puma 2.2.0 → 8.0.2

data/ext/puma_http11/puma_http11.c

@@ -1,29 +1,19 @@
 /**
  * Copyright (c) 2005 Zed A. Shaw
  * You can redistribute it and/or modify it under the same terms as Ruby.
+ * License 3-clause BSD
  */
 
 #define RSTRING_NOT_MODIFIED 1
 
 #include "ruby.h"
-#include "ext_help.h"
+#include "ruby/encoding.h"
 #include <assert.h>
 #include <string.h>
+#include <ctype.h>
 #include "http11_parser.h"
 
-#ifndef MANAGED_STRINGS
-
-#ifndef RSTRING_PTR
-#define RSTRING_PTR(s) (RSTRING(s)->ptr)
-#endif
-#ifndef RSTRING_LEN
-#define RSTRING_LEN(s) (RSTRING(s)->len)
-#endif
-
-#define rb_extract_chars(e, sz) (*sz = RSTRING_LEN(e), RSTRING_PTR(e))
-#define rb_free_chars(e) /* nothing */
-
-#endif
+#define ARRAY_SIZE(x) (sizeof(x)/sizeof(x[0]))
 
 static VALUE eHttpParserError;
 
@@ -34,33 +24,50 @@ static VALUE global_request_method;
 static VALUE global_request_uri;
 static VALUE global_fragment;
 static VALUE global_query_string;
-static VALUE global_http_version;
+static VALUE global_server_protocol;
 static VALUE global_request_path;
 
 /** Defines common length and error messages for input length validation. */
-#define DEF_MAX_LENGTH(N,length) const size_t MAX_##N##_LENGTH = length; const char *MAX_##N##_LENGTH_ERR = "HTTP element " # N  " is longer than the " # length " allowed length."
+#define QUOTE(s) #s
+#define EXPAND_MAX_LENGTH_VALUE(s) QUOTE(s)
+#define DEF_MAX_LENGTH(N,length) const size_t MAX_##N##_LENGTH = length; const char *MAX_##N##_LENGTH_ERR = "HTTP element " # N  " is longer than the " EXPAND_MAX_LENGTH_VALUE(length) " allowed length (was %d)"
 
 /** Validates the max length of given input and throws an HttpParserError exception if over. */
-#define VALIDATE_MAX_LENGTH(len, N) if(len > MAX_##N##_LENGTH) { rb_raise(eHttpParserError, "%s", MAX_##N##_LENGTH_ERR); }
+#define VALIDATE_MAX_LENGTH(len, N) if(len > MAX_##N##_LENGTH) { rb_raise(eHttpParserError, MAX_##N##_LENGTH_ERR, len); }
 
 /** Defines global strings in the init method. */
-#define DEF_GLOBAL(N, val)   global_##N = rb_str_new2(val); rb_global_variable(&global_##N)
-
+static inline void DEF_GLOBAL(VALUE *var, const char *cstr)
+{
+  rb_global_variable(var);
+  *var = rb_enc_interned_str_cstr(cstr, rb_utf8_encoding());
+}
 
 /* Defines the maximum allowed lengths for various input elements.*/
+#ifndef PUMA_REQUEST_URI_MAX_LENGTH
+#define PUMA_REQUEST_URI_MAX_LENGTH (1024 * 12)
+#endif
+
+#ifndef PUMA_REQUEST_PATH_MAX_LENGTH
+#define PUMA_REQUEST_PATH_MAX_LENGTH (8192)
+#endif
+
+#ifndef PUMA_QUERY_STRING_MAX_LENGTH
+#define PUMA_QUERY_STRING_MAX_LENGTH (1024 * 10)
+#endif
+
 DEF_MAX_LENGTH(FIELD_NAME, 256);
 DEF_MAX_LENGTH(FIELD_VALUE, 80 * 1024);
-DEF_MAX_LENGTH(REQUEST_URI, 1024 * 12);
+DEF_MAX_LENGTH(REQUEST_URI, PUMA_REQUEST_URI_MAX_LENGTH);
 DEF_MAX_LENGTH(FRAGMENT, 1024); /* Don't know if this length is specified somewhere or not */
-DEF_MAX_LENGTH(REQUEST_PATH, 1024);
-DEF_MAX_LENGTH(QUERY_STRING, (1024 * 10));
+DEF_MAX_LENGTH(REQUEST_PATH, PUMA_REQUEST_PATH_MAX_LENGTH);
+DEF_MAX_LENGTH(QUERY_STRING, PUMA_QUERY_STRING_MAX_LENGTH);
 DEF_MAX_LENGTH(HEADER, (1024 * (80 + 32)));
 
 struct common_field {
-	const size_t len;
-	const char *name;
+  const size_t len;
+  const char *name;
   int raw;
-	VALUE value;
+  VALUE value;
 };
 
 /*
@@ -71,60 +78,45 @@ struct common_field {
 static struct common_field common_http_fields[] = {
 # define f(N) { (sizeof(N) - 1), N, 0, Qnil }
 # define fr(N) { (sizeof(N) - 1), N, 1, Qnil }
-	f("ACCEPT"),
-	f("ACCEPT_CHARSET"),
-	f("ACCEPT_ENCODING"),
-	f("ACCEPT_LANGUAGE"),
-	f("ALLOW"),
-	f("AUTHORIZATION"),
-	f("CACHE_CONTROL"),
-	f("CONNECTION"),
-	f("CONTENT_ENCODING"),
-	fr("CONTENT_LENGTH"),
-	fr("CONTENT_TYPE"),
-	f("COOKIE"),
-	f("DATE"),
-	f("EXPECT"),
-	f("FROM"),
-	f("HOST"),
-	f("IF_MATCH"),
-	f("IF_MODIFIED_SINCE"),
-	f("IF_NONE_MATCH"),
-	f("IF_RANGE"),
-	f("IF_UNMODIFIED_SINCE"),
-	f("KEEP_ALIVE"), /* Firefox sends this */
-	f("MAX_FORWARDS"),
-	f("PRAGMA"),
-	f("PROXY_AUTHORIZATION"),
-	f("RANGE"),
-	f("REFERER"),
-	f("TE"),
-	f("TRAILER"),
-	f("TRANSFER_ENCODING"),
-	f("UPGRADE"),
-	f("USER_AGENT"),
-	f("VIA"),
-	f("X_FORWARDED_FOR"), /* common for proxies */
-	f("X_REAL_IP"), /* common for proxies */
-	f("WARNING")
+  f("ACCEPT"),
+  f("ACCEPT_CHARSET"),
+  f("ACCEPT_ENCODING"),
+  f("ACCEPT_LANGUAGE"),
+  f("ALLOW"),
+  f("AUTHORIZATION"),
+  f("CACHE_CONTROL"),
+  f("CONNECTION"),
+  f("CONTENT_ENCODING"),
+  fr("CONTENT_LENGTH"),
+  fr("CONTENT_TYPE"),
+  f("COOKIE"),
+  f("DATE"),
+  f("EXPECT"),
+  f("FROM"),
+  f("HOST"),
+  f("IF_MATCH"),
+  f("IF_MODIFIED_SINCE"),
+  f("IF_NONE_MATCH"),
+  f("IF_RANGE"),
+  f("IF_UNMODIFIED_SINCE"),
+  f("KEEP_ALIVE"), /* Firefox sends this */
+  f("MAX_FORWARDS"),
+  f("PRAGMA"),
+  f("PROXY_AUTHORIZATION"),
+  f("RANGE"),
+  f("REFERER"),
+  f("TE"),
+  f("TRAILER"),
+  f("TRANSFER_ENCODING"),
+  f("UPGRADE"),
+  f("USER_AGENT"),
+  f("VIA"),
+  f("X_FORWARDED_FOR"), /* common for proxies */
+  f("X_REAL_IP"), /* common for proxies */
+  f("WARNING")
 # undef f
 };
 
-/*
- * qsort(3) and bsearch(3) improve average performance slightly, but may
- * not be worth it for lack of portability to certain platforms...
- */
-#if defined(HAVE_QSORT_BSEARCH)
-/* sort by length, then by name if there's a tie */
-static int common_field_cmp(const void *a, const void *b)
-{
-  struct common_field *cfa = (struct common_field *)a;
-  struct common_field *cfb = (struct common_field *)b;
-  signed long diff = cfa->len - cfb->len;
-  return diff ? diff : memcmp(cfa->name, cfb->name, cfa->len);
-}
-#endif /* HAVE_QSORT_BSEARCH */
-
 static void init_common_fields(void)
 {
   unsigned i;
@@ -133,36 +125,18 @@ static void init_common_fields(void)
   memcpy(tmp, HTTP_PREFIX, HTTP_PREFIX_LEN);
 
   for(i = 0; i < ARRAY_SIZE(common_http_fields); cf++, i++) {
+    rb_global_variable(&cf->value);
     if(cf->raw) {
-      cf->value = rb_str_new(cf->name, cf->len);
+      cf->value = rb_enc_interned_str(cf->name, cf->len, rb_utf8_encoding());
     } else {
       memcpy(tmp + HTTP_PREFIX_LEN, cf->name, cf->len + 1);
-      cf->value = rb_str_new(tmp, HTTP_PREFIX_LEN + cf->len);
+      cf->value = rb_enc_interned_str(tmp, HTTP_PREFIX_LEN + cf->len, rb_utf8_encoding());
     }
-    rb_global_variable(&cf->value);
   }
-
-#if defined(HAVE_QSORT_BSEARCH)
-  qsort(common_http_fields,
-        ARRAY_SIZE(common_http_fields),
-        sizeof(struct common_field),
-        common_field_cmp);
-#endif /* HAVE_QSORT_BSEARCH */
 }
 
 static VALUE find_common_field_value(const char *field, size_t flen)
 {
-#if defined(HAVE_QSORT_BSEARCH)
-  struct common_field key;
-  struct common_field *found;
-  key.name = field;
-  key.len = (signed long)flen;
-  found = (struct common_field *)bsearch(&key, common_http_fields,
-                                         ARRAY_SIZE(common_http_fields),
-                                         sizeof(struct common_field),
-                                         common_field_cmp);
-  return found ? found->value : Qnil;
-#else /* !HAVE_QSORT_BSEARCH */
   unsigned i;
   struct common_field *cf = common_http_fields;
   for(i = 0; i < ARRAY_SIZE(common_http_fields); i++, cf++) {
@@ -170,20 +144,21 @@ static VALUE find_common_field_value(const char *field, size_t flen)
       return cf->value;
   }
   return Qnil;
-#endif /* !HAVE_QSORT_BSEARCH */
 }
 
-void http_field(http_parser* hp, const char *field, size_t flen,
+static int is_ows(const char c) {
+    return c == ' ' || c == '\t';
+}
+
+static void http_field(puma_parser* hp, const char *field, size_t flen,
                                  const char *value, size_t vlen)
 {
-  VALUE v = Qnil;
   VALUE f = Qnil;
+  VALUE v;
 
   VALIDATE_MAX_LENGTH(flen, FIELD_NAME);
   VALIDATE_MAX_LENGTH(vlen, FIELD_VALUE);
 
-  v = rb_str_new(value, vlen);
-
   f = find_common_field_value(field, flen);
 
   if (f == Qnil) {
@@ -198,13 +173,29 @@ void http_field(http_parser* hp, const char *field, size_t flen,
     memcpy(hp->buf, HTTP_PREFIX, HTTP_PREFIX_LEN);
     memcpy(hp->buf + HTTP_PREFIX_LEN, field, flen);
 
-    f = rb_str_new(hp->buf, new_size);
+    f = rb_enc_interned_str(hp->buf, new_size, rb_utf8_encoding());
+  }
+
+  while (vlen > 0 && is_ows(value[vlen - 1])) vlen--;
+  while (vlen > 0 && is_ows(value[0])) {
+      vlen--;
+      value++;
   }
 
-  rb_hash_aset(hp->request, f, v);
+  /* check for duplicate header */
+  v = rb_hash_aref(hp->request, f);
+
+  if (v == Qnil) {
+      v = rb_str_new(value, vlen);
+      rb_hash_aset(hp->request, f, v);
+  } else {
+      /* if duplicate header, normalize to comma-separated values */
+      rb_str_cat2(v, ", ");
+      rb_str_cat(v, value, vlen);
+  }
 }
 
-void request_method(http_parser* hp, const char *at, size_t length)
+static void request_method(puma_parser* hp, const char *at, size_t length)
 {
   VALUE val = Qnil;
 
@@ -212,7 +203,7 @@ void request_method(http_parser* hp, const char *at, size_t length)
   rb_hash_aset(hp->request, global_request_method, val);
 }
 
-void request_uri(http_parser* hp, const char *at, size_t length)
+static void request_uri(puma_parser* hp, const char *at, size_t length)
 {
   VALUE val = Qnil;
 
@@ -222,7 +213,7 @@ void request_uri(http_parser* hp, const char *at, size_t length)
   rb_hash_aset(hp->request, global_request_uri, val);
 }
 
-void fragment(http_parser* hp, const char *at, size_t length)
+static void fragment(puma_parser* hp, const char *at, size_t length)
 {
   VALUE val = Qnil;
 
@@ -232,7 +223,7 @@ void fragment(http_parser* hp, const char *at, size_t length)
   rb_hash_aset(hp->request, global_fragment, val);
 }
 
-void request_path(http_parser* hp, const char *at, size_t length)
+static void request_path(puma_parser* hp, const char *at, size_t length)
 {
   VALUE val = Qnil;
 
@@ -242,7 +233,7 @@ void request_path(http_parser* hp, const char *at, size_t length)
   rb_hash_aset(hp->request, global_request_path, val);
 }
 
-void query_string(http_parser* hp, const char *at, size_t length)
+static void query_string(puma_parser* hp, const char *at, size_t length)
 {
   VALUE val = Qnil;
 
@@ -252,51 +243,74 @@ void query_string(http_parser* hp, const char *at, size_t length)
   rb_hash_aset(hp->request, global_query_string, val);
 }
 
-void http_version(http_parser* hp, const char *at, size_t length)
+static void server_protocol(puma_parser* hp, const char *at, size_t length)
 {
   VALUE val = rb_str_new(at, length);
-  rb_hash_aset(hp->request, global_http_version, val);
+  rb_hash_aset(hp->request, global_server_protocol, val);
 }
 
 /** Finalizes the request header to have a bunch of stuff that's
   needed. */
 
-void header_done(http_parser* hp, const char *at, size_t length)
+static void header_done(puma_parser* hp, const char *at, size_t length)
 {
   hp->body = rb_str_new(at, length);
 }
 
 
-void HttpParser_free(void *data) {
-  TRACE();
+static void HttpParser_mark(void *ptr) {
+  puma_parser *hp = ptr;
+  rb_gc_mark_movable(hp->request);
+  rb_gc_mark_movable(hp->body);
+}
 
-  if(data) {
-    xfree(data);
-  }
+static size_t HttpParser_size(const void *ptr) {
+  return sizeof(puma_parser);
 }
 
-void HttpParser_mark(http_parser* hp) {
-  if(hp->request) rb_gc_mark(hp->request);
-  if(hp->body) rb_gc_mark(hp->body);
+static void HttpParser_compact(void *ptr) {
+  puma_parser *hp = ptr;
+  hp->request = rb_gc_location(hp->request);
+  hp->body = rb_gc_location(hp->body);
 }
 
-VALUE HttpParser_alloc(VALUE klass)
+static const rb_data_type_t HttpParser_data_type = {
+    .wrap_struct_name = "Puma::HttpParser",
+    .function = {
+      .dmark = HttpParser_mark,
+      .dfree = RUBY_TYPED_DEFAULT_FREE,
+      .dsize = HttpParser_size,
+      .dcompact = HttpParser_compact,
+    },
+    .flags = RUBY_TYPED_FREE_IMMEDIATELY,
+};
+
+static VALUE HttpParser_alloc(VALUE klass)
 {
-  http_parser *hp = ALLOC_N(http_parser, 1);
-  TRACE();
+  puma_parser *hp = ALLOC_N(puma_parser, 1);
   hp->http_field = http_field;
   hp->request_method = request_method;
   hp->request_uri = request_uri;
   hp->fragment = fragment;
   hp->request_path = request_path;
   hp->query_string = query_string;
-  hp->http_version = http_version;
+  hp->server_protocol = server_protocol;
   hp->header_done = header_done;
   hp->request = Qnil;
 
-  http_parser_init(hp);
+  puma_parser_init(hp);
 
-  return Data_Wrap_Struct(klass, HttpParser_mark, HttpParser_free, hp);
+  return TypedData_Wrap_Struct(klass, &HttpParser_data_type, hp);
+}
+
+static inline puma_parser *HttpParser_unwrap(VALUE self)
+{
+  puma_parser *http;
+  TypedData_Get_Struct(self, puma_parser, &HttpParser_data_type, http);
+  if (http == NULL) {
+    rb_raise(rb_eArgError, "%s", "NULL http_parser found");
+  }
+  return http;
 }
 
 /**
@@ -305,11 +319,10 @@ VALUE HttpParser_alloc(VALUE klass)
  *
  * Creates a new parser.
  */
-VALUE HttpParser_init(VALUE self)
+static VALUE HttpParser_init(VALUE self)
 {
-  http_parser *http = NULL;
-  DATA_GET(self, http_parser, http);
-  http_parser_init(http);
+  puma_parser *http = HttpParser_unwrap(self);
+  puma_parser_init(http);
 
   return self;
 }
@@ -322,11 +335,10 @@ VALUE HttpParser_init(VALUE self)
  * Resets the parser to it's initial state so that you can reuse it
  * rather than making new ones.
  */
-VALUE HttpParser_reset(VALUE self)
+static VALUE HttpParser_reset(VALUE self)
 {
-  http_parser *http = NULL;
-  DATA_GET(self, http_parser, http);
-  http_parser_init(http);
+  puma_parser *http = HttpParser_unwrap(self);
+  puma_parser_init(http);
 
   return Qnil;
 }
@@ -339,13 +351,12 @@ VALUE HttpParser_reset(VALUE self)
  * Finishes a parser early which could put in a "good" or bad state.
  * You should call reset after finish it or bad things will happen.
  */
-VALUE HttpParser_finish(VALUE self)
+static VALUE HttpParser_finish(VALUE self)
 {
-  http_parser *http = NULL;
-  DATA_GET(self, http_parser, http);
-  http_parser_finish(http);
+  puma_parser *http = HttpParser_unwrap(self);
+  puma_parser_finish(http);
 
-  return http_parser_is_finished(http) ? Qtrue : Qfalse;
+  return puma_parser_is_finished(http) ? Qtrue : Qfalse;
 }
 
 
@@ -366,32 +377,28 @@ VALUE HttpParser_finish(VALUE self)
  * the parsing from that position.  It needs all of the original data as well
  * so you have to append to the data buffer as you read.
  */
-VALUE HttpParser_execute(VALUE self, VALUE req_hash, VALUE data, VALUE start)
+static VALUE HttpParser_execute(VALUE self, VALUE req_hash, VALUE data, VALUE start)
 {
-  http_parser *http = NULL;
+  puma_parser *http = HttpParser_unwrap(self);
   int from = 0;
   char *dptr = NULL;
   long dlen = 0;
 
-  DATA_GET(self, http_parser, http);
-
   from = FIX2INT(start);
-  dptr = rb_extract_chars(data, &dlen);
+  RSTRING_GETMEM(data, dptr, dlen);
 
   if(from >= dlen) {
-    rb_free_chars(dptr);
     rb_raise(eHttpParserError, "%s", "Requested start is after data buffer end.");
   } else {
     http->request = req_hash;
-    http_parser_execute(http, dptr, dlen, from);
+    puma_parser_execute(http, dptr, dlen, from);
 
-    rb_free_chars(dptr);
-    VALIDATE_MAX_LENGTH(http_parser_nread(http), HEADER);
+    VALIDATE_MAX_LENGTH(puma_parser_nread(http), HEADER);
 
-    if(http_parser_has_error(http)) {
-      rb_raise(eHttpParserError, "%s", "Invalid HTTP format, parsing fails.");
+    if(puma_parser_has_error(http)) {
+      rb_raise(eHttpParserError, "%s", "Invalid HTTP format, parsing fails. Are you trying to open an SSL connection to a non-SSL Puma?");
     } else {
-      return INT2FIX(http_parser_nread(http));
+      return INT2FIX(puma_parser_nread(http));
     }
   }
 }
@@ -404,12 +411,11 @@ VALUE HttpParser_execute(VALUE self, VALUE req_hash, VALUE data, VALUE start)
  *
  * Tells you whether the parser is in an error state.
  */
-VALUE HttpParser_has_error(VALUE self)
+static VALUE HttpParser_has_error(VALUE self)
 {
-  http_parser *http = NULL;
-  DATA_GET(self, http_parser, http);
+  puma_parser *http = HttpParser_unwrap(self);
 
-  return http_parser_has_error(http) ? Qtrue : Qfalse;
+  return puma_parser_has_error(http) ? Qtrue : Qfalse;
 }
 
 
@@ -419,12 +425,11 @@ VALUE HttpParser_has_error(VALUE self)
  *
  * Tells you whether the parser is finished or not and in a good state.
  */
-VALUE HttpParser_is_finished(VALUE self)
+static VALUE HttpParser_is_finished(VALUE self)
 {
-  http_parser *http = NULL;
-  DATA_GET(self, http_parser, http);
+  puma_parser *http = HttpParser_unwrap(self);
 
-  return http_parser_is_finished(http) ? Qtrue : Qfalse;
+  return puma_parser_is_finished(http) ? Qtrue : Qfalse;
 }
 
 
@@ -435,10 +440,9 @@ VALUE HttpParser_is_finished(VALUE self)
  * Returns the amount of data processed so far during this processing cycle.  It is
  * set to 0 on initialize or reset calls and is incremented each time execute is called.
  */
-VALUE HttpParser_nread(VALUE self)
+static VALUE HttpParser_nread(VALUE self)
 {
-  http_parser *http = NULL;
-  DATA_GET(self, http_parser, http);
+  puma_parser *http = HttpParser_unwrap(self);
 
   return INT2FIX(http->nread);
 }
@@ -449,31 +453,34 @@ VALUE HttpParser_nread(VALUE self)
  *
  * If the request included a body, returns it.
  */
-VALUE HttpParser_body(VALUE self) {
-  http_parser *http = NULL;
-  DATA_GET(self, http_parser, http);
+static VALUE HttpParser_body(VALUE self) {
+  puma_parser *http = HttpParser_unwrap(self);
 
   return http->body;
 }
 
-void Init_io_buffer(VALUE puma);
+#ifdef HAVE_OPENSSL_BIO_H
 void Init_mini_ssl(VALUE mod);
+#endif
 
-void Init_puma_http11()
+RUBY_FUNC_EXPORTED void Init_puma_http11(void)
 {
+#ifdef HAVE_RB_EXT_RACTOR_SAFE
+  rb_ext_ractor_safe(true);
+#endif
 
   VALUE mPuma = rb_define_module("Puma");
   VALUE cHttpParser = rb_define_class_under(mPuma, "HttpParser", rb_cObject);
 
-  DEF_GLOBAL(request_method, "REQUEST_METHOD");
-  DEF_GLOBAL(request_uri, "REQUEST_URI");
-  DEF_GLOBAL(fragment, "FRAGMENT");
-  DEF_GLOBAL(query_string, "QUERY_STRING");
-  DEF_GLOBAL(http_version, "HTTP_VERSION");
-  DEF_GLOBAL(request_path, "REQUEST_PATH");
+  DEF_GLOBAL(&global_request_method, "REQUEST_METHOD");
+  DEF_GLOBAL(&global_request_uri, "REQUEST_URI");
+  DEF_GLOBAL(&global_fragment, "FRAGMENT");
+  DEF_GLOBAL(&global_query_string, "QUERY_STRING");
+  DEF_GLOBAL(&global_server_protocol, "SERVER_PROTOCOL");
+  DEF_GLOBAL(&global_request_path, "REQUEST_PATH");
 
-  eHttpParserError = rb_define_class_under(mPuma, "HttpParserError", rb_eIOError);
   rb_global_variable(&eHttpParserError);
+  eHttpParserError = rb_define_class_under(mPuma, "HttpParserError", rb_eStandardError);
 
   rb_define_alloc_func(cHttpParser, HttpParser_alloc);
   rb_define_method(cHttpParser, "initialize", HttpParser_init, 0);
@@ -486,6 +493,7 @@ void Init_puma_http11()
   rb_define_method(cHttpParser, "body", HttpParser_body, 0);
   init_common_fields();
 
-  Init_io_buffer(mPuma);
+#ifdef HAVE_OPENSSL_BIO_H
   Init_mini_ssl(mPuma);
+#endif
 }