100 Commits

This Branch

This Branch

All Branches

Author	SHA1	Message	Date
Jordan Harband	882ed79ece	[Docs] threat model: document the mirror payload/metadata trust boundary ... Payloads (binaries, plus source tarballs that nvm compiles and runs at install time) are trusted by construction, since nvm exists to build and run them; parsed metadata (index.tab, SHASUMS, LTS codenames) is not, and must never reach a command evaluator, an awk/sed program body, or an unvalidated filesystem path. This records why the existing version-string and checksum hardening exists, and scopes a malicious payload from a configured mirror as out of scope: no privilege boundary is crossed.	2026-07-15 13:44:04 -07:00
Jordan Harband	d83da5d75b	[actions] add tests-alpine: a green fast suite and a musl-binary regression matrix ... The full suite cannot run on Alpine: the install-based suites pin ancient Node (0.10.x, io.js) that has no musl binary and cannot source-compile on musl. Run the fast unit suite instead (mirroring the ubuntu runner: non-root via su-exec, passwordless sudo, a PTY, no system node), plus a binary-only regression matrix that installs every (Alpine, Node) pair with a real unofficial musl binary via `nvm install -b`: x64 back to node 8.17.0 on old Alpine, arm64 at the v20.20.1/v22.21.1/v24.9.0 floors on modern Alpine.	2026-07-14 23:18:07 -07:00
Jordan Harband	438009f54e	[Tests] make the fast suite pass on Alpine ... `nvm_get_arch`: skip the smartos arm64/armv8l cases, which have no uname mock and fall through to the real host (adding a musl suffix on Alpine). `nvm_get_arch_unofficial`: skip the glibc chroot (fixed /lib64 loader, sudo); the musl mapping is already covered by the `nvm_get_arch alpine` test. `nvm_install_no_progress_bar`: skip below the musl floor, since v0.12.18 has no musl binary and hardcodes the glibc tarball URL.	2026-07-14 23:17:01 -07:00
Jordan Harband	f1e7a84cb5	[Robustness] avoid a set -u error when $_ is unset ... Sourcing nvm.sh under `set -u` on a POSIX shell that leaves `$_` unset (e.g. busybox ash or dash on Alpine, where the parent shell does not export it) aborted at `NVM_SCRIPT_SOURCE="$_"`. Default to empty; the only consumer already falls back to `$0`.	2026-07-14 23:16:39 -07:00
Jordan Harband	86c937436d	[Robustness] nvm_get_arch: only apply the Alpine musl suffix when the OS is linux ... The `/etc/alpine-release` check applied the -musl suffix regardless of the resolved OS. In practice Alpine is always linux, so this is behavior-neutral, but it also makes the `nvm_get_arch` unit test hermetic: its mocked smartos/osx cases no longer pick up a real host's Alpine marker.	2026-07-14 23:10:33 -07:00
Jordan Harband	c7d5d7748b	[Tests] run every remaining test file: exec bits, fixes; add check-exec to CI ... urchin only executes test files that have the executable bit, silently skipping the rest, and `npm run test:check-exec` would catch non-executable ones, but no workflow ran it - so several test files have never actually run. Make all of them executable, and fix the ones that were broken: - the io.js source-install tests, skipped since 2016: io.js requires python 2 and gcc <= 5, so they now skip on toolchains that can not compile it, and a new gcc:4.9-container CI job runs them unmodified - resolving, downloading, checksumming, and compiling io.js for real. New `fake source` variants exercise nvm's source pipeline everywhere - cache lookup, offline resolution, extraction, configure, make, make install, activation - via a tiny fake source tree seeded into the download cache and installed with `--offline`. Also strengthen the `nvm run | grep` assertions with `--silent`, since the version banner used to satisfy the grep even when the built binary could not run, and give the final assertions the `die` they always meant to call. - `nvm uninstall ... incorrect file permissions fails nicely`, parked in 2025 as failing: `sudo npm` resets PATH, so node 5's npm crashed under the system node. The permissions check only trips on files that are neither writable nor self-owned, and that precondition only needs `sudo touch` - not npm, jspm, or the network - so use that, against a fake installed version, and skip only where passwordless sudo does not exist. - `nvm install --offline`: `nvm ls | tail -1` always grabbed an alias line, never a version; use `nvm_ls` and skip its trailing `system <version>` line. - `nvm exec ... help should not parse`: in a BRE, `[options]` is a bracket expression, so the grep could never match node's literal `Usage: node [options]` output. - `nvm use ... nvmrc containing not installed version`: current output once again matches its expectations verbatim; no changes needed. Rewrite `test:check-exec` in POSIX sh - npm runs scripts with /bin/sh, which is dash on Ubuntu, where the previous `[[`/`$'\n'` bashisms would break; match by extension, so that filenames containing dots can not dodge the check; and disable git's `core.quotePath` mangling of the non-ASCII test filenames. Then wire it into the lint workflow.	2026-07-09 22:40:07 -07:00
Jordan Harband	4e48294bd6	[Fix] nvm_ls_cached: include source-cached and io.js artifacts ... `--offline` version resolution could only see cached node binaries: `nvm_ls_cached` listed `.cache/bin` with a `node-` prefix filter, so cached source tarballs, and everything io.js, could never resolve offline. List both cache kinds, for both flavors, and update the pinned test expectations to match.	2026-07-09 22:35:15 -07:00
Jordan Harband	1a39786c8a	[Tests] nvm_offline_version: make it executable; pin nvm_ls_cached cache visibility ... The test was born without its executable bit, and urchin silently skips non-executable files, so it has never run. Also pin what `nvm_ls_cached` can currently see: only bin-cached node artifacts - bin-cached io.js, and src-cached anything, are invisible to it, and thus to `--offline` version resolution.	2026-07-09 22:35:15 -07:00
Jordan Harband	ebada63259	[Tests] aliases: cover the regex-metachar cycle fix; clean up fixtures in teardown ... The cycle scenarios all pass with the old grep-based detection too; the bug the `case`-based detection actually fixes is resolved names interpolated into the grep pattern as regexes, where resolving axb -> a.b falsely reported ∞ because the pattern `a.b` matches the seen name `axb`. Pin both directions: the fixed false positive, and a genuine cycle through a metachar name. Also correct the space-name scenario comment (the old anchored grep passed that scenario; only token-delimited seen-storage would not), and register every new fixture name in the suite teardown, so a mid-test failure can not leak aliases into later tests.	2026-07-09 15:58:41 -07:00
Jordan Harband	dd61ae4bc8	[Robustness] nvm_alias: explicitly succeed when the alias file is unreadable ... The sed/awk pipeline exits with awk's status, so an existing-but-unreadable alias file produces empty output with status 0 - but only by accident, alongside sed's read error on stderr. Make that contract explicit: a nonzero status here would flip `nvm_ensure_default_set` from "default is already set" to recreating it, silently overwriting a write-only default alias.	2026-07-09 15:43:56 -07:00
Jordan Harband	59cf9a4ef6	[actions] nvm-install-test: turn the expected v0.40.0 failures green ... The two `v0.40.0` + `nvmrc` jobs fail by design (https://github.com/nvm-sh/nvm/issues/3405), but job-level `continue-on-error` still displays them as failures on every run, which reads like the contributor broke something. Tolerate failure only in the steps that exercise the bug, scoped to that matrix combination, and add a final step that fails the job when the expected failure did not happen - so those jobs are green exactly when v0.40.0 misbehaves as documented, and turn red as a signal to remove this handling if v0.40.0 ever starts passing. This also restores strict failure semantics to the passing `v0.40.0` + `no nvmrc` jobs, which the job-level `continue-on-error` was needlessly masking.	2026-07-09 17:01:32 -07:00
Jordan Harband	21d33ef04a	[Tests] nvm_ls handles hash in pattern: fix cleanup, isolate NVM_DIR, cover the sed path ... - `cleanup` unset TEST_DIR before `rm -rf "${TEST_DIR-}"`, so the temp dir (with an executable fake `node`) was never removed, and urchin executes any executable file it finds under the test dir on the next local run. - the preexisting `foo#bar` assertions ran against the ambient NVM_DIR; now that `#` patterns comment-strip to `foo`, a real local alias named `foo` would resolve and break them, so all assertions now run against an isolated NVM_DIR. - the multiline content used a full x.y.z version, which takes nvm_ls's explicit-version fast path and never reaches the find/sed pipeline where the newline actually broke sed; a partial version exercises that path, and matching on "unterminated" covers both the BSD and GNU sed error wordings.	2026-07-08 23:14:23 -07:00
Jordan Harband	9142a92cdc	[Fix] nvm_download, nvm_get_latest, install: only select a downloader that exists as an executable ... `nvm_has` matches shell functions and aliases, but downloads now run via `command`, which skips them - a `curl` shell function with no curl binary on the PATH would select the curl path and fail with exit 127, instead of falling back to an available wget executable. The new `nvm_has_executable` helper resolves names the same way `command` does, so downloader selection and execution agree.	2026-07-08 13:03:16 -07:00
Jordan Harband	04fef13bdc	[Fix] use command to bypass curl/wget shell functions and aliases ... zsh (and interactive bash with `expand_aliases`) bakes a preexisting `curl` alias into nvm's function bodies at source time, and shell functions named `curl`/`wget` shadow the binaries at call time - either one breaks downloads. Prefixing invocations with `command` bypasses both: here, `nvm_download`'s dispatch, `nvm_curl_version`, `nvm_curl_libz_support`, and the wget branch of `nvm_get_latest`; the remaining bare `curl` invocations in `nvm_get_latest` and the install script are prefixed in a followup commit. The tests that previously mocked curl/wget as shell functions now install fake executables on PATH instead, via a shared `make_fake_curl` helper in `test/common.sh`, and a new test asserts the bypass. Refs #2923	2026-07-08 13:02:41 -07:00
Jordan Harband	e2e612e722	[Tests] installation_node: only match top-level global packages	2026-07-02 15:49:01 -05:00
Jordan Harband	041a57747a	[Tests] installation_node: use a published is-nan version ... `is-nan@1.0.0` was published on 2014-07-05 and unpublished minutes later (the registry's `time` map still lists it, but `versions` jumps from `0.0.0` to `1.0.1`), so `npm install -g is-nan@1.0.0` fails with `ETARGET`, and the regression test added in ce157343 fails deterministically in every shell. See https://github.com/nvm-sh/nvm/actions/runs/28407118533	2026-07-02 14:35:57 -05:00
Jordan Harband	419ff4efec	[actions] update workflows	2026-06-29 17:33:31 -05:00
Jordan Harband	7a7908ffa1	[Dev Deps] update doctoc, semver	2026-06-29 16:43:55 -05:00
Jordan Harband	ce15734300	[Fix] nvm install: migrate packages/set alias when target is already installed ... When `nvm install <target>` found that `<target>` was already installed, the post-`nvm use` steps (`--reinstall-packages-from`, default packages, and `--alias`/`--default`) were gated on `[ $EXIT_CODE -ne 0 ]`. Since that branch is only entered when `nvm use` succeeded (EXIT_CODE == 0), those conditions were always false, so the steps were silently skipped. The fresh-install branch correctly uses `-eq 0`; mirror that here so `--reinstall-packages-from` actually migrates global packages (and the alias/default get set) when the target version already exists. Includes a regression test covering the already-installed path. Fixes #3858	2026-06-29 16:39:19 -05:00
Jordan Harband	ee4fa818d2	[Docs] document nvm reinstall-packages as a standalone command ... The standalone `nvm reinstall-packages <version>` command can migrate global npm packages between already-installed versions, but it was only mentioned in `nvm --help`, not in the README. Add a section covering it. Refs #3858	2026-06-29 14:51:15 -05:00
Jordan Harband	a6ec739430	[Docs] document current and clarify .nvmrc fallback in help/README ... - Help text for `nvm install`/`use`/`exec`/`run`/`which` now ends with "Uses .nvmrc if version is omitted; otherwise errors." so it's clear that omitting the version is not a free fallback. - Help signatures for `use`/`exec`/`run` now show `[current | <version>]` to mirror `nvm which` and document that `current` is accepted. - `nvm current` description now spells out that it resolves via `$PATH` and is not affected by `.nvmrc`. - README: add `current` to the list of special aliases, with the same caveat. The `.nvmrc` section now states that `nvm use`/`install`/`which` exit with status 127 when neither a version nor an `.nvmrc` resolves, notes the (current) `exec`/`run` fallback as undefined behavior, and points readers at `current` for the explicit "active node" use case. Refs #3755	2026-06-10 15:16:24 -07:00
Jordan Harband	718e880890	[Fix] focused error on missing/invalid args for several subcommands ... Previously a number of subcommands dumped the entire `nvm --help` output (~100 lines) when arguments were missing or invalid, drowning the real error. Replace each dump with a short, command-specific usage block that names the expected syntax and points to `nvm --help` for full help. The exit code (127) is unchanged. Affected subcommands: - `nvm install` (no version + no .nvmrc) - `nvm use` (version unresolvable) - `nvm run` (no version + no .nvmrc) - `nvm which` (no version + no .nvmrc) - `nvm cache` (unknown subcommand) - `nvm uninstall` (wrong arg count) - `nvm unalias` (wrong arg count) - `nvm install-latest-npm` (wrong arg count) - `nvm reinstall-packages` / `nvm copy-packages` (wrong arg count) The catch-all unknown-subcommand handler still dumps full help, since in that case the user has no narrower context to be reminded about. Refs #3755	2026-06-10 15:16:23 -07:00
Jordan Harband	ed4dbdfdd5	[Fix] nvm_rc_version: clarify error when no version and no .nvmrc ... Previously the message read "No .nvmrc file found", which obscured the fact that the user also did not pass a version. The new wording names both halves of the actual problem. Refs #3755	2026-06-10 15:16:23 -07:00
Jordan Harband	a9933f77a6	[Fix] nvm exec/nvm run: warn when no version and no .nvmrc ... Currently these commands silently fall back to the active node version when neither a version argument nor an `.nvmrc` resolves, making them invisibly dependent on shell state and impossible to script predictably (see #3755). Print a stderr deprecation warning in this case (suppressed by `--silent`) and continue with the active node version, so existing callers keep working. The follow-up change will turn this into a hard error; pass `current` explicitly (e.g. `nvm exec current node ...`) to silence the warning and lock in the new behavior now. Refs #3755	2026-06-10 15:16:23 -07:00
Jordan Harband	7079a5d61c	[actions] nodejs.org PR: sync the fork before branching, and surface create-ref errors	2026-06-04 14:52:41 -07:00
Jordan Harband	306f04a90a	[actions] nodejs.org PR: fix force boolean and grab the latest tag on dispatch ... The branch-reset step sent `force` as a string via `gh api -f "force=true"`, which the Git refs API rejects ("not a boolean", HTTP 422), failing the release run. Use `-F` so it is sent as a boolean. Also resolve the version from the latest `vX.Y.Z` tag (via the tags API) rather than `releases/latest`, so a manual `workflow_dispatch` with no input works even before the GitHub release object for a freshly pushed tag is published.	2026-06-04 09:54:01 -07:00
Jordan Harband	77fca77ddd	[actions] fall back to the canonical nvmrc submodule when a fork lacks its own ... Upstream's `.gitmodules` uses a relative submodule url (`../nvmrc.git`), which resolves against the superproject's origin: on a fork without its own `nvmrc` fork that is `<owner>/nvmrc`, which 404s and fails `actions/checkout` with `submodules: true`. Keep `submodules: true` (so a fork's own `nvmrc` is used when present), but mark checkout `continue-on-error` and, only when it failed, re-point the submodule at `nvm-sh/nvmrc` and update.	2026-06-04 09:22:04 -07:00
Jordan Harband	1889911f08	v0.40.5	2026-06-04 09:33:05 -07:00
Jordan Harband	70fb4ede6b	[Fix] nvm_download_artifact: reject version strings with disallowed characters ... The mirror-supplied (untrusted) version flows into download URLs, filesystem paths, and the checksum awk match. Reject any version outside the node/io.js grammar (`[0-9A-Za-z._+-]`) before it is used. A blocklist of metacharacters is used rather than a strict semver allowlist so RCs, nightlies, v8-canary, and io.js versions still install. Completes the remediation of GHSA-3c52-35h2-gfmm.	2026-06-03 13:09:07 -07:00
Jordan Harband	90bb88748b	[Fix] nvm_get_checksum: pass the tarball name to awk as data, not program text ... The awk program string-interpolated the slug (which embeds the untrusted, mirror-supplied version) into its source, so a crafted version such as `v1"==$2){system("touch${IFS}/tmp/x")}#` was executed by awk's `system()`. Pass the value via `-v tarball=...` so awk treats it as data and never as code. See GHSA-3c52-35h2-gfmm (a second injection sink fed by the same untrusted version field that `nvm_download`'s eval was; the source-install path reaches this during a normal `nvm install <version>`).	2026-06-03 13:11:26 -07:00
Jordan Harband	6d870d182c	[Fix] nvm_download: avoid eval so mirror-supplied version strings can't inject commands ... `nvm_download` built a curl/wget command string and ran it with `eval`. The download URLs embed the version string taken from the mirror's `index.tab`, which is untrusted. Wrapping each argument in double quotes inside the `eval` does not prevent command substitution, so a version field such as `v1$(touch /tmp/proof)` was executed by the shell. This bypassed the earlier quoting hardening in 0ce8f5a. Pass every argument as a literal argv element instead of constructing a string for `eval`, on both the curl and wget paths, so URL arguments are never re-parsed by the shell. The wget flag translation is now done per-argument with a POSIX `set --` loop rather than `sed` over the joined string. The auth header is sanitized and added once, before invoking the downloader.	2026-06-02 17:41:44 -07:00
Jordan Harband	d264b796a3	[Fix] nvm_download: send a well-formed Authorization header on the wget path ... The wget path passed `NVM_AUTH_HEADER` as the raw header line (e.g. `--header "Bearer secret-token"`), omitting the `Authorization:` header name that the curl path includes. Per the documented usage (`NVM_AUTH_HEADER="Bearer secret-token"`) the value is the credential, so wget was sending a malformed header. Prefix it with `Authorization: ` to match the curl path.	2026-06-02 17:40:50 -07:00
Jordan Harband	84079dbff3	[Tests] install_nvm_from_git: stop git background gc/maintenance racing with cleanup ... The test's `rm -rf "$NVM_DIR"` intermittently failed with `rm: cannot remove...: Directory not empty` and aborted under `set -e`. Cause: after the clone/fetch, git can spawn a detached background gc/maintenance process that keeps writing into the clone dir while `rm -rf` runs (a concurrent writer makes the final rmdir fail). It is not egress- or version-related (it reproduces with all endpoints allowed), and it is environment-timing dependent (recently became consistent on the GitHub runners). Disable git's background work for the test (gc.autoDetach / maintenance.auto) so all git operations finish synchronously, and retry the rm once as a safety net.	2026-06-03 20:12:26 -07:00
Jordan Harband	61c9ff7371	[Tests] install_nvm_from_git: fix malformed test command (missing space before ]) ... `[ "${head_ref}" != "${avoid_ref}"]` is missing the space before the closing bracket, so the shell prints `[: missing ']'` and the avoid_ref assertion never actually runs (it is inside an `if` condition, so the error was non-fatal and silently disabled the check)	2026-06-03 14:53:48 -07:00
Jordan Harband	16f6ddc3d5	[actions] set per-job permissions in the nvm install workflow ... CodeQL (actions/missing-workflow-permissions) flagged the matrix, test, and finisher jobs of nvm-install-test.yml for not declaring permissions. Every other test workflow already sets least-privilege permissions per job; add them here to match: `contents: read` for the matrix and test jobs, `contents: none` for the no-op finisher.	2026-06-03 14:09:02 -07:00
Jordan Harband	c4d9239cbb	[actions] allow DockerHub's CloudFront CDN so image pulls aren't blocked ... harden-runner runs with `egress-policy: block`, and the allow-list only included `production.cloudflare.docker.com`. DockerHub serves image blobs from either its Cloudflare or its CloudFront CDN; when a pull was routed to CloudFront (`production.cloudfront.docker.com`) the connection was dropped, causing `error pulling image configuration: ... connect: connection refused` and exit 125 in the xenial, installation_node, and fast (httpbin) suites. Allow both CDNs.	2026-06-03 13:29:38 -07:00
Jordan Harband	dd13968c1e	[Tests] reduce CI flakiness from transient Docker registry failures ... Container-based suites and the `nvm_download` httpbin check hard-fail whenever DockerHub is briefly unreachable (observed: `dial tcp ...:443: connect: connection refused` while pulling images), even though the change under test is fine. This is unrelated to any test logic. - tests-xenial / tests-installation-node: retry the `docker pull` up to 5 times before `docker run`, mirroring the existing apt-get retry - `nvm_download` test: retry the httpbin pull and skip the auth-header checks (rather than fail) when the image cannot be pulled or run, and make cleanup tolerant of a missing container.	2026-06-03 13:12:31 -07:00
Jordan Harband	53855417eb	[Fix] avoid an unbound variable ... Fixes #3820	2026-05-15 23:08:37 -07:00
Jordan Harband	1f70aea0b0	[Docs] fix --offline help line alignment	2026-05-05 17:16:31 -07:00
Jordan Harband	d200a21559	[Fix] nvm_normalize_lts: only reject uppercase for LTS names, not regular aliases ... The lowercase check was in the `*` catch-all branch of the `case` statement, rejecting any alias name with uppercase characters. This prevented creating or reading uppercase aliases like `TESTY`. The check should only apply to `lts/*` patterns, since LTS codenames are always lowercase. Fixes #3764. Bug introduced in https://github.com/nvm-sh/nvm/commit/9fb9dec710e2c8d3aecb24c2a7c9f9fb45b1435b as part of fixing #3417.	2026-03-26 10:18:40 -07:00
Jordan Harband	3e37cfb9c4	[Tests] fix 4 test failures ... - `nvm-exec` test: expect "Found .nvmrc" message in output, since `nvm_rc_version` now outputs it to stdout via fd 3 redirection (ef162036) - `nvm_install_binary_nosource`: fix exit code capture by running the command directly instead of inside a subshell with `echo $?` (05d78477) - `nvm_iojs_version_has_solaris_binary`: bare versions like `v3.3.1` (without `iojs-` prefix) are node versions and should be rejected. The old tests relied on the buggy comparison that let them through (53e6244a) - `nvm_get_arch_unofficial`: copy `uname` into the chroot. The old test passed only because the unconditional `NVM_ARCH=x64-musl` masked the missing binary, but the `case` fix now requires a real arch to match (39e71eab)	2026-03-24 12:44:21 -07:00
Jordan Harband	ae31eceecc	[meta] run npm run doctoc	2026-03-24 11:50:40 -07:00
Jordan Harband	25f836f81b	[Fix] install.sh: check mkdir return codes ... If directory creation fails (e.g., permissions), the script would continue and fail with confusing errors later. Fail early with a clear error message instead. Bugs introduced in https://github.com/nvm-sh/nvm/commit/68bf93514bdf5d2852ce74819350bed99eb472a6, https://github.com/nvm-sh/nvm/commit/6cee20a0717298658dd222efbc94616d96d5224c, and https://github.com/nvm-sh/nvm/commit/703babe60a9fa09d55fbc67107a07beba09ef090.	2026-03-14 14:44:24 -07:00
Jordan Harband	581317d05c	[Fix] install.sh: fix POSIX compliance, printf format strings, and profile detection ... - Use `=` instead of `==` for string comparison (POSIX compliance) - Use `printf '%b'` instead of variable as format string (prevents `%` characters in paths from being interpreted as format specifiers) - Fix `TRIED_PROFILE` to reference `PROFILE` instead of `NVM_PROFILE` which is known to be empty at that point Bugs introduced in https://github.com/nvm-sh/nvm/commit/a24ff3e605d72ce9480b2fad2d4c93f0093fa306, https://github.com/nvm-sh/nvm/commit/b6f1c156da0a998442c1c0b638082390118085ce, and https://github.com/nvm-sh/nvm/commit/a461a0fffcd9163726f4d652ca974d97a1097356 (PR https://github.com/nvm-sh/nvm/pull/1605).	2026-03-14 14:41:11 -07:00
Jordan Harband	018958cc47	[Fix] nvm which: show alias name in infinite loop error message ... `${2}` was empty because positional parameters had been consumed by `shift` in the argument parsing loop. Use `${provided_version}` which holds the resolved alias name. Bug introduced in https://github.com/nvm-sh/nvm/commit/1c00753fd9c3ae5e0bb3f4992c5e283ef5bf5085.	2026-03-14 14:37:57 -07:00
Jordan Harband	33fda56d2a	[Fix] nvm uninstall: fix alias cleanup glob expansion ... The `*` glob was inside double quotes, preventing shell expansion. `nvm_grep -l` received a literal `*` filename instead of the list of alias files, so aliases pointing to uninstalled versions were never cleaned up. Bug introduced in https://github.com/nvm-sh/nvm/commit/7807a9f09e54a14f9b493c6eeda0022489cda496.	2026-03-14 14:34:06 -07:00
Jordan Harband	32863e1052	[Fix] nvm debug: use default empty values for potentially unset variables ... These variables may be unset, which would cause errors in shells with `set -u` (nounset) enabled. Bugs introduced in https://github.com/nvm-sh/nvm/commit/b805de80cb8058f09fecb11786d4d2aec0dac35c and https://github.com/nvm-sh/nvm/commit/49e446dac18ce8b1267107849e6be6fe99390543.	2026-03-14 14:33:35 -07:00
Jordan Harband	53e6244aa6	[Fix] nvm_iojs_version_has_solaris_binary: fix comparison to detect non-iojs versions ... Missing `_` prefix on the right side of the comparison meant the guard clause that rejects non-iojs versions almost never matched, allowing non-iojs versions to fall through. Bug introduced in https://github.com/nvm-sh/nvm/commit/2d692d9d7863844d9e5c499faf423b350235f72d / #854.	2026-03-14 14:30:54 -07:00
Jordan Harband	bda39c7c04	[Fix] nvm_download_artifact: fix error propagation from subshells ... `return` inside `(...)` subshells only exits the subshell, not the calling function. Errors in mkdir, download, and checksum verification were silently ignored. Use `{ ...; }` brace groups instead. Bug introduced in https://github.com/nvm-sh/nvm/commit/ba3ad8e460b88aab281d203276b354ce3c9df52f.	2026-03-14 14:30:31 -07:00
Jordan Harband	05d784772c	[Fix] nvm_install_binary: return failure when binary download fails with -b ... When `nosource=1` (the `-b` flag) and binary download fails, the function returned 0 (success) instead of a non-zero exit code, masking the installation failure. Bug introduced in https://github.com/nvm-sh/nvm/commit/4fdef427e4b1d563570ba5992bdbe7d7e23f025b / #2439.	2026-03-14 14:29:56 -07:00
Jordan Harband	39e71eab49	[Fix] nvm_get_arch: only apply musl suffix on x64 Alpine ... Alpine detection unconditionally set `x64-musl` regardless of actual architecture, which would be incorrect on ARM-based Alpine containers. Bug introduced in https://github.com/nvm-sh/nvm/commit/ef7fc2f2c06ad75fe7fbabf28d427561ae7b007d / #3212. Fixes #3616.	2026-03-14 12:53:14 -07:00
Jordan Harband	a5de7236d6	[Fix] nvm_get_arch: add command prefix to uname call ... Other `uname` calls in the file use `command uname` to bypass any user aliases or functions. This one was missing it. Bug introduced in https://github.com/nvm-sh/nvm/commit/779a34e6a9adcb6528e63d595a7a8838747e7915 / #2469.	2026-03-14 10:16:48 -07:00
Jordan Harband	d2fca5f0ec	[Fix] nvm_resolve_local_alias: avoid using variable as printf format string ... Using a variable as the format string means `%` characters in alias names would be interpreted as format specifiers. Use `%b` format with the variable as an argument to safely interpret `\n` escapes. Bug introduced in https://github.com/nvm-sh/nvm/commit/9b91734f0b6a210f0c6655b0cd25ea288d9ae376.	2026-03-14 10:16:25 -07:00
Jordan Harband	3c48015986	[Fix] nvm_get_mirror: fix awk URL validation to actually reject invalid URLs ... The awk expression `$0 ~ "regex"` as a bare statement in the action block evaluates the match but doesn't affect the exit code. awk always prints the line and exits 0, making the validation a no-op. Bug introduced in https://github.com/nvm-sh/nvm/commit/b1fa143dd8cbebd9847972c08fb383646ca00642.	2026-03-14 09:40:43 -07:00
Jordan Harband	a937cb595e	[Fix] nvm_ls_remote_combined: propagate iojs remote listing failures ... `return $A || $B` only evaluates the first argument, since `return` always succeeds. The io.js exit code was never checked, silently swallowing remote listing failures. Bug introduced in https://github.com/nvm-sh/nvm/commit/ea127846290b1126556e256031383d171a4e38d8 / #616.	2026-03-14 09:39:20 -07:00
Jordan Harband	6f428a10bb	[Fix] nvm install: fix nvm err typo to nvm_err for -s/-b conflict ... The error message for using `-s` and `-b` together was calling `nvm err` (invoking nvm with subcommand "err") instead of the `nvm_err` helper function, causing the error message to never be displayed and instead showing the help text with exit code 127. Bug introduced in https://github.com/nvm-sh/nvm/commit/4fdef427e4b1d563570ba5992bdbe7d7e23f025b / #2439	2026-03-14 09:38:04 -07:00
Jordan Harband	ef1620361a	[Refactor] nvm_rc_version: use fd 3 instead of exported env var for multiple return	2026-03-12 15:32:17 -04:00
Jordan Harband	a27a8b7da8	[Fix] nvm alias: fix colors not showing by default ... Colors were lost because `nvm_has_colors` checks `[ -t 1 ]`, which is false inside the `(...) | sort` pipeline in `nvm_list_aliases`. Evaluate `nvm_has_colors` before the pipe and propagate via `NVM_HAS_COLORS`, matching the approach used by `nvm_print_versions`. Bug introduced in https://github.com/nvm-sh/nvm/commit/35212c134652197c779dffd42b9567a270355987.	2026-03-13 16:13:41 -04:00
Jordan Harband	59bd32be6b	[New] nvm install --offline: install from cache without network access ... Add `--offline` flag to `nvm install` that resolves versions using only locally installed versions and cached downloads. No network calls are made. New helper functions `nvm_ls_cached` and `nvm_offline_version` scan `$NVM_DIR/.cache/bin/` for previously downloaded tarballs. In offline mode, `nvm_download_artifact` returns cached tarballs directly without checksum verification or download attempts. The curl/wget requirement is skipped when `--offline` is set. Supports `--lts` via locally stored LTS alias files.	2026-03-13 16:13:19 -04:00
Jordan Harband	14d01c6877	[Tests] add try/try_err helpers; convert tests to use them ... Add `try` and `try_err` helper functions to `test/common.sh` that capture stdout/stderr and exit code from a single invocation, eliminating duplicate command executions in tests. Convert all existing tests that used the `OUTPUT`/`EXIT_CODE` double-invocation pattern to use the new helpers. Also fixes a pre-existing bug in the `nvm_die_on_prefix` test where ASCII apostrophes were used instead of U+2019 to match nvm.sh output.	2026-03-13 15:26:07 -04:00
Jordan Harband	4c556a19b0	[actions] add workflow to update nodejs.org nvm version ... Ref: https://github.com/nodejs/nodejs.org/issues/8628	2026-02-19 21:27:42 -08:00
Jordan Harband	62387b8f92	v0.40.4	2026-01-29 14:12:03 -08:00
Jordan Harband	242d997da5	[actions] add permissions to GHA workflow	2026-01-29 13:08:52 -08:00
Jordan Harband	6d761baef8	[Tests] add retry logic	2026-01-27 20:59:30 -08:00
Jordan Harband	973840565e	[Tests] add a timeout	2026-01-27 18:09:39 -08:00
Jordan Harband	b1dd81097f	[actions] add a finisher to lint	2026-01-27 17:29:59 -08:00
Jordan Harband	4c3edc5e56	[actions] migrate Travis CI tests to GitHub Actions	2026-01-27 10:31:07 -08:00
Jordan Harband	35212c1346	[Fix] nvm_has_colors: also check if stdout is a terminal	2026-01-27 10:31:07 -08:00
Jordan Harband	d2f93c1c8e	[Tests] mock node in nvm_die_on_prefix test	2026-01-26 22:34:24 -08:00
Jordan Harband	f9dbad9832	[Tests] clean up .nvmrc after nvm-exec test	2026-01-26 22:34:07 -08:00
Jordan Harband	c451e7a4c8	[Tests] set $_ before sourcing nvm.sh in fast tests ... nvm.sh uses `NVM_SCRIPT_SOURCE="$_"` to detect its source location. Adding `: nvm.sh` before each source line ensures `$_` is set correctly, preventing breakage when the previous command (e.g., `set -ex`) overwrites it.	2026-01-26 21:41:57 -08:00
Jordan Harband	9ba422e561	[Fix] nvm_strip_path: avoid gawk-specific RT variable for mawk compatibility	2026-01-26 22:33:58 -08:00
Jordan Harband	38ffc713a7	[Fix] nvm_get_default_packages: use portable awk patterns ... Replace POSIX `[[:space:]]` character class with `[ \t]` for mawk compatibility on Ubuntu 16.04.	2026-01-26 16:59:11 -08:00
Jordan Harband	aa1d829102	[Fix] nvm_install_source: explicitly set SHELL=/bin/sh for make ... Old Node.js versions have Makefiles with unquoted glob patterns like `rm -f *.o` that fail in zsh's strict glob mode. By passing SHELL=/bin/sh to make, we ensure POSIX-compliant shell behavior regardless of what shell nvm is running in.	2026-01-26 15:07:01 -08:00
Jordan Harband	26066c10f4	[Fix] prevent sed errors when pattern contains # ... When `.nvmrc` or alias files contained comments (lines with `#`), the `#` character could end up in the search pattern passed to sed, causing "unterminated regular expression" errors because `#` is used as the sed address delimiter. This commit fixes the issue in two places: 1. `nvm_alias`: Strip comments from alias file contents before returning them, and trim trailing whitespace 2. `nvm_ls`: Escape `#` characters in SEARCH_PATTERN so they're treated as literal characters in the sed address Fixes #3761	2026-01-19 22:58:43 -08:00
Jordan Harband	80cc4cdd1b	[actions] fix WSL tests: properly handle Debian apt sources issue ... The Debian WSL image has stale apt sources pointing to ftp.debian.org which no longer hosts bullseye-backports. The previous fix ran sed after the first setup-wsl call, but that call already failed because additional-packages triggers apt-get update internally. For wsl_matrix (Debian, Ubuntu): - Ubuntu: install with additional-packages directly (no apt issue) - Debian: install without packages first, fix sources.list with sed, then run apt-get update/upgrade/install manually with retries - Exclude Debian and Ubuntu-20.04 --lts since node v24+ has exec format errors on WSL1 For wsl_matrix_unofficial (Alpine): - Remove the Debian-specific fix entirely (Alpine uses apk, not apt) See https://github.com/Vampire/setup-wsl/issues/76	2026-01-19 22:11:41 -08:00
Jordan Harband	630a01f2e0	[actions] fix workflow permissions; add codeQL	2025-12-03 12:05:43 -08:00
Jordan Harband	b7b7e949aa	[Dev Deps] update markdown-link-check, semver	2025-12-03 12:02:29 -08:00
Jordan Harband	e4e34ec7f9	[Dev Deps] update markdown-link-check, semver	2025-11-10 10:07:02 +07:00
Jordan Harband	44a3cdb9b8	[actions] temporary fix for WSL issue ... See https://github.com/Vampire/setup-wsl/issues/76	2025-09-05 09:20:08 -07:00
Jordan Harband	34ace599a2	[actions] add new github asset domain to harden-runner	2025-09-05 13:19:27 -07:00
Jordan Harband	18e638e42b	[Fix] sh lacks -O; thanks shellcheck	2025-09-05 13:19:27 -07:00
Jordan Harband	1aab8b2d5a	[meta] update security policy; add IRP	2025-06-23 16:11:48 -07:00
Jordan Harband	977563e97d	v0.40.3	2025-04-23 16:34:31 -07:00
Jordan Harband	ffec9fec72	v0.40.2 ... New Stuff - `install.sh`: add `$ZDOTDIR` to zsh search (#3458) Fixes - `reinstall-packages`: do not reinstall corepack (#3544) - avoid bash-specific syntax (#3499) - `install-latest-npm`: npm v11 is out - `nvm_install_latest_npm`: avoid unbound variable (#3447) - give a more helpful message when `lts` alias is mistakenly used (#3441) - `nvm ls`, `nvm alias`, `nvm install`: error when an LTS name is invalid - `nvm_normalize_lts`: error when an LTS name is not lowercase (#3417) Documentation - [readme] update link - [readme] fix `--no-use` example (#3479) - [readme] update copyright notice (#3507) - [readme] note zsh-nvm's AUTO_USE option (#2670) - [readme] add note about reloading zshrc after editing (#3052) - [readme] Update shell profile file install notes (#2241) - [readme] add docker tips (#2318) - [readme] remove `avn` from readme (#3469) - [readme] fnm -> nvm.fish (#2541) Refactors - prefer `case` over if/else chains - combine `sed -e` invocations/arguments Tests - `nvm exec`/`nvm run`: add `--silent` tests (#1259) - [actions] release test needs git tags - migrate `installation_iojs` test suite to GitHub Actions (#3476) - Migrate slow test suite from Travis CI (#3470) - temporarily skip this failing travis test to unblock progress - [actions] TOC: use latest LTS node - `install.sh`: clean up `nvm_detect_profile` tests - `nvm_detect_profile`: refactor (#3467) - run urchin tests on pull requests (#3466) - update mocks - ensure that unit tests use only mocked LTS names - [actions] use `node/install` instead of `node/run` Meta - disable blank issues - update issue template - add DCO (#3456) - Rename .github/ISSUE_TEMPLATE.md to .github/ISSUE_TEMPLATE/ISSUE_TEMPLATE.md (#3454)	2025-03-11 13:39:40 -07:00
Jordan Harband	df9ac58f8b	[Fix] reinstall-packages: do not reinstall corepack ... Fixes #3544	2025-03-11 13:16:58 -07:00
Jordan Harband	8dbc22f0e9	[readme] update link	2025-03-11 12:46:19 -07:00
Jordan Harband	5c12680a62	[Dev Deps] update markdown-link-check	2025-03-11 12:43:42 -07:00
Jordan Harband	f0f8ed12f8	[Dev Deps] update semver	2025-02-25 18:39:49 -08:00
Jordan Harband	759f70f196	[Refactor] prefer case over if/else chains	2025-02-04 22:45:35 -08:00
Jordan Harband	06a9179309	[Refactor] combine sed -e invocations/arguments	2025-02-04 15:55:10 -08:00
Jordan Harband	74eb396099	[Dev Deps] update markdown-link-check	2025-02-04 12:37:59 -08:00
Jordan Harband	e5521cfd3c	[actions] release test needs git tags	2025-02-04 08:44:48 -08:00
Jordan Harband	1d39e35bd3	[Tests] temporarily skip this failing travis test to unblock progress	2025-02-03 10:50:19 -08:00
Jordan Harband	19f452ba0f	[Fix] avoid bash-specific syntax ... Fixes #3499	2024-12-19 22:21:09 -08:00
Jordan Harband	b9b0ea8020	[Fix] install-latest-npm: npm v11 is out	2024-12-16 10:03:18 -08:00
Jordan Harband	dae1427f7e	[actions] TOC: use latest LTS node	2024-11-22 21:44:36 -10:00
Jordan Harband	0641363102	[Tests] install.sh: clean up nvm_detect_profile tests	2024-11-08 10:52:41 +00:00
Jordan Harband	9659af6c16	[Tests] nvm_detect_profile: refactor	2024-11-12 13:31:36 -08:00
Jordan Harband	d648a3b1ba	[meta] add DCO ... imo this shouldn‘t be necessary, as it should be implied by the act of making a PR. Following the example in https://github.com/expressjs/express/pull/6048 See https://github.com/openjs-foundation/project-status/issues/2	2024-11-05 17:51:39 +00:00