package)/(default)/(widgetcommand) doc ids declare no invocation-style -name today and render the id as before. Real command ids render byte-identical (full args testsuite passes with no pin updates). Six new synopsis.test pins: (script) full/summary/multiform/ surfaces, no-name constructed id, autodef unchanged, real-id byte-identical. Goal tiers: index entry archived to GOALS-archive.md, detail file moved to goals/archive/ (G-143's pending-tense Related line rewritten to reflect achievement + follow-through pointer: make.tcl's @form -synopsis overrides are now optional, retirement under G-143 or a follow-on edit). Docs: @form -synopsis directive known-case text, synopsis -help, src/modules/AGENTS.md ergonomics bullet. Project 0.39.5 -> 0.39.6 (patch: latent capability, no shipped definition renders differently today). Bootsupport refresh (make.tcl modules + bootsupport): snapshot gains args-0.13.0.tm and picks up two lagging source syncs - mix/commandset::layout 0.2.0 -> 0.3.0, mix/commandset::project 0.4.0 -> 0.4.1, and the shellfilter 0.2.4 G-145 comment; superseded copies pruned. Assisted-by: harness=opencode; primary-model=openrouter/moonshotai/kimi-k3; api-location=openrouter.ai |
1 week ago |
|
Julian Noble
|
24bc272c93 |
G-012 achieved: inert VCS-config template payloads + layout gitignore refresh
Layout .gitignore payloads are now stored inert as gitignore.in and
materialized to .gitignore at project generation, ending the
self-censoring-template hazard (live nested .gitignore files silently
untracked template content in git while fossil managed it fine):
- punk::mix::commandset::layout 0.3.0: layout_materialize_renames map;
layout_stage_chain renames inert payloads after composing all overlay
layers (.anti markers keep targeting store names); layout_materialize
withholds the in-place fast path from inert-payload folders so the
store is never renamed; inert+live name conflict errors loudly
- make.tcl thin-layout sync refreshes every vendor/punk layout's
gitignore.in from the canonical repo-root .gitignore (punkcheck-tracked
sync_layouts events); workflow text updated
- the four vintage payload variants eliminated as drift: all payloads
now byte-identical mirrors of root; modpod copy carried and its stale
.gitignore pruned
- materialize.test +5 G-012 characterization tests (25/25; mix subtree
71 pass / 1 known skip)
Verified end-to-end: generation from each affected layout yields a
byte-identical .gitignore (othersample .anti case yields none); root
edits propagate to payloads via make.tcl libs in both directions; the
|
1 week ago |
|
Julian Noble
|
15a7e27acb |
G-145 achieved: ansistrip transform drops 'clear' op - remnant-free piped help
The Tcl core delivers 'clear' to a transform stack before EVERY write op
(output-buffer flush-down; tclIORTrans.c ReflectOutput, gated on METH_CLEAR -
intentional and test-pinned, but under-documented: doc/transchan.n and TIP 230
scope clear to seek/read-side buffers). ::punkboot::ansistrip::transchan's
clear handler ran 'dict unset carry $chanid', discarding a split sequence's
held ESC tail at every write-chunk boundary: the next chunk's ESC-less
remainder ('0;1m', '1m') passed through unstripped as orphan fragment text in
piped usage tables, while the dropped ESC bytes kept the G-113 zero-ESC pin
green. The transform is write-only with no read-side state, so its
contract-conformant clear is a no-op - it no longer declares the op, the
per-write calls never fire, and the split-sequence carry survives every chunk
boundary. finalize still drops an end-of-stream partial sequence. No change in
textblock/punk::ansi (pre-transform stream verified well-formed during
diagnosis; exonerated).
Pin: maketclcolour.test gains maketcl_colour_help_tables_zero_remnants -
piped 'tclsh src/make.tcl help' for all 20 declared SUMMARIES subjects plus
bare 'help'/'-help', asserting exit 0, zero ESC bytes, zero orphan fragments
(two-step detector: strip complete sequences, then scan for [0-9;]+m not
preceded by ESC or an open bracket - a raw scan false-positives on ';1m' inside complete
multi-parameter SGR). Failing-first evidence: pre-fix the pin names exactly
the 5 fragmenting invocations (help, -help, help libs, help info, help
bootsupport); the 22-invocation NO_COLOR+PUNK_FORCE_COLOR pre-transform
corpus carries no legit text matching the detector.
Verified (tclsh 8.7a6, win32-x86_64): all 22 invocations byte-level clean;
PUNK_FORCE_COLOR=1 output intact (1177 complete sequences, 0 orphans); the 3
G-113 colour-policy pins pass unchanged; shell/testsuites/punkexe family (15
files, 102 tests) and shell/*** (127 tests) green, warnings pre-existing only;
'make.tcl packages' verified, layout + modpod make.tcl copies synced by the
build (punkcheck-managed outputs batched here per the carve-out).
Bookkeeping: goal flipped active -> achieved 2026-08-01, detail file archived
(goals/archive/G-145-piped-usage-ansi-remnants.md), G-056 Notes gains the
exoneration pointer; src/AGENTS.md colour bullet + src/tests/shell/AGENTS.md
suite description updated; project version 0.39.4 (patch) with CHANGELOG
entry.
Assisted-by: harness=opencode; primary-model=opencode/kimi-k3; api-location=unknown
|
1 week ago |
|
Julian Noble
|
320add60ad |
make.tcl argdoc restyle: braced -& definitions with tstr placeholders; centred Descriptions
Convert every ::punkboot::argdoc punk::args::define to the braced file-style
block form established for the shell/tool definitions in 7ec7da4a: -& record
continuations, -summary/-help pulled from SUMMARIES/HELPTEXTS via tstr
placeholders, shared OPT_* option fragments interpolated as ${$OPT_...}
records. The SUBOPTS/OPT_SYNOPSES/VALUES_SYNOPSES/SUBVALUES tables and the
define loop are unrolled into 16 per-subcommand blocks; buildsuite/help and
the top-level definition follow the same idiom (help gains a HELPTEXTS entry;
the top-level body moves to argdoc's TOPLEVEL_HELP variable). @normalize is
dropped: -help bodies are display fields (punk::args G-046 deferred
expansion), so the HELPTEXTS block indentation now reaches the rendered
Description verbatim - every 'make.tcl help <subcommand>' Description renders
centred like 'make.tcl help tool' instead of left-aligned.
No parsing or interface changes: option/values specs, synopses, exit codes,
prefix resolution, unknown-flag/subcommand errors, PUNKBOOT_PLAIN degraded
mode and the capability probes all verified unchanged (before/after usage
captures diffed for all 20 help subjects; shell/tool byte-identical). Narrow
usage tables widen slightly to fit the indented Description.
KITNAME_CHOICEPART persists as a namespace variable because the kitname
records interpolate it at (lazy) definition resolve time.
Docs: src/AGENTS.md make.tcl bullets updated (braced-def contract, per-block
maintenance rule); src/modules/AGENTS.md G-045 subsection re-points the
@normalize exemplar to punk.tm only and records argdoc as the braced -& +
tstr exemplar. punkshell 0.39.3 + CHANGELOG entry. Thin-layout make.tcl
copies resynced via 'make.tcl libs -confirm 0' (basic, project-0.1, modpod
templates copy - byte-identical).
Assisted-by: harness=claude; primary-model=claude-fable-5; api-location=anthropic.com
|
1 week ago |
|
Julian Noble
|
7ec7da4a5a |
make.tcl tool build: stamp installed exe mtime so the bin freshness state can clear
tool_install_state judges bin/<tool>.exe current/stale by comparing its
mtime against the newest file anywhere in the tool tree. Two layers kept
that exe mtime frozen at the original link time: a zig cache-hit rebuild
does not rewrite the zig-out artifact, and Tcl's 'file copy -force'
preserves the source mtime (Windows CopyFile semantics). Once any
non-build-input file (e.g. PROVENANCE.md, edited after the G-128/G-135
doc passes) was newer than that link time, the tool was reported 'stale'
permanently - repeated 'make.tcl tool build' runs (field-observed
2026-08-01) reinstalled a timestamp-identical artifact and never cleared
it.
Fix: after the install copy, stamp the installed exe with the install
time ('file mtime $exe [clock seconds]'), making the state measure what
it means - installed since the last change in the tree. Also self-heals
mtime churn from branch switches/fresh clones on the next build.
punkshell 0.39.2 + CHANGELOG entry (patch: build-tooling bug fix).
Thin-layout make.tcl copies resynced via 'make.tcl libs -confirm 0'
(basic, project-0.1, modpod templates copy - byte-identical).
Assisted-by: harness=claude; primary-model=claude-fable-5; api-location=anthropic.com
|
1 week ago |
|
Julian Noble
|
7cb0dc1333 |
make.tcl: remove deprecated 'project'/'vfs' aliases (G-112 transition window closed)
The G-112 stage-true rename (0.23.0) kept 'project' -> bakehouse and
'vfs' -> bake as transitional aliases until the 8.6 arc completed; that
arc landed with G-099 and the new shape is settled. All alias plumbing
is removed from src/make.tcl: known_commands, the pre-dispatch mapping
switch, the punk::args definitions (SUMMARIES/HELPTEXTS/SUBOPTS/
SUBVALUES/VALUES_SYNOPSES), the "deprecated aliases" help group, the
plain-help block, the workflow-text mention and the degraded-mode
kitname collection. 'vfs' now gets an unknown-subcommand usage error;
'project' prefix-resolves to the read-only 'projectversion' check under
punk::args unambiguous-prefix matching (PUNKBOOT_PLAIN degraded mode
rejects both). Also trimmed a pre-existing trailing space (line 8990)
surfaced by git diff --check via the verbatim layout copies.
Docs: src/README.md, ARCHITECTURE.md, src/AGENTS.md (alias mentions
removed; also corrected the stale layout-sync trigger claim - the
thin-layout sync runs in modules/libs/packages/bakehouse, not
bootsupport), root AGENTS.md sync-step subcommand list ('project' ->
'bakehouse'). Historical records (CHANGELOG 0.23.0 entry, goals
archives, GOALS-archive) deliberately untouched.
punkshell 0.39.1 + CHANGELOG entry (make.tcl interface change - patch
per root AGENTS.md versioning policy).
Build outputs batched per src/AGENTS.md: thin-layout make.tcl copies
resynced via 'make.tcl libs -confirm 0' (basic, project-0.1, modpod
templates copy - all byte-identical to src/make.tcl), plus the lagging
store->modpod catch-up the same run performed (modpod layout copy:
mapvfs.config removed, mapvfs.toml + vfs/README.md added).
Assisted-by: harness=claude; primary-model=claude-fable-5; api-location=anthropic.com
|
1 week ago |
|
Julian Noble
|
d953ce5c07 |
G-127 increment 4: docs + records; punkshell 0.39.0
- bin/AGENTS.md: "Cross-target kit outputs (bin/kits/<platform>/)" contract section (locations, coexistence, ignore coverage, relocations + orphan note) - src/AGENTS.md: target-keyed output bullet beside the G-122 host/target material (locations, name-spanning selection, %platform% payload axis) - make.tcl workflow text per its update contract: [K9] output split, [K8] untracked additions, .vfs.toml %platform% line; bakelist helptext + report header name the kits/<platform>/ tier (both renders verified, width ok) - punkproject.toml 0.38.0 -> 0.39.0 + CHANGELOG entry (minor bump: new declaration capability + relocated cross-target outputs) - goal detail Progress: G-023 reconciliation stated (versioned outputs compose with the platform tier; release gate checks per-kit deploy dir), punkshell902 + verify-ix86 relocation records finalized, punk9linux migration decision recorded (declared where a governed tier exists, hand-curated tcltls/tdom/ thread drop-ins pending G-116/G-004) Assisted-by: harness=claude; primary-model=claude-fable-5; api-location=anthropic.com |
1 week ago |
|
Julian Noble
|
0d80cf6771 |
G-024 + G-115 achieved: mapvfs.toml kit mapping + declarative per-.vfs payloads (punkshell 0.38.0)
G-024 - kit mapping converted to toml (tomlish-parsed), same-day activation->achieved: - src/runtime/mapvfs.toml: [kit.<name>] / [group.<name>] / [scheme.<name>] tables (header comment = user-facing spec). All 12 active line-format mappings migrated (order preserved - identical bakelist rows vs the legacy parse); historical comments carried over; mapvfs.config deleted here AND in the project-0.1 layout (converted to mapvfs.toml there). - make.tcl reader family under the unchanged G-121 model surface: mapvfs_locate (toml > legacy; PUNK_MAPVFS_CONFIG env override = characterization seam), mapvfs_parse dispatching on extension, strict mapvfs_parse_toml, deprecated mapvfs_parse_config fallback (un-migrated generated projects keep baking, with a NOTE; toml wins when both files exist). Shared toml helpers hoisted to punkboot::lib (inline toml_untag copies removed). - Strict entry-named validation: unknown keys/types/schemes, conflicting targets, duplicate outputs, parse failures fatal (exit 3 pre-build). Full-bake unresolvables split by nature: missing vfs folder (repo content) fatal per entry; missing runtime file (environment - part-populated stores are legitimate) entry-named recapped BUILD-WARNING + skip. - Groups: @<group> selection in bake/bakelist, group/default/scheme notes columns + detail lines, groups listed in errors and argdoc choices. verify-ix86 wraps the G-129 kits; bake_default=false excludes entries from full bakes. - G-023 versioned scheme declared live ([scheme.punk8]/[scheme.punk9], group versioned, bake_default=false pending G-023 activation): parse-time expansion from punkproject.toml to punk<gen>-<version> / punk<gen>-dev / release-gated punk<gen>; kit loop creates the plain name when absent and never overwrites it on a normal bake. punk9-dev baked end-to-end; punk9 gate skip demonstrated. - Consumer dedupe fix: several entries may share one runtime on one vfs - kit-output enumeration and the kit-loop runtimes list process each distinct runtime once (latent multiplication in the legacy format, unreachable there). G-115 - declarative .vfs composition with drop-in preservation, same-day activation->achieved: - Sibling declarations src/vfs/<name>.vfs.toml ([payload.<name>]: source, source_root src|packages, target, supersedes, replace) materialized by the new VFSPAYLOAD phase (bakehouse/bake/vfslibs; selective bakes narrow; strict entry-named errors). Drop-in-wins precedence via punkcheck -overwrite synced-targets: undeclared or modified-since-install files preserved and reported (changed-but-preserved NOTE = files_skipped minus sources_unchanged); supersedes/replace are the explicit exceptions; an undeclared .vfs is untouched. - vendorlib_vfs.toml relationship settled: folded/superseded - all six install entries migrated into per-.vfs files for the 13 participating folders (first materialization run reproduced every payload with zero tracked-content change); file deleted; a leftover copy warns and is ignored. - Demonstration kit punkdeclare (group declare-demo, bake_default=false, smokerequire udp+tcllibc): VCS carries only the boot fauxlink; the whole lib_tcl9 payload (tcludp vendor tree + tcllibc consent-gated packages tier) rematerializes from the declaration - clean-tree rebuild smokes pass in-kit, and drop-in survival was proven live (a foreign pkgIndex.tcl and an undeclared note both survived; the foreign file genuinely won until removed, caught by the G-133 smoke probe as designed). Payload ignored in both VCS (.gitignore + derived ignore-glob; ignore-sync verification clean). - Docs: src/vfs/README.md (format + precedence spec), src/vfs + src/runtime + src + bin AGENTS.md, ARCHITECTURE.md (+lint clean), workflow text and summaries/helptexts/argdoc; layout store seeds the convention (vfs README + mapvfs.toml; machinery travels via the established make.tcl sync). Verification: legacy-vs-toml identical kit rows; punkluck86 force-rebuilt under both formats content-identical (identical 1.73MB runtime+stamp prefix, identical 2993-member payloads; whole-artifact byte identity is not a pipeline property - consecutive same-config rebuilds differ in zip timestamp metadata, pre-existing); punkexe suite 98 tests 0 failures incl. 5 new G-024 characterizations (maketclbakelist.test: @group filter + PUNK_MAPVFS_CONFIG fixture tests) and updated pins (payloadcheck declared list + punkdeclare; platform detail wording). Goals: both flipped achieved 2026-07-31 with evidence in the detail files; GOALS.md entries moved to GOALS-archive.md; live-tier reference sweep applied (G-004/G-019/G-023/G-025/G-026/G-035/G-065/G-127/G-131/G-137/G-141) including the G-127 mapvfs-seam retirement note (PUNK_MAPVFS_CONFIG) and a direct G-137<->G-127 bridge replacing the archived G-024 one. Detail files move to goals/archive/ in the follow-up pure-rename commit. Also included as found: the developer's concurrent G-142 goal addition (GOALS.md entry + goals/G-142-punkbin-listing-manifests.md), committed unmodified for git/fossil coherence - not agent-authored. Assisted-by: harness=claude; primary-model=claude-fable-5; api-location=anthropic.com |
1 week ago |
|
Julian Noble
|
a71d7bd377 |
buildsuite artifact identity fields default from punkproject.toml (punkshell 0.37.0)
G-117/G-123 lineage: project, project_url and packager in emitted punkbin-artifact.toml records (runtime family + library tier, both suites) now default from the enclosing project's punkproject.toml [project] table via a bounded walk-up from the build root (common.punkprojectField) - a copied suite reads its own project's manifest. Chains: project: -Dproject > manifest name > "punkshell" (was hardcoded); project_url: -Dprojecturl > manifest url > "unrecorded"; packager: -Dpackager > PUNKBIN_PACKAGER > manifest packager > git identity > fossil user default (new fossil-only fallback) > "unrecorded". Verified end-to-end: suite_tcl90 kit-family + library emissions and the suite_tcl86 library emission carry the manifest values, -D overrides win, family_check green. Root punkproject.toml gains an active url; dev project.new's seeded manifest now carries commented url/packager placeholders documenting the consumption site (punk::mix::commandset::project 0.4.1). Assisted-by: harness=opencode; primary-model=huggingface/moonshotai/Kimi-K3; api-location=unknown |
1 week ago |
|
Julian Noble
|
670ab78dd8 |
G-123 increment 5: consent surface + e2e demonstration + acceptance walk (punkshell 0.36.0)
fetch now surfaces the artifact's build-origin class in its report
("provenance: class=<c>" after sidecar retrieval, table-aware read,
both payloads, re-wrapped, pinned in runtimecmd_provenance.test) -
the acceptance's consent-surface clause.
END-TO-END DEMONSTRATION (live, fully reverted): fixture punkbin
served via httpfixture with a suffixless testplat-x86_64 runtime; the
committed bin/punk-runtime.cmd fetched fakert-suite-r1 (-platform +
-trust-server; sha1 ok, sidecar + provenance line) and 'use -platform'
materialized the working copy; a temporary mapvfs entry wired it to a
bakelist row: runtime file bin/runtime/testplat-x86_64/fakert-suite
(present), target=testplat-x86_64 - with the .exe-named negative half
(runtime=missing) captured first. Config reverted, tier removed, tree
verified clean. Live canonical evidence also captured: unattended
fetch of tclsh9.0.5-punk-r2.exe from the real origin ran gate-free
(published r2 hash matched).
Full battery: provenance 5/5, freshness + checkfile 12/12,
runtimebash_wsl PASS, dtplite 8/8 (after its separate 0.35.1
pre-existing-fallout fix), roundtrip byte-identical under the punk
baseline runner. Acceptance walk recorded item-by-item in the goal
Progress - every item evidence-satisfied; user-gated follow-throughs
(punkbin push + classification review; index Scope ps1-path
correction) recorded as outside acceptance.
Assisted-by: harness=claude; primary-model=claude-fable-5; api-location=anthropic.com
|
1 week ago |
|
Julian Noble
|
e26489fd57 |
dtplite fallback: deployed lib_tcl9 tree replaces the retired vendored tcllib (punkshell 0.35.1)
G-139's vendored-tree removal (
|
1 week ago |
|
Julian Noble
|
643ac3aa37 |
G-123 increment 4: provenance/gate characterization + bare-name sidecar fetch (punkshell 0.35.0)
Payload fix, both payloads (found designing the coverage): fetch only retrieved beside-toml sidecars for -r<N> family names, so a fetched bare-named third-party artifact arrived WITHOUT the retroactive record its class display depends on. fetch now tries the sidecar for ANY name; absence stays tolerated (no-basis degradation unchanged). bin/punk-runtime.cmd re-wrapped. NEW binscripts suite runtimecmd_provenance.test (5 tests, both payload routes, httpfixture non-native tier testplat-x86_64: suite-built -r1 family artifact + bare third-party artifact with a retroactive-style schema-v2 sidecar + a record-less artifact): server-trust gate refusal from the non-canonical origin (canonical url + -trust-server named, no artifact lands), -trust-server flag fetch including sidecar retrieval, PUNKBIN_TRUST_SERVER=1 unattended form, class= tags in list -remote + plain list (third-party tagged, suite-built quiet, record-less untagged), and info schema-v2 fields + provenance_class row with the flat 'class = runtime' ordering-caveat pin and record-row parity between payloads. All green: new suite 5/5; pinned checkfile + freshness 12/12; roundtrip byte-identical under the tclsh9.0.5-punk baseline runner. Assisted-by: harness=claude; primary-model=claude-fable-5; api-location=anthropic.com |
1 week ago |
|
Julian Noble
|
bc22def1dc |
G-123 increment 3: punk-runtime server-trust consent gate (punkshell 0.34.0)
fetch - the executable-retrieving act - now gates at entry when the origin is non-canonical (PUNKBIN_URL override): the explicit -trust-server flag or PUNKBIN_TRUST_SERVER=1 env (unattended form) is the acknowledgement, one vocabulary with make.tcl libfetch's gate (G-139). The canonical punkbin origin gates nothing (G-058's unattended tclsfe constraint fetch is unchanged); metadata-only actions (list -remote, platforms -remote) never gate; never an interactive prompt. Refusal wording mirrors libfetch's. ps1: -trust-server is pre-scanned before parameter binding (hyphenated dynamic-parameter names are fragile under the PS 5.1 binder the windows wrap routes through). Trap recorded in the goal Progress: a ps1 comment line must never BEGIN with the word 'requires' - PowerShell parses '#requires ...' as the #Requires statement and fails the whole file (found by the powershell.exe 5.1 smoke). runtimecmd_freshness.test: env seam gains PUNKBIN_TRUST_SERVER=1 + the matching teardown unset (the fixture is a non-canonical origin, so its fetch-driving tests need the unattended acknowledgement) - flagged as a test-contract edit in the goal Progress; every existing assertion unchanged and passing. Gate refusal characterization lands with the increment-4 fixture coverage. Payload growth pushed template label :exit_multishell onto a 512-byte cmd label-scan boundary - checkfile caught it as designed; fixed via a +16-byte spacer line (the documented knob). bin/punk-runtime.cmd re-wrapped. Verification: gate smokes in both payloads (refusal + both acknowledgement forms; ps1 under powershell.exe 5.1); checkfile 0 ERROR; pinned tests all PASS (checkfile + freshness 12/12, roundtrip byte-identical under the tclsh9.0.5-punk baseline runner). Assisted-by: harness=claude; primary-model=claude-fable-5; api-location=anthropic.com |
1 week ago |
|
Julian Noble
|
2af2364947 |
G-123 increment 2: punk-runtime provenance display + bash info -platform fix (punkshell 0.33.0)
Both payloads gain a TABLE-AWARE [provenance] class reader
(provenance_class() awk helper / Get-PunkRuntimeProvenanceClass) - the
flat first-wins field scans deliberately keep reporting [artifact] class
per the documented schema-v2 ordering caveat. Display: compact
class=third-party|local tag in the plain-list metadata summary and in
list -remote's sparse identity column (suite-built stays quiet; LOCAL
record basis only - a listing does no per-row remote fetches, toml-less
and remote-only rows keep degrading as no-basis rows); info gains the
v2 fields builder/source_url/upstream_ref/retrieved in the flat table
plus a derived provenance_class row, disagreement-checked; help texts
updated to the v1/v2 wording.
bash defect fixed: 'info -platform <p>' was advertised in both help
texts but silently ignored - 'info' was missing from the option-scan
action list (punk-runtime.bash case arm). The per-name report now
honours the tier argument like the ps1 payload (proven by the
tier-named not-found path).
bin/punk-runtime.cmd re-wrapped via the documented multishell
invocation under punk905. Verification: bash -n clean; staged v2
third-party fixture smoke shows parity output in both payloads
('[tcl=9.9.9 class=third-party]' + provenance_class row);
runtimecmd_checkfile PASS (no 512B label crossing after payload
growth); runtimecmd_freshness PASS (13 fixture tests, both routes);
runtimecmd_roundtrip PASS byte-identical under the tclsh9.0.5-punk
baseline runner (plain native tclsh lacks struct::stack for tomlish -
runner-choice trap noted in the goal Progress).
Assisted-by: harness=claude; primary-model=claude-fable-5; api-location=anthropic.com
|
1 week ago |
|
Julian Noble
|
4f753264b2 |
punk-runtime: strip CR before sha1sums lookups in the bash payload (punkshell 0.32.3)
fetch and list's local-file comparison anchored runtime names at end-of-line without stripping CR first, so a CRLF-published server sha1sums.txt aborted fetch with "Unable to locate hash" on strict grep platforms (GNU and BSD grep; cygwin grep tolerates CRLF, and the ps1 payload is immune via Get-Content line splitting). Both lookups now share the tr -d '\r' form list -remote already used. bin/punk-runtime.cmd regenerated via the scriptwrap machinery; the re-wrap invocation in src/scriptapps/AGENTS.md is corrected to the working -force 1 -askme 0 form found during regeneration. Assisted-by: harness=opencode; primary-model=huggingface/moonshotai/Kimi-K3; api-location=unknown |
1 week ago |
|
Julian Noble
|
7fc15cd5ad |
G-140 achieved: linux lib-tier emission - cross-built tcllibc for punk9linux (punkshell 0.32.2)
suite_tcl90: critcl_zig.config gains the linux-x86_64-zig cross target (zig cc -target x86_64-linux-gnu.2.17 - the dynamic-glibc floor decision, recorded in the goal file and fed to G-105). build905.zig gains the tcllibc-linux step (critcl driver on the windows suite shell, stubs linkage, no linux host; uuid.tcl excluded matching upstream's own native-linux package shape) gated by the new tools/elf_gate.tcl (G-133-equivalent ELF64/LSB/ET_DYN/x86_64 structural check - the load-smoke stand-in for cross lanes). library-artifacts emits lib/linux-x86_64/tcllibc-tcl9-r<N>.zip in the same run: target = linux-x86_64, build_host_platform = win32-x86_64, full provenance set, deliberately no [tests] section. build_id seeds now include the tier path so same-named artifacts in different target tiers get distinct identity digests (both suites' tool copies kept byte-identical). Step lists (build905.zig -Dsteps default, suite.tcl) and README document the lane and invocation. Consumption: libpackages.toml declares tcllibc-tcl9-linux (target linux-x86_64); vendorlib_vfs.toml install.tcllibc_tcl9_linux switches punk9linux.vfs/lib_tcl9 - replace retires the hand-dropped tcllibc, supersedes removes the stray tcllibc2.0 - so the last provenance-less accelerator hand-drops in any kit payload are gone. punkshell902 baked (G-133 payload arch scan: 6/6 match target linux-x86_64) and WSL-verified: materialized-tier accel smoke plus baked-kit smokes (default paths + a pinned variant proving the kit's own lib_tcl9 artifact tree loads with the accelerator engaged). Finding recorded: the tclkit-902 runtime bundles its own lib/tcllibc2.0 which wins default precedence - the origin of the retired stray folder name. Determinism: back-to-back re-emission byte-identical for all three artifacts; the zig Run session-env cache finding extended (FOSSIL_HOME and invocation mode churn the key; the linux ELF .so observed bit-stable across critcl re-runs, unlike the windows PE). Goal flipped achieved with evidence in the detail file; G-105 notes carry the libc decision as prior art. Docs: buildsuites AGENTS.md child index, ARCHITECTURE.md buildsuites paragraph, CHANGELOG + punkproject 0.32.2. Assisted-by: harness=claude; primary-model=claude-fable-5; api-location=anthropic.com |
1 week ago |
|
Julian Noble
|
d50115517f |
libfetch characterization test + isolation seams (punkshell 0.32.1)
make.tcl libfetch gains env-only isolation seams - PUNK_LIBFETCH_CONFIG (alternate declarations file) and PUNK_LIBFETCH_PACKAGES (alternate tier root), the PUNK_RUNTIME_PLATFORM idiom - so fixture/mirror runs never read the real declarations or write the real bin/packages tier. Neither seam bypasses the -trust-server consent gate. New shell/testsuites/punkexe/maketcllibfetch.test (piped maketcl* harness + testsupport/httpfixture.tcl + a file:// mirror; the canonical origin is never contacted) pins the G-139 behaviours that previously had only manual session evidence: the server-trust gate refusing non-canonical origins with exit 3 before any network access or tier write, -serverurl-over-PUNKBIN_URL precedence, consented fetch/verify/materialize (sha1sums + zip + sidecar + installed-shape tree with embedded record), idempotent re-run vs -force, declared-revision-change re-materialization keyed to the tree record (with the sidecar-not-listed note lane), sha1-MISMATCH rejection without residue, missing-from-server-sha1sums failure, and the no-config no-op. 8/8 pass in both runner modes (singleproc + -jobs); full punkexe subtree green (93 runnable, 0 failed); the real bin/packages tier verified untouched. Also refreshes the stale src/runtime/libpackages.toml header (r1 IS published as of 2026-07-30; the G-139 reference now points at the archive) and documents the seams there and in the libfetch help text. Assisted-by: harness=claude; primary-model=claude-fable-5; api-location=anthropic.com |
1 week ago |
|
Julian Noble
|
47cfa58520 |
G-139 achieved 2026-07-30: flip + archive record + reference sweep (0.32.0)
Acceptance met in full (walk in the detail file's Progress): libfetch with sha1 verification + beside-toml metadata + server-trust consent proven LIVE against the canonical origin post-publication; PACKAGES_tcl<N> libs-step materialization with punkcheck provenance; vendorlib_vfs.toml declarations (source_root=packages + replace) with supersedes removing tcllib1.21 incl. nested tcllibc 0.4 + md5c from the 8.6-family kits; punk86 kit-hosted runtests vs the same-day baseline IDENTICAL (zero differences to disposition); vendored trees removed in their own commit after the proven bake ( |
1 week ago |
|
Julian Noble
|
e7e2ec81f4 |
G-139 phase 2 increment 1: PACKAGES_tcl<N> deployment + vfslibs packages-root declarations (0.31.1)
make.tcl libs/packages/bakehouse gain the PACKAGES_tcl<N> phase: materialized punkbin lib-tier packages under bin/packages/<target>/tcl<N>/ deploy into lib_tcl<N>/allplatforms + lib_tcl<N>/<this platform> with punkcheck provenance, mirroring the VENDORLIB_tcl<N> shape (embedded punkbin-artifact.toml records ride into the deployed trees; absent tier = silent per-item no-op with a libfetch pointer). vfslibs declaration schema extended (vendorlib_vfs.toml header documents both): source_root = "packages" resolves an entry's source against the bin/packages tier (missing source errors name 'make.tcl libfetch'); replace = true clean-slates a SAME-NAMED existing target folder before install (supersedes deliberately cannot name the folder being installed - needed because the pre-G-139 tcl9 hand-drops nested md5c + tcllibc INSIDE tcllib2.0). Declarations added for the switchover set (per-kit, never blanket - the trees that carried hand-drops at 2026-07-30): 8 win32 tcl9 kit vfs trees get tcllib2.0-tcl9 + tcllibc-tcl9 (supersedes the stray-named tcllibc2.0); punk9linux (linux target) gets only platform-neutral tcllib2.0 (its linux-built tcllibc siblings stay - a linux lib-tier emission is future work); punk86/punk8win/punk8_statictwapi upgrade tcllib1.21 (+nested tcllibc 0.4 + md5c, removed with it via supersedes) to tcllib2.0-tcl8 + tcllibc-tcl8; punk86bawt refreshes its bare tcllibc only. Verified: make.tcl libs deployed the tier into lib_tcl8|9 (allplatforms + win32-x86_64) with embedded records present; vendored mirrors coexist in the deployed trees until the sequenced retirement. Thin-layout make.tcl sync copies updated by the run (punkcheck-managed outputs, batched per the carve-out). Project 0.31.1 + CHANGELOG. Assisted-by: harness=claude; primary-model=claude-fable-5; api-location=anthropic.com |
1 week ago |
|
Julian Noble
|
11200dbf8a |
G-139 phase 1: make.tcl libfetch + bin/packages input tier (project 0.31.0)
New 'make.tcl libfetch' subcommand (argdoc'd; workflow text + plain help updated): fetches the punkbin LIB-TIER artifacts declared in the new src/runtime/libpackages.toml ([artifact.<name>] name/target; the four G-138 r1 artifacts, both generations) from <origin>/lib/<target>/ into the untracked bin/packages/<target>/ input tier - zip + .toml sidecar, each sha1-verified against the server's per-target sha1sums.txt (local tier copy kept, the punk-runtime store idiom) - then materializes each verified zip via punk::zip::unzip (crc-verified, mtimes restored, accelerator-eligible) into bin/packages/<target>/tcl<N>/<pkgfolder>/ (generation subdirs: both generations install identically-named folders). Idempotent: present artifacts re-verify and skip; a materialized tree is current while its embedded record names the declared artifact (the G-138 self-description payoff); -force overrides. Consent keyed to server trust (the G-123 posture): the canonical punkbin origin (punk-runtime raw-base convention) gates nothing; any other origin (-serverurl or env PUNKBIN_URL) refuses before network access without the explicit -trust-server flag - never an interactive prompt. file:// origins served natively (mirrors/fixtures); http(s) transport: Tcl http (+tls) when loadable, else curl, else PowerShell. The dispatch block sits before the build-command whitelist gate per the standalone-command pattern and exits itself. Emission-side amendment (both suites' library_artifacts.tcl): sidecar tomls now ride the per-tier staging sha1sums.txt, matching the punkbin repo's build_sha1sums.tcl coverage so staging-derived mirrors serve verifiable sidecars. Verified against a file:// mirror assembled from both suites' staging: trust-gate refusal, consented fetch of all 4 artifacts (sha1-verified), materialization of all 4 trees, fully idempotent re-run, and a package-load proof from the tier (md5 2.0.9 + tcllibc accel=1 under the plain family kit, ifneeded paths in bin/packages). bin/packages is already ignored in both VCS via the /bin/* rules. Docs: bin/AGENTS.md tier section, src/runtime/AGENTS.md libpackages.toml bullet. Project version 0.31.0 + CHANGELOG (new make.tcl subcommand = product surface). Remaining G-139 work recorded in the goal's Progress section. Assisted-by: harness=claude; primary-model=claude-fable-5; api-location=anthropic.com |
1 week ago |
|
Julian Noble
|
4aae330b31 |
G-134 achieved: flip, archive record - archive-relative output is a checked contract; punkshell 0.30.7
Implemented and achieved same day. punkboot::utils 0.6.0 gains kit_offsetstyle_report (argdoc'd thin classifier over punk::zip::archive_info), synced via modules + bootsupport (one stale dead-pid .punkcheck.lock removed per the recorded recovery) and promoted into _vfscommon. make.tcl: guarded getter, post-assembly probe beside the smoke-require block (recapped BUILD-WARNING only on a FILE-relative attached payload; plain/none/unreadable silent; NOTE on stale bootsupport), and the 'check' row (ACTIVE (advisory) + contract lines). Characterization: offsetstyle.test 7/7 (mkzip -runtime/-offsettype hermetic fixtures for plain/archive/file, none on text + metakit-magic binary, unreadable detail, src/_build baseline sweep - no kit image probes file); maketcl_check_offsetstyle_pin added (payload-check suite 4/4). Full utils suite 69/69, punkexe suite 81/0, live punkluck86 bake silent as the acceptance's baseline clause requires. Docs: src/AGENTS.md pin bullet with does/does-not-guarantee, bin/AGENTS.md deployed-kit section, ARCHITECTURE.md check-family bullet (architecture_lint clean). Reference sweep: no live-tier references beyond the index entry; goals_xref shows no pairs bridged solely by G-134. Archive move follows as a pure-rename commit. Assisted-by: harness=claude; primary-model=claude-fable-5; api-location=anthropic.com |
2 weeks ago |
|
Julian Noble
|
1f8af99805 |
G-128: punkbin publication route settled, staged and documented; punkshell 0.30.6
Distribution convention (recorded in src/tools/punkres/PROVENANCE.md): the tool publishes to the punkbin tools tier as win32-x86_64/tools/punkres-x86_64-windows-<ver>.exe (arch-os-version beside the zig archives; immutable), integrity via that tier's sha1sums.txt. 0.3.1 STAGED as punkbin commit 1b471fc (sha1 166cd176..., built by the tool step from vendored 0.3.1/f0af880, all other sha1sums SAME) - push is maintainer-gated per punkbin AGENTS.md, and agents never push. Route documented in bin/AGENTS.md; the seam's skip notice names both remedies (build from vendored source, or fetch the prebuilt artifact). Demonstrated in the program-files checkout: artifact placed as bin/punkres.exe reports 0.3.1 and make.tcl tool list reads it CURRENT against the vendored tree. Reproducibility note recorded: zig builds differ across invocations in the PE-header timestamp region - the published sha1 pins THE artifact. Goal Progress: increment 1f; the only remaining acceptance item is the maintainer push making the fetch URL live - achieved flip flagged for confirmation at that point. punkres.test 8/8 re-verified. Assisted-by: harness=claude; primary-model=claude-fable-5; api-location=anthropic.com |
2 weeks ago |
|
Julian Noble
|
c0263ff934 |
punk-getzig: spaced-path + Windows PowerShell 5 fixes (launcher and payload); punkshell 0.30.5
User-reported: './punk-getzig.cmd' from a spaced checkout failed with 'C:\repo\jn\program: The term ... is not recognized' while punk-runtime worked. Root cause: punk-getzig_wrap.toml still used the pwsh '-c' command-string launcher, which tokenizes a spaced script path - punk-runtime received the 'cmd.exe /c powershell ... -File' fix during G-119 (its wrap toml carries the comment) and punk-getzig never did. Adopting the same launcher surfaced three latent payload issues, all fixed in src/scriptapps/bin/punk-getzig.ps1: - Resolve-Path -RelativeBasePath is pwsh 7.4+ - replaced with the Push-Location + -Relative spelling (PS5-compatible) - Invoke-WebRequest without -UseBasicParsing NullReferences on PS5 (IE parsing engine); all three call sites fixed, with progress suppressed around the large distribution download (PS5 -OutFile slowdown) - Invoke-Expression "$zigexe version" splits spaced paths under ANY powershell (latent even on pwsh) - both sites now use the call operator Payload growth pushed :exit_multishell across a 512-byte boundary; the punk-runtime-style '(512B spacer)' comment added to punk-getzig.bash and the polyglot regenerated (checkfile ERROR-free). Verified in the 'C:/repo/jn/program files/punkshell' checkout: full fetch (community-mirror latency ordering, local punkbin mirror, minisign verify, extract, ZIG VERSION: 0.16.0) and the already-installed branch; main-repo argless run clean. Sources and regenerated bin/punk-getzig.cmd committed together per the wrap workflow (no roundtrip pin exists for punk-getzig - candidate noted in the completion report). Assisted-by: harness=claude; primary-model=claude-fable-5; api-location=anthropic.com |
2 weeks ago |
|
Julian Noble
|
0e0a64eb23 |
punk-runtime fetch: materialization hint when the working name is absent; punkshell 0.30.4
Closes the observed UX gap (G-128 detail, program-files checkout findings): fetch reports an active runtime, but bakes reference the suffixless WORKING name that only 'use' materializes - so a first fetch left bakelist reading runtime=missing with nothing pointing at the next step. Both payloads (ps1 + bash, parity wording) now print note: bakes and projects reference <working> - materialize it with: punk-runtime use <artifact> after fetching an -r<N> artifact whose working name is absent, and stay silent once it exists (the downloads-never-bind rule is unchanged - fetch still materializes nothing). Sources edited under src/scriptapps/bin and bin/punk-runtime.cmd regenerated via scriptwrap multishell (-force 1; checkfile ERROR-free, 4 normal possibly-bogus warnings) - committed together per the wrap workflow; runtimecmd_roundtrip byte-identity passes. New pin runtimecmd_fetch_materialize_hint in runtimecmd_freshness.test (11/11, ~+5s: hint present on both payloads via the fixture server, absent after use via the cmd route). Live-verified against the real punkbin server from the spaced test checkout: hint fires on the no-download path too, silent with the working copy present. Assisted-by: harness=claude; primary-model=claude-fable-5; api-location=anthropic.com |
2 weeks ago |
|
Julian Noble
|
c6d50b3b2a |
G-128: fleet renamed to PUNKSHELL via punkres; punkres 0.3.0 -drop-opaque-overlay for mingw stub symbol tables; punkshell 0.30.3
Forced fleet rebake (user-directed): all ten win32 kits now carry the embedded PUNKSHELL group icon via punkres, verified per kit (punkres list, sidecar icon_group=PUNKSHELL + embedding=punkres, extract-ico byte-identical spot check) and booting (incl. metakit punksys 8.6.13 / punk86 8.6.17 and punk902z 9.0.2). punkmagic excluded (runtime absent - keeps its old icon and _build cache); punkshell902 linux = not applicable. The first fleet pass surfaced a real shape: mingw-built runtimes (tclsh902z, tclkit-win64-dyn family) keep COFF symbol/debug tables AFTER their sections, the extraction head carries them, and punkres correctly refused those stubs as data-classified overlay. Measured against twapi-era artifacts: Windows' EndUpdateResource has always silently STRIPPED exactly those bytes at stub-stamp time (leaving PointerToSymbolTable dangling) and the kits work - so punkres 0.3.0 (upstream dca8634, re-vendored, 33/33) encodes the distinction: -drop-opaque-overlay DISCARDS expendable trailing data and zeroes the orphaned symbol-table header fields, vs -allow-opaque-overlay which RELOCATES a real self-locating payload; conflicting flags rejected; drop is inert on zip overlays. The seam's punkres arm passes -drop (twapi-arm-equivalent stub hygiene; bin/AGENTS.md documents it). Recovery note: a 10-minute foreground timeout killed the first 3-kit rerun during deploy - stale src/_build/.punkcheck.lock removed after a dead-pid check per the recorded procedure; the completion bake deployed the already-wrapped kits. Full punkexe suite 83 tests 0 failed; goals_lint + architecture_lint clean. Assisted-by: harness=claude; primary-model=claude-fable-5; api-location=anthropic.com |
2 weeks ago |
|
Julian Noble
|
bd65183061 |
G-128: seam names the kit group icon PUNKSHELL (icon-derived, both arms); punkshell 0.30.2
RT_GROUP_ICON name policy (user decision 2026-07-29): kit_icon_process derives icon_group from the icon FILE's uppercased rootname (RC convention) - punkshell.ico -> PUNKSHELL for the default icon and per-kit overrides alike - and passes it to whichever mechanism embeds. punkres arm uses set-icon -group (punkres 0.2.0); kit_icon_embed_twapi gains the matching optional group_override (name only - language adoption unchanged) so the arms stay in parity. Replaces the retained stub-lineage names (punk91 "SFE", suite kits "TCLSH"); already-deployed kits keep theirs until their next natural rebuild (the policy is not in the punkcheck source set). Sidecar gains icon_group (recorded whenever the icon file exists); bin/AGENTS.md format doc + ARCHITECTURE.md updated, architecture_lint clean. Live-verified on punkluck86 rebakes: punkres arm and twapi arm (TCLLIBPATH route) both emit RT_GROUP_ICON "PUNKSHELL" lang 1033, extracted icons byte-identical between arms; sidecar icon_group=PUNKSHELL; final deployed state punkres-stamped and booting. Full punkexe suite 83 tests 0 failed. Assisted-by: harness=claude; primary-model=claude-fable-5; api-location=anthropic.com |
2 weeks ago |
|
Julian Noble
|
5e9f1c75fc |
G-128: seam selects punkres when built; characterization tests; punkshell 0.30.1
src/make.tcl kit icon seam (G-057): mechanism selection is now punkres-first - kiticon::punkres_available (memoized; binary presence at bin/punkres(.exe) is the selection signal) + kit_icon_embed_punkres (exec wrapper; punkres verifies its own output). twapi arm unchanged as fallback; when NEITHER serves the skip notice combines both reasons and names the 'make.tcl tool build punkres' remedy. Sidecar embedding value gains 'punkres' (bin/AGENTS.md format doc, ARCHITECTURE.md kit-icon bullet updated; architecture_lint clean). Live-bake fallback matrix on punkluck86 (win32-ix86 = PE32 arm live): punkres present -> embedded via punkres, kit boots + smoke OK; punkres absent + twapi loadable -> via twapi; both absent -> unavailable with combined notice, embedding none (exact pre-G-128 degradation). Live parity: both arms' baked kits listing-identical (codepage host-ACP noise finding recorded in the goal detail) with byte-identical extracted icons. New src/tests/shell/testsuites/punkexe/punkres.test (7/7 under the tclsh9.0.5-punk runner; self-gates on bin/punkres + store fixture): overlay- unsafe refusal exit 3 with untouched input, consented file-relative shift with convention preserved, payload intact via punk::zip members + crc-verified extraction, icon replaced (extract-ico == punkshell.ico), idempotent re-stamp byte-identical, make.tcl tool info punkres provenance records. Full punkexe suite 83 tests 0 failed (no collateral). Goal detail: Notes findings (codepage noise, root-vendorlib twapi fragility, nested tclsfe payload, metakit move-safety -> G-101 pointer updated) + Progress increment 1 with the two remaining acceptance items (non-windows-host e2e, punkbin publication route) and one acceptance-wording flag for the user (metakit 'kit' shape vs 'zip overlay' clause). punkshell 0.30.1 + CHANGELOG. Assisted-by: harness=claude; primary-model=claude-fable-5; api-location=anthropic.com |
2 weeks ago |
|
Julian Noble
|
5a8a62547b |
G-057: kit icon step - resource sidecar every target, twapi PE embedding
Every kit a bake builds now gets a <kitname>.resources.toml sidecar (build folder + deployed beside bin/<kit>) recording the build-time icon choice: schema-1 flat TOML in the G-135 tolerant family - kit, target, kit_type, icon_origin default|override, icon_source (repo-relative), icon_hash, provenance source+hash lifted from the icon's assetorigin sidecar (the chain reaches the SVG master), embedding, embedding_status embedded|not-applicable|unavailable|failed, embedding_reason. Byte-stable (no timestamps, write-only-if-changed); named "resources" so the parked RT_VERSION follow-on extends the same file. Format doc: bin/AGENTS.md. win32-target kits get the icon embedded as RT_ICON/RT_GROUP_ICON via the twapi resource-update APIs - the tcl-sfe sfe-0.2.tm mechanism (Ashok P. Nadkarni) generalized: enumerate all existing icon/group entries via extract_resources, delete ALL, write ids 1..N plus one group under the first pre-existing group's name+lang (fallback 1/1033) in a single update transaction - idempotent by construction. Ordering is stub-first: a PER-KIT copy of the extraction phase's payload-free raw runtime prefix is stamped BEFORE payload attach, so the resource rewrite can never corrupt an overlay; all three assemblers (zipfs mkimg, zipcat/concat, sdx wrap) consume the stamped copy; runtime store originals untouched. ::punkboot::kit_icon_process is the single internal entry point with mechanism selection inside (a G-128 portable stamper becomes a substitution). Skip taxonomy: non-PE target / runtimeless .kit = NOT APPLICABLE; cookfs, missing icon, missing raw prefix, cross-host (non-windows process family), twapi-unloadable = UNAVAILABLE with distinct named reasons; embed errors = recapped BUILD-WARNING + status failed. Sidecar written in every case. Default icon joins the kit's punkcheck source set (icon changes rebuild kits); sha256 self-contained (bootsupport has sha1/md5 only), memoized per run. Verified (evidence in the goal detail file Progress): full bake embeds all 10 buildable win32 kits (zip x64+ix86, sdx metakit trio) with resource inspection byte-matching the source ico (punk91, punk86, punk91ix86 inspected; group name SFE preserved); scratch-copy boots for zip, metakit and a stamped concat image (androwish exe-path mount); override round-trip via punk9wintk903.vfs (solid ico embedded + recorded, then reverted); linux cross-target punkshell902 sidecar with not-applicable; mingw tclsh86 bake takes the concat path with the distinct twapi-unloadable notice; vendored twapi 5.0b1 arm proven under tclsh9.0.5-punk; double-restamp converges. punk9_beta embedded+built but deploy still blocked by the known bin file lock (stamped product waits in src/_build). make.tcl workflow text gains the icon step + sidecar output; punkshell 0.29.0 -> 0.30.0 (minor: kits now wear the punkshell icon; new shipped sidecar) with CHANGELOG entry. G-057 detail file: ordering + override + sidecar + seam decisions recorded with tcl-sfe attribution; cross-host known-shortcoming + G-128 remedy recorded per acceptance. Assisted-by: harness=claude; primary-model=claude-fable-5; api-location=anthropic.com |
2 weeks ago |
|
Julian Noble
|
768dab399f |
G-033: proj: package-mode scope prefix - visitor binary resolves dev/src against the cwd's project
punk_main.tcl: an optional proj: prefix on the package-mode string
(e.g 'proj:internal-src') rebinds the dev/src path blocks to the punk
project containing the current directory instead of the executable's
own project. Root discovery is ::punkboot::proj_root_find, a lean
boot mirror of punk::repo::find_project/is_project_root in Tcl
builtins only (git/fossil control markers via 'file exists', unwise-
path exclusion, ./src plus src/modules|src/vfs|src/scriptapps|
punkproject.toml candidate check; .git accepted as file OR dir for
worktree checkouts - recorded divergence). The proc stays defined
post-boot like the G-129 mount procs. The prefix sits outside the
ordered dash-list, whose block order remains the same-version
tie-break dial; an invalid mode remainder is not consumed (whole arg
falls through as subcommand/script as today); bare 'proj:' reads as
internal and warns. Discovery outcome is always reported on stderr
(detected root + effective precedence - never a silent rebind), a
failed walk-up warns and leaves dev/src unbound (no exe-relative
fallback rebind), and proj: without a dev/src block warns it has no
effect. Under proj:, dev resolves <root>/modules{,_tclN} and
<root>/lib{,_tclN} only; src feeds the discovered root into the
existing src-mode machinery unchanged (tm paths, auto_path, package
prefer latest, inline #modpod scanner). Exe-relative src/dev launches
are untouched.
punk::args::moduledoc::punkexe 0.1.1: the packagemode help text
drafted in the goal file is now live as the packagemode leader of
(script)::punkexe (@leaders max 2), with proj: wording in the @cmd
overview; rendering/parse wiring remains G-032.
Verified on a rebaked punk91 kit, standalone copy outside any source
tree: proj:internal-src from tomlish (fossil root, builds no
executable) loads tomlish 999999.0a1.0 from its src/modules with
kit-first ties; proj:src flips the ordering; from punkshell the
visited-root modpod scan registers 6 modpods; no-project and
no-root-using-block warnings fire and proceed; bin/punk91 src
unchanged.
Docs: bin/AGENTS.md launch-package-modes proj: contract,
src/vfs/AGENTS.md punk_main contract line, ARCHITECTURE.md entry
point. punkshell 0.29.0 (minor - new launch invocation) + CHANGELOG.
Claude-Session: https://claude.ai/code/session_0156PuejSCGjgeGb7jiABrDU
Assisted-by: harness=claude; primary-model=claude-fable-5; api-location=anthropic.com
|
2 weeks ago |
|
Julian Noble
|
df6642e189 |
punk::zip 0.3.0: punkzip extraction accelerator + parity suite (G-126); 0.28.4
The final G-126 acceptance clauses:
- punk::zip::accelerator (auto|none|<path>): auto probes
env(PUNKZIP_EXE) then a punkzip beside [info nameofexecutable];
resolution cached until reconfigured
- unzip hands whole-archive member WRITING to the accelerator only when
the call is one it serves identically (globs {*}, no excludes,
default -overwrite/-mtime/-verify, ascii member names) and re-stamps
mtimes with punk::zip's local-time convention afterwards, so the two
engines produce identical trees; preflight refusals, member selection
and the returned names always come from punk::zip's own reader (the
G-124 floor, unchanged); accelerator failure falls back to pure Tcl
silently; last_unzip_engine/last_accelerator_note expose the per-call
decision
- parity suite zipaccel.test (9 tests): identical names, bytes and
mtimes through both engines over the G-124 shapes and whole-kit
extraction of the file-relative tclsh90b4_piperepl.exe (841 members);
listings engine-independent; punkzip porcelain agrees with punk::zip
member dicts on every shared field (mtime excluded by design - UTC vs
local convention); pure path proven when the binary is absent;
selective calls proven to run pure. Implicit directories (kits store
no dir entries) are excluded from mtime comparison - neither engine
stamps them
- recorded benchmark (in the goal detail): accelerator fixed cost
~9-13ms, crossover ~15-20 members, material (>= x1.7) from ~50;
x2.3-2.6 at kit scale under a native tclsh and x3.8-4.9 kit-hosted
(punk91.exe 21.9s -> 5.7s) - the bake-from-kit case the goal
motivated
- full runtests at recorded baseline under tclsh9.0.5-punk: 1186/1211,
1 failed = the pre-existing exec-14.3 baseline
Assisted-by: harness=claude; primary-model=claude-fable-5; api-location=anthropic.com
|
2 weeks ago |
|
Julian Noble
|
d76a7907ea |
make.tcl tool subcommand: build vendored zig tools into bin/ (G-126); 0.28.3
New surface for the vendored first-party build tools under src/tools:
make.tcl tool list|info|build|test ?<toolname> ...? ?-test 0|1?
- discovery is convention-based (a src/tools directory carrying
build.zig is a tool), so the G-128 tool will appear with zero
make.tcl edits; per-tool records are read tolerantly from
build.zig.zon (version, minimum_zig_version floor) and PROVENANCE.md
(upstream, vendored commit)
- list/info report installed-binary state (current|stale|absent vs
bin/<name><exe-suffix>) and the resolved toolchain
- toolchain resolution: PUNK_ZIG=<path> overrides; otherwise
bin/tools/zig* is scanned and the LOWEST release satisfying the floor
wins - deterministic as newer toolchains get unpacked beside the
pinned one; dev builds count only when their base version exceeds
the floor
- build runs the tree's own 'zig build test' as an EXIT-CODE gate
(expected stderr warnings do not fail it), then builds ReleaseSafe
and installs to bin/; nothing installs when the gate or build fails;
-test 0 skips the gate; build caches are deliberately left in place
between runs for rebuild speed (git- and fossil-ignored)
- zig stays OPTIONAL: packages/bake never require the step; without a
suitable toolchain, list reports the state and build/test exit
nonzero with bin/punk-getzig.cmd guidance
- full subcommand accompaniments per src/AGENTS.md: punk::args argdoc
(SUMMARIES/HELPTEXTS/bespoke passthrough definition), SUBGROUPS
("vendored tools"), known_commands, degraded-mode dispatch, plain
help text, bootsupport-staleness exemption, workflow text (new
OUT-OF-BAND SUBSYSTEMS section also covering buildsuite), project
version 0.28.3 + CHANGELOG entry
- piped characterization (zig-independent):
src/tests/shell/testsuites/punkexe/maketcltool.test - 4/4 green
- verified live: 'tool build punkzip' gated on the vendored tree's
130/130 suite and installed bin/punkzip.exe (ReleaseSafe, v2.3.1,
1.1MB) over the stale Nov-2024 v2.1.0 binary; list state flipped
stale -> current; PUNKBOOT_PLAIN degraded dispatch and tabled help
verified
G-126 Progress updated: the acceptance's build-step clause is
satisfied. Remaining: punk::zip fast path, parity suite, benchmark.
Assisted-by: harness=claude; primary-model=claude-fable-5; api-location=anthropic.com
|
2 weeks ago |
|
Julian Noble
|
583dd5e0b2 |
G-133: bake payload/target consistency checks - advisory arch scan + smoke-require probe
- punkboot::utils 0.5.0: binary_arch_classify (PE Machine field, ELF e_machine,
Mach-O thin+universal - honest unknowns), platform_expected_binary,
platform_discriminated_segment (canonical <os>-<cpu> tokens + vendor spellings,
extensible namespace variables), vfs_binary_arch_report (per-directory exemption,
exempt subtrees counted unread)
- make.tcl: advisory binary-arch scan at the G-125 gate seam (recapped
BUILD-WARNINGs naming file/found-arch/kit-target, capped at 8 per kit with honest
total; cross-target kits included); post-build smoke-require probe running the
freshly built artifact via its tclsh subcommand with drained stdin (failures
recapped naming kit/package/actual error; cross-target skip with stated reason;
undeclared kits run nothing new); mapvfs.config entries accept a 5th
smoke-require element (mapvfs_parse/mapvfs_kit_outputs carry smokerequire);
'check' reports scan ACTIVE/UNAVAILABLE + declared smoke matrix; workflow K11
- mapvfs.config: smoke-requires declared - punkluck86 {Thread} (the 2026-07-27
incident construction), punk91ix86 {Thread iocp}, punkshell902 {Thread}
(cross-target skip demonstrator)
- tests: binaryarch.test (generated header fixtures only - no committed binaries;
punkluck86 case reproduced; iocp pair + win-x64 exemptions; real-tree sweep with
known-real findings filtered), maketclpayloadcheck.test (piped check
characterization, ESC-free per G-113)
- docs: src/AGENTS.md + bin/AGENTS.md state what the checks do and do NOT
guarantee; src/runtime/AGENTS.md entry grammar; ARCHITECTURE.md bake section
- REAL FINDING on the scan's first sweep: zint.dll 2.13.0 in punk8win.vfs
lib_tcl8/ is 32-bit (PE i386, confirmed with file(1)) and can never load in the
x64 tcl8 kits punk86/punkbi/punksys that carry it - those bakes warn until the
payload is fixed
- punkshell 0.28.2
Assisted-by: harness=claude; primary-model=claude-fable-5; api-location=anthropic.com
|
2 weeks ago |
|
Julian Noble
|
f4d1091d61 |
win32-ix86 verification kits repointed at minimal payloads punk8min.vfs / punk9min.vfs
The full punk8win/punk9win_for_tkruntime payloads shipped x64-only binary packages into the 32-bit kits (user-observed): a higher-versioned x64 Thread 2.8.12 shadowed the LUCK runtime's working 32-bit 2.8.5 on plain package require and broke the repl (tsv machinery), and x64-only dlls (tcludp etc) failed with "couldn't load from current filesystem". Measured: forcing -exact Thread 2.8.5 loaded fine, proving the backport's zipfs dll loading works and the failure was purely wrong-arch version shadowing. New minimal payloads: both startup fauxlinks (root main.tcl + the undroidwish app/main.tcl hook) plus directory fauxlinks to MULTI-ARCH packages only - those carrying a win32-ix86 platform subdir: iocp-2.0.2 (both majors, from punk8win.vfs / punk9magicsplat.vfs) and twapi-5.0b1 (tcl8, from punk86.vfs). No binary duplication in VCS (G-004); the pure-tcl module/lib set arrives via the dll-free _vfscommon.vfs overlay. Verified: both kits boot, resolve punk::args from _vfscommon, load Thread (2.8.5 own-arch on LUCK / 3.0.5 static on sfe), iocp 2.0.2 (ix86 dll), twapi (4.7.2 stable beats the linked 5.0b1 beta under prefer-stable on LUCK; 5.2.0 static on sfe), udp 1.0.11 on LUCK (own archive copy, no longer shadowed); piped shell sessions evaluate cleanly on both with zero tsv-unavailable noise. Remaining specimen quirk (out of payload scope): punkluck86 segfaults at process exit AFTER a completed shell session (exit 139) - repl-teardown x 2019-era runtime interaction; the tclsh subcommand exits 0 with all four packages loaded. punkexe subtree 69 tests 0 fail. Claude-Session: https://claude.ai/code/session_01UEgomWq6kA6c4A8GswqqGW Assisted-by: harness=claude; primary-model=claude-fable-5; api-location=anthropic.com |
2 weeks ago |
|
Julian Noble
|
191d8ff490 |
G-129: kit boot derives its payload mount from the runtime's mount table (punkshell 0.28.0)
Boot (punk_main.tcl + project_main.tcl, identical copies):
- zipfs presence keys on tcl::zipfs::mount - present in every supported zipfs
generation - not tcl::zipfs::root (8.7/9-era; absent from the androwish/
undroidwish 8.6 backport, whose ::zipfs ensemble also lacks a root subcommand).
- New factored ::punkboot procs (zipfs_mount_pairs, zipfs_kit_mountbase,
path_within) derive where this executable's attached archive actually mounted:
the mount-table entry whose archive file is [info nameofexecutable] (the
backport reports it without the windows drive prefix - handled), fallback the
mount containing the boot script. //zipfs:/app stays the expected modern
answer, not an assumption. Procs remain defined post-boot for tests.
- A mounted archive that cannot be attributed to the executable is reported on
stderr instead of silently configuring no internal paths (the old failure shape).
- The starkit quick-search line [zipfs root]/app/tcl_library (an error on the
backport) uses the derived base; internal_paths gains the derived base when no
zipfs volume root covers it; the internal-only simulate-kit branch recognises a
script under the derived mount as kit-internal.
make.tcl (necessary companion fixes - Scope extension flagged in the goal file):
- Runtime capability probe keys has_zipfs on tcl::zipfs::mount (was root, which
refused zip assembly for the backport; the block also rode a body-less
'if {![catch ..]}' that only worked because piped-stdin tclsh continues past
command errors).
- merge_over resolves relative fauxlink targets against the link file's own
directory (the fauxlink module's symlink semantics) - identical for the
pre-existing root-level links, required for the new nested one.
Payload + matrix:
- punk8win.vfs gains an app/main.tcl fauxlink beside root main.tcl: the
undroidwish family auto-runs app/main.tcl INSIDE the attached archive
(embedded /app/main.tcl string in the exe) and ignores a root main.tcl.
Inert in the metakit kits sharing the payload (punksys/punk86/punkbi verified).
- mapvfs.config: punkluck86 (tclsh8.6.10-luck-zip) + punk91ix86
(tclsh9.1b0-tclsfe), both zip kits targeting win32-ix86.
Tests + docs:
- New shell/testsuites/punkexe/kitmountpoint.test (4 tests): backport exe-path
attribution, modern-32 //zipfs:/app, baked punkexe regression pin,
unattributed-mount stderr report + no foreign paths. Self-gates on the
untracked verification runtimes.
- bin/AGENTS.md "Kit-wrappable runtime requirements (G-129)" + src/AGENTS.md
pointer, src/vfs/AGENTS.md boot contract, src/tests/shell/AGENTS.md suite doc,
ARCHITECTURE.md boot-chain bullet.
- Thin-layout + templates-modpod copies synced (make.tcl modules; _config boot
mains refreshed into the project-0.1 store layout per the
|
2 weeks ago |
|
Julian Noble
|
df83b21a24 |
bootsupport: replace mounted modpods by delete-then-place, and explain the windows catch-all (punkshell 0.27.4)
'make.tcl bootsupport' failed three times this session replacing the zip-based
punk::mix::templates modpod, always with:
error copying "<src>" to "<dst>": invalid argument
Root cause, demonstrated rather than inferred. A zipfs mount MEMORY-MAPS its archive
(tclZipfs.c CreateFileMappingW + MapViewOfFile), and windows refuses to overwrite a file
with a user-mapped section open - ERROR_USER_MAPPED_FILE (1224). Any holder blocks it: a
running punk shell that loaded the modpod, a src-mode session, or the build itself.
Reproduced directly: the same file copies fine unmounted, fails while zipfs-mounted, and
copies fine again once unmounted.
The message is a dead end BY CONSTRUCTION, which is why two earlier attempts at
characterising this (in commit messages and in the archived G-124 detail file - first "a
modpod make.tcl has itself mounted", then "a transient lock") were both wrong in different
ways. Tcl_WinConvertError maps only Win32 codes 0..267 and sends everything above them to
errorTable[1], which is EINVAL - so EVERY high-numbered windows error surfaces as "invalid
argument", naming neither cause nor remedy. Byte-identical code in 8.6 and 9.1b1.
FIX. ::punkboot::replace_possibly_mapped_file tries the ordinary copy and, on failure,
falls back to DELETE-THEN-PLACE, which windows does permit: unlinking a mapped file leaves
the holder reading its own mapping while new content lands at the name (measured: delete
succeeds while mounted, the mount keeps serving its old content afterwards). Replacement
content is staged to a sibling <target>.punkboot-new first, so a mid-sequence failure can
never leave the target missing; a 'broken' return - unlinked and unrestorable - is reported
as a build failure naming the file to restore by hand. This replaces a narrower prior
workaround that only recognised the already-identical case and otherwise recorded FAILED.
DIAGNOSIS. ::punkboot::mapped_file_hint fires when a genuine failure carries the catch-all
message and explains what it actually means and which process class to look for, instead of
leaving "invalid argument" bare. It stays silent for unrelated errors.
Verified on the real failure: the bootsupport run that has failed all session now reports
"(in-place overwrite refused - replaced by delete-then-place; target was memory-mapped, e.g
a mounted modpod)" and completes, with no hand-copy and no rerun. Helper-level checks cover
the unmapped path (plain copy, unchanged), the mapped path (recovers, content correct,
holder's mapping intact, no temp left behind) and both hint branches. Full suite 1167 tests,
1146 passed, 1 failure - core/tcl exec-14.3, the documented baseline.
Documented in src/bootsupport/AGENTS.md, including the general warning that 'invalid
argument' from ANY windows file operation in Tcl means "some windows error above 267" and
must never be read at face value.
Also drafts G-132 (user-approved): the errno catch-all is upstream's defect, not ours, and
deserves reporting. The goal is deliberately shaped so an AGENT DRAFTS AND THE DEVELOPER
SUBMITS - no account use, no web form, no API - following the G-039 precedent, with the
submission-ready text staged in TEMP_REFERENCE and the finding, reproducer, eventual ticket
URL and disposition kept in the tracked detail file, because TEMP_REFERENCE is one
'git clean -xdf' from gone. The report's claim is deliberately narrow: the errno conversion
is plainly wrong, while the related fact that 'file copy -force' over a mapped file succeeds
on unix and fails on windows rides along as context rather than as a second demand.
punkshell's workaround is explicitly not removed if upstream fixes it - an upstream fix
dates it rather than deletes it.
Assisted-by: harness=claude; primary-model=claude-opus-5[1m]; api-location=anthropic.com
|
2 weeks ago |
|
Julian Noble
|
176f2ad702 |
punk-runtime list/use stop offering support files as runtimes (punkshell 0.27.3)
'bin/punk-runtime.cmd list -platform win32-ix86' listed a LUCK.url shortcut and a tclsh8.6.10-luck-zip.cfg build request as if they were selectable runtimes. Both payloads already filtered support files - .txt/.toml/.tm/.tmp/.log - so the mechanism existed and only the recognised set was too narrow. It was also duplicated across NINE filter sites in three syntaxes (a powershell array, a bash case pattern, a bash grep -E regex), and two of them had already drifted: the "not a selectable runtime" guard omitted .log and matched *_BUILDCOPY without a trailing wildcard. The set is now defined ONCE per payload - $PunkRuntimeSupportExt / Test-PunkRuntimeSupportName in the ps1, punk_support_ext_re / is_punk_support_name in the bash - and every site consults it, so the two payloads cannot drift apart. Their listings are required to agree byte-for-byte and now do so by construction. Recognised support files, by category: docs/text (.txt .md .html .htm .pdf .url), config/metadata (.toml .json .xml .yml .yaml .ini .cfg), dev leftovers (.tm .tmp .log .bak .old), checksums/signatures (.sha1 .sha256 .asc .sig) and archives (.zip .tar .gz .bz2 .xz .7z), plus any *_BUILDCOPY* name. Anything else is still treated as a runtime candidate - deliberately, so a stray binary is visible rather than silently hidden. Verified: the win32-ix86 listing drops from 6 entries to the 4 real runtimes; the bash helpers produce the identical four (the polyglot routes msys bash to the ps1 payload on windows, so the bash side was exercised directly). Re-wrapped per bin/AGENTS.md with checkfile ERROR-free; runtimecmd_roundtrip (byte-identity), runtimecmd_checkfile, runtimecmd_freshness (10/10, both payloads against the fixture server) and runtimebash_wsl (8/8) all pass, as does the full suite at its documented baseline. bin/runtime/README.md now states what a platform folder is FOR (a store tier keyed by the platform a runtime is for, not by this machine) and what is tolerated in it, with the recognised-support-file table and the warning that anything else will be offered as a runtime. It is force-added to git: /bin/* is ignored and a ! negation cannot re-include a file inside an excluded directory, so git add -f is the only mechanism - recorded in the .fossil-settings/AGENTS.md force-tracked one-off list, whose fossil counterpart is an explicit 'fossil add' at the next catch-up checkin. Also records in G-101 a survey of TEMP_REFERENCE/androwish for the zipfs-on-8.6 question the developer raised (non-contract findings; pointer added to TEMP_REFERENCE/AGENTS.md per its contract so the survey is not repeated). The undroidwish zipfs backport is ADDITIVE - generic/zipfs.c (6586 lines) + zipfs.h + a library/zipfs1.0 dir + one object line in the win and unix Makefiles - so it is carryable as a suite patch rather than a fork; it is C rather than C++ like mk4tcl; Tcl_StaticPackage registration is gated on ZIPFS_IN_TCL / ZIPFS_IN_TK, i.e. designed to compile into tcl or tk; and it is Christian Werner's, under the tree's Tcl-style license.terms. Its capability delta against Tcl 9 zipfs is recorded (no 'root', no ::zipfs ensemble, and not registered in child or thread interps), as is the exe-path mount convention that motivates G-129. Provenance of our specimen is established: LUCK is a CGI repacker that selects packages out of a prebuilt vanillatclsh binary's ZIP, so bin/runtime/win32-ix86/tclsh8.6.10-luck-zip.exe is a repack of androwish's own build - a working existence proof of zipfs-on-8.6 rather than a claim. Three open questions are recorded rather than guessed: whether it applies to 8.6.18 (the tree is 8.6.10), whether it builds under the pinned zig, and where the boot-time self-mount hook lives - stock tclAppInit calls Tclzipfs_Init only under #ifdef TCL_TEST, so it is in the vanillatclsh/undroidwish application main, which is also the hook a dual-container autodetecting boot would need. Assisted-by: harness=claude; primary-model=claude-opus-5[1m]; api-location=anthropic.com |
2 weeks ago |
|
Julian Noble
|
20ee103120 |
boot-precondition gate learns the exe-path-mount tcl library convention (punkshell 0.27.2)
The G-125 gate refuses a kit whose merged .vfs supplies no tcl library. It knew two conventions - tcl_library/ for zipfs-attached kits and lib/tcl<major>.<minor>/ for starkit-style ones - which covered all ten kit trees this project builds. A third-party runtime under evaluation (Lean Undroidwish Construction Kit, 32-bit windows, Tcl 8.6.10) uses a third: its attached archive mounts at the EXECUTABLE'S OWN PATH rather than at //zipfs:/app, so [info library] is <exe>/tcl8.6 and the library sits at tcl8.6/ in the root of the tree. The gate reported that runtime unbootable - a false refusal on a directory that is unmistakably a complete Tcl library (auto.tcl, clock.tcl, encoding/, history.tcl, init.tcl, package.tcl ...). Found while investigating whether our bake infrastructure can handle that kit; the androwish/undroidwish zipfs backport for 8.6 mounts this way generally, so it is a family of runtimes, not one artifact. punkboot::utils 0.3.0 -> 0.4.0 adds tcl<M>.<m>/ at the vfs root as a third recognised location; the report's 'checked' list and the not-found reason name all three, as do 'make.tcl check' and the workflow K10 key note. Glob note worth keeping: the root tcl[0-9]* pattern also matches the tcl8/ and tcl9/ MODULE trees this project's own kits carry. They hold no init.tcl, so the existing qualification test rejects them without needing a narrower pattern - now pinned by bootlib_module_tree_not_a_library so a later tightening of the glob has to argue with a test rather than look like an obvious cleanup. Verified: the LUCK runtime's extracted tree now reports ok=1 locations=tcl8.6, and all ten assembled src/_build/*.vfs trees still pass with unchanged detected locations (tcl_library for the six zipfs kits, lib/tcl8.6 for punk86/punkbi/punksys, lib/tcl9.0 for punkshell902). Full suite 1167 tests, 1146 passed, 1 failure - core/tcl exec-14.3, the documented baseline. One of the gate's own punkexe pins failed first time round and was right to: re-wrapping the 'check' report split "...that / cannot initialise" across lines. The pattern now matches across the break rather than the assertion being loosened. Note this does NOT make such a runtime bakeable yet - src/vfs/_config/punk_main.tcl still probes zipfs with 'info commands tcl::zipfs::root' (a 9-era command the 8.6 backport lacks) and assumes the //zipfs:/app mount point, so a kit built on it would not find its own payload. That is separate work; this change only stops the gate refusing a tree that has a perfectly good library. Assisted-by: harness=claude; primary-model=claude-opus-5[1m]; api-location=anthropic.com |
2 weeks ago |
|
Julian Noble
|
ec781d7019 |
punk::libunknown 0.2.3: pkgIndex.tcl scripts run in the global namespace again (punkshell 0.27.1)
A third-party 8.6 kit under evaluation (Lean Undroidwish Construction Kit, 32-bit windows)
could not load its own bundled twapi 4.7.2 inside punkshell:
P% package require twapi
invalid command name "twapi::set_scriptdir"
("package ifneeded twapi 4.7.2" script)
twapi's pkgIndex.tcl defines 'namespace eval twapi { proc set_scriptdir ... }' when the
INDEX is sourced, and its ifneeded script calls that command at require time. The command
existed - as ::punk::libunknown::twapi::set_scriptdir. punk::libunknown 0.2.0 moved index
sourcing out of 'namespace eval ::' into source_pkgindex, a proc of the punk::libunknown
namespace, to stop each index's helper variables leaking into the global namespace and to
stop a user global named 'dir' being clobbered. That fixed the frame, but a proc body
executes in ITS OWN namespace, so an index's RELATIVE 'namespace eval foo' started
creating ::punk::libunknown::foo instead of ::foo. Stock tclPkgUnknown is a proc in ::,
so indexes have always had global resolution; the ifneeded script, evaluated later at
global scope, then looked for a command that was never there.
The body now runs via apply with :: as its namespace argument, which gives both
properties at once: a local frame (index helper variables still discarded, user 'dir'
still safe) and the global resolution context indexes are written against.
Not twapi-specific: any index that defines commands or namespaces at source time for its
own ifneeded scripts to use hits this, and the pattern is common in third-party indexes.
punkshell's own vendored twapi 5.0b1 happens to avoid it (self-contained 'apply' ifneeded
scripts), which is why the tree never surfaced it.
Verified on the LUCK 8.6.10 runtime end-to-end - 'package require twapi' at the punkshell
prompt now returns 4.7.2 with ::twapi::scriptdir correctly set - and pinned generically by
pkgindex_relative_namespace_is_global in the discovery suite, whose new fixture copies
twapi's pattern so no third-party kit is needed. The pin was proven to bite: driven
against the pre-fix body the same fixture fails with 'invalid command name' and leaves
::punk::libunknown::pklu_nsindex behind; against 0.2.3 it resolves and leaves nothing.
Full suite 1165 tests, 1144 passed, 1 failure - core/tcl exec-14.3, the documented
baseline.
Manual-version bump per src/modules/AGENTS.md: file renamed, Meta line, manpage_begin and
provide-block updated, version-history line appended. The rename sweep converted the two
goal Scope references (G-035, G-109) from the exact filename to the glob form
src/modules/punk/libunknown-*.tm that GOALS.md's own Scope-authoring guidance recommends,
so the reference stops drifting at every bump.
Also records in G-035 the mechanism behind its "avoid mixing .tm and pkgIndex.tcl
provision" folklore, which this investigation identified (non-contract Notes). Stock
::tcl:™️:UnknownHandler returns early when $satisfied is set and therefore never falls
through to ::tclPkgUnknown, so a require satisfied from a .tm suppresses the whole library
scan. Measured with a synthetic fixture on both 8.6.10 and 9.0.3: with mixpkg 1.0 as a .tm
and mixpkg 2.0 in an auto_path library under 'package prefer latest', stock returns 1.0
and reports only 1.0 from 'package versions', runs no index side effect and registers no
sibling package; the libunknown chain returns 2.0 and does all three. libunknown already
neutralises it - its handler still computes $satisfied but the early return is commented
out - and that unconditional fallthrough is exactly what the epoch cache pays for. The
folklore is therefore substantiated for stock Tcl with a named failure mode (silent
older-version selection) and neutralised under the punkshell chain, with stock behaviour
still reachable in any interp that has not run punk::libunknown::init.
Assisted-by: harness=claude; primary-model=claude-opus-5[1m]; api-location=anthropic.com
|
2 weeks ago |
|
Julian Noble
|
b7c632b670 |
G-125 achieved: a kit that cannot boot is not deployed (punkshell 0.27.0)
The kit machinery already knew when it had failed to extract anything from a source
runtime, and under G-122 that notice became a recapped BUILD-WARNING - but the build
still assembled an artifact, DELETED the deployed kit and copied the new one over it,
closing running instances of the kit first. Field-observed 2026-07-26: an msys2-hosted
bake of punk91 produced a 49,501,792-byte bin/punk91.exe that failed at startup with
"application-specific initialization failed: Cannot find a usable init.tcl", by which
point the previously working kit was already gone. G-124 removed the particular cause;
this removes the class. A build tool should not make a working executable worse.
A bake now refuses any kit whose merged .vfs supplies no tcl library: the kit is listed
under FAILED KITS with a reason naming the cause and NOTHING is written - no
src/_build/<kit>, no deploy, and the previously deployed bin/<kit> is left byte-identical.
The gate sits immediately after the _vfscommon.vfs + <kit>.vfs merges and before the
kit-type assembly switch, the last point at which no build product exists; a refusal
appends to failed_kits, ends the punkcheck event FAILED and continues, so the punkcheck
records stay consistent with what is on disk. The merged src/_build/<kit>.vfs tree is
deliberately left behind - it is the evidence for the refusal.
It gates on the BOOT PRECONDITION, not on the extraction step having run. Some .vfs
folders legitimately supply their own tcl library (src/vfs/punk8_statictwapi.vfs,
punk9test.vfs and mkzipfix.vfs do today), and those must keep building - so the check
reads the merged tree. It is structural and executes nothing, which also makes it valid
for cross-target kits this host could never run, and costs 0.63ms per kit (measured over
200 iterations against the real src/_build/punk905.exe.vfs), so it runs unconditionally.
The predicate is punkboot::utils::vfs_boot_library_report (0.2.0 -> 0.3.0), called through
a guarded require modelled on the existing get_src_provenance_warnings: a stale or missing
bootsupport snapshot degrades the gate to a NOTE rather than failing every kit. Both
branches were observed live - before the bootsupport propagation 'make.tcl check' reported
UNAVAILABLE, after it ACTIVE. Scope was extended at activation (user-approved) to put the
predicate in that module rather than private to the build script, because a gate that
cannot be exercised cannot be characterized, and punkboot::utils is already where
make.tcl's provenance gate helper lives.
What qualifies as a tcl library: tcl_library/init.tcl (zipfs-attached kits) or
lib/tcl<major>.<minor>/init.tcl (starkit-style), AND at least one companion file beside it
(tm.tcl, package.tcl, auto.tcl, clock.tcl, history.tcl, word.tcl) or an encoding/
directory. The companion test is load-bearing: every punkshell kit carries
lib/BWidget1.10.1/init.tcl, so a check that looked for any init.tcl anywhere would pass a
kit with no tcl library at all. lib/tcllib2.0 is correctly not matched by the lib/tcl[0-9]*
glob. Verified against all 10 assembled trees in src/_build - both conventions are
represented (tcl_library for punk902z/punk905/punk91/punk9_beta/punk9bi_beta/punkmagic,
lib/tcl8.6 for punk86/punkbi/punksys, lib/tcl9.0 for punkshell902).
The no-extraction BUILD-WARNING was reworded from "will not initialise unless src/vfs/<x>
supplies one" - a consequence the build then ignored - to "will FAIL the boot-precondition
gate unless src/vfs/<x> supplies one". It stays a warning rather than becoming the failure,
because a .vfs that supplies its own library needs no extraction.
Verified 2026-07-27 end-to-end against a temporary fixture kit (a mapvfs entry pointing a
zip kit at a payload-free runtime, plus a vfs with no tcl library), with sentinel files
standing in for a previously deployed kit and a previous build product. Across the failing
run bin/punkgatefixture.exe (sha1 2a0700eb...) and src/_build/punkgatefixture.exe (sha1
12c50567...) were byte-identical with unchanged mtimes, and the kit was reported under
FAILED KITS with the cause named. Adding tcl_library/{init.tcl,tm.tcl} to that same fixture
vfs - extraction still yielding nothing - made it build and deploy normally, which is the
other half of the criterion. The fixture was removed afterwards and mapvfs.config restored
byte-identical to its backup. Real-kit bake through the gate: punk905 builds, deploys and
boots.
Characterization: 12 tests in modules/punkboot/utils/testsuites/utils/bootlibrary.test
(both conventions, the companion-file requirement and its BWidget decoy, near-miss
reporting, missing/empty trees, and a sweep asserting every assembled src/_build/*.vfs
tree still passes so the gate cannot fail kits that boot today) and 2 in
shell/testsuites/punkexe/maketclbootgate.test (the ACTIVE/UNAVAILABLE report, ESC-free per
the G-113 piped policy). Existing punkexe suites pass unchanged.
Documented in src/AGENTS.md with the remedy when a kit is refused, in ARCHITECTURE.md
beside the provenance gates, and in the embedded 'make.tcl workflow' data flow as key note
K10 (verified under the default and PUNKBOOT_PLAIN=1 paths, within the 100-column budget).
The achieved flip archives the entry and the detail file, and sweeps the live tier: G-127
(its output location inherits the never-written guarantee as long as the gate stays
upstream of the artifact write, and its payload/target mismatch reporting should share this
vocabulary), G-028 (the sibling "why bin/<kit> was not updated" surface in the same deploy
step) and G-101 (a new container type must land its library where the gate looks, or extend
the location list).
Assisted-by: harness=claude; primary-model=claude-opus-5[1m]; api-location=anthropic.com
|
2 weeks ago |
|
Julian Noble
|
78d9bf238a |
G-124 achieved: dependency-free zip reading in punk::zip (punkshell 0.26.0)
punk::zip could write a zip and split an executable-prefixed one, but nothing in the tree
could READ a zip without zipfs, vfs::zip or tcllib's zipfile::decode - a system package
punkshell does not vendor, with exactly one consumer: make.tcl's zipfs-less kit
extraction. Measured at drafting: msys2's /usr/bin/tclsh8.6 has no tcllib, so a bake
driven from that host could not lift the runtime's own zip (its tcl_library) into the kit
and the artifact did not boot.
punk::zip 0.1.1 -> 0.2.0 adds three commands, stock Tcl only. archive_info reports where
the archive begins inside its host file and which offset convention it uses; members
walks the central directory and returns one dict per entry without decompressing
anything; unzip extracts all or part of it with every member's crc32 verified,
directories materialized, and the writer's 2MB whole-member-vs-streaming threshold
mirrored. All three accept a plain .zip or a zip attached to an executable or script
prefix, so "what is inside this kit / runtime / modpod" is now answerable on a runtime
without zipfs.
The offset derivation is stated once and shared: offsetbase = cdiroffset - diroffset.
Positive means an external preamble with archive-relative offsets, zero means the offsets
are file-relative (or there is no preamble); dataoffset is the split point either way.
Because every surface reads the ORIGINAL file at that base, the file-relative shape stops
being a special case - the broken split-off intermediate is never produced.
tclsh90b4_piperepl.exe is the store's one file-relative artifact: the old
split-then-tcllib path still fails on it with exactly "Bad zip file. Bad closure.", and
it now extracts all 841 members crc-verified in 735ms.
extract_preamble consumes that machinery instead of carrying its own scan, and two
defects went with the refactor: it read 28 bytes of a 30-byte local file header, and it
derived the file-relative base from the FIRST central directory record only (its own
'#todo! loop through all cdr file headers'). The shared walk takes the minimum over ALL
records and then validates that a PK\3\4 really sits there. Its five puts stdout debug
lines are gone.
Refusals are decided for every selected member before any file is opened, so an
unsupported archive cannot leave a half-populated directory: encrypted entries, zip64
size/offset fields, unknown compression methods, crc mismatch (partial file removed) and
member paths that would escape the target are each named. Store and deflate only,
mirroring the writer's own primitives.
make.tcl's zipfs-less extraction now calls punk::zip::unzip on the original runtime and no
longer creates the extracted_<runtime>.zip intermediate. No punkshell code requires
zipfile::decode any more. Two categories of hit remain and are deliberate: tcllib's own
zip/decode.tcl shipped as a LIBRARY inside some .vfs payloads, and one guarded optional
probe inside the vendored third-party ooxml1.10/ooxml.tcl.
First tests punk::zip has ever had (src/tests/modules/punk/zip/), covering the three
archive shapes from one source tree, the introspection output, the named refusals, and
the mkzip -> read round trip - the first coverage that what punkshell WRITES is readable.
Verification, all 2026-07-26. Reference host for the dependency-free claim: msys2
tclsh8.6 8.6.12, probed as zipfs / vfs::zip / zipfile::decode all ABSENT - suite driven
there harness-free, 28 passed, 2 skipped (the zipfs-gated cross-checks only), 0 failed.
30/30 under Tcl 9.0.3, 28 + 2 skipped under native 8.6. Full source-tree suite after the
change: 1142 tests, 1 failure - core/tcl exec-14.3, the documented baseline. Independent
cross-check: 868 members of tclsh905.exe extracted byte-identical against the same
archive mounted with tcl zipfs. Bake demonstration: 'bake -confirm 0 punk905' driven by
the msys2 tclsh8.6 (host=msys-x86_64, target=win32-x86_64) produced
extracted_tclsh905.exe/{tcl_library,lib} with no intermediate .zip, and the deployed kit
boots with tcl_library at //zipfs:/app/tcl_library, 95 encodings, msgcat 1.7.1, http
2.10.2 - and reads itself with the new reader.
Trap worth keeping: zlib stream inflate's 'get' returns only what the stream has already
produced, so one get per put silently truncates a large member to the first few chunks
(measured 327680 of 5048890 bytes on both 8.6.12 and 9.0.3). The inner drain loop is
load-bearing, and the >2MB streaming test exists to catch its removal.
Standing question answered and pinned: zipfs does NOT misidentify punk::zip::mkzip
directories. It keys on the trailing slash (which punk::zip::walk always emits), not the
stored permission bits, exactly as the current cli-999999.0a1.0.tm wording says - the
older, less precise wording survives only in frozen module snapshots under
src/vfs/mkzipfix.vfs and src/vfs/project.vfs. The real difference from zipfs mkzip is
that zipfs writes no directory entries at all.
The achieved flip archives the entry and the detail file, and sweeps the live tier: G-034
(the reader is its extract-rather-than-mount alternative, plus the measured finding that
make.tcl on 8.6 cannot mount its own templates modpod), G-101 (the read half of a
zipfs-less 8.6 container is solved), G-125 (the gate now has a named failure reason to
report), G-126 (the member-dict keys and fixture set its parity suite must reproduce, and
a back-pointer to G-128 whose only recorded bridge was this goal), G-128 (archive_info
answers "is this artifact safe to move the overlay of?" in one call).
One acceptance clause is reported for review rather than claimed silently: "NO
zipfile::decode requirement remains anywhere in the shipped tree" is satisfied for
punkshell's own code, with the vendored-third-party residue named above.
The archived detail file's diff is a rename plus the flip's evidence write-up.
Assisted-by: harness=claude; primary-model=claude-opus-5[1m]; api-location=anthropic.com
|
2 weeks ago |
|
Julian Noble
|
04d2a51b67 |
G-122 achieved: make.tcl bakes by target platform, not host personality (punkshell 0.25.0)
Everything a bake EMITS is now keyed by the artifact's TARGET platform rather than the driving tclsh's personality: the bin/runtime/<tier> store a runtime is read from, .exe suffixing of runtime files and kit outputs, presence checks, and whether the pre-deploy process sweep uses tasklist/taskkill or ps/kill. Host semantics - copy commands, path handling, filesystem case rules, prompts - stay keyed to the host. The default target is the host's platform canon EXCEPT for a cygwin-family host (an msys2/cygwin-runtime tclsh, which reports tcl_platform(platform) unix on windows), which targets win32-x86_64. Such a host now drives the identical kit set, names and store addressing as a native tclsh; before, it reported every kit runtime=missing against a nonexistent mingw64-x86_64 store. mapvfs.config entries take an optional 4th element naming the kit's target platform, so a runtime kept in another platform's tier is a first-class mapping entry. The linux tclkit entry now declares linux-x86_64 and bakes to a suffixless bin/punkshell902. Zip-type kits assemble without zipfs in the driving tcl: when tcl::zipfs::mkimg is absent (any 8.6) the image is built by raw-runtime split + punk::zip::mkzip + concatenation - the same helper the zipcat type uses. Both mount identically (archive-start-relative offsets). Fixes exposed by the above: the runtime capability probe looked for a suffixless filename and so reported find-fail for every runtime on windows; with real capabilities arriving, the extraction try-list was found to test capability presence rather than value, which would send a plain zipfs runtime down the kit path where a failed sdx mksplit replaces raw_runtime with the un-split original; BUILDCOPY names now follow the runtime's own filename; runtime map keys strip only .exe (file rootname would eat the last dotted segment of tclsh9.0.5); and native windows command lines run through a guard suppressing msys2's posix-path argument rewriting - without it taskkill /PID <n> reached taskkill as taskkill C:/<msysroot>/PID <n>, so the sweep could find a running kit but never replace it. punk::platform folds the msys2 family (mingw64/mingw32/ucrt64/clang*) onto one stable 'msys' tag and CYGWIN_NT onto 'cygwin' - msys2's uname renders that token from MSYSTEM, so one binary otherwise canonizes differently per launching shell. New normalize_os; cygwin-x86_64 added to the platform table; 'help platforms' documents the fold. make.tcl check reports the derivation on one line; bakelist rows carry target=<platform> for non-default targets and the detail block names the tier; workflow text gains key note [K9]. New characterization suite maketclplatform.test (5 tests, two self-gating on discovering and probing a real msys/cygwin tclsh). Full suite 1112 tests: 1093 passed, 18 skipped, 1 failed = the documented exec-14.3 baseline. Verified: msys2 tclsh8.6 bakelist output identical to native Tcl 9; a changed punk91 baked and deployed from that host with the sweep killing a running instance msys ps cannot see; a real cross-target linux kit baked from windows to a 24MB ELF; a zipfs-less 8.6 bake producing a bootable zip kit. Assisted-by: harness=claude; primary-model=claude-opus-5[1m]; api-location=anthropic.com |
2 weeks ago |
|
Julian Noble
|
d9061b90ce |
bakelist/bake: loud missing-store-folder diagnosis; punk::platform into bootsupport (punkshell 0.24.1)
Field-driven follow-ons to G-121 from driving make.tcl with an msys/cygwin-runtime tclsh (scoop msys2 tclsh8.6, tcl_platform(platform)=unix): - bakelist: when the derived per-platform runtime store folder does not exist, a stderr WARNING now shows the platform-canon derivation (tcl_platform os/platform -> canon, observed mingw64-x86_64 under a MINGW64_NT MSYSTEM env) and the header line carries (FOLDER MISSING) - every row reading runtime=missing now self- diagnoses its non-per-entry cause. bake's "No executable runtimes found" exit gains the same note when the folder is absent. Native-host output unchanged (maketclbakelist.test 4 PASS). - make.tcl shell 'help platforms' degraded to "punk::platform package not available" on ALL hosts (not msys-specific): the module was absent from the bootsupport snapshot. Added punk::platform to src/bootsupport/modules/include_modules.config (pure Tcl, deps only Tcl's own platform package); verified the full canon table renders via a piped 'help platforms' under tclsh90 src/make.tcl shell. - G-013 Notes: recorded the msys raw-mode field finding (cursor placed at row 1 after each command under the msys pty class) as a raw-mode-default blocker dimension - console/repl work, deliberately outside the make.tcl host/target catering goal being drafted. Claude-Session: https://claude.ai/code/session_01Jz7wkUsknJzyuJ3tgMaL2t Assisted-by: harness=claude; primary-model=claude-fable-5; api-location=anthropic.com |
2 weeks ago |
|
Julian Noble
|
0abd07ecd7 |
G-121: make.tcl bakelist + selective bake (punkshell 0.24.0)
One parsed-mapping model (::punkboot::lib::mapvfs_parse / mapvfs_kit_outputs / mapvfs_match_outputs) now backs the kit surfaces - the bake machinery, the new bakelist report and the bake/bakelist argdoc kit-name choices all consume the model, never the file format, so the G-024 toml conversion can swap the reader underneath. - bakelist ?kitname ...? (informational SUBGROUP): kit name, type, runtime with presence, vfs folder, deployed state of bin/<kit> vs src/_build/<kit> (current|stale|absent|nobuild; size/mtime fast path, byte-compare when in doubt), anomalies as trailing key=value notes (runtime=missing, vfs=missing, rtrev=r<cur><r<max> via the shared runtime_materialization_check core). Name filtering doubles as per-kit detail (resolved paths/sizes/mtimes). - bake ?kitname ...?: selective kit assembly - unknown names error before any build listing the configured names (flag-order hint included); selected-but- unbuildable kits (missing runtime/vfs) fail fast; runtime BUILDCOPY/caps probe/vfs cksums/vfslibs phase all narrow to the selected kits so other kits' _build/bin artifacts and punkcheck records stay untouched. Bare bake unchanged (verified full-set run). Deprecated alias vfs mirrors the values. - argdocs: SUBVALUES/VALUES_SYNOPSES mechanism, kitname choices enumerated from the mapping at define time (-choicerestricted 0, degrade-safe), block-form help values; SUMMARIES/HELPTEXTS/SUBOPTS/SUBGROUPS/known_commands/plain gethelp/workflow text updated per the update contract; degraded-mode (PUNKBOOT_PLAIN) dispatch collects trailing kit names. - mapvfs.config todo comment retired to point at G-121. - piped characterization: src/tests/shell/testsuites/punkexe/maketclbakelist.test (full-report row/vocabulary + ESC-free, filtered detail, unknown-name errors on both surfaces with the no-build guarantee) - 4 PASS; maketclcolour.test still 3 PASS. - selective-refresh verification (recorded in the goal file): forced punk91 rebuild + deploy via 'bake -confirm 0 punk91'; stat diffs over src/_build and bin show only punk91.exe, punk91.exe.vfs, raw_tclsfe-x64.exe and bin/punk91.exe changed; .punkcheck diffs mention only punk91's records (negative grep for all other kit/vfs names clean). - docs: src/AGENTS.md work-guidance bullet; ARCHITECTURE.md build-entry subcommand list refreshed (bake/bakehouse naming + bakelist, stale in-flux G-112 mention dropped); CHANGELOG 0.24.0 + punkproject.toml minor bump. Claude-Session: https://claude.ai/code/session_01Jz7wkUsknJzyuJ3tgMaL2t Assisted-by: harness=claude; primary-model=claude-fable-5; api-location=anthropic.com |
2 weeks ago |