tor-browser

The Tor Browser
git clone https://git.dasho.dev/tor-browser.git
Log | Files | Refs | README | LICENSE

extradata.h (2438B)

      1 // © 2017 and later: Unicode, Inc. and others.
      2 // License & terms of use: http://www.unicode.org/copyright.html
      3 
      4 // extradata.h
      5 // created: 2017jun04 Markus W. Scherer
      6 // (pulled out of n2builder.cpp)
      7 
      8 // Write mappings and compositions in compact form for Normalizer2 "extra data",
      9 // the data that does not fit into the trie itself.
     10 
     11 #ifndef __EXTRADATA_H__
     12 #define __EXTRADATA_H__
     13 
     14 #include "unicode/utypes.h"
     15 
     16 #if !UCONFIG_NO_NORMALIZATION
     17 
     18 #include "unicode/errorcode.h"
     19 #include "unicode/unistr.h"
     20 #include "unicode/utf16.h"
     21 #include "hash.h"
     22 #include "norms.h"
     23 #include "toolutil.h"
     24 #include "utrie2.h"
     25 #include "uvectr32.h"
     26 
     27 U_NAMESPACE_BEGIN
     28 
     29 class ExtraData : public Norms::Enumerator {
     30 public:
     31    ExtraData(Norms &n, UBool fast);
     32 
     33    void rangeHandler(UChar32 start, UChar32 end, Norm &norm) override;
     34 
     35    UnicodeString maybeNoMappingsOnly;
     36    UnicodeString maybeNoMappingsAndCompositions;
     37    UnicodeString maybeYesCompositions;
     38    UnicodeString yesYesCompositions;
     39    UnicodeString yesNoMappingsAndCompositions;
     40    UnicodeString yesNoMappingsOnly;
     41    UnicodeString noNoMappingsCompYes;
     42    UnicodeString noNoMappingsCompBoundaryBefore;
     43    UnicodeString noNoMappingsCompNoMaybeCC;
     44    UnicodeString noNoMappingsEmpty;
     45 
     46 private:
     47    /**
     48     * Requires norm.hasMapping().
     49     * Returns the offset of the "first unit" from the beginning of the extraData for c,
     50     * not from the beginning of the dataString.
     51     * That is the same as the length of the optional data
     52     * for the raw mapping and the ccc/lccc word.
     53     */
     54    int32_t writeMapping(UChar32 c, const Norm &norm, UnicodeString &dataString);
     55    /** Returns the full offset into the dataString of the "first unit" for c. */
     56    int32_t writeNoNoMapping(UChar32 c, const Norm &norm,
     57                             UnicodeString &dataString, Hashtable &previousMappings);
     58    UBool setNoNoDelta(UChar32 c, Norm &norm) const;
     59    /** Requires norm.combinesFwd(). */
     60    void writeCompositions(UChar32 c, const Norm &norm, UnicodeString &dataString);
     61    void writeExtraData(UChar32 c, Norm &norm);
     62 
     63    UBool optimizeFast;
     64    Hashtable previousNoNoMappingsCompYes;  // If constructed in runtime code, pass in UErrorCode.
     65    Hashtable previousNoNoMappingsCompBoundaryBefore;
     66    Hashtable previousNoNoMappingsCompNoMaybeCC;
     67    Hashtable previousNoNoMappingsEmpty;
     68 };
     69 
     70 U_NAMESPACE_END
     71 
     72 #endif // #if !UCONFIG_NO_NORMALIZATION
     73 
     74 #endif  // __EXTRADATA_H__