uri_parser 0.0.1

data/ext/uri_parser/url_canon_stdstring.h

@@ -0,0 +1,134 @@
+// Copyright 2007, Google Inc.
+// All rights reserved.
+//
+// Redistribution and use in source and binary forms, with or without
+// modification, are permitted provided that the following conditions are
+// met:
+//
+//     * Redistributions of source code must retain the above copyright
+// notice, this list of conditions and the following disclaimer.
+//     * Redistributions in binary form must reproduce the above
+// copyright notice, this list of conditions and the following disclaimer
+// in the documentation and/or other materials provided with the
+// distribution.
+//     * Neither the name of Google Inc. nor the names of its
+// contributors may be used to endorse or promote products derived from
+// this software without specific prior written permission.
+//
+// THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+// "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+// LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+// A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+// OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+// SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+// LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+// DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+// THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+// (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+// OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+
+// This header file defines a canonicalizer output method class for STL
+// strings. Because the canonicalizer tries not to be dependent on the STL,
+// we have segregated it here.
+
+#ifndef GOOGLEURL_SRC_URL_CANON_STDSTRING_H__
+#define GOOGLEURL_SRC_URL_CANON_STDSTRING_H__
+
+#include <string>
+#include "url_canon.h"
+
+namespace url_canon {
+
+// Write into a std::string given in the constructor. This object does not own
+// the string itself, and the user must ensure that the string stays alive
+// throughout the lifetime of this object.
+//
+// The given string will be appended to; any existing data in the string will
+// be preserved. The caller should reserve() the amount of data in the string
+// they expect to be written. We will resize if necessary, but that's slow.
+//
+// Note that when canonicalization is complete, the string will likely have
+// unused space at the end because we make the string very big to start out
+// with (by |initial_size|). This ends up being important because resize
+// operations are slow, and because the base class needs to write directly
+// into the buffer.
+//
+// Therefore, the user should call Complete() before using the string that
+// this class wrote into.
+class StdStringCanonOutput : public CanonOutput {
+ public:
+  StdStringCanonOutput(std::string* str)
+      : CanonOutput(),
+        str_(str) {
+    cur_len_ = static_cast<int>(str_->size());  // Append to existing data.
+    str_->resize(str_->capacity());
+    buffer_ = &(*str_)[0];
+    buffer_len_ = static_cast<int>(str_->size());
+  }
+  virtual ~StdStringCanonOutput() {
+    // Nothing to do, we don't own the string.
+  }
+
+  // Must be called after writing has completed but before the string is used.
+  void Complete() {
+    str_->resize(cur_len_);
+    buffer_len_ = cur_len_;
+  }
+
+  virtual void Resize(int sz) {
+    str_->resize(sz);
+    buffer_ = &(*str_)[0];
+    buffer_len_ = sz;
+  }
+
+ protected:
+  std::string* str_;
+};
+
+// An extension of the Replacements class that allows the setters to use
+// standard strings.
+//
+// The strings passed as arguments are not copied and must remain valid until
+// this class goes out of scope.
+template<typename STR>
+class StdStringReplacements :
+    public url_canon::Replacements<typename STR::value_type> {
+ public:
+  void SetSchemeStr(const STR& s) {
+    this->SetScheme(s.data(),
+                    url_parse::Component(0, static_cast<int>(s.length())));
+  }
+  void SetUsernameStr(const STR& s) {
+    this->SetUsername(s.data(),
+                      url_parse::Component(0, static_cast<int>(s.length())));
+  }
+  void SetPasswordStr(const STR& s) {
+    this->SetPassword(s.data(),
+                      url_parse::Component(0, static_cast<int>(s.length())));
+  }
+  void SetHostStr(const STR& s) {
+    this->SetHost(s.data(),
+                  url_parse::Component(0, static_cast<int>(s.length())));
+  }
+  void SetPortStr(const STR& s) {
+    this->SetPort(s.data(),
+                  url_parse::Component(0, static_cast<int>(s.length())));
+  }
+  void SetPathStr(const STR& s) {
+    this->SetPath(s.data(),
+                  url_parse::Component(0, static_cast<int>(s.length())));
+  }
+  void SetQueryStr(const STR& s) {
+    this->SetQuery(s.data(),
+                   url_parse::Component(0, static_cast<int>(s.length())));
+  }
+  void SetRefStr(const STR& s) {
+    this->SetRef(s.data(),
+                 url_parse::Component(0, static_cast<int>(s.length())));
+  }
+};
+
+}  // namespace url_canon
+
+#endif  // GOOGLEURL_SRC_URL_CANON_STDSTRING_H__
+

data/ext/uri_parser/url_canon_stdurl.cc

@@ -0,0 +1,211 @@
+// Copyright 2007, Google Inc.
+// All rights reserved.
+//
+// Redistribution and use in source and binary forms, with or without
+// modification, are permitted provided that the following conditions are
+// met:
+//
+//     * Redistributions of source code must retain the above copyright
+// notice, this list of conditions and the following disclaimer.
+//     * Redistributions in binary form must reproduce the above
+// copyright notice, this list of conditions and the following disclaimer
+// in the documentation and/or other materials provided with the
+// distribution.
+//     * Neither the name of Google Inc. nor the names of its
+// contributors may be used to endorse or promote products derived from
+// this software without specific prior written permission.
+//
+// THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+// "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+// LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+// A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+// OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+// SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+// LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+// DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+// THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+// (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+// OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+
+// Functions to canonicalize "standard" URLs, which are ones that have an
+// authority section including a host name.
+
+#include "url_canon.h"
+#include "url_canon_internal.h"
+
+namespace url_canon {
+
+namespace {
+
+template<typename CHAR, typename UCHAR>
+bool DoCanonicalizeStandardURL(const URLComponentSource<CHAR>& source,
+                               const url_parse::Parsed& parsed,
+                               CharsetConverter* query_converter,
+                               CanonOutput* output,
+                               url_parse::Parsed* new_parsed) {
+  // Scheme: this will append the colon.
+  bool success = CanonicalizeScheme(source.scheme, parsed.scheme,
+                                    output, &new_parsed->scheme);
+
+  // Authority (username, password, host, port)
+  bool have_authority;
+  if (parsed.username.is_valid() || parsed.password.is_valid() ||
+      parsed.host.is_nonempty() || parsed.port.is_valid()) {
+    have_authority = true;
+
+    // Only write the authority separators when we have a scheme.
+    if (parsed.scheme.is_valid()) {
+      output->push_back('/');
+      output->push_back('/');
+    }
+
+    // User info: the canonicalizer will handle the : and @.
+    success &= CanonicalizeUserInfo(source.username, parsed.username,
+                                    source.password, parsed.password,
+                                    output,
+                                    &new_parsed->username,
+                                    &new_parsed->password);
+
+    success &= CanonicalizeHost(source.host, parsed.host,
+                                output, &new_parsed->host);
+
+    // Host must not be empty for standard URLs.
+    if (!parsed.host.is_nonempty())
+      success = false;
+
+    // Port: the port canonicalizer will handle the colon.
+    int default_port = DefaultPortForScheme(
+        &output->data()[new_parsed->scheme.begin], new_parsed->scheme.len);
+    success &= CanonicalizePort(source.port, parsed.port, default_port,
+                                output, &new_parsed->port);
+  } else {
+    // No authority, clear the components.
+    have_authority = false;
+    new_parsed->host.reset();
+    new_parsed->username.reset();
+    new_parsed->password.reset();
+    new_parsed->port.reset();
+    success = false;  // Standard URLs must have an authority.
+  }
+
+  // Path
+  if (parsed.path.is_valid()) {
+    success &= CanonicalizePath(source.path, parsed.path,
+                                output, &new_parsed->path);
+  } else if (have_authority ||
+             parsed.query.is_valid() || parsed.ref.is_valid()) {
+    // When we have an empty path, make up a path when we have an authority
+    // or something following the path. The only time we allow an empty
+    // output path is when there is nothing else.
+    new_parsed->path = url_parse::Component(output->length(), 1);
+    output->push_back('/');
+  } else {
+    // No path at all
+    new_parsed->path.reset();
+  }
+
+  // Query
+  CanonicalizeQuery(source.query, parsed.query, query_converter,
+                    output, &new_parsed->query);
+
+  // Ref: ignore failure for this, since the page can probably still be loaded.
+  CanonicalizeRef(source.ref, parsed.ref, output, &new_parsed->ref);
+
+  return success;
+}
+
+}  // namespace
+
+
+// Returns the default port for the given canonical scheme, or PORT_UNSPECIFIED
+// if the scheme is unknown.
+int DefaultPortForScheme(const char* scheme, int scheme_len) {
+  int default_port = url_parse::PORT_UNSPECIFIED;
+  switch (scheme_len) {
+    case 4:
+      if (!strncmp(scheme, "http", scheme_len))
+        default_port = 80;
+      break;
+    case 5:
+      if (!strncmp(scheme, "https", scheme_len))
+        default_port = 443;
+      break;
+    case 3:
+      if (!strncmp(scheme, "ftp", scheme_len))
+        default_port = 21;
+      else if (!strncmp(scheme, "wss", scheme_len))
+        default_port = 443;
+      break;
+    case 6:
+      if (!strncmp(scheme, "gopher", scheme_len))
+        default_port = 70;
+      break;
+    case 2:
+      if (!strncmp(scheme, "ws", scheme_len))
+        default_port = 80;
+      break;
+  }
+  return default_port;
+}
+
+bool CanonicalizeStandardURL(const char* spec,
+                             int spec_len,
+                             const url_parse::Parsed& parsed,
+                             CharsetConverter* query_converter,
+                             CanonOutput* output,
+                             url_parse::Parsed* new_parsed) {
+  return DoCanonicalizeStandardURL<char, unsigned char>(
+      URLComponentSource<char>(spec), parsed, query_converter,
+      output, new_parsed);
+}
+
+bool CanonicalizeStandardURL(const char16* spec,
+                             int spec_len,
+                             const url_parse::Parsed& parsed,
+                             CharsetConverter* query_converter,
+                             CanonOutput* output,
+                             url_parse::Parsed* new_parsed) {
+  return DoCanonicalizeStandardURL<char16, char16>(
+      URLComponentSource<char16>(spec), parsed, query_converter,
+      output, new_parsed);
+}
+
+// It might be nice in the future to optimize this so unchanged components don't
+// need to be recanonicalized. This is especially true since the common case for
+// ReplaceComponents is removing things we don't want, like reference fragments
+// and usernames. These cases can become more efficient if we can assume the
+// rest of the URL is OK with these removed (or only the modified parts
+// recanonicalized). This would be much more complex to implement, however.
+//
+// You would also need to update DoReplaceComponents in url_util.cc which
+// relies on this re-checking everything (see the comment there for why).
+bool ReplaceStandardURL(const char* base,
+                        const url_parse::Parsed& base_parsed,
+                        const Replacements<char>& replacements,
+                        CharsetConverter* query_converter,
+                        CanonOutput* output,
+                        url_parse::Parsed* new_parsed) {
+  URLComponentSource<char> source(base);
+  url_parse::Parsed parsed(base_parsed);
+  SetupOverrideComponents(base, replacements, &source, &parsed);
+  return DoCanonicalizeStandardURL<char, unsigned char>(
+      source, parsed, query_converter, output, new_parsed);
+}
+
+// For 16-bit replacements, we turn all the replacements into UTF-8 so the
+// regular codepath can be used.
+bool ReplaceStandardURL(const char* base,
+                        const url_parse::Parsed& base_parsed,
+                        const Replacements<char16>& replacements,
+                        CharsetConverter* query_converter,
+                        CanonOutput* output,
+                        url_parse::Parsed* new_parsed) {
+  RawCanonOutput<1024> utf8;
+  URLComponentSource<char> source(base);
+  url_parse::Parsed parsed(base_parsed);
+  SetupUTF16OverrideComponents(base, replacements, &utf8, &source, &parsed);
+  return DoCanonicalizeStandardURL<char, unsigned char>(
+      source, parsed, query_converter, output, new_parsed);
+}
+
+}  // namespace url_canon

data/ext/uri_parser/url_common.h

@@ -0,0 +1,48 @@
+// Copyright 2010, Google Inc.
+// All rights reserved.
+//
+// Redistribution and use in source and binary forms, with or without
+// modification, are permitted provided that the following conditions are
+// met:
+//
+//     * Redistributions of source code must retain the above copyright
+// notice, this list of conditions and the following disclaimer.
+//     * Redistributions in binary form must reproduce the above
+// copyright notice, this list of conditions and the following disclaimer
+// in the documentation and/or other materials provided with the
+// distribution.
+//     * Neither the name of Google Inc. nor the names of its
+// contributors may be used to endorse or promote products derived from
+// this software without specific prior written permission.
+//
+// THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+// "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+// LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+// A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+// OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+// SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+// LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+// DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+// THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+// (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+// OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+
+#ifndef GOOGLEURL_SRC_URL_COMMON_H__
+#define GOOGLEURL_SRC_URL_COMMON_H__
+
+#if !defined(GURL_IMPLEMENTATION)
+#define GURL_IMPLEMENTATION 0
+#endif
+
+#if defined(WIN32) && defined(GURL_DLL)
+#if GURL_IMPLEMENTATION
+#define GURL_API __declspec(dllexport)
+#else
+#define GURL_API __declspec(dllimport)
+#endif
+#else
+#define GURL_API
+#endif
+
+#endif  // GOOGLEURL_SRC_URL_COMMON_H__
+

data/ext/uri_parser/url_file.h

@@ -0,0 +1,108 @@
+// Copyright 2007, Google Inc.
+// All rights reserved.
+//
+// Redistribution and use in source and binary forms, with or without
+// modification, are permitted provided that the following conditions are
+// met:
+//
+//     * Redistributions of source code must retain the above copyright
+// notice, this list of conditions and the following disclaimer.
+//     * Redistributions in binary form must reproduce the above
+// copyright notice, this list of conditions and the following disclaimer
+// in the documentation and/or other materials provided with the
+// distribution.
+//     * Neither the name of Google Inc. nor the names of its
+// contributors may be used to endorse or promote products derived from
+// this software without specific prior written permission.
+//
+// THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+// "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+// LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+// A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+// OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+// SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+// LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+// DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+// THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+// (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+// OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+
+// Provides shared functions used by the internals of the parser and
+// canonicalizer for file URLs. Do not use outside of these modules.
+
+#ifndef GOOGLEURL_SRC_URL_FILE_H__
+#define GOOGLEURL_SRC_URL_FILE_H__
+
+#include "url_parse_internal.h"
+
+namespace url_parse {
+
+#ifdef WIN32
+
+// We allow both "c:" and "c|" as drive identifiers.
+inline bool IsWindowsDriveSeparator(char16 ch) {
+  return ch == ':' || ch == '|';
+}
+inline bool IsWindowsDriveLetter(char16 ch) {
+  return (ch >= 'A' && ch <= 'Z') || (ch >= 'a' && ch <= 'z');
+}
+
+#endif  // WIN32
+
+// Returns the index of the next slash in the input after the given index, or
+// spec_len if the end of the input is reached.
+template<typename CHAR>
+inline int FindNextSlash(const CHAR* spec, int begin_index, int spec_len) {
+  int idx = begin_index;
+  while (idx < spec_len && !IsURLSlash(spec[idx]))
+    idx++;
+  return idx;
+}
+
+#ifdef WIN32
+
+// Returns true if the start_offset in the given spec looks like it begins a
+// drive spec, for example "c:". This function explicitly handles start_offset
+// values that are equal to or larger than the spec_len to simplify callers.
+//
+// If this returns true, the spec is guaranteed to have a valid drive letter
+// plus a colon starting at |start_offset|.
+template<typename CHAR>
+inline bool DoesBeginWindowsDriveSpec(const CHAR* spec, int start_offset,
+                                      int spec_len) {
+  int remaining_len = spec_len - start_offset;
+  if (remaining_len < 2)
+    return false;  // Not enough room.
+  if (!IsWindowsDriveLetter(spec[start_offset]))
+    return false;  // Doesn't start with a valid drive letter.
+  if (!IsWindowsDriveSeparator(spec[start_offset + 1]))
+    return false;  // Isn't followed with a drive separator.
+  return true;
+}
+
+// Returns true if the start_offset in the given text looks like it begins a
+// UNC path, for example "\\". This function explicitly handles start_offset
+// values that are equal to or larger than the spec_len to simplify callers.
+//
+// When strict_slashes is set, this function will only accept backslashes as is
+// standard for Windows. Otherwise, it will accept forward slashes as well
+// which we use for a lot of URL handling.
+template<typename CHAR>
+inline bool DoesBeginUNCPath(const CHAR* text,
+                             int start_offset,
+                             int len,
+                             bool strict_slashes) {
+  int remaining_len = len - start_offset;
+  if (remaining_len < 2)
+    return false;
+
+  if (strict_slashes)
+    return text[start_offset] == '\\' && text[start_offset + 1] == '\\';
+  return IsURLSlash(text[start_offset]) && IsURLSlash(text[start_offset + 1]);
+}
+
+#endif  // WIN32
+
+}  // namespace url_parse
+
+#endif  // GOOGLEURL_SRC_URL_FILE_H__