[Downloads] Introduce GUIDs for downloads.

components/history/core/browser/download_database.cc

 // Copyright (c) 2012 The Chromium Authors. All rights reserved.
 // Use of this source code is governed by a BSD-style license that can be
 // found in the LICENSE file.
 
 #include "components/history/core/browser/download_database.h"
 
 #include <limits>
 #include <string>
 #include <vector>
 
@@ skipping 168 matching lines @@
 bool DownloadDatabase::MigrateDownloadedByExtension() {
   return EnsureColumnExists("by_ext_id", "VARCHAR NOT NULL DEFAULT \"\"") &&
          EnsureColumnExists("by_ext_name", "VARCHAR NOT NULL DEFAULT \"\"");
 }
 
 bool DownloadDatabase::MigrateDownloadValidators() {
   return EnsureColumnExists("etag", "VARCHAR NOT NULL DEFAULT \"\"") &&
          EnsureColumnExists("last_modified", "VARCHAR NOT NULL DEFAULT \"\"");
 }
 
+bool DownloadDatabase::MigrateHashHttpMethodAndGenerateGuids() {
+  if (!EnsureColumnExists("guid", "VARCHAR NOT NULL DEFAULT ''") ||
+      !EnsureColumnExists("hash", "BLOB NOT NULL DEFAULT X''") ||
+      !EnsureColumnExists("http_method", "VARCHAR NOT NULL DEFAULT ''"))
+    return false;
+
+  // Generate GUIDs for each download. The GUID is generated thusly:
+  //
+  //    XXXXXXXX-RRRR-RRRR-RRRR-RRRRRRRRRRRR
+  //    \__  __/ \___________  ____________/
+  //       \/                \/
+  //       |          Random hex digits
+  //       |
+  //       Hex representation of 32-bit download ID.
+  //
+  // The 96 random bits provide entropy while the 32-bits from the download ID
+  // ensure that the generated identifiers will at least be unique amongst the
+  // download rows in the unlikely event there's a collision in the 96 entropy
+  // bits.
+  //
+  // This GUID generation scheme is only used for migrated download rows and
+  // assumes that the likelihood of a collision with a GUID generated via
+  // base::GenerateGUID() will be vanishingly small.
+  const char kMigrateGuidsQuery[] =
+      "UPDATE downloads SET guid = printf"
+      "(\"%08X-%s-%s-%s-%s\", id, hex(randomblob(2)), hex(randomblob(2)),"
+      " hex(randomblob(2)), hex(randomblob(6)))";
+  return GetDB().Execute(kMigrateGuidsQuery);
+}
+
 bool DownloadDatabase::InitDownloadTable() {
   const char kSchema[] =
       "CREATE TABLE downloads ("
       "id INTEGER PRIMARY KEY,"             // Primary key.
+      "guid VARCHAR NOT NULL,"              // GUID.
       "current_path LONGVARCHAR NOT NULL,"  // Current disk location
       "target_path LONGVARCHAR NOT NULL,"   // Final disk location
       "start_time INTEGER NOT NULL,"        // When the download was started.
       "received_bytes INTEGER NOT NULL,"    // Total size downloaded.
       "total_bytes INTEGER NOT NULL,"       // Total size of the download.
       "state INTEGER NOT NULL,"             // 1=complete, 4=interrupted
       "danger_type INTEGER NOT NULL,"       // Danger type, validated.
       "interrupt_reason INTEGER NOT NULL,"  // DownloadInterruptReason
+      "hash BLOB NOT NULL,"                 // Raw SHA-256 hash of contents.
       "end_time INTEGER NOT NULL,"          // When the download completed.
       "opened INTEGER NOT NULL,"            // 1 if it has ever been opened
                                             // else 0
       "referrer VARCHAR NOT NULL,"          // HTTP Referrer
+      "http_method VARCHAR NOT NULL,"       // HTTP method.
       "by_ext_id VARCHAR NOT NULL,"         // ID of extension that started the
                                             // download
       "by_ext_name VARCHAR NOT NULL,"       // name of extension
       "etag VARCHAR NOT NULL,"              // ETag
       "last_modified VARCHAR NOT NULL,"     // Last-Modified header
       "mime_type VARCHAR(255) NOT NULL,"    // MIME type.
       "original_mime_type VARCHAR(255) NOT NULL)";  // Original MIME type.
 
   const char kUrlChainSchema[] =
       "CREATE TABLE downloads_url_chains ("
@@ skipping 35 matching lines @@
   return 1 + static_cast<uint32_t>(
                  std::max(static_cast<int64_t>(kInvalidDownloadId),
                           select_max_id.ColumnInt64(0)));
 }
 
 bool DownloadDatabase::DropDownloadTable() {
   return GetDB().Execute("DROP TABLE downloads");
 }
 
 void DownloadDatabase::QueryDownloads(std::vector<DownloadRow>* results) {
+  SCOPED_UMA_HISTOGRAM_TIMER("Download.Database.QueryDownloadDuration");
   EnsureInProgressEntriesCleanedUp();
 
   results->clear();
   std::set<uint32_t> ids;
 
   std::map<uint32_t, DownloadRow*> info_map;
 
-  sql::Statement statement_main(GetDB().GetCachedStatement(SQL_FROM_HERE,
-      "SELECT id, current_path, target_path, "
-      "mime_type, original_mime_type, "
-      "start_time, received_bytes, "
-      "total_bytes, state, danger_type, interrupt_reason, end_time, opened, "
-      "referrer, by_ext_id, by_ext_name, etag, last_modified "
+  sql::Statement statement_main(GetDB().GetCachedStatement(
+      SQL_FROM_HERE,
+      "SELECT id, guid, current_path, target_path, mime_type, "
+      "original_mime_type, start_time, received_bytes, total_bytes, state, "
+      "danger_type, interrupt_reason, hash, end_time, opened, referrer, "
+      "http_method, by_ext_id, by_ext_name, etag, last_modified "
       "FROM downloads ORDER BY start_time"));
 
   while (statement_main.Step()) {
     scoped_ptr<DownloadRow> info(new DownloadRow());
     int column = 0;
 
     // SQLITE does not have unsigned integers, so explicitly handle negative
     // |id|s instead of casting them to very large uint32s, which would break
     // the max(id) logic in GetNextDownloadId().
     int64_t signed_id = statement_main.ColumnInt64(column++);
     info->id = IntToDownloadId(signed_id);
+    info->guid = statement_main.ColumnString(column++);
     info->current_path = ColumnFilePath(statement_main, column++);
     info->target_path = ColumnFilePath(statement_main, column++);
     info->mime_type = statement_main.ColumnString(column++);
     info->original_mime_type = statement_main.ColumnString(column++);
     info->start_time =
         base::Time::FromInternalValue(statement_main.ColumnInt64(column++));
     info->received_bytes = statement_main.ColumnInt64(column++);
     info->total_bytes = statement_main.ColumnInt64(column++);
     int state = statement_main.ColumnInt(column++);
     info->state = IntToDownloadState(state);
     if (info->state == DownloadState::INVALID)
       UMA_HISTOGRAM_COUNTS("Download.DatabaseInvalidState", state);
     info->danger_type =
         IntToDownloadDangerType(statement_main.ColumnInt(column++));
     info->interrupt_reason =
         IntToDownloadInterruptReason(statement_main.ColumnInt(column++));
+    statement_main.ColumnBlobAsString(column++, &info->hash);
     info->end_time =
         base::Time::FromInternalValue(statement_main.ColumnInt64(column++));
     info->opened = statement_main.ColumnInt(column++) != 0;
     info->referrer_url = GURL(statement_main.ColumnString(column++));
+    info->http_method = statement_main.ColumnString(column++);
     info->by_ext_id = statement_main.ColumnString(column++);
     info->by_ext_name = statement_main.ColumnString(column++);
     info->etag = statement_main.ColumnString(column++);
     info->last_modified = statement_main.ColumnString(column++);
 
     // If the record is corrupted, note that and drop it.
     // http://crbug.com/251269
     DroppedReason dropped_reason = DROPPED_REASON_MAX;
     if (signed_id <= static_cast<int64_t>(kInvalidDownloadId)) {
       // SQLITE doesn't have unsigned integers.
@@ skipping 69 matching lines @@
     } else {
       // Copy the contents of the stored info.
       results->push_back(*row);
     }
     delete row;
     it->second = NULL;
   }
 }
 
 bool DownloadDatabase::UpdateDownload(const DownloadRow& data) {
+  // UpdateDownload() is called fairly frequently.
+  SCOPED_UMA_HISTOGRAM_TIMER("Download.Database.UpdateDownloadDuration");
   EnsureInProgressEntriesCleanedUp();
 
   DCHECK_NE(kInvalidDownloadId, data.id);
   if (data.state == DownloadState::INVALID) {
     NOTREACHED();
     return false;
   }
   if (data.danger_type == DownloadDangerType::INVALID) {
     NOTREACHED();
     return false;
   }
 
-  sql::Statement statement(GetDB().GetCachedStatement(SQL_FROM_HERE,
+  sql::Statement statement(GetDB().GetCachedStatement(
+      SQL_FROM_HERE,
       "UPDATE downloads "
       "SET current_path=?, target_path=?, "
       "mime_type=?, original_mime_type=?, "
       "received_bytes=?, state=?, "
-      "danger_type=?, interrupt_reason=?, end_time=?, total_bytes=?, "
+      "danger_type=?, interrupt_reason=?, hash=?, end_time=?, total_bytes=?, "
       "opened=?, by_ext_id=?, by_ext_name=?, etag=?, last_modified=? "
       "WHERE id=?"));
   int column = 0;
   BindFilePath(statement, data.current_path, column++);
   BindFilePath(statement, data.target_path, column++);
   statement.BindString(column++, data.mime_type);
   statement.BindString(column++, data.original_mime_type);
   statement.BindInt64(column++, data.received_bytes);
   statement.BindInt(column++, DownloadStateToInt(data.state));
   statement.BindInt(column++, DownloadDangerTypeToInt(data.danger_type));
   statement.BindInt(column++,
                     DownloadInterruptReasonToInt(data.interrupt_reason));
+  statement.BindBlob(column++, data.hash.data(), data.hash.size());
   statement.BindInt64(column++, data.end_time.ToInternalValue());
   statement.BindInt64(column++, data.total_bytes);
   statement.BindInt(column++, (data.opened ? 1 : 0));
   statement.BindString(column++, data.by_ext_id);
   statement.BindString(column++, data.by_ext_name);
   statement.BindString(column++, data.etag);
   statement.BindString(column++, data.last_modified);
   statement.BindInt(column++, DownloadIdToInt(data.id));
 
   return statement.Run();
 }
 
 void DownloadDatabase::EnsureInProgressEntriesCleanedUp() {
   if (in_progress_entry_cleanup_completed_)
     return;
 
   sql::Statement statement(GetDB().GetCachedStatement(SQL_FROM_HERE,
       "UPDATE downloads SET state=?, interrupt_reason=? WHERE state=?"));
   statement.BindInt(0, DownloadStateToInt(DownloadState::INTERRUPTED));
   statement.BindInt(
       1, DownloadInterruptReasonToInt(download_interrupt_reason_crash_));
   statement.BindInt(2, DownloadStateToInt(DownloadState::IN_PROGRESS));
 
   statement.Run();
   in_progress_entry_cleanup_completed_ = true;
 }
 
 bool DownloadDatabase::CreateDownload(const DownloadRow& info) {
   DCHECK_NE(kInvalidDownloadId, info.id);
+  DCHECK(!info.guid.empty());
+  SCOPED_UMA_HISTOGRAM_TIMER("Download.Database.CreateDownloadDuration");
   EnsureInProgressEntriesCleanedUp();
 
   if (info.url_chain.empty())
     return false;
 
   if (info.state == DownloadState::INVALID)
     return false;
 
   if (info.danger_type == DownloadDangerType::INVALID)
     return false;
 
   {
     sql::Statement statement_insert(GetDB().GetCachedStatement(
         SQL_FROM_HERE,
         "INSERT INTO downloads "
-        "(id, current_path, target_path, "
-        " mime_type, original_mime_type, "
-        " start_time, "
-        " received_bytes, total_bytes, state, danger_type, interrupt_reason, "
-        " end_time, opened, referrer, by_ext_id, by_ext_name, etag, "
-        " last_modified) "
-        "VALUES (?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?)"));
+        "(id, guid, current_path, target_path, mime_type, original_mime_type, "
+        " start_time, received_bytes, total_bytes, state, danger_type, "
+        " interrupt_reason, hash, end_time, opened, referrer, http_method, "
+        " by_ext_id, by_ext_name, etag, last_modified) "
+        "VALUES (?, ?, ?, ?, ?, ?, ?, ?, ?, ?, "
+        "        ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, "
+        "        ?)"));
 
     int column = 0;
     statement_insert.BindInt(column++, DownloadIdToInt(info.id));
+    statement_insert.BindString(column++, info.guid);
     BindFilePath(statement_insert, info.current_path, column++);
     BindFilePath(statement_insert, info.target_path, column++);
     statement_insert.BindString(column++, info.mime_type);
     statement_insert.BindString(column++, info.original_mime_type);
     statement_insert.BindInt64(column++, info.start_time.ToInternalValue());
     statement_insert.BindInt64(column++, info.received_bytes);
     statement_insert.BindInt64(column++, info.total_bytes);
     statement_insert.BindInt(column++, DownloadStateToInt(info.state));
     statement_insert.BindInt(column++,
                              DownloadDangerTypeToInt(info.danger_type));
     statement_insert.BindInt(
         column++, DownloadInterruptReasonToInt(info.interrupt_reason));
+    statement_insert.BindBlob(column++, info.hash.data(), info.hash.size());
     statement_insert.BindInt64(column++, info.end_time.ToInternalValue());
     statement_insert.BindInt(column++, info.opened ? 1 : 0);
     statement_insert.BindString(column++, info.referrer_url.spec());
+    statement_insert.BindString(column++, info.http_method);
     statement_insert.BindString(column++, info.by_ext_id);
     statement_insert.BindString(column++, info.by_ext_name);
     statement_insert.BindString(column++, info.etag);
     statement_insert.BindString(column++, info.last_modified);
     if (!statement_insert.Run()) {
       // GetErrorCode() returns a bitmask where the lower byte is a more general
       // code and the upper byte is a more specific code. In order to save
       // memory, take the general code, of which there are fewer than 50. See
       // also sql/connection.cc
       // http://www.sqlite.org/c3ref/c_abort_rollback.html
@@ skipping 66 matching lines @@
 size_t DownloadDatabase::CountDownloads() {
   EnsureInProgressEntriesCleanedUp();
 
   sql::Statement statement(GetDB().GetCachedStatement(SQL_FROM_HERE,
       "SELECT count(*) from downloads"));
   statement.Step();
   return statement.ColumnInt(0);
 }
 
 }  // namespace history