From df93c3df9bc4e21d8a11fc5b223de73dc7a32fc2 Mon Sep 17 00:00:00 2001 From: adamXbot <111877622+adamXbot@users.noreply.github.com> Date: Fri, 9 Oct 2026 23:54:03 +1100 Subject: [PATCH] Document --no-color and the live progress board for preview and upgrade privacycommand preview and privacycommand upgrade gained a --no-color option after 0.2.1, matching the one on audit, and on a terminal they now draw a live per-app progress board on stderr instead of a spinner. List the flag in both option tables, in the same wording as the audit table, and say what the board shows and when it appears: one row per app with its current step, download progress for --fetch and upgrade, each app's result as it finishes, cleared before the report prints, and only when stderr is a terminal so piped output and --json are unchanged. Both option lists were checked against the built CLI's preview --help and upgrade --help. llms-full.txt is regenerated. Co-Authored-By: Claude Fable 5.1 --- cli.mdx | 11 +++++++++++ llms-full.txt | 11 +++++++++++ 2 files changed, 22 insertions(+) diff --git a/cli.mdx b/cli.mdx index a41b7c5..21f970d 100644 --- a/cli.mdx +++ b/cli.mdx @@ -180,6 +180,12 @@ it: privacycommand preview --fetch firefox ``` +On a terminal, `preview` draws a live progress board on stderr while it +works: one row per app with the analysis step it's on — with `--fetch`, how +many MB of the download have arrived — and then its result as it finishes. +The board clears before the report prints. It only appears when stderr is a +terminal, so piped output and `--json` are unchanged. + After the apps, it lists what it couldn't analyse: casks that install no app (command-line tools, fonts and the like) with the reason, and your outdated formulae, so you can see everything `brew upgrade` would touch. Without @@ -194,6 +200,7 @@ formulae, so you can see everything `brew upgrade` would touch. Without | `--max-risk ` | Mark each cask cleared or held for review against a risk limit, and exit 3 if anything is held — see [below](#upgrading-within-a-risk-limit) | | `--only-noteworthy` | Suppress apps with nothing to say | | `--min-tier ` | Only report at or above a risk tier — `low`, `medium`, `high` or `critical`. Anything else exits 2 | +| `--no-color` | Plain output. Setting `NO_COLOR` in the environment does the same | | `--json` | Machine-readable output | @@ -250,6 +257,9 @@ compares its risk with the limit — a tier (`low`, `medium`, `high`, leaves them held without asking, and is implied when stdin isn't a terminal and with `--json`. +On a terminal it shows the same live progress board as `preview --fetch`: each +cask's download in MB, then its analysis, then its result. + Casks that install no app, and outdated formulae, are neither analysed nor upgraded. The footer lists them with the `brew` command to run yourself. @@ -260,6 +270,7 @@ upgraded. The footer lists them with the `brew` command to run yourself. | `--no-input` | With `--max-risk`: don't ask about held casks | | `--greedy` | Include casks that normally update themselves | | `--min-tier `, `--only-noteworthy` | As for `preview`. Not with `--max-risk`, which judges every cask | +| `--no-color` | Plain output. Setting `NO_COLOR` in the environment does the same | | `--json` | Machine-readable output; `brew`'s own output goes to stderr | | Exit code | Meaning | diff --git a/llms-full.txt b/llms-full.txt index 02946f1..e883f20 100644 --- a/llms-full.txt +++ b/llms-full.txt @@ -1222,6 +1222,12 @@ it: privacycommand preview --fetch firefox ``` +On a terminal, `preview` draws a live progress board on stderr while it +works: one row per app with the analysis step it's on — with `--fetch`, how +many MB of the download have arrived — and then its result as it finishes. +The board clears before the report prints. It only appears when stderr is a +terminal, so piped output and `--json` are unchanged. + After the apps, it lists what it couldn't analyse: casks that install no app (command-line tools, fonts and the like) with the reason, and your outdated formulae, so you can see everything `brew upgrade` would touch. Without @@ -1236,6 +1242,7 @@ formulae, so you can see everything `brew upgrade` would touch. Without | `--max-risk ` | Mark each cask cleared or held for review against a risk limit, and exit 3 if anything is held — see [below](#upgrading-within-a-risk-limit) | | `--only-noteworthy` | Suppress apps with nothing to say | | `--min-tier ` | Only report at or above a risk tier — `low`, `medium`, `high` or `critical`. Anything else exits 2 | +| `--no-color` | Plain output. Setting `NO_COLOR` in the environment does the same | | `--json` | Machine-readable output | > **Note** @@ -1292,6 +1299,9 @@ compares its risk with the limit — a tier (`low`, `medium`, `high`, leaves them held without asking, and is implied when stdin isn't a terminal and with `--json`. +On a terminal it shows the same live progress board as `preview --fetch`: each +cask's download in MB, then its analysis, then its result. + Casks that install no app, and outdated formulae, are neither analysed nor upgraded. The footer lists them with the `brew` command to run yourself. @@ -1302,6 +1312,7 @@ upgraded. The footer lists them with the `brew` command to run yourself. | `--no-input` | With `--max-risk`: don't ask about held casks | | `--greedy` | Include casks that normally update themselves | | `--min-tier `, `--only-noteworthy` | As for `preview`. Not with `--max-risk`, which judges every cask | +| `--no-color` | Plain output. Setting `NO_COLOR` in the environment does the same | | `--json` | Machine-readable output; `brew`'s own output goes to stderr | | Exit code | Meaning |