feat(pipeline): the pure pipeline, checked against its goldens

The modules that turn a frozen range into passes and passes into a
verdict: :Schema, :Tools (the DENY, git and base lists, a leaf, so the
config and the pool both read them), :Plan, :Artifact, :Inventory,
:Dispositions, :State, :Normalize, :Stream, :Prompt, :Config, and the URL
half of :PgWire. Units, jobs, results, the config and the claude stream
stay JSON from end to end, so every key keeps its place; :Text gives JSON
values truthiness, get, ==, membership and Str(), and throws Dyn::Error
under a name for each kind of unexpected data.

Exactness where the bytes are seen:
- the stream classifiers are hand-written scanners with case-insensitive
  folds (the Kelvin sign is a k), Unicode \b and \s, and `.` stopping at
  \n; the SQL call detection takes `<exe> sql` with any path before it;
- config problems carry the JSON scanner's and the UTF-8 decoder's own
  texts, positions included;
- a float total is Neumaier's compensated sum, and the build passes
  -ffp-contract=off: -march=native would fuse a*b+c;
- `crafter-review inventory` and `config` print what the goldens record,
  byte for byte (the fixture and 14 configs, broken ones included); a
  message names this binary as the command to run.

tests/Goldens: the recorded answers over the fixtures next to them (the
commits fixture, a rich versioned series with CRLF, a folded subject,
Markdown headings in and out of fences, the code-key rewrites); the suite
compares 2304 results as DumpFile writes them, so a key out of order fails
too. All agree.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
This commit is contained in:
Jorijn van der Graaf 2026-10-31 06:48:55 +01:00
commit 741e3a656d
Signed by: jorijnvdgraaf
GPG key ID: 2937E59CDCC1BCFB
70 changed files with 30565 additions and 21 deletions

View file

@ -0,0 +1,116 @@
// SPDX-License-Identifier: AGPL-3.0-only
// SPDX-FileCopyrightText: Copyright (C) 2026 Catcrafts®
module Crafter.Review:Artifact_impl;
import :Artifact;
import :Text;
import :Crypto;
import :Platform;
import :Process;
import std;
namespace fs = std::filesystem;
namespace Crafter::Review::Artifact {
namespace {
std::string ReadBytes(const fs::path& path) {
std::optional<std::string> bytes = Platform::ReadFile(path);
if (!bytes) throw std::runtime_error(std::format("{}: cannot read", path.string()));
return std::move(*bytes);
}
// Four decimal digits and a '-' at `at`, as (\d{4})-: the number.
std::optional<std::int64_t> FourDigits(std::string_view name, std::size_t at) {
std::size_t end = at;
for (std::int32_t i = 0; i < 4; ++i) {
if (end >= name.size()) return std::nullopt;
const auto [c, used] = Dyn::CodePointAt(name, end);
if (!Dyn::IsDecimal(c)) return std::nullopt;
end += used;
}
if (end >= name.size() || name[end] != '-') return std::nullopt;
return Dyn::ParseInt(name.substr(at, end - at));
}
}
std::string Sha256File(const fs::path& path) {
return Sha256Hex(ReadBytes(path));
}
std::string ArtifactHash(std::string_view mode, const fs::path& path) {
if (mode == "reply") return Sha256File(path);
std::string all;
for (const std::string& name : PatchFiles(path)) all += ReadBytes(path / name);
return Sha256Hex(all);
}
std::vector<std::string> PatchFiles(const fs::path& path) {
std::vector<std::string> out;
for (const fs::directory_entry& entry : fs::directory_iterator(path)) {
std::string name = entry.path().filename().string();
if (name.ends_with(".patch")) out.push_back(std::move(name));
}
std::ranges::sort(out);
return out;
}
std::optional<std::int64_t> PatchNumber(std::string_view name) {
if (name.starts_with('v')) {
const std::size_t digits = Dyn::DecimalRun(name, 1);
if (digits > 0 && 1 + digits < name.size() && name[1 + digits] == '-') {
if (const std::optional<std::int64_t> k = FourDigits(name, 2 + digits)) return k;
}
}
return FourDigits(name, 0);
}
std::map<std::int64_t, std::string> PatchTexts(const fs::path& path) {
std::map<std::int64_t, std::string> out;
for (const std::string& name : PatchFiles(path)) {
const std::optional<std::int64_t> k = PatchNumber(name);
if (!k) throw Dyn::Error("AttributeError", "'NoneType' object has no attribute 'group'");
out[*k] = Dyn::ReadText(path / name);
}
return out;
}
std::optional<std::string> PatchHead(const fs::path& path) {
const std::vector<std::string> files = PatchFiles(path);
if (files.empty()) return std::nullopt;
// open(errors='replace').readline().split(): the first line, universal
// newlines and all.
const std::string text = Dyn::UniversalNewlines(Dyn::DecodeReplace(ReadBytes(path / files.back())));
const std::vector<std::string> first = Dyn::Split(std::string_view(text).substr(0, text.find('\n')));
if (first.size() > 1 && first[0] == "From" && Dyn::Length(first[1]) == 40) return first[1];
return std::nullopt;
}
std::string Git(const fs::path& tree, std::span<const std::string> args) {
std::vector<std::string> argv = {"git", "-C", tree.string()};
argv.insert(argv.end(), args.begin(), args.end());
const Process::Captured run = Process::Run(argv);
if (run.returnCode != 0) {
const std::string& said = run.err.empty() ? run.out : run.err;
throw Dyn::Error("RuntimeError", std::format("git -C {} {}: {}", tree.string(), Dyn::Join(args, " "),
Dyn::Strip(Dyn::UniversalNewlines(Dyn::DecodeReplace(said)))));
}
return Dyn::UniversalNewlines(Dyn::DecodeReplace(run.out));
}
bool IsToplevel(const fs::path& tree) {
try {
const std::string top = Dyn::Strip(Git(tree, std::array<std::string, 2>{"rev-parse", "--show-toplevel"}));
return fs::weakly_canonical(fs::absolute(top)) == fs::weakly_canonical(fs::absolute(tree));
} catch (const std::exception&) {
return false;
}
}
std::string TreeHead(const fs::path& tree) {
return Dyn::Strip(Git(tree, std::array<std::string, 2>{"rev-parse", "HEAD"}));
}
std::string TreeDirty(const fs::path& tree) {
return Dyn::Strip(Git(tree, std::array<std::string, 3>{"status", "--porcelain", "--untracked-files=no"}));
}
}