Class: Hyperion::CParser

Inherits:

Object

• Object
• Hyperion::CParser

Defined in:

ext/hyperion_http/parser.c

Constant Summary

PREINTERNED_HEADERS =

rb_aHeaderTable

Class Method Summary

• .build_access_line(format_sym, rb_ts, rb_method, rb_path, rb_query, rb_status, rb_duration, rb_remote, rb_http_version) ⇒ Object

  Hyperion::CParser.build_access_line(format, ts, method, path, query, status, duration_ms, remote_addr, http_version) -> String.

• .build_access_line_colored(format_sym, rb_ts, rb_method, rb_path, rb_query, rb_status, rb_duration, rb_remote, rb_http_version) ⇒ Object

  Hyperion::CParser.build_access_line_colored(format, ts, method, path, query, status, duration_ms, remote_addr, http_version) -> String.

• .build_env(env, request) ⇒ Object

  Hyperion::CParser.build_env(env, request) -> env.

• .build_response_head(rb_status, rb_reason, rb_headers, rb_body_size, rb_keep_alive, rb_date) ⇒ Object

  Hyperion::CParser.build_response_head(status, reason, headers, body_size, keep_alive, date_str) -> String.

• .chunked_body_complete?(rb_buffer, rb_body_start) ⇒ Boolean

  Hyperion::CParser.chunked_body_complete?(buffer, body_start) -> [complete?, end_offset].

• .parse_cookie_header(rb_cookie) ⇒ Object

  Hyperion::CParser.parse_cookie_header(cookie_str) -> Hash.

• .upcase_underscore(rb_name) ⇒ Object

  Hyperion::CParser.upcase_underscore(name) -> “HTTP_<UPCASED_UNDERSCORED>”.

Instance Method Summary

• #parse(buffer) ⇒ Object

  parse(buffer) -> [Request, end_offset].

Class Method Details

.build_access_line(format_sym, rb_ts, rb_method, rb_path, rb_query, rb_status, rb_duration, rb_remote, rb_http_version) ⇒ Object

Hyperion::CParser.build_access_line(format, ts, method, path, query,

status, duration_ms, remote_addr,
http_version) -> String

Hand-rolled access-log line builder used by Hyperion::Logger#access on the hot path. The Ruby version allocates 1-2 throwaway Strings per line; this builds the line into a stack scratch buffer (with rb_str_buf overflow for extreme cases) and returns a single Ruby String. ~10× faster on the common case, which closes the perf gap between log_requests on/off.

‘format` is :text or :json (Symbol). The format strings here mirror Logger#build_access_text / #build_access_json byte-for-byte (no colour —the C builder is only used when @colorize is false, i.e. non-TTY production deployments where access logs are the highest-volume log line).

String inputs are passed through verbatim. Access logs are best-effort structured output, not a security boundary; CRLF in path/remote_addr would be a log-injection nuisance but cannot escalate. Status (int) and duration_ms (double/Numeric) go through snprintf, which is type-safe.

# File 'ext/hyperion_http/parser.c', line 953

static VALUE cbuild_access_line(VALUE self,
                                VALUE format_sym, VALUE rb_ts, VALUE rb_method,
                                VALUE rb_path, VALUE rb_query, VALUE rb_status,
                                VALUE rb_duration, VALUE rb_remote,
                                VALUE rb_http_version) {
    (void)self;
    Check_Type(rb_ts, T_STRING);
    Check_Type(rb_method, T_STRING);
    Check_Type(rb_path, T_STRING);
    Check_Type(rb_http_version, T_STRING);

    int is_json = (TYPE(format_sym) == T_SYMBOL) &&
                  (SYM2ID(format_sym) == rb_intern("json"));

    int status     = NUM2INT(rb_status);
    double dur_ms  = NUM2DBL(rb_duration);

    int has_query  = !NIL_P(rb_query) && RSTRING_LEN(rb_query) > 0;
    int has_remote = !NIL_P(rb_remote) && RSTRING_LEN(rb_remote) > 0;

    /* 1 KiB initial buffer covers the vast majority of access-log lines
     * (timestamp + level + path + status + addr ~= 200 bytes). rb_str_cat
     * grows on overflow.
     *
     * We use a CAT_LIT macro for literal-string appends so the compiler
     * computes length via sizeof — manual byte counts on hand-rolled
     * literal lengths are an off-by-one waiting to happen. */
#define CAT_LIT(b, s) rb_str_cat((b), (s), (long)(sizeof(s) - 1))

    VALUE buf = rb_str_buf_new(512);

    if (is_json) {
        /* Prefix: {"ts":"...","level":"info","source":"hyperion","message":"request", */
        CAT_LIT(buf, "{\"ts\":\"");
        rb_str_cat(buf, RSTRING_PTR(rb_ts), RSTRING_LEN(rb_ts));
        CAT_LIT(buf, "\",\"level\":\"info\",\"source\":\"hyperion\",\"message\":\"request\",");
        CAT_LIT(buf, "\"method\":\"");
        rb_str_cat(buf, RSTRING_PTR(rb_method), RSTRING_LEN(rb_method));
        CAT_LIT(buf, "\",\"path\":\"");
        rb_str_cat(buf, RSTRING_PTR(rb_path), RSTRING_LEN(rb_path));
        CAT_LIT(buf, "\"");

        if (has_query) {
            CAT_LIT(buf, ",\"query\":\"");
            rb_str_cat(buf, RSTRING_PTR(rb_query), RSTRING_LEN(rb_query));
            CAT_LIT(buf, "\"");
        }

        char num[64];
        int n = snprintf(num, sizeof(num), ",\"status\":%d,\"duration_ms\":%g,",
                         status, dur_ms);
        rb_str_cat(buf, num, n);

        if (has_remote) {
            CAT_LIT(buf, "\"remote_addr\":\"");
            rb_str_cat(buf, RSTRING_PTR(rb_remote), RSTRING_LEN(rb_remote));
            CAT_LIT(buf, "\",");
        } else {
            CAT_LIT(buf, "\"remote_addr\":null,");
        }

        CAT_LIT(buf, "\"http_version\":\"");
        rb_str_cat(buf, RSTRING_PTR(rb_http_version), RSTRING_LEN(rb_http_version));
        CAT_LIT(buf, "\"}\n");
    } else {
        /* text: "<ts> INFO  [hyperion] message=request method=... path=... [query=...] status=... duration_ms=... remote_addr=... http_version=...\n" */
        rb_str_cat(buf, RSTRING_PTR(rb_ts), RSTRING_LEN(rb_ts));
        CAT_LIT(buf, " INFO  [hyperion] message=request method=");
        rb_str_cat(buf, RSTRING_PTR(rb_method), RSTRING_LEN(rb_method));
        CAT_LIT(buf, " path=");
        rb_str_cat(buf, RSTRING_PTR(rb_path), RSTRING_LEN(rb_path));

        if (has_query) {
            /* Mirror Logger#quote_if_needed: quote if value contains
             * whitespace, '"', or '='. Hot path skips quoting. */
            const char *q_ptr = RSTRING_PTR(rb_query);
            long q_len = RSTRING_LEN(rb_query);
            int need_quote = 0;
            for (long j = 0; j < q_len; j++) {
                char c = q_ptr[j];
                if (c == ' ' || c == '\t' || c == '\n' || c == '\r' ||
                    c == '"' || c == '=') {
                    need_quote = 1;
                    break;
                }
            }
            if (need_quote) {
                /* Defer to Ruby's String#inspect for correct quoting. */
                VALUE quoted = rb_funcall(rb_query, rb_intern("inspect"), 0);
                CAT_LIT(buf, " query=");
                rb_str_cat(buf, RSTRING_PTR(quoted), RSTRING_LEN(quoted));
            } else {
                CAT_LIT(buf, " query=");
                rb_str_cat(buf, q_ptr, q_len);
            }
        }

        char num[80];
        /* Use %g to match the existing Ruby format which interpolates
         * Float#to_s (no fixed precision). Status is an int. */
        int n = snprintf(num, sizeof(num), " status=%d duration_ms=%g remote_addr=",
                         status, dur_ms);
        rb_str_cat(buf, num, n);

        if (has_remote) {
            rb_str_cat(buf, RSTRING_PTR(rb_remote), RSTRING_LEN(rb_remote));
        } else {
            CAT_LIT(buf, "nil");
        }

        CAT_LIT(buf, " http_version=");
        rb_str_cat(buf, RSTRING_PTR(rb_http_version), RSTRING_LEN(rb_http_version));
        CAT_LIT(buf, "\n");
    }

    return buf;
}

.build_access_line_colored(format_sym, rb_ts, rb_method, rb_path, rb_query, rb_status, rb_duration, rb_remote, rb_http_version) ⇒ Object

Hyperion::CParser.build_access_line_colored(format, ts, method, path, query,

status, duration_ms, remote_addr,
http_version) -> String

TTY-coloured variant of build_access_line. The text path wraps the level label with ANSI escape “e[32mINFO e[0m” so a developer running Hyperion in a terminal sees a green INFO tag. The :json branch is identical to the non-coloured builder — JSON access lines are machine-readable and never carry ANSI escapes.

Lifted from cbuild_access_line above; the only divergence is the level label injection in the text branch. We deliberately duplicate the text format rather than templating, because the text body is short and a single function with a colour flag would compile to the same code with an extra branch in the hot loop.

# File 'ext/hyperion_http/parser.c', line 1088

static VALUE cbuild_access_line_colored(VALUE self,
                                        VALUE format_sym, VALUE rb_ts,
                                        VALUE rb_method, VALUE rb_path,
                                        VALUE rb_query, VALUE rb_status,
                                        VALUE rb_duration, VALUE rb_remote,
                                        VALUE rb_http_version) {
    (void)self;
    Check_Type(rb_ts, T_STRING);
    Check_Type(rb_method, T_STRING);
    Check_Type(rb_path, T_STRING);
    Check_Type(rb_http_version, T_STRING);

    int is_json = (TYPE(format_sym) == T_SYMBOL) &&
                  (SYM2ID(format_sym) == rb_intern("json"));

    int status     = NUM2INT(rb_status);
    double dur_ms  = NUM2DBL(rb_duration);

    int has_query  = !NIL_P(rb_query) && RSTRING_LEN(rb_query) > 0;
    int has_remote = !NIL_P(rb_remote) && RSTRING_LEN(rb_remote) > 0;

#define CAT_LIT(b, s) rb_str_cat((b), (s), (long)(sizeof(s) - 1))

    VALUE buf = rb_str_buf_new(512);

    if (is_json) {
        /* JSON output is identical to the non-coloured path — ANSI escapes
         * have no place in a structured log record. */
        CAT_LIT(buf, "{\"ts\":\"");
        rb_str_cat(buf, RSTRING_PTR(rb_ts), RSTRING_LEN(rb_ts));
        CAT_LIT(buf, "\",\"level\":\"info\",\"source\":\"hyperion\",\"message\":\"request\",");
        CAT_LIT(buf, "\"method\":\"");
        rb_str_cat(buf, RSTRING_PTR(rb_method), RSTRING_LEN(rb_method));
        CAT_LIT(buf, "\",\"path\":\"");
        rb_str_cat(buf, RSTRING_PTR(rb_path), RSTRING_LEN(rb_path));
        CAT_LIT(buf, "\"");

        if (has_query) {
            CAT_LIT(buf, ",\"query\":\"");
            rb_str_cat(buf, RSTRING_PTR(rb_query), RSTRING_LEN(rb_query));
            CAT_LIT(buf, "\"");
        }

        char num[64];
        int n = snprintf(num, sizeof(num), ",\"status\":%d,\"duration_ms\":%g,",
                         status, dur_ms);
        rb_str_cat(buf, num, n);

        if (has_remote) {
            CAT_LIT(buf, "\"remote_addr\":\"");
            rb_str_cat(buf, RSTRING_PTR(rb_remote), RSTRING_LEN(rb_remote));
            CAT_LIT(buf, "\",");
        } else {
            CAT_LIT(buf, "\"remote_addr\":null,");
        }

        CAT_LIT(buf, "\"http_version\":\"");
        rb_str_cat(buf, RSTRING_PTR(rb_http_version), RSTRING_LEN(rb_http_version));
        CAT_LIT(buf, "\"}\n");
    } else {
        /* text: "<ts> \e[32mINFO \e[0m [hyperion] message=request method=..." */
        rb_str_cat(buf, RSTRING_PTR(rb_ts), RSTRING_LEN(rb_ts));
        CAT_LIT(buf, " \x1b[32mINFO \x1b[0m [hyperion] message=request method=");
        rb_str_cat(buf, RSTRING_PTR(rb_method), RSTRING_LEN(rb_method));
        CAT_LIT(buf, " path=");
        rb_str_cat(buf, RSTRING_PTR(rb_path), RSTRING_LEN(rb_path));

        if (has_query) {
            const char *q_ptr = RSTRING_PTR(rb_query);
            long q_len = RSTRING_LEN(rb_query);
            int need_quote = 0;
            for (long j = 0; j < q_len; j++) {
                char c = q_ptr[j];
                if (c == ' ' || c == '\t' || c == '\n' || c == '\r' ||
                    c == '"' || c == '=') {
                    need_quote = 1;
                    break;
                }
            }
            if (need_quote) {
                VALUE quoted = rb_funcall(rb_query, rb_intern("inspect"), 0);
                CAT_LIT(buf, " query=");
                rb_str_cat(buf, RSTRING_PTR(quoted), RSTRING_LEN(quoted));
            } else {
                CAT_LIT(buf, " query=");
                rb_str_cat(buf, q_ptr, q_len);
            }
        }

        char num[80];
        int n = snprintf(num, sizeof(num), " status=%d duration_ms=%g remote_addr=",
                         status, dur_ms);
        rb_str_cat(buf, num, n);

        if (has_remote) {
            rb_str_cat(buf, RSTRING_PTR(rb_remote), RSTRING_LEN(rb_remote));
        } else {
            CAT_LIT(buf, "nil");
        }

        CAT_LIT(buf, " http_version=");
        rb_str_cat(buf, RSTRING_PTR(rb_http_version), RSTRING_LEN(rb_http_version));
        CAT_LIT(buf, "\n");
    }

    return buf;
}

.build_env(env, request) ⇒ Object

Hyperion::CParser.build_env(env, request) -> env

Phase 3a (1.7.1) — populate the Rack env hash with REQUEST_METHOD, PATH_INFO, QUERY_STRING, HTTP_VERSION, SERVER_PROTOCOL, CONTENT_TYPE, CONTENT_LENGTH, and HTTP_<UPCASED_UNDERSCORED> for every parsed header.

The Ruby caller (Hyperion::Adapter::Rack#build_env) sets the rest of the Rack-required keys (rack.input, REMOTE_ADDR, SERVER_NAME/PORT, …) since those need a StringIO from a pool and a peer-address split. The header loop is the bytewise-bound piece and the only thing worth pulling into C — moving the full env build would mean threading the pool, host splitter, and version constant through the FFI boundary for ~no extra win.

Returns the same env Hash (callers can either chain or ignore).

# File 'ext/hyperion_http/parser.c', line 1478

static VALUE cbuild_env(VALUE self, VALUE env, VALUE request) {
    (void)self;
    Check_Type(env, T_HASH);

    /* Read Request ivars directly — Request is a frozen value object set
     * up in initialize; no risk of stale reads, no method-dispatch cost. */
    VALUE method       = rb_ivar_get(request, id_iv_method);
    VALUE path         = rb_ivar_get(request, id_iv_path);
    VALUE query_string = rb_ivar_get(request, id_iv_query_string);
    VALUE http_version = rb_ivar_get(request, id_iv_http_version);
    VALUE headers      = rb_ivar_get(request, id_iv_headers);

    rb_hash_aset(env, rb_kREQUEST_METHOD,  method);
    rb_hash_aset(env, rb_kPATH_INFO,       path);
    rb_hash_aset(env, rb_kQUERY_STRING,    query_string);
    rb_hash_aset(env, rb_kSERVER_PROTOCOL, http_version);
    rb_hash_aset(env, rb_kHTTP_VERSION,    http_version);

    if (TYPE(headers) == T_HASH) {
        rb_hash_foreach(headers, build_env_iter, env);
    }

    return env;
}

.build_response_head(rb_status, rb_reason, rb_headers, rb_body_size, rb_keep_alive, rb_date) ⇒ Object

Hyperion::CParser.build_response_head(status, reason, headers, body_size,

keep_alive, date_str) -> String

Builds the HTTP/1.1 response head:

"HTTP/1.1 <status> <reason>\r\n"
"<lowercased-key>: <value>\r\n" for each user header (except
  content-length / connection — we always set these from the framing
  args below, mirroring the rc16 Ruby behaviour where the normalized
  hash is overridden in place).
"content-length: <body_size>\r\n"
"connection: <close|keep-alive>\r\n"
"date: <date_str>\r\n"  (only if user headers didn't include 'date')
"\r\n"

Header values containing CR/LF raise ArgumentError (response-splitting guard). Bypasses Ruby Hash#each + per-line String#<< allocation; the status line, framing headers, and join slices live in C buffers.

2.13-B — three CPU savings over the rc17 baseline:

1. Common (status, reason) pairs hit a static table of pre-baked
   "HTTP/1.1 NNN <reason>\r\n" lines — one rb_str_cat replaces the
   per-request snprintf + reason-cat + CRLF-cat triple.
2. Header iteration uses rb_hash_foreach instead of
   `rb_funcall(:keys)` + per-key `rb_hash_aref` — eliminates the
   keys-Array allocation and the N hash lookups per call.
3. Per-key downcase result + "<lc>: " prefix is cached on the
   input frozen String's identity (capped at 64 entries; a
   misbehaving app emitting unique keys per request just falls
   back to the slow path on overflow). For the canonical Rack-3
   app emitting `'content-type' / 'cache-control' / ...` from
   frozen literals, every header lookup is a single st hit.
4. (key, value) full-line cache: both sides are frozen-literal
   Strings (e.g. `'cache-control' => 'no-store'`) — entire
   "<lc-key>: <value>\r\n" line is one rb_str_cat after the first
   request populates the cache. Capped at 256 entries.

# File 'ext/hyperion_http/parser.c', line 818

static VALUE cbuild_response_head(VALUE self, VALUE rb_status, VALUE rb_reason,
                                  VALUE rb_headers, VALUE rb_body_size,
                                  VALUE rb_keep_alive, VALUE rb_date) {
    (void)self;
    Check_Type(rb_headers, T_HASH);
    Check_Type(rb_reason, T_STRING);
    Check_Type(rb_date, T_STRING);

    int status     = NUM2INT(rb_status);
    long body_size = NUM2LONG(rb_body_size);
    int keep_alive = RTEST(rb_keep_alive);

    /* body_size == -1 is the chunked-encoding sentinel; any other
     * negative value is a programming error (likely an integer
     * underflow in a caller). Reject early with a clear message
     * rather than silently treating -2 / -42 as chunked. */
    if (body_size < -1) {
        rb_raise(rb_eArgError,
                 "body_size must be >= 0 (or -1 for chunked sentinel), got %ld",
                 body_size);
    }

    /* body_size == -1 is the chunked-encoding sentinel (from
     * hyperion_build_response_head_chunked).  In this mode we emit
     * "transfer-encoding: chunked\r\n" instead of "content-length: N\r\n"
     * and suppress any user-supplied content-length / transfer-encoding
     * headers (RFC 7230 §3.3.3 — they are mutually exclusive). */
    int is_chunked = (body_size == -1);

    /* Most heads fit in 1 KiB; rb_str_cat grows on demand. */
    VALUE buf = rb_str_buf_new(1024);

    /* Status line: pre-baked when (status, reason) is one of the well-known
     * pairs in `Hyperion::ResponseWriter::REASONS`; falls back to
     * `snprintf("HTTP/1.1 %d ", status)` + reason-cat for unknowns. */
    const struct status_line *sline =
        lookup_status_line(status, RSTRING_PTR(rb_reason), RSTRING_LEN(rb_reason));
    if (sline != NULL) {
        rb_str_cat(buf, sline->bytes, sline->len);
    } else {
        char status_line_buf[48];
        int n = snprintf(status_line_buf, sizeof(status_line_buf), "HTTP/1.1 %d ", status);
        rb_str_cat(buf, status_line_buf, n);
        rb_str_cat(buf, RSTRING_PTR(rb_reason), RSTRING_LEN(rb_reason));
        rb_str_cat(buf, "\r\n", 2);
    }

    /* Iterate user headers — lowercase key, validate value, skip framing.
     * Threaded through rb_hash_foreach so we can reuse the per-key
     * downcase cache and skip the per-call `keys` Array allocation.
     * is_chunked is threaded through state so build_head_each can drop
     * user-supplied transfer-encoding and content-length in chunked mode. */
    build_head_state_t state = { buf, 0, is_chunked };
    rb_hash_foreach(rb_headers, build_head_each, (VALUE)&state);

    /* Framing headers — always emitted.
     * Non-chunked: content-length uses a hand-rolled itoa rather than
     *   snprintf (vfprintf was 1 % of CPU on the CPU-JSON profile).
     * Chunked: transfer-encoding: chunked (no content-length — RFC 7230 §3.3.3). */
    if (is_chunked) {
        rb_str_cat(buf, "transfer-encoding: chunked\r\n", 28);
    } else {
        char itoa_scratch[24];
        int cl_off = itoa_positive_decimal(body_size, itoa_scratch, (int)sizeof(itoa_scratch));
        rb_str_cat(buf, "content-length: ", 16);
        rb_str_cat(buf, itoa_scratch + cl_off, sizeof(itoa_scratch) - cl_off);
        rb_str_cat(buf, "\r\n", 2);
    }

    if (keep_alive) {
        rb_str_cat(buf, "connection: keep-alive\r\n", 24);
    } else {
        rb_str_cat(buf, "connection: close\r\n", 19);
    }

    if (!state.has_date) {
        rb_str_cat(buf, "date: ", 6);
        rb_str_cat(buf, RSTRING_PTR(rb_date), RSTRING_LEN(rb_date));
        rb_str_cat(buf, "\r\n", 2);
    }

    /* End of head */
    rb_str_cat(buf, "\r\n", 2);

    return buf;
}

.chunked_body_complete?(rb_buffer, rb_body_start) ⇒ Boolean

Hyperion::CParser.chunked_body_complete?(buffer, body_start)

-> [complete?, end_offset]

Walks chunked-transfer framing in ‘buffer` starting at byte offset `body_start`. Returns a 2-element array:

[true,  end_offset] — chunked body fully buffered; end_offset is the
                      byte just after the trailer CRLF (where pipelined
                      bytes from a follow-on request would begin).
[false, last_safe]  — body is not yet complete; last_safe is the
                      furthest cursor we successfully advanced to,
                      useful as a hint for incremental parsing.

Mirrors Connection#chunked_body_complete? in pure Ruby — see lib/hyperion/ connection.rb. Trailing whitespace after the size token (e.g. “5 ; extrn”) is permitted as a permissive parse to match the upstream Ruby ‘.strip`.

Returns:

• (Boolean)

# File 'ext/hyperion_http/parser.c', line 1267

static VALUE cchunked_body_complete(VALUE self, VALUE rb_buffer, VALUE rb_body_start) {
    (void)self;
    Check_Type(rb_buffer, T_STRING);

    const char *data = RSTRING_PTR(rb_buffer);
    long len         = RSTRING_LEN(rb_buffer);
    long cursor      = NUM2LONG(rb_body_start);

    if (cursor < 0 || cursor > len) {
        rb_raise(rb_eArgError, "body_start out of range");
    }

    long last_safe = cursor;
    VALUE result   = rb_ary_new_capa(2);

    while (1) {
        /* Find the next CRLF starting at cursor. */
        long line_end = -1;
        for (long i = cursor; i + 1 < len; i++) {
            if (data[i] == '\r' && data[i + 1] == '\n') {
                line_end = i;
                break;
            }
        }
        if (line_end < 0) {
            rb_ary_push(result, Qfalse);
            rb_ary_push(result, LONG2NUM(last_safe));
            RB_GC_GUARD(rb_buffer);
            return result;
        }

        /* Parse the size token: hex digits up to ';' or whitespace, optional
         * chunk extension after ';' which we ignore wholesale. */
        long tok_start = cursor;
        long tok_end   = line_end;
        for (long i = cursor; i < line_end; i++) {
            if (data[i] == ';') { tok_end = i; break; }
        }
        /* Trim leading/trailing ASCII whitespace from the token. */
        while (tok_start < tok_end &&
               (data[tok_start] == ' ' || data[tok_start] == '\t')) {
            tok_start++;
        }
        while (tok_end > tok_start &&
               (data[tok_end - 1] == ' ' || data[tok_end - 1] == '\t')) {
            tok_end--;
        }
        if (tok_end <= tok_start) {
            /* Empty size token — incomplete frame. */
            rb_ary_push(result, Qfalse);
            rb_ary_push(result, LONG2NUM(last_safe));
            RB_GC_GUARD(rb_buffer);
            return result;
        }

        /* Validate + decode hex. */
        unsigned long size = 0;
        for (long i = tok_start; i < tok_end; i++) {
            unsigned char c = (unsigned char)data[i];
            unsigned int digit;
            if (c >= '0' && c <= '9') {
                digit = c - '0';
            } else if (c >= 'a' && c <= 'f') {
                digit = 10 + (c - 'a');
            } else if (c >= 'A' && c <= 'F') {
                digit = 10 + (c - 'A');
            } else {
                /* Non-hex byte: incomplete/malformed. Match the Ruby
                 * regex `/\A\h+\z/` semantics — return false, advance no
                 * further. The caller will read more bytes and retry. */
                rb_ary_push(result, Qfalse);
                rb_ary_push(result, LONG2NUM(last_safe));
                RB_GC_GUARD(rb_buffer);
                return result;
            }
            size = (size << 4) | digit;
        }

        cursor = line_end + 2;

        if (size == 0) {
            /* Final chunk — walk trailer headers until we hit "\r\n\r\n"
             * (i.e. an empty trailer line directly after the size line). */
            while (1) {
                long nl = -1;
                for (long i = cursor; i + 1 < len; i++) {
                    if (data[i] == '\r' && data[i + 1] == '\n') {
                        nl = i;
                        break;
                    }
                }
                if (nl < 0) {
                    rb_ary_push(result, Qfalse);
                    rb_ary_push(result, LONG2NUM(last_safe));
                    RB_GC_GUARD(rb_buffer);
                    return result;
                }
                if (nl == cursor) {
                    /* Empty line — body complete. */
                    rb_ary_push(result, Qtrue);
                    rb_ary_push(result, LONG2NUM(nl + 2));
                    RB_GC_GUARD(rb_buffer);
                    return result;
                }
                cursor = nl + 2;
            }
        }

        /* Need cursor + size + 2 bytes (chunk data + trailing CRLF). */
        if ((unsigned long)(len - cursor) < size + 2) {
            rb_ary_push(result, Qfalse);
            rb_ary_push(result, LONG2NUM(last_safe));
            RB_GC_GUARD(rb_buffer);
            return result;
        }

        cursor += (long)size + 2;
        last_safe = cursor;
    }
}

.parse_cookie_header(rb_cookie) ⇒ Object

Hyperion::CParser.parse_cookie_header(cookie_str) -> Hash

Phase 3b (1.7.1) — split a single Cookie header value into its { “name” => “value” } pairs.

Standard format: “name1=val1; name2=val2; name3=val3”. Leading/trailing ASCII whitespace is trimmed around each pair and around each key. Empty values are valid. Pairs without ‘=` are skipped (RFC 6265 calls them ignorable). Repeated names are last-wins —middlewares that need RFC-strict merge can override.

Cookies are NOT URL-decoded by spec; values are opaque octets. We leave them verbatim. The returned Hash is mutable so the caller can extend it (e.g. for session-cookie hot-swaps).

# File 'ext/hyperion_http/parser.c', line 1518

static VALUE cparse_cookie_header(VALUE self, VALUE rb_cookie) {
    (void)self;
    Check_Type(rb_cookie, T_STRING);

    VALUE result = rb_hash_new();

    const char *src = RSTRING_PTR(rb_cookie);
    long src_len    = RSTRING_LEN(rb_cookie);
    long i = 0;

    while (i < src_len) {
        /* Skip leading whitespace and stray semicolons. */
        while (i < src_len && (src[i] == ' ' || src[i] == '\t' ||
                               src[i] == ';')) {
            i++;
        }
        if (i >= src_len) break;

        /* Pair runs to next ';' (or end of string). */
        long pair_start = i;
        while (i < src_len && src[i] != ';') i++;
        long pair_end = i;

        /* Trim trailing whitespace inside the pair. */
        while (pair_end > pair_start &&
               (src[pair_end - 1] == ' ' || src[pair_end - 1] == '\t')) {
            pair_end--;
        }
        if (pair_end == pair_start) continue;

        /* Find '=' inside [pair_start, pair_end). */
        long eq = -1;
        for (long j = pair_start; j < pair_end; j++) {
            if (src[j] == '=') { eq = j; break; }
        }
        if (eq < 0) continue; /* malformed — no '=' — skip per RFC 6265. */

        /* Trim trailing ws on key (between pair_start and eq). */
        long key_end = eq;
        while (key_end > pair_start &&
               (src[key_end - 1] == ' ' || src[key_end - 1] == '\t')) {
            key_end--;
        }
        if (key_end == pair_start) continue; /* empty name — skip. */

        /* Skip leading ws on value (between eq+1 and pair_end). */
        long val_start = eq + 1;
        while (val_start < pair_end &&
               (src[val_start] == ' ' || src[val_start] == '\t')) {
            val_start++;
        }

        VALUE key = rb_str_new(src + pair_start, key_end - pair_start);
        VALUE val = rb_str_new(src + val_start,  pair_end - val_start);
        rb_hash_aset(result, key, val);
    }

    RB_GC_GUARD(rb_cookie);
    return result;
}

.upcase_underscore(rb_name) ⇒ Object

Hyperion::CParser.upcase_underscore(name) -> “HTTP_<UPCASED_UNDERSCORED>”

Single-allocation replacement for ‘“HTTP_#’_’)”‘. Hot path on the Rack adapter: every uncached request header (any `X-*` custom header) hits this on every request, and the Ruby version spawns three String allocations (the upcase result, the tr result, and the “HTTP_…” interpolation) plus a per-byte loop in tr.

We allocate one Ruby String of length 5 + name.bytesize, fill it in a single byte loop, return it. ASCII letters get OR’d with 0x20 inverted (i.e. cleared bit 5 to upcase ‘a’..‘z’); ‘-’ becomes ‘_’; everything else passes through (header names are ASCII per RFC 9110, but multi-byte UTF-8 bytes pass through bytewise unmolested rather than crashing).

Encoding is set to US-ASCII because Ruby’s String#upcase on an ASCII-only input returns a US-ASCII string, and the env-key lookup downstream is encoding-agnostic anyway.

# File 'ext/hyperion_http/parser.c', line 1215

static VALUE cupcase_underscore(VALUE self, VALUE rb_name) {
    (void)self;
    Check_Type(rb_name, T_STRING);

    const char *src = RSTRING_PTR(rb_name);
    long src_len    = RSTRING_LEN(rb_name);

    /* Single allocation: 5 prefix bytes + N source bytes. */
    VALUE out = rb_str_new(NULL, 5 + src_len);
    char *dst = RSTRING_PTR(out);

    dst[0] = 'H';
    dst[1] = 'T';
    dst[2] = 'T';
    dst[3] = 'P';
    dst[4] = '_';

    for (long i = 0; i < src_len; i++) {
        unsigned char c = (unsigned char)src[i];
        if (c >= 'a' && c <= 'z') {
            dst[5 + i] = (char)(c - 32);
        } else if (c == '-') {
            dst[5 + i] = '_';
        } else {
            dst[5 + i] = (char)c;
        }
    }

    rb_enc_associate(out, rb_usascii_encoding());
    /* Keep rb_name live across the loop above. RSTRING_PTR returns an
     * interior pointer that becomes invalid if the GC moves the source
     * String — unlikely on this tight path, but cheap insurance. */
    RB_GC_GUARD(rb_name);
    return out;
}

Instance Method Details

#parse(buffer) ⇒ Object

parse(buffer) -> [Request, end_offset]

Parse one complete HTTP/1.1 request from ‘buffer`. If buffer doesn’t yet contain a complete request, raise ParseError(“incomplete”). For pipelined input, end_offset is the byte boundary of the first request — Connection carries the rest forward.

# File 'ext/hyperion_http/parser.c', line 373

static VALUE cparser_parse(VALUE self, VALUE buffer) {
    Check_Type(buffer, T_STRING);
    (void)self;

    parser_state_t s;
    state_init(&s);

    llhttp_t parser;
    llhttp_init(&parser, HTTP_REQUEST, &settings);
    parser.data = &s;

    const char *data = RSTRING_PTR(buffer);
    size_t len = (size_t)RSTRING_LEN(buffer);

    enum llhttp_errno err = llhttp_execute(&parser, data, len);

    /* Custom error flags (set inside callbacks) take precedence. */
    if (s.parse_error == 2) {
        rb_raise(rb_eUnsupportedError, "%s", s.error_message);
    }
    if (s.parse_error == 1) {
        rb_raise(rb_eParseError, "%s", s.error_message);
    }

    if (err == HPE_PAUSED_UPGRADE) {
        rb_raise(rb_eUnsupportedError, "Upgrade not supported");
    }
    if (err != HPE_OK && err != HPE_PAUSED) {
        const char *reason = llhttp_get_error_reason(&parser);
        rb_raise(rb_eParseError, "llhttp: %s",
                 (reason && *reason) ? reason : llhttp_errno_name(err));
    }

    if (!s.message_complete) {
        rb_raise(rb_eParseError, "incomplete request");
    }

    /* Compute end_offset. We pause inside on_message_complete, so
     * llhttp_get_error_pos returns the byte just after the message
     * boundary — exactly the carry-over offset we want. */
    size_t consumed;
    if (err == HPE_PAUSED) {
        const char *epos = llhttp_get_error_pos(&parser);
        consumed = epos ? (size_t)(epos - data) : len;
    } else {
        consumed = len;
    }

    /* 2.4-B (S1): Qnil-to-empty-String coercion for fields that the
     * llhttp callbacks never touched (e.g. zero-length URL, GET with
     * no body, HTTP/1.0 with no version detail). The frozen empty
     * String is shared across every nil-coerced field — no allocation. */
    VALUE method        = NIL_P(s.method)        ? rb_kEMPTY_STR : s.method;
    VALUE path          = NIL_P(s.path)          ? rb_kEMPTY_STR : s.path;
    VALUE query_string  = NIL_P(s.query_string)  ? rb_kEMPTY_STR : s.query_string;
    VALUE http_version  = NIL_P(s.http_version)  ? rb_kHTTP_1_1  : s.http_version;
    VALUE body          = NIL_P(s.body)          ? rb_kEMPTY_STR : s.body;

    /* Build the Request. */
    VALUE kwargs = rb_hash_new();
    rb_hash_aset(kwargs, ID2SYM(id_method_kw),       method);
    rb_hash_aset(kwargs, ID2SYM(id_path_kw),         path);
    rb_hash_aset(kwargs, ID2SYM(id_query_string_kw), query_string);
    rb_hash_aset(kwargs, ID2SYM(id_http_version_kw), http_version);
    rb_hash_aset(kwargs, ID2SYM(id_headers_kw),      s.headers);
    rb_hash_aset(kwargs, ID2SYM(id_body_kw),         body);

    VALUE args[1] = { kwargs };
    VALUE request = rb_funcallv_kw(rb_cRequest, id_new, 1, args, RB_PASS_KEYWORDS);

    return rb_ary_new_from_args(2, request, ULONG2NUM((unsigned long)consumed));
}