|
| 1 | +# compose2pod 0.3.1 — the `image` mount type |
| 2 | + |
| 3 | +A focused follow-up to 0.3.0: the long-form `volumes` `type: image` mount, which |
| 4 | +closes the last long-form volume form Podman can express but compose2pod did not |
| 5 | +yet parse. |
| 6 | + |
| 7 | +## Feature |
| 8 | + |
| 9 | +- **`type: image` volume mounts.** A long-form entry `{type: image, source, |
| 10 | + target, read_only, image: {subpath}}` compiles to `podman run --mount |
| 11 | + type=image,source=<ref>,target=<path>[,subpath=<abs>]`, mounting an image's |
| 12 | + rootfs into the container. Measured against `docker compose config` v5.1.2 and |
| 13 | + Podman 6.0.1: |
| 14 | + - `source` (the image reference) is **required** — Podman refuses a sourceless |
| 15 | + image mount, so compose2pod does too (a rule-two narrowing; Docker allows it |
| 16 | + optional at config). |
| 17 | + - `read_only` is accepted but **inert** — an image mount is always read-only, |
| 18 | + and Podman rejects both a `readonly` and an `rw` option on it, so no `ro` is |
| 19 | + emitted (matching what Docker does with the field). |
| 20 | + - `image: {subpath}` must be an **absolute** path — Podman requires it for an |
| 21 | + image mount (a relative `subpath` is rejected at run time), unlike a `volume` |
| 22 | + `subpath`, which may be relative. A `${VAR}` is carved out. |
| 23 | + |
| 24 | +## Docs |
| 25 | + |
| 26 | +- The README's supported-subset section, stale since before the 0.3.0 parity |
| 27 | + work, now leads with the measured parity property (compose2pod refuses every |
| 28 | + document `docker compose config` refuses, verified by the differential |
| 29 | + conformance harness) and reflects the full 0.3.0 subset. |
| 30 | + |
| 31 | +## Downstream |
| 32 | + |
| 33 | +- `type: image` compose files that previously raised now compile. Nothing else |
| 34 | + changes — the short form, the other long-form types, and every existing key |
| 35 | + behave identically. |
| 36 | +- With `image` landed, the honest subset is complete for its scope: every |
| 37 | + remaining refusal of a form Docker accepts is either a permanent rule-two limit |
| 38 | + (Podman cannot form the flag) or a genuinely niche gap (a Windows drive-letter |
| 39 | + volume source on a Linux-targeting tool), each measured and documented in |
| 40 | + `architecture/supported-subset.md`, `planning/deferred.md`, and |
| 41 | + `planning/decisions/`. |
| 42 | + |
| 43 | +## Internals |
| 44 | + |
| 45 | +- Tests at 100% line coverage (enforced); `ruff select=ALL`, `ty`, `eof-fixer` |
| 46 | + clean; the conformance harness (real Docker CLI) and integration harness (real |
| 47 | + Podman) both green — the integration suite mounts an image's rootfs on real |
| 48 | + Podman, and the final review re-measured every emitted `--mount` option against |
| 49 | + Podman 6.0.1, catching (and closing) a relative-`subpath` value Podman rejects |
| 50 | + at run time. |
0 commit comments