Home | History | Annotate | Download | only in strings 
      1 // Copyright (c) 2011 The Chromium Authors. All rights reserved.
      2 // Use of this source code is governed by a BSD-style license that can be
      3 // found in the LICENSE file.
      4 
      5 #ifndef BASE_STRINGS_UTF_OFFSET_STRING_CONVERSIONS_H_
      6 #define BASE_STRINGS_UTF_OFFSET_STRING_CONVERSIONS_H_
      7 
      8 #include <string>
      9 #include <vector>
     10 
     11 #include "base/base_export.h"
     12 #include "base/strings/string16.h"
     13 #include "base/strings/string_piece.h"
     14 
     15 namespace base {
     16 
     17 // Like the conversions in utf_string_conversions.h, but also takes one or more
     18 // offsets (|offset[s]_for_adjustment|) into the source strings, each offset
     19 // will be adjusted to point at the same logical place in the result strings.
     20 // If this isn't possible because an offset points past the end of the source
     21 // strings or into the middle of a multibyte sequence, the offending offset will
     22 // be set to string16::npos. |offset[s]_for_adjustment| may be NULL.
     23 BASE_EXPORT bool UTF8ToUTF16AndAdjustOffset(const char* src,
     24                                             size_t src_len,
     25                                             string16* output,
     26                                             size_t* offset_for_adjustment);
     27 BASE_EXPORT bool UTF8ToUTF16AndAdjustOffsets(
     28     const char* src,
     29     size_t src_len,
     30     string16* output,
     31     std::vector<size_t>* offsets_for_adjustment);
     32 
     33 BASE_EXPORT string16 UTF8ToUTF16AndAdjustOffset(const base::StringPiece& utf8,
     34                                                 size_t* offset_for_adjustment);
     35 BASE_EXPORT string16 UTF8ToUTF16AndAdjustOffsets(
     36     const base::StringPiece& utf8,
     37     std::vector<size_t>* offsets_for_adjustment);
     38 
     39 BASE_EXPORT std::string UTF16ToUTF8AndAdjustOffset(
     40     const base::StringPiece16& utf16,
     41     size_t* offset_for_adjustment);
     42 BASE_EXPORT std::string UTF16ToUTF8AndAdjustOffsets(
     43     const base::StringPiece16& utf16,
     44     std::vector<size_t>* offsets_for_adjustment);
     45 
     46 // Limiting function callable by std::for_each which will replace any value
     47 // which is equal to or greater than |limit| with npos.
     48 template <typename T>
     49 struct LimitOffset {
     50   explicit LimitOffset(size_t limit)
     51     : limit_(limit) {}
     52 
     53   void operator()(size_t& offset) {
     54     if (offset >= limit_)
     55       offset = T::npos;
     56   }
     57 
     58   size_t limit_;
     59 };
     60 
     61 // Stack object which, on destruction, will update a vector of offsets based on
     62 // any supplied adjustments.  To use, declare one of these, providing the
     63 // address of the offset vector to adjust.  Then Add() any number of Adjustments
     64 // (each Adjustment gives the |original_offset| of a substring and the lengths
     65 // of the substring before and after transforming).  When the OffsetAdjuster
     66 // goes out of scope, all the offsets in the provided vector will be updated.
     67 class BASE_EXPORT OffsetAdjuster {
     68  public:
     69   struct BASE_EXPORT Adjustment {
     70     Adjustment(size_t original_offset,
     71                size_t original_length,
     72                size_t output_length);
     73 
     74     size_t original_offset;
     75     size_t original_length;
     76     size_t output_length;
     77   };
     78 
     79   explicit OffsetAdjuster(std::vector<size_t>* offsets_for_adjustment);
     80   ~OffsetAdjuster();
     81 
     82   void Add(const Adjustment& adjustment);
     83 
     84  private:
     85   void AdjustOffset(std::vector<size_t>::iterator offset);
     86 
     87   std::vector<size_t>* offsets_for_adjustment_;
     88   std::vector<Adjustment> adjustments_;
     89 };
     90 
     91 }  // namespace base
     92 
     93 #endif  // BASE_STRINGS_UTF_OFFSET_STRING_CONVERSIONS_H_
     94