Blog

I blamed encoding/json for an allocation spike in an API client. That accusation was directionally correct but not particularly useful. I was asking it to decode into interface{}, inspect maps, convert numbers, and then copy everything into structs. The package was faithfully performing the work I had requested twice.

Here was the convenient version:

var raw interface{}
if err := json.Unmarshal(body, &raw); err != nil {
	return err
}

m := raw.(map[string]interface{})
id := int64(m["id"].(float64))
name := m["name"].(string)

Without a concrete destination, JSON objects become map[string]interface{}, arrays become []interface{}, strings become strings, booleans become bools, null becomes nil, and numbers become float64. Every assertion is another opportunity to panic. Large integer identifiers can also lose precision when represented as floating point.

A concrete struct gives the decoder much better instructions:

type record struct {
	ID   int64  `json:"id"`
	Name string `json:"name"`
}

var r record
if err := json.Unmarshal(body, &r); err != nil {
	return err
}

The decoder scans the input, matches object keys to exported struct fields, allocates strings and nested values as needed, and uses reflection to assign converted values. Struct field metadata is cached internally, so it is not rediscovering every field from first principles on every object. Reflection still has a cost, but my intermediate tree was the larger mistake.

I measured a reduced case with a 40-element array. The exact numbers depend on machine and revision, but the shape was stable:

BenchmarkInterface-4    5000    356000 ns/op    60400 B/op    1012 allocs/op
BenchmarkStruct-4      10000    181000 ns/op    14500 B/op     126 allocs/op

The benchmark reset the destination each iteration and used testing.B. It was not a production trace, but it proved that decoding into a generic tree and translating it was expensive enough to stop doing.

There are a few details behind apparently simple field matching. A json tag overrides the field name. A tag value of - ignores the field. Embedded fields participate in selection, with conflicts resolved according to the package’s rules. Only exported fields are candidates. Case-insensitive matches are accepted, though I prefer exact wire names because ambiguity is not a feature I need.

Missing and null are another source of false confidence. If a field is absent, unmarshaling leaves the existing Go value alone. If I reuse a struct, stale data can survive:

r := record{ID: 99, Name: "old"}
json.Unmarshal([]byte(`{"name":"new"}`), &r)
fmt.Printf("%+v\n", r)

The output still contains ID:99. I decode each independent document into a fresh value unless merging is intentional. For optional scalar fields, a pointer can distinguish missing or null from a present zero only if that distinction is actually part of the application contract.

For streams, json.Decoder avoids reading the whole input before decoding:

dec := json.NewDecoder(resp.Body)
for {
	var r record
	if err := dec.Decode(&r); err == io.EOF {
		break
	} else if err != nil {
		return err
	}
	consume(r)
}

It buffers beyond the current value, so code must not assume the underlying reader sits exactly at a JSON boundary afterward. Decoder.UseNumber is useful when decoding unknown shapes and preserving number text; json.Number can then be parsed deliberately rather than silently becoming float64.

Custom UnmarshalJSON methods are the escape hatch for special wire forms. I use them sparingly. They hide work behind ordinary decoding and are easy to make recursive by calling json.Unmarshal into the same type. An alias type avoids that recursion, but a separate wire struct is often clearer.

Syntax errors carry an offset, which I include in diagnostics without returning the entire untrusted document. Type errors differ: valid JSON may contain a string where the destination expects an integer. Other fields may already have been assigned when decoding fails, so I treat any error as invalidating the whole destination rather than using a partially filled struct.

Unknown object fields are ignored by default. That helps forwards-compatible clients but hurts configuration, where a typo should fail. In this Go version I make strict configuration explicit by decoding an object into map[string]json.RawMessage, deleting recognized keys as I decode them, and rejecting leftovers. RawMessage also helps when one discriminator selects the concrete shape of a payload. I decode the small envelope first, then decode only its raw payload into the chosen type.

Decoder convenience does not impose resource limits. Before unmarshaling an HTTP body I bound what I read, and for a stream I bound the number and size of accepted records at the protocol level. A syntactically valid JSON array can still be an excellent memory exhaustion device.

The direct opinion I took from this exercise is simple: interface{} is not a schema. If I know the shape of a response, I write it down as Go types. This improves errors, numeric behavior, allocations, and the next reader’s chances. encoding/json was not slow because reflection is cursed. It was slow because I asked for a completely generic representation and then complained that it was completely generic.

I had a polling client that became steadily slower. The server was fine, DNS was fine, and my diagnostic method of staring at it was producing limited returns.

The client checked only the status:

resp, err := http.Get(url)
if err != nil {
	return err
}
defer resp.Body.Close()
if resp.StatusCode != http.StatusOK {
	return fmt.Errorf("status: %s", resp.Status)
}

Closing the body is necessary, but for persistent HTTP connections it is not always sufficient. If I do not read the response body to EOF, the transport may be unable to reuse that TCP connection. My server returned a small explanatory body for an error status; I ignored it and paid for a new connection on the next poll.

For responses whose contents I deliberately discard, I now do this:

defer resp.Body.Close()
if _, err := io.Copy(ioutil.Discard, resp.Body); err != nil {
	return err
}

The transport owns a pool of persistent connections keyed by destination. Reading through EOF lets its body wrapper observe completion and return the connection to the idle pool. Close still matters because it releases resources on every exit path.

This is not permission to drain an unbounded hostile response. If I only need to preserve reuse for a known-small error body, I put a limit around it and accept that an oversized body may force the connection closed:

r := io.LimitReader(resp.Body, 4<<10)
_, err = io.Copy(ioutil.Discard, r)

The same ownership rule applies on success. The caller that receives *http.Response generally owns Body and must close it. A helper that consumes the body should close it itself and return decoded data instead. Splitting those responsibilities is how bodies remain open.

I verified reuse by wrapping Transport.Dial with a counter. Ten requests to the same server produced ten dials before the fix and one after it. That is a crude instrument, but better than inferring connection behavior from elapsed time. A server can close persistent connections itself, so one dial is not a universal expected value.

There is a related server-side rule: a handler should not assume unread request bytes are harmless. net/http often manages request bodies for handlers, but large or malformed input still needs explicit size limits. Connection reuse is valuable; reading arbitrary quantities merely to obtain it is not.

After draining the tiny responses, repeated requests settled onto the same few connections instead of continually dialing. It was not an exotic kernel problem. It was one unread paragraph of error text, patiently converting itself into sockets.

I wrote a small validator that accepted interface{} and walked fields marked validate:"required". The happy path worked, so I naturally declared victory before trying a pointer, a nil pointer, or anything that was not a struct.

The first version began with this:

v := reflect.ValueOf(x)
for i := 0; i < v.NumField(); i++ {
	// inspect v.Field(i)
}

Passing &User{} produced reflect: call of reflect.Value.NumField on ptr Value. Recovering the panic would hide the real question: what inputs does this validator promise to accept? I chose structs and non-nil pointers to structs, with ordinary errors for everything else.

func validate(x interface{}) error {
	v := reflect.ValueOf(x)
	if !v.IsValid() {
		return errors.New("cannot validate nil")
	}
	if v.Kind() == reflect.Ptr {
		if v.IsNil() {
			return errors.New("cannot validate nil pointer")
		}
		v = v.Elem()
	}
	if v.Kind() != reflect.Struct {
		return fmt.Errorf("cannot validate %s", v.Kind())
	}

	t := v.Type()
	for i := 0; i < v.NumField(); i++ {
		if t.Field(i).Tag.Get("validate") != "required" {
			continue
		}
		if isZero(v.Field(i)) {
			return fmt.Errorf("%s is required", t.Field(i).Name)
		}
	}
	return nil
}

The harder policy is what “required” means. For this validator it rejects empty strings, numeric zero, false, and nil pointers, maps, slices, and interfaces. That makes false invalid, so a required boolean is usually a design smell; a pointer can represent the distinct states absent, false, and true.

I ignore unexported fields. A general validator should not turn another package’s private representation into public input policy.

Nested structs are not automatically validated. A field gets nested validation only when its tag asks for it; otherwise adding a private detail to a child type could unexpectedly change the parent’s contract. Nil optional pointers remain acceptable, while a required nil pointer fails at its own field path.

Errors report the Go field path as well as the failed rule. A bare “required” is compact but useless once validation descends into two nested structs with similarly named fields.

Reflection removes field-by-field plumbing, but the useful part is still the contract: accepted top-level shapes, the meaning of “required,” whether nesting is opt-in, and useful error paths. Once those choices are explicit, the reflective code can stay small and unsurprising.

I wanted one struct to describe both its JSON name and a tiny validation rule. The obvious declaration was:

type User struct {
	Name string `json:"name" validate:"required"`
}

My first parser split the raw tag on spaces and colons. It worked until quoting became interesting, which took approximately seven minutes. reflect.StructTag already understands the convention:

t := reflect.TypeOf(User{})
f, _ := t.FieldByName("Name")
fmt.Println(f.Tag.Get("json"))
fmt.Println(f.Tag.Get("validate"))

Output:

name
required

Tags are metadata attached to fields in the type. Reflection exposes them; the compiler does not enforce what validate:"required" means. Misspelling the key gives an empty string, indistinguishable from an explicitly empty value when using Get.

Only exported fields are useful to packages such as encoding/json, because reflection cannot generally set unexported fields from another package. I also avoid turning tags into a miniature programming language. Once a rule needs conditions, database access, or cross-field state, ordinary Go is clearer.

One more trap: reflect.TypeOf on a pointer yields a pointer type. To inspect its fields I use t.Elem(). Reflection is precise, but it is not especially interested in guessing what I meant.

On the way to the Moon, the Apollo crew and ground control needed precise information about the position, speed, and acceleration of the stack. Getting that information was a feat of engineering involving very clever solutions that deserve a post of their own, but the easiest piece of the puzzle was the position.

To figure out where exactly the Apollo was at any given moment, the crew used a guidance computer designed by the MIT along with a sextant, not much unlike the ones used for centuries by sea captains.

The CM pilot would pick a specific star (or planet sometimes) that could be placed within view of the sextant optics, he’d then align the sextant viewport with the chosen star, and then tell the guidance computer to calculate the position by telling it which star he had just pointed the sextant to.

The astronauts had a table of stars they could use, each accompanied by an octal code that could be fed to the computer. In this list, along with well-known stars—such as Sirius (15), Rigel (12), and the Sun (46)—were three oddly named stars.

While helping with the development of this system, the crew of Apollo I decided to have a little fun and added references to themselves by changing the names of some stars to parts of their own names spelled backwards:

• Navi (03) for Gus Grisson’s middle name (Ivan)
• Regor (17) for Roger Chaffee
• Dnoces (20) for Edward White II.

In reality, Navi was γ Cassiopeiae, Regor was γ Velorum, and Dnoces was ι Ursae Majoris.

After the crew lost their lives in a tragic fire—the first fallen in the American space manned program—the people at NASA kept those names in all their documentation and literature.