tvl-depot/absl/strings/strip.cc
Abseil Team cf6ab6bb2b Changes imported from Abseil "staging" branch:
- b527a3e4b36b644ac424e3c525b1cd393f6f6c40 Fix some typos in the usage examples by Jorg Brown <jorg@google.com>
  - 82be4a9adf3bb0ddafc0d46274969c99afffe870 Fix typo in optional.h comment. by Abseil Team <absl-team@google.com>
  - d6ee63bf8fc51fba074c23b33cebc28c808d7f07 Remove internal-only identifiers from code. by Daniel Katz <katzdm@google.com>
  - f9c3ad2f0d73f53b21603638af8b4bed636e79f4 Use easier understandable names for absl::StartsWith and ... by Abseil Team <absl-team@google.com>
  - 7c16c14fefee89c927b8789d6043c4691bcffc9b Add -Wno-missing-prototypes back to the LLVM copts. by Derek Mauro <dmauro@google.com>
  - 2f4b7d2e50c7023240242f1e15db60ccd7e8768d IWYU | absl/strings by Juemin Yang <jueminyang@google.com>
  - a99cbcc1daa34a2d6a2bb26de275e05173cc77e9 IWYU | absl/type by Juemin Yang <jueminyang@google.com>
  - 12e1146d0fc76c071d7e0ebaabb62f0a984fae66 Use LLVM_FLAGS and LLVM_TEST_FLAGS when --compiler=llvm. by Derek Mauro <dmauro@google.com>
  - cd6bea616abda558d0bace5bd77455662a233688 IWYU | absl/debugging by Juemin Yang <jueminyang@google.com>
  - d9a7382e59d46a8581b6b7a31cd5a48bb89326e9 IWYU | absl/synchronization by Juemin Yang <jueminyang@google.com>
  - 07ec7d6d5a4a666f4183c5d0ed9c342baa7b24bc IWYU | absl/numeric by Juemin Yang <jueminyang@google.com>
  - 12bfe40051f4270f8707e191af5652f83f2f750c Remove the RoundTrip{Float,Double}ToBuffer routines from ... by Jorg Brown <jorg@google.com>
  - eeb4fd67c9d97f66cb9475c3c5e51ab132f1c810 Adds conversion functions for converting between absl/tim... by Greg Miller <jgm@google.com>
  - 59a2108d05d4ea85dc5cc11e49b2cd2335d4295a Change Substitute to use %.6g formatting rather than 15/1... by Jorg Brown <jorg@google.com>
  - 394becb48e0fcd161642cdaac5120d32567e0ef8 IWYU | absl/meta by Juemin Yang <jueminyang@google.com>
  - 1e5da6e8da336699b2469dcf6dda025b9b0ec4c9 Rewrite atomic_hook.h to not use std::atomic<T*> under Wi... by Greg Falcon <gfalcon@google.com>

GitOrigin-RevId: b527a3e4b36b644ac424e3c525b1cd393f6f6c40
Change-Id: I14e331d91c956ef045ac7927091a9f179716de0c
2017-09-25 10:12:29 -04:00

268 lines
8.4 KiB
C++

// Copyright 2017 The Abseil Authors.
//
// Licensed under the Apache License, Version 2.0 (the "License");
// you may not use this file except in compliance with the License.
// You may obtain a copy of the License at
//
// http://www.apache.org/licenses/LICENSE-2.0
//
// Unless required by applicable law or agreed to in writing, software
// distributed under the License is distributed on an "AS IS" BASIS,
// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
// See the License for the specific language governing permissions and
// limitations under the License.
// This file contains functions that remove a defined part from the std::string,
// i.e., strip the std::string.
#include "absl/strings/strip.h"
#include <algorithm>
#include <cstring>
#include <string>
#include "absl/strings/ascii.h"
#include "absl/strings/string_view.h"
// ----------------------------------------------------------------------
// ReplaceCharacters
// Replaces any occurrence of the character 'remove' (or the characters
// in 'remove') with the character 'replace_with'.
// ----------------------------------------------------------------------
void ReplaceCharacters(char* str, size_t len, absl::string_view remove,
char replace_with) {
for (char* end = str + len; str != end; ++str) {
if (remove.find(*str) != absl::string_view::npos) {
*str = replace_with;
}
}
}
void ReplaceCharacters(std::string* s, absl::string_view remove, char replace_with) {
for (char& ch : *s) {
if (remove.find(ch) != absl::string_view::npos) {
ch = replace_with;
}
}
}
bool StripTrailingNewline(std::string* s) {
if (!s->empty() && (*s)[s->size() - 1] == '\n') {
if (s->size() > 1 && (*s)[s->size() - 2] == '\r')
s->resize(s->size() - 2);
else
s->resize(s->size() - 1);
return true;
}
return false;
}
// ----------------------------------------------------------------------
// Misc. stripping routines
// ----------------------------------------------------------------------
void StripCurlyBraces(std::string* s) {
return StripBrackets('{', '}', s);
}
void StripBrackets(char left, char right, std::string* s) {
std::string::iterator opencurly = std::find(s->begin(), s->end(), left);
while (opencurly != s->end()) {
std::string::iterator closecurly = std::find(opencurly, s->end(), right);
if (closecurly == s->end()) return;
opencurly = s->erase(opencurly, closecurly + 1);
opencurly = std::find(opencurly, s->end(), left);
}
}
void StripMarkupTags(std::string* s) {
std::string::iterator output = std::find(s->begin(), s->end(), '<');
std::string::iterator input = output;
while (input != s->end()) {
if (*input == '<') {
input = std::find(input, s->end(), '>');
if (input == s->end()) break;
++input;
} else {
*output++ = *input++;
}
}
s->resize(output - s->begin());
}
std::string OutputWithMarkupTagsStripped(const std::string& s) {
std::string result(s);
StripMarkupTags(&result);
return result;
}
ptrdiff_t TrimStringLeft(std::string* s, absl::string_view remove) {
size_t i = 0;
while (i < s->size() && memchr(remove.data(), (*s)[i], remove.size())) {
++i;
}
if (i > 0) s->erase(0, i);
return i;
}
ptrdiff_t TrimStringRight(std::string* s, absl::string_view remove) {
size_t i = s->size(), trimmed = 0;
while (i > 0 && memchr(remove.data(), (*s)[i - 1], remove.size())) {
--i;
}
if (i < s->size()) {
trimmed = s->size() - i;
s->erase(i);
}
return trimmed;
}
// Unfortunately, absl::string_view does not have erase, so we've to replicate
// the implementation with remove_prefix()/remove_suffix()
ptrdiff_t TrimStringLeft(absl::string_view* s, absl::string_view remove) {
size_t i = 0;
while (i < s->size() && memchr(remove.data(), (*s)[i], remove.size())) {
++i;
}
if (i > 0) s->remove_prefix(i);
return i;
}
ptrdiff_t TrimStringRight(absl::string_view* s, absl::string_view remove) {
size_t i = s->size(), trimmed = 0;
while (i > 0 && memchr(remove.data(), (*s)[i - 1], remove.size())) {
--i;
}
if (i < s->size()) {
trimmed = s->size() - i;
s->remove_suffix(trimmed);
}
return trimmed;
}
// ----------------------------------------------------------------------
// Various removal routines
// ----------------------------------------------------------------------
ptrdiff_t strrm(char* str, char c) {
char* src;
char* dest;
for (src = dest = str; *src != '\0'; ++src)
if (*src != c) *(dest++) = *src;
*dest = '\0';
return dest - str;
}
ptrdiff_t memrm(char* str, ptrdiff_t strlen, char c) {
char* src;
char* dest;
for (src = dest = str; strlen-- > 0; ++src)
if (*src != c) *(dest++) = *src;
return dest - str;
}
ptrdiff_t strrmm(char* str, const char* chars) {
char* src;
char* dest;
for (src = dest = str; *src != '\0'; ++src) {
bool skip = false;
for (const char* c = chars; *c != '\0'; c++) {
if (*src == *c) {
skip = true;
break;
}
}
if (!skip) *(dest++) = *src;
}
*dest = '\0';
return dest - str;
}
ptrdiff_t strrmm(std::string* str, const std::string& chars) {
size_t str_len = str->length();
size_t in_index = str->find_first_of(chars);
if (in_index == std::string::npos) return str_len;
size_t out_index = in_index++;
while (in_index < str_len) {
char c = (*str)[in_index++];
if (chars.find(c) == std::string::npos) (*str)[out_index++] = c;
}
str->resize(out_index);
return out_index;
}
// ----------------------------------------------------------------------
// StripDupCharacters
// Replaces any repeated occurrence of the character 'dup_char'
// with single occurrence. e.g.,
// StripDupCharacters("a//b/c//d", '/', 0) => "a/b/c/d"
// Return the number of characters removed
// ----------------------------------------------------------------------
ptrdiff_t StripDupCharacters(std::string* s, char dup_char, ptrdiff_t start_pos) {
if (start_pos < 0) start_pos = 0;
// remove dups by compaction in-place
ptrdiff_t input_pos = start_pos; // current reader position
ptrdiff_t output_pos = start_pos; // current writer position
const ptrdiff_t input_end = s->size();
while (input_pos < input_end) {
// keep current character
const char curr_char = (*s)[input_pos];
if (output_pos != input_pos) // must copy
(*s)[output_pos] = curr_char;
++input_pos;
++output_pos;
if (curr_char == dup_char) { // skip subsequent dups
while ((input_pos < input_end) && ((*s)[input_pos] == dup_char))
++input_pos;
}
}
const ptrdiff_t num_deleted = input_pos - output_pos;
s->resize(s->size() - num_deleted);
return num_deleted;
}
// ----------------------------------------------------------------------
// TrimRunsInString
// Removes leading and trailing runs, and collapses middle
// runs of a set of characters into a single character (the
// first one specified in 'remove'). Useful for collapsing
// runs of repeated delimiters, whitespace, etc. E.g.,
// TrimRunsInString(&s, " :,()") removes leading and trailing
// delimiter chars and collapses and converts internal runs
// of delimiters to single ' ' characters, so, for example,
// " a:(b):c " -> "a b c"
// "first,last::(area)phone, ::zip" -> "first last area phone zip"
// ----------------------------------------------------------------------
void TrimRunsInString(std::string* s, absl::string_view remove) {
std::string::iterator dest = s->begin();
std::string::iterator src_end = s->end();
for (std::string::iterator src = s->begin(); src != src_end;) {
if (remove.find(*src) == absl::string_view::npos) {
*(dest++) = *(src++);
} else {
// Skip to the end of this run of chars that are in 'remove'.
for (++src; src != src_end; ++src) {
if (remove.find(*src) == absl::string_view::npos) {
if (dest != s->begin()) {
// This is an internal run; collapse it.
*(dest++) = remove[0];
}
*(dest++) = *(src++);
break;
}
}
}
}
s->erase(dest, src_end);
}
// ----------------------------------------------------------------------
// RemoveNullsInString
// Removes any internal \0 characters from the std::string.
// ----------------------------------------------------------------------
void RemoveNullsInString(std::string* s) {
s->erase(std::remove(s->begin(), s->end(), '\0'), s->end());
}