mirror of
https://github.com/apple/swift.git
synced 2025-12-14 20:36:38 +01:00
252 lines
7.2 KiB
C++
252 lines
7.2 KiB
C++
//===--- StringExtras.cpp - String Utilities ------------------------------===//
|
|
//
|
|
// This source file is part of the Swift.org open source project
|
|
//
|
|
// Copyright (c) 2014 - 2015 Apple Inc. and the Swift project authors
|
|
// Licensed under Apache License v2.0 with Runtime Library Exception
|
|
//
|
|
// See http://swift.org/LICENSE.txt for license information
|
|
// See http://swift.org/CONTRIBUTORS.txt for the list of Swift project authors
|
|
//
|
|
//===----------------------------------------------------------------------===//
|
|
//
|
|
// This file implements utilities for working with words and camelCase
|
|
// names.
|
|
//
|
|
//===----------------------------------------------------------------------===//
|
|
#include "swift/Basic/StringExtras.h"
|
|
#include "swift/Basic/Optional.h"
|
|
#include "clang/Basic/CharInfo.h"
|
|
#include "llvm/ADT/SmallString.h"
|
|
#include "llvm/ADT/SmallVector.h"
|
|
#include <algorithm>
|
|
|
|
using namespace swift;
|
|
using namespace camel_case;
|
|
|
|
PrepositionKind swift::getPrepositionKind(StringRef word) {
|
|
#define DIRECTIONAL_PREPOSITION(Word) \
|
|
if (word.equals_lower(#Word)) \
|
|
return PK_Directional;
|
|
#define PREPOSITION(Word) \
|
|
if (word.equals_lower(#Word)) \
|
|
return PK_Nondirectional;
|
|
#include "Prepositions.def"
|
|
|
|
return PK_None;
|
|
}
|
|
|
|
bool swift::isLinkingVerb(StringRef word) {
|
|
#define LINKING_VERB(Word) \
|
|
if (word.equals_lower(#Word)) \
|
|
return true;
|
|
#include "LinkingVerbs.def"
|
|
|
|
return false;
|
|
}
|
|
|
|
void WordIterator::computeNextPosition() const {
|
|
assert(Position < String.size() && "Already at end of string");
|
|
|
|
unsigned i = Position, n = String.size();
|
|
|
|
// Treat _ as a word on its own. Don't coalesce.
|
|
if (String[i] == '_') {
|
|
NextPosition = i + 1;
|
|
NextPositionValid = true;
|
|
return;
|
|
}
|
|
|
|
// Skip over any uppercase letters at the beginning of the word.
|
|
while (i < n && clang::isUppercase(String[i]))
|
|
++i;
|
|
|
|
// If there was more than one uppercase letter, this is an
|
|
// acronym.
|
|
if (i - Position > 1) {
|
|
// If we hit the end of the string, that's it. Otherwise, this
|
|
// word ends before the last uppercase letter if the next word is alphabetic
|
|
// (URL_Loader) or after the last uppercase letter if it's not (UTF_8).
|
|
NextPosition = (i == n || !clang::isLowercase(String[i])) ? i : i-1;
|
|
NextPositionValid = true;
|
|
return;
|
|
}
|
|
|
|
// Skip non-uppercase letters.
|
|
while (i < n && !clang::isUppercase(String[i]) && String[i] != '_')
|
|
++i;
|
|
|
|
NextPosition = i;
|
|
NextPositionValid = true;
|
|
}
|
|
|
|
void WordIterator::computePrevPosition() const {
|
|
assert(Position > 0 && "Already at beginning of string");
|
|
|
|
unsigned i = Position;
|
|
|
|
// While we see non-uppercase letters, keep moving back.
|
|
while (i > 0 && !clang::isUppercase(String[i-1]) && String[i-1] != '_')
|
|
--i;
|
|
|
|
// If we found any lowercase letters, this was a normal camel case
|
|
// word (not an acronym).
|
|
if (i < Position) {
|
|
// If we hit the beginning of the string, that's it. Otherwise, this
|
|
// word starts with an uppercase letter if the next word is alphabetic
|
|
// (URL_Loader) or after the last uppercase letter if it's not (UTF_8).
|
|
PrevPosition = i;
|
|
if (i != 0 && clang::isLowercase(String[i]) && String[i-1] != '_')
|
|
--PrevPosition;
|
|
PrevPositionValid = true;
|
|
return;
|
|
}
|
|
|
|
// Treat _ as a word on its own. Don't coalesce.
|
|
if (String[i-1] == '_') {
|
|
PrevPosition = i - 1;
|
|
PrevPositionValid = true;
|
|
return;
|
|
}
|
|
|
|
// There were no lowercase letters, so this is an acronym. Keep
|
|
// skipping uppercase letters.
|
|
while (i > 0 && clang::isUppercase(String[i-1]))
|
|
--i;
|
|
|
|
PrevPosition = i;
|
|
PrevPositionValid = true;
|
|
}
|
|
|
|
StringRef camel_case::getFirstWord(StringRef string) {
|
|
if (string.empty())
|
|
return "";
|
|
|
|
return *WordIterator(string, 0);
|
|
}
|
|
|
|
StringRef camel_case::getLastWord(StringRef string) {
|
|
if (string.empty())
|
|
return "";
|
|
|
|
return *--WordIterator(string, string.size());
|
|
}
|
|
|
|
bool camel_case::sameWordIgnoreFirstCase(StringRef word1, StringRef word2) {
|
|
if (word1.size() != word2.size())
|
|
return false;
|
|
|
|
if (clang::toLowercase(word1[0]) != clang::toLowercase(word2[0]))
|
|
return false;
|
|
|
|
return word1.substr(1) == word2.substr(1);
|
|
}
|
|
|
|
bool camel_case::startsWithIgnoreFirstCase(StringRef word1, StringRef word2) {
|
|
if (word1.size() < word2.size())
|
|
return false;
|
|
|
|
if (clang::toLowercase(word1[0]) != clang::toLowercase(word2[0]))
|
|
return false;
|
|
|
|
return word1.substr(1) == word2.substr(1, word1.size() - 1);
|
|
}
|
|
|
|
StringRef camel_case::toLowercaseWord(StringRef string,
|
|
SmallVectorImpl<char> &scratch) {
|
|
if (string.empty())
|
|
return string;
|
|
|
|
// Already lowercase.
|
|
if (!clang::isUppercase(string[0]))
|
|
return string;
|
|
|
|
// Acronym doesn't get lowercased.
|
|
if (string.size() > 1 && clang::isUppercase(string[1]))
|
|
return string;
|
|
|
|
// Lowercase the first letter, append the rest.
|
|
scratch.clear();
|
|
scratch.push_back(clang::toLowercase(string[0]));
|
|
scratch.append(string.begin() + 1, string.end());
|
|
return StringRef(scratch.data(), scratch.size());
|
|
}
|
|
|
|
StringRef camel_case::toSentencecase(StringRef string,
|
|
SmallVectorImpl<char> &scratch) {
|
|
if (string.empty())
|
|
return string;
|
|
|
|
// Can't be uppercased.
|
|
if (!clang::isLowercase(string[0]))
|
|
return string;
|
|
|
|
// Uppercase the first letter, append the rest.
|
|
scratch.clear();
|
|
scratch.push_back(clang::toUppercase(string[0]));
|
|
scratch.append(string.begin() + 1, string.end());
|
|
return StringRef(scratch.data(), scratch.size());
|
|
}
|
|
|
|
StringRef camel_case::dropPrefix(StringRef string) {
|
|
|
|
unsigned firstLower = 0, n = string.size();
|
|
|
|
if (n < 4)
|
|
return string;
|
|
|
|
for (; firstLower < n; ++firstLower) {
|
|
if (!clang::isUppercase(string[firstLower]))
|
|
break;
|
|
}
|
|
|
|
if (firstLower == n)
|
|
return string;
|
|
|
|
if (firstLower >= 3 && firstLower <= 4)
|
|
return string.substr(firstLower - 1);
|
|
|
|
return string;
|
|
}
|
|
|
|
StringRef camel_case::appendSentenceCase(SmallVectorImpl<char> &buffer,
|
|
StringRef string) {
|
|
// Trivial case: empty string.
|
|
if (string.empty())
|
|
return StringRef(buffer.data(), buffer.size());
|
|
|
|
// Uppercase the first letter, append the rest.
|
|
buffer.push_back(clang::toUppercase(string[0]));
|
|
buffer.append(string.begin() + 1, string.end());
|
|
return StringRef(buffer.data(), buffer.size());
|
|
}
|
|
|
|
size_t camel_case::findWord(StringRef string, StringRef word) {
|
|
assert(!word.empty());
|
|
assert(clang::isUppercase(word[0]));
|
|
|
|
// Scan forward until we find the word as a complete word.
|
|
size_t startingIndex = 0;
|
|
while (true) {
|
|
size_t index = string.find(word, startingIndex);
|
|
if (index == StringRef::npos)
|
|
return StringRef::npos;
|
|
|
|
// If any of the following checks fail, we want to start searching
|
|
// past the end of the match. (This assumes that the word doesn't
|
|
// end with a prefix of itself, e.g. "LikeableLike".)
|
|
startingIndex = index + word.size();
|
|
|
|
// We assume that we don't have to check if the match starts a new
|
|
// word in the string.
|
|
|
|
// If we find the word, check whether it's a valid match.
|
|
StringRef suffix = string.substr(index);
|
|
if (!suffix.empty() && clang::isLowercase(suffix[0]))
|
|
continue;
|
|
|
|
return index;
|
|
}
|
|
}
|
|
|