chromium/net/tools/tld_cleanup/tld_cleanup.cc

// Copyright 2012 The Chromium Authors
// Use of this source code is governed by a BSD-style license that can be
// found in the LICENSE file.

// This command-line program converts an effective-TLD data file in UTF-8 from
// the format provided by Mozilla to the format expected by Chrome.  This
// program generates an intermediate file which is then used by gperf to
// generate a perfect hash map.  The benefit of this approach is that no time is
// spent on program initialization to generate the map of this data.
//
// Running this program finds "effective_tld_names.dat" in the expected location
// in the source checkout and generates "effective_tld_names.gperf" next to it.
//
// Any errors or warnings from this program are recorded in tld_cleanup.log.
//
// In particular, it
//  * Strips blank lines and comments, as well as notes for individual rules.
//  * Strips a single leading and/or trailing dot from each rule, if present.
//  * Logs a warning if a rule contains '!' or '*.' other than at the beginning
//    of the rule.  (This also catches multiple ! or *. at the start of a rule.)
//  * Logs a warning if GURL reports a rule as invalid, but keeps the rule.
//  * Canonicalizes each rule's domain by converting it to a GURL and back.
//  * Adds explicit rules for true TLDs found in any rule.
//  * Marks entries in the file between "// ===BEGIN PRIVATE DOMAINS==="
//    and "// ===END PRIVATE DOMAINS===" as private.

#include "base/at_exit.h"
#include "base/command_line.h"
#include "base/files/file_path.h"
#include "base/files/file_util.h"
#include "base/i18n/icu_util.h"
#include "base/logging.h"
#include "base/path_service.h"
#include "base/process/memory.h"
#include "net/tools/tld_cleanup/tld_cleanup_util.h"

int main(int argc, const char* argv[]) { … }