feedkit/config/load.go

package config

import (
	"fmt"
	"os"
	"sort"
	"strings"

	"gopkg.in/yaml.v3"
)

// Load reads the YAML config file from disk, decodes it into Config,
// and then validates it.
//
// Important behaviors:
// - Uses yaml.Decoder with KnownFields(true) to catch typos like "srouces:".
// - Returns a validation error that (usually) contains multiple issues at once.
func Load(path string) (*Config, error) {
	if strings.TrimSpace(path) == "" {
		path = "config.yml"
	}

	raw, err := os.ReadFile(path)
	if err != nil {
		return nil, fmt.Errorf("config.Load: read %q: %w", path, err)
	}

	var cfg Config

	dec := yaml.NewDecoder(strings.NewReader(string(raw)))
	dec.KnownFields(true) // strict mode for struct fields

	if err := dec.Decode(&cfg); err != nil {
		return nil, fmt.Errorf("config.Load: parse YAML %q: %w", path, err)
	}

	// Optional: ensure there isn't a second YAML document accidentally appended.
	var extra any
	if err := dec.Decode(&extra); err == nil {
		return nil, fmt.Errorf("config.Load: %q contains multiple YAML documents; expected exactly one", path)
	}

	if err := cfg.Validate(); err != nil {
		return nil, err
	}

	return &cfg, nil
}

// Validate checks whether the config is internally consistent and safe to run.
//
// This is intentionally DOMAIN-AGNOSTIC validation:
// - required fields are present
// - names are unique
// - durations are > 0
// - routes reference defined sinks
//
// We DO NOT enforce domain-specific constraints like "allowed kinds" or
// "NWS requires a user-agent". Those belong in the domain module (weatherfeeder).
func (c *Config) Validate() error {
	var m multiError

	// ---------- sources ----------
	if len(c.Sources) == 0 {
		m.Add(fieldErr("sources", "must contain at least one source"))
	} else {
		seen := map[string]bool{}
		for i, s := range c.Sources {
			path := fmt.Sprintf("sources[%d]", i)

			// Name
			if strings.TrimSpace(s.Name) == "" {
				m.Add(fieldErr(path+".name", "is required"))
			} else {
				if seen[s.Name] {
					m.Add(fieldErr(path+".name", fmt.Sprintf("duplicate source name %q (source names must be unique)", s.Name)))
				}
				seen[s.Name] = true
			}

			// Driver
			if strings.TrimSpace(s.Driver) == "" {
				m.Add(fieldErr(path+".driver", "is required (e.g. openmeteo_observation, rss_feed, ...)"))
			}

			// Every
			if s.Every.Duration <= 0 {
				m.Add(fieldErr(path+".every", "must be a positive duration (e.g. 15m, 1m, 30s)"))
			}

			// Kind (optional but if present must be non-empty after trimming)
			if s.Kind != "" && strings.TrimSpace(s.Kind) == "" {
				m.Add(fieldErr(path+".kind", "cannot be blank (omit it entirely, or provide a non-empty string)"))
			}

			// Params can be nil; that's fine.
		}
	}

	// ---------- sinks ----------
	sinkNames := map[string]bool{}
	if len(c.Sinks) == 0 {
		m.Add(fieldErr("sinks", "must contain at least one sink"))
	} else {
		for i, s := range c.Sinks {
			path := fmt.Sprintf("sinks[%d]", i)

			if strings.TrimSpace(s.Name) == "" {
				m.Add(fieldErr(path+".name", "is required"))
			} else {
				if sinkNames[s.Name] {
					m.Add(fieldErr(path+".name", fmt.Sprintf("duplicate sink name %q (sink names must be unique)", s.Name)))
				}
				sinkNames[s.Name] = true
			}

			if strings.TrimSpace(s.Driver) == "" {
				m.Add(fieldErr(path+".driver", "is required (stdout|file|postgres|rabbitmq|...)"))
			}

			// Params can be nil; that's fine.
		}
	}

	// ---------- routes ----------
	// Routes are optional. If provided, validate shape + references.
	for i, r := range c.Routes {
		path := fmt.Sprintf("routes[%d]", i)

		if strings.TrimSpace(r.Sink) == "" {
			m.Add(fieldErr(path+".sink", "is required"))
		} else if !sinkNames[r.Sink] {
			m.Add(fieldErr(path+".sink", fmt.Sprintf("references unknown sink %q (define it under sinks:)", r.Sink)))
		}

		if len(r.Kinds) == 0 {
			// You could relax this later (e.g. empty == "all kinds"), but for now
			// keeping it strict prevents accidental "route does nothing".
			m.Add(fieldErr(path+".kinds", "must contain at least one kind"))
		} else {
			for j, k := range r.Kinds {
				kpath := fmt.Sprintf("%s.kinds[%d]", path, j)
				if strings.TrimSpace(k) == "" {
					m.Add(fieldErr(kpath, "kind cannot be empty"))
				}
			}
		}
	}

	return m.Err()
}

// ---- error helpers ----

// fieldErr produces consistent "path: message" errors.
func fieldErr(path, msg string) error {
	return fmt.Errorf("%s: %s", path, msg)
}

// multiError collects many errors and returns them as one error.
// This makes config iteration much nicer: you fix several things per run.
type multiError struct {
	errs []error
}

func (m *multiError) Add(err error) {
	if err == nil {
		return
	}
	m.errs = append(m.errs, err)
}

func (m *multiError) Err() error {
	if len(m.errs) == 0 {
		return nil
	}
	// Sort for stable output (useful in tests and when iterating).
	sort.Slice(m.errs, func(i, j int) bool {
		return m.errs[i].Error() < m.errs[j].Error()
	})
	return m
}

func (m *multiError) Error() string {
	var b strings.Builder
	b.WriteString("config validation failed:\n")
	for _, e := range m.errs {
		b.WriteString("  - ")
		b.WriteString(e.Error())
		b.WriteString("\n")
	}
	return strings.TrimRight(b.String(), "\n")
}