- Write a subset of the tab required for bottles as an annotation.
- Add option on new bottle creation to skip writing tab into bottle
and instead add it (and other useful metadata) to bottle JSON.
- Read formula information and tab from bottle JSON.
- Write prettier JSON to disk.
- Don't write `HEAD` to tab; this duplicates `HOMEBREW_VERSION`.
- Allow `brew bottle` to use `--json` to generate JSON files from a
local bottle file.
- make the description more generic/correct
- use "internet archive" over "archive"
- move some logic to a new `GitHubReleases` class (for consistency)
- remove some obvious comments
- extract out and move some constants
> Bootsnap is a library that plugs into Ruby, with optional support
> for ActiveSupport and YAML, to optimize and cache expensive
> computations.
https://github.com/Shopify/bootsnap
For our case that translates to "repeated calls to `brew` have
reductions in the time spend `require`ing speeding up the process
boot time".
For example:
```
$ hyperfine --warmup=2 "unset HOMEBREW_BOOTSNAP; brew info wget" "export HOMEBREW_BOOTSNAP=1; brew info wget"
Benchmark #1: unset HOMEBREW_BOOTSNAP; brew info wget
Time (mean ± σ): 2.417 s ± 0.032 s [User: 659.0 ms, System: 855.5 ms]
Range (min … max): 2.382 s … 2.464 s 10 runs
Benchmark #2: export HOMEBREW_BOOTSNAP=1; brew info wget
Time (mean ± σ): 1.862 s ± 0.064 s [User: 425.3 ms, System: 566.8 ms]
Range (min … max): 1.736 s … 1.952 s 10 runs
Summary
'export HOMEBREW_BOOTSNAP=1; brew info wget' ran
1.30 ± 0.05 times faster than 'unset HOMEBREW_BOOTSNAP; brew info wget'
```
We were previously only looking at the `cellar` value when pouring
bottles and ignoring the `prefix` and (implicit) `repository`.
Actually look at these values and set the defaults for each platform.
Also, when we're relocating to create or pour bottles when `prefix`
and `repository` are equal then skip relocating the `repository` and
always use references to the `prefix` instead.
Fixes#9453
The first attempt to build and pull a formula (hello) for linux in homebrew-core
resulted in a wrong cellar line being added to the formula's bottle block.
How to test/debug this, using the 4 bottles that where built for hello:
brew bottle --merge --debug hello--2.10_1.mojave.bottle.json hello--2.10_1.x86_64_linux.bottle.json hello--2.10_1.big_sur.bottle.json hello--2.10_1.catalina.bottle.json
This command would add create the following bottle block:
bottle do
cellar "/home/linuxbrew/.linuxbrew/Cellar"
sha256 "a0af7dcbb5c83f6f3f7ecd507c2d352c1a018f894d51ad241ce8492fa598010f" => :big_sur
sha256 "5334dd344986e46b2aa4f0471cac7b0914bd7de7cb890a34415771788d03f2ac" => :catalina
sha256 "22948764d8f8d7be4870ff92dae64d986eb63a9150b219c20fff87d1a6aa93d6" => :mojave
sha256 "702dc7f78444d2f4f1c19324be654bcbb8b99dd0e9ce26c3e2fbc3b6464a189f" => :x86_64_linux
end
After the change in this PR, the result is the following:
bottle do
sha256 "a0af7dcbb5c83f6f3f7ecd507c2d352c1a018f894d51ad241ce8492fa598010f" => :big_sur
sha256 "5334dd344986e46b2aa4f0471cac7b0914bd7de7cb890a34415771788d03f2ac" => :catalina
sha256 "22948764d8f8d7be4870ff92dae64d986eb63a9150b219c20fff87d1a6aa93d6" => :mojave
sha256 "702dc7f78444d2f4f1c19324be654bcbb8b99dd0e9ce26c3e2fbc3b6464a189f" => :x86_64_linux
end
The brew bottle --merge code will pick the most common cellar line between the 4 bottles, by order of priority:
- non-relocatable (fixed cellar path)
- cellar :any
- cellar :any_skip_relocation
In the case of the hello bottle, the 3 mac bottles are "cellar :any_skip_relocation", and the linux bottle
is non-relocatable. So the linux bottle wins and the code correctly determines that the 4 bottles should
be non-relocatable.
In that case, the /home/linuxbrew/.linuxbrew/Cellar path is defined as cellar, and by convention
we do not write that out to the formula file, hence the cellar path check that needs to be modified
in this PR.
This PR also fixes the same situation for mac ARM cellar paths
- Output `brew doctor` and `brew install` messages noting this configuration is (currently) unsupported and encourage use of Rosetta instead
- Output Rosetta 2 usage in `brew config` on ARM (whether in Rosetta 2 or not)
- Check the architecture of (newly installed) dependencies and ensure they are using the correct architecture.
- Don't allow installing macOS Intel Homebrew in macOS ARM Homebrew default prefix (and vice versa
- Actually write out the architecture of dependencies to the tab rather than generating and throwing them away
- Set and document the expected default prefix for macOS Intel Homebrew, macOS ARM Homebrew (`/opt/homebrew`) and Homebrew on Linux
While we're here:
- Don't say Big Sur is a prerelease version but still make it clear we
don't support it (yet).
- Don't reference non-existent IRC channel
Align the logic in `cleanup_portable_ruby` with that in `ruby.sh`.
Co-authored-by: Maxim Belkin <maxim.belkin@gmail.com>
Co-authored-by: Mike McQuaid <mike@mikemcquaid.com>
Refactor the CLI::Args module so it doesn't have different paths to
check arguments depending on whether the arguments have been parsed or
not. Instead, set the values we need from the global ARGV at
first, global initialisation time where they will be thrown away when
the actual arguments are parsed.
To do this some other general refactoring was needed:
- more methods made private when possible
- e.g. `HEAD?` used consistently instead of `head` before arguments
are parsed.
- formula options are only parsed after named arguments are extracted
- This also required auto-fixes for Layout/EmptyLinesAroundBlockBody and
Layout/InconsistentIndentation once the auto-fixer had got rid of the
"redundant begin"s.
This means that any new environment variables or changes to `bin/brew`
or `brew.sh` will be used in the new process. This also allows the
removal of various fallbacks from autoupdates from old versions.
