add_vendor_origins.py

#!/usr/bin/env python3

import json
import os


def direct_dep(pkg):
    for d in {
            "github.com/Bowery/prompt",
            "github.com/apache/thrift",
            "github.com/cespare/xxhash",
            "github.com/dchest/safefile",
            "github.com/elastic/apm-agent-go",
            "github.com/elastic/beats",
            "github.com/elastic/go-elasticsearch",
            "github.com/go-sourcemap/sourcemap",
            "github.com/gofrs/uuid",
            "github.com/gogo",
            "github.com/golang/groupcache/lru",
            "github.com/golang/lint",
            "github.com/google/addlicense",
            "github.com/google/go-github/github",
            "github.com/google/go-querystring",
            "github.com/google/pprof",
            "github.com/google/shlex",
            "github.com/grpc-ecosystem/grpc-gateway",
            "github.com/hashicorp/golang-lru",
            "github.com/haya14busa/errorformat",
            "github.com/haya14busa/reviewdog",
            "github.com/jaegertracing/jaeger",
            "github.com/kabukky/httpscerts",
            "github.com/kardianos/govendor",
            "github.com/mattn/go-shellwords",
            "github.com/open-telemetry/opentelemetry-collector",
            "github.com/opentracing/opentracing-go",
            "github.com/patrickmn/go-cache",
            "github.com/pierrre/gotestcover",
            "github.com/reviewdog/errorformat",
            "github.com/reviewdog/reviewdog",
            "github.com/ryanuber/go-glob",
            "github.com/santhosh-tekuri/jsonschema",
            "github.com/sergi/go-diff/diffmatchpatch",
            "github.com/stretchr/testify",
            "github.com/t-yuki/gocover-cobertura",
            "github.com/uber/tchannel-go",
            "github.com/xanzy/go-gitlab",
            "github.com/yudai/gojsondiff",
            "go.elastic.co/apm",
            "golang.org/x/lint",
            "golang.org/x/net/netutil",
            "golang.org/x/sync/errgroup",
            "golang.org/x/tools/cmd/benchcmp",
            "golang.org/x/tools/cmd/goimports",
            "golang.org/x/tools/go/ast/astutil",
            "golang.org/x/tools/imports",
            "golang.org/x/tools/internal/fastwalk",
            "golang.org/x/tools/internal/gopathwalk",
    }:
        if pkg.startswith(d):
            return True


def main():
    with open("vendor/vendor.json") as f:
        vendor = json.load(f)

    with open("_beats/vendor/vendor.json") as f:
        beats_vendor = json.load(f)
        beats_packages = [p["path"] for p in beats_vendor["package"]]

    packages = []

    for package in vendor["package"]:
        origin = package.get("origin")
        path = package["path"]

        pkg_vendor_path = os.path.join("vendor", path)

        if not os.path.exists(pkg_vendor_path):
            print(f"{path} is missing, removing...")
            continue

        if "origin" in package:
            if pkg_vendor_path not in package["origin"]:
                print(f"{path} origin {origin} is incorrect")
        else:
            if direct_dep(path):
                pass
            elif path in beats_packages:
                print(f"{path} has no origin but pulled in from beats")
                package["origin"] = f"github.com/elastic/beats/vendor/{path}"
            else: # not direct_dep(path):
                print(f"{path} missing origin but it's not from beats")
        packages.append(package)

    vendor["package"] = packages

    with open("vendor/vendor.json", "w") as f:
        json.dump(vendor, f, indent="\t", sort_keys=True)
        f.write("\n")


if __name__ == "__main__":
    main()