Chromium Code Reviews
chromiumcodereview-hr@appspot.gserviceaccount.com (chromiumcodereview-hr) | Please choose your nickname with Settings | Help | Chromium Project | Gerrit Changes | Sign out
(189)

Unified Diff: net/base/registry_controlled_domains/registry_controlled_domain.cc

Issue 197183002: Reduce footprint of registry controlled domain table (Closed) Base URL: https://chromium.googlesource.com/chromium/src.git@master
Patch Set: Removed shebang and execution bits Created 6 years, 7 months ago
Use n/p to move between diff chunks; N/P to move between comments. Draft comments are only viewable by you.
Jump to:
View side-by-side diff with in-line comments
Download patch
Index: net/base/registry_controlled_domains/registry_controlled_domain.cc
diff --git a/net/base/registry_controlled_domains/registry_controlled_domain.cc b/net/base/registry_controlled_domains/registry_controlled_domain.cc
index 56d5ed9d6a176864f6f4beb9af06d9d9cfafd485..057159ed7308e447d3bcf05019800b7c2b729794 100644
--- a/net/base/registry_controlled_domains/registry_controlled_domain.cc
+++ b/net/base/registry_controlled_domains/registry_controlled_domain.cc
@@ -53,26 +53,151 @@
#include "url/gurl.h"
#include "url/url_parse.h"
-#include "effective_tld_names.cc"
-
namespace net {
namespace registry_controlled_domains {
namespace {
+#include "effective_tld_names-inc.cc"
+
+// See make_dafsa.py for documentation of the generated dafsa byte array.
+const unsigned char* g_graph = kDafsa;
+size_t g_graph_length = sizeof(kDafsa);
+
+const int kNotFound = -1;
const int kExceptionRule = 1;
const int kWildcardRule = 2;
const int kPrivateRule = 4;
-const FindDomainPtr kDefaultFindDomainFunction = Perfect_Hash::FindDomain;
+// Read next offset from pos.
+// Returns true if an offset could be read, false otherwise.
+bool GetNextOffset(const unsigned char** pos, const unsigned char* end,
+ const unsigned char** offset) {
+ if (*pos == end)
+ return false;
+
+ // When reading an offset the byte array must always contain at least
+ // three more bytes to consume. First the offset to read, then a node
+ // to skip over and finally a destination node. No object can be smaller
+ // than one byte.
+ CHECK_LT(*pos + 2, end);
+ size_t bytes_consumed;
+ switch (**pos & 0x60) {
+ case 0x60: // Read three byte offset
+ *offset += (((*pos)[0] & 0x1F) << 16) | ((*pos)[1] << 8) | (*pos)[2];
+ bytes_consumed = 3;
+ break;
+ case 0x40: // Read two byte offset
+ *offset += (((*pos)[0] & 0x1F) << 8) | (*pos)[1];
+ bytes_consumed = 2;
+ break;
+ default:
+ *offset += (*pos)[0] & 0x3F;
+ bytes_consumed = 1;
+ }
+ if ((**pos & 0x80) != 0) {
+ *pos = end;
+ } else {
+ *pos += bytes_consumed;
+ }
+ return true;
+}
+
+// Check if byte at offset is last in label.
+bool IsEOL(const unsigned char* offset, const unsigned char* end) {
+ CHECK_LT(offset, end);
+ return (*offset & 0x80) != 0;
+}
+
+// Check if byte at offset matches first character in key.
+// This version matches characters not last in label.
+bool IsMatch(const unsigned char* offset, const unsigned char* end,
+ const char* key) {
+ CHECK_LT(offset, end);
+ return *offset == *key;
+}
+
+// Check if byte at offset matches first character in key.
+// This version matches characters last in label.
+bool IsEndCharMatch(const unsigned char* offset, const unsigned char* end,
+ const char* key) {
+ CHECK_LT(offset, end);
+ return *offset == (*key | 0x80);
+}
-// 'stringpool' is defined as a macro by the gperf-generated
-// "effective_tld_names.cc". Provide a real constant value for it instead.
-const char* const kDefaultStringPool = stringpool;
-#undef stringpool
+// Read return value at offset.
+// Returns true if a return value could be read, false otherwise.
+bool GetReturnValue(const unsigned char* offset, const unsigned char* end,
+ int* return_value) {
+ CHECK_LT(offset, end);
+ if ((*offset & 0xE0) == 0x80) {
+ *return_value = *offset & 0x0F;
+ return true;
+ }
+ return false;
+}
-FindDomainPtr g_find_domain_function = kDefaultFindDomainFunction;
-const char* g_stringpool = kDefaultStringPool;
+// Lookup a domain key in a byte array generated by make_dafsa.py.
+// The rule type is returned if key is found, otherwise kNotFound is returned.
+int LookupString(const unsigned char* graph, size_t length, const char* key,
+ size_t key_length) {
+ const unsigned char* pos = graph;
+ const unsigned char* end = graph + length;
+ const unsigned char* offset = pos;
+ const char* key_end = key + key_length;
+ while (GetNextOffset(&pos, end, &offset)) {
+ // char <char>+ end_char offsets
+ // char <char>+ return value
+ // char end_char offsets
+ // char return value
+ // end_char offsets
+ // return_value
+ bool did_consume = false;
+ if (key != key_end && !IsEOL(offset, end)) {
+ // Leading <char> is not a match. Don't dive into this child
+ if (!IsMatch(offset, end, key))
+ continue;
+ did_consume = true;
+ ++offset;
+ ++key;
+ // Possible matches at this point:
+ // <char>+ end_char offsets
+ // <char>+ return value
+ // end_char offsets
+ // return value
+ // Remove all remaining <char> nodes possible
+ while (!IsEOL(offset, end) && key != key_end) {
+ if (!IsMatch(offset, end, key))
+ return kNotFound;
+ ++key;
+ ++offset;
+ }
+ }
+ // Possible matches at this point:
+ // end_char offsets
+ // return_value
+ // If one or more <char> elements were consumed, a failure
+ // to match is terminal. Otherwise, try the next node.
+ if (key == key_end) {
+ int return_value;
+ if (GetReturnValue(offset, end, &return_value))
+ return return_value;
+ // The DAFSA guarantees that if the first char is a match, all
+ // remaining char elements MUST match if the key is truly present.
+ if (did_consume)
+ return kNotFound;
+ continue;
+ }
+ if (!IsEndCharMatch(offset, end, key)) {
+ if (did_consume)
+ return kNotFound; // Unexpected
+ continue;
+ }
+ ++key;
+ pos = ++offset; // Dive into child
+ }
+ return kNotFound; // No match
+}
size_t GetRegistryLengthImpl(
const std::string& host,
@@ -105,46 +230,40 @@ size_t GetRegistryLengthImpl(
return 0; // This can't have a registry + domain.
while (1) {
const char* domain_str = host.data() + curr_start;
- int domain_length = host_check_len - curr_start;
- const DomainRule* rule = g_find_domain_function(domain_str, domain_length);
-
- // We need to compare the string after finding a match because the
- // no-collisions of perfect hashing only refers to items in the set. Since
- // we're searching for arbitrary domains, there could be collisions.
- // Furthermore, if the apparent match is a private registry and we're not
- // including those, it can't be an actual match.
- if (rule) {
- bool do_check = !(rule->type & kPrivateRule) ||
- private_filter == INCLUDE_PRIVATE_REGISTRIES;
- if (do_check && base::strncasecmp(domain_str,
- g_stringpool + rule->name_offset,
- domain_length) == 0) {
- // Exception rules override wildcard rules when the domain is an exact
- // match, but wildcards take precedence when there's a subdomain.
- if (rule->type & kWildcardRule && (prev_start != std::string::npos)) {
- // If prev_start == host_check_begin, then the host is the registry
- // itself, so return 0.
- return (prev_start == host_check_begin) ?
- 0 : (host.length() - prev_start);
- }
+ size_t domain_length = host_check_len - curr_start;
+ int type = LookupString(g_graph, g_graph_length, domain_str, domain_length);
+ bool do_check =
+ type != kNotFound && (!(type & kPrivateRule) ||
+ private_filter == INCLUDE_PRIVATE_REGISTRIES);
+
+ // If the apparent match is a private registry and we're not including
+ // those, it can't be an actual match.
+ if (do_check) {
+ // Exception rules override wildcard rules when the domain is an exact
+ // match, but wildcards take precedence when there's a subdomain.
+ if (type & kWildcardRule && (prev_start != std::string::npos)) {
+ // If prev_start == host_check_begin, then the host is the registry
+ // itself, so return 0.
+ return (prev_start == host_check_begin) ? 0
+ : (host.length() - prev_start);
+ }
- if (rule->type & kExceptionRule) {
- if (next_dot == std::string::npos) {
- // If we get here, we had an exception rule with no dots (e.g.
- // "!foo"). This would only be valid if we had a corresponding
- // wildcard rule, which would have to be "*". But we explicitly
- // disallow that case, so this kind of rule is invalid.
- NOTREACHED() << "Invalid exception rule";
- return 0;
- }
- return host.length() - next_dot - 1;
+ if (type & kExceptionRule) {
+ if (next_dot == std::string::npos) {
+ // If we get here, we had an exception rule with no dots (e.g.
+ // "!foo"). This would only be valid if we had a corresponding
+ // wildcard rule, which would have to be "*". But we explicitly
+ // disallow that case, so this kind of rule is invalid.
+ NOTREACHED() << "Invalid exception rule";
+ return 0;
}
-
- // If curr_start == host_check_begin, then the host is the registry
- // itself, so return 0.
- return (curr_start == host_check_begin) ?
- 0 : (host.length() - curr_start);
+ return host.length() - next_dot - 1;
}
+
+ // If curr_start == host_check_begin, then the host is the registry
+ // itself, so return 0.
+ return (curr_start == host_check_begin) ? 0
+ : (host.length() - curr_start);
}
if (next_dot >= host_check_len) // Catches std::string::npos as well.
@@ -264,10 +383,16 @@ size_t GetRegistryLength(
return GetRegistryLengthImpl(canon_host, unknown_filter, private_filter);
}
-void SetFindDomainFunctionAndStringPoolForTesting(FindDomainPtr function,
- const char* stringpool) {
- g_find_domain_function = function ? function : kDefaultFindDomainFunction;
- g_stringpool = stringpool ? stringpool : kDefaultStringPool;
+void SetFindDomainGraph() {
+ g_graph = kDafsa;
+ g_graph_length = sizeof(kDafsa);
+}
+
+void SetFindDomainGraph(const unsigned char* domains, size_t length) {
+ CHECK(domains);
+ CHECK_NE(length, 0);
+ g_graph = domains;
+ g_graph_length = length;
}
} // namespace registry_controlled_domains

Powered by Google App Engine
This is Rietveld 408576698