Annotation of parser3/src/include/pa_string.h, revision 1.144.2.5

1.41      paf         1: /** @file
1.43      paf         2:        Parser: string class decl.
                      3: 
1.144     paf         4:        Copyright (c) 2001, 2003 ArtLebedev Group (http://www.artlebedev.com)
1.124     paf         5:        Author: Alexandr Petrosian <paf@design.ru> (http://paf.design.ru)
1.1       paf         6: */
                      7: 
                      8: #ifndef PA_STRING_H
                      9: #define PA_STRING_H
1.140     paf        10: 
1.144.2.5! paf        11: static const char* IDENT_STRING_H="$Date: 2003/01/24 08:19:09 $";
1.1       paf        12: 
1.15      paf        13: #include "pa_pool.h"
1.4       paf        14: #include "pa_types.h"
1.144.2.2  paf        15: #include "pa_array.h"
1.4       paf        16: 
1.144.2.5! paf        17: // forwards
        !            18: 
        !            19: class String;
        !            20: 
        !            21: typedef object_ptr<String> StringPtr;
        !            22: typedef object_ptr<const String> ConstStringPtr;
        !            23: const ConstStringPtr ConstStringPtrZero(0);
        !            24: 
        !            25: class Table;
        !            26: typedef object_ptr<Table> TablePtr;
        !            27: class SQL_Connection;
        !            28: class Dictionary;
        !            29: 
        !            30: // helpers
        !            31: 
1.9       paf        32: #ifndef NO_STRING_ORIGIN
1.33      paf        33: #      define STRING_APPEND_PARAMS \
                     34:                const char *src, size_t size,  \
1.119     paf        35:                uchar lang, \
1.33      paf        36:                const char *file, uint line
1.54      paf        37: /// appends piece to String  @see String::real_append
                     38: #      define APPEND(src, size, lang, file, line) \
                     39:                real_append(src, size, lang, file, line)
1.9       paf        40: #else
1.33      paf        41: #      define STRING_APPEND_PARAMS \
                     42:                const char *src, \
                     43:                size_t size, \
1.119     paf        44:                uchar lang
1.54      paf        45: /// appends piece to String  @see String::real_append
                     46: #      define APPEND(src, size, lang, file, line) \
                     47:                real_append(src, size, lang)
                     48: #endif
1.83      paf        49: /// appends clean piece to String @see String::real_append
1.54      paf        50: #define APPEND_CLEAN(src, size, file, line) \
1.62      paf        51:        APPEND(src, size, String::UL_CLEAN, file, line)
1.83      paf        52: /// appends piece to String as-is @see String::real_append
                     53: #define APPEND_AS_IS(src, size, file, line) \
                     54:        APPEND(src, size, String::UL_AS_IS, file, line)
1.48      paf        55: /// appends tainted piece to String  @see String::real_append
1.54      paf        56: #define APPEND_TAINTED(src, size, file, line) \
1.62      paf        57:        APPEND(src, size, String::UL_TAINTED, file, line)
1.48      paf        58: /// handy: appends const char* piece to String  @see String::real_append
1.83      paf        59: #define        APPEND_CONST(src) APPEND_AS_IS(src, 0, 0, 0)
1.9       paf        60: 
1.144.2.2  paf        61: struct String_fragment { 
1.144.2.5! paf        62:        uchar/*Untaint_lang*/ lang; ///< untaint flag, later untaint language
1.144.2.2  paf        63:        const char *ptr;  ///< pointer to the start
                     64:        size_t size;  ///< length
1.41      paf        65: 
1.144.2.2  paf        66: #ifndef NO_STRING_ORIGIN
                     67:        /// all String pieces hold information of where they come from
                     68:        struct Origin {
                     69:                const char *file;  ///< macros file name | load file name | sql query text
                     70:                uint line; ///< file line no | record no
                     71:        } origin;  ///< origin
                     72: #endif
                     73: };
                     74: 
                     75: /** 
                     76:        String which knows the language of all it's fragments.
1.41      paf        77: 
                     78:        All pieces remember 
                     79:        - the file and its line they are from [can be turned off by NO_STRING_ORIGIN]
                     80:        - whether they are tainted or not, 
                     81:          and the language which should be used to detaint them
                     82: */
1.48      paf        83: 
1.144.2.2  paf        84: class String: public Array<String_fragment> {
                     85:        size_t fsize;
1.144.2.5! paf        86:        int count() const; // hiding fro accidental use instead of size()
1.144.2.2  paf        87: public:
1.1       paf        88: 
1.106     parser     89:        /** piece is tainted or not. the language to use when detaint
                     90:                remember to change String_Untaint_lang_name @ untaint.C along
                     91:        */
1.27      paf        92:        enum Untaint_lang {
1.72      paf        93:                UL_UNSPECIFIED=0, ///< zero value handy for hash lookup @see untaint_lang_name2enum
1.122     paf        94:                // these two must go before others, there are checks for >UL_AS_IS
1.62      paf        95:                UL_CLEAN, ///< clean
1.122     paf        96:                UL_AS_IS,     ///< leave all characters intact
                     97: 
1.48      paf        98:                UL_PASS_APPENDED,
1.41      paf        99:                        /**<
                    100:                                leave language built into string being appended.
                    101:                                just a flag, that value not stored
                    102:                        */
1.122     paf       103:                UL_TAINTED,  ///< tainted, untaint language as assigned later 
                    104:                // untaint languages. assigned by ^untaint[lang]{...}
1.103     parser    105:                UL_FILE_SPEC, ///< file specification
1.74      paf       106:                UL_HTTP_HEADER,    ///< text in HTTP response header
                    107:                UL_MAIL_HEADER,    ///< text in mail header
1.48      paf       108:                UL_URI,       ///< text in uri
                    109:                UL_TABLE,     ///< ^table:set body
                    110:                UL_SQL,       ///< ^table:sql body
                    111:                UL_JS,        ///< JavaScript code
1.102     parser    112:                UL_XML,         ///< ^dom:set xml
1.48      paf       113:                UL_HTML,      ///< HTML code (for editing)
1.119     paf       114:                UL_OPTIMIZE_BIT = 0x80  ///< flag, requiring cstr whitespace optimization
1.27      paf       115:        };
                    116: 
1.8       paf       117: public:
1.144.2.5! paf       118:        static const ConstStringPtr undefined_origin;
1.8       paf       119: 
1.144.2.1  paf       120:        //static String& OnPool(Pool& apool, const char *local_src=0, size_t src_size=0, bool tainted=false);
                    121:        String(const char *src=0, size_t src_size=0, bool tainted=false);
1.14      paf       122:        String(const String& src);
1.144.2.2  paf       123:        bool is_empty() const { return !size(); }
                    124:        size_t size() const { return fsize; }
1.69      paf       125:        /// convert to C string. if 'lang' known, forcing 'lang' to it
1.144.2.5! paf       126:        CharPtr cstr(Untaint_lang lang=UL_AS_IS, 
1.82      paf       127:                SQL_Connection *connection=0,
1.133     paf       128:                Charset *cstr_charset=0, const char *cstr_charset_name=0) const {
1.88      parser    129: 
1.144.2.5! paf       130:                CharPtr result(new char[cstr_bufsize(lang, connection, cstr_charset)]);
        !           131:                char *eol=store_to(result.get(), lang, connection, cstr_charset, cstr_charset_name);
1.50      paf       132:                *eol=0;
                    133:                return result;
                    134:        }
1.117     paf       135:        char *cstr_debug_origins() const;
1.108     parser    136:        /// puts pieces to buf
1.144.2.5! paf       137:        void serialize(Pool& pool, size_t prolog_size,  char *& buf, size_t& buf_size) const;
1.108     parser    138:        /// appends pieces from buf to self
1.132     paf       139:        bool deserialize(size_t prolog_size, void *buf, size_t buf_size, const char *file);
1.46      paf       140:        /** append fragment
1.83      paf       141:                @see APPEND_AS_IS, APPEND_CLEAN, APPEND_TAINTED, APPEND_CONST
1.46      paf       142:        */
1.9       paf       143:        String& real_append(STRING_APPEND_PARAMS);
1.44      paf       144:        /// @return <0 ==0 or >0 depending on comparison result
1.62      paf       145:        int cmp (int& partial, const String& src, 
1.72      paf       146:                size_t this_offset=0, Untaint_lang lang=UL_UNSPECIFIED) const;
1.56      paf       147:        bool operator < (const String& src) const {     int p; return cmp(p, src)<0; }
                    148:        bool operator > (const String& src) const {     int p; return cmp(p, src)>0; }
                    149:        bool operator <= (const String& src) const { int p; return cmp(p, src)<=0; }
                    150:        bool operator >= (const String& src) const { int p; return cmp(p, src)>=0; }
1.26      paf       151:        bool operator == (const String& src) const {
                    152:                if(size()!=src.size()) // can speed up in trivial case
                    153:                        return false;
1.56      paf       154:                int p; return cmp(p, src)==0;
1.26      paf       155:        }
1.56      paf       156:        bool operator != (const String& src) const { int p; return cmp(p, src)!=0; }
1.26      paf       157: 
1.50      paf       158:        /**
                    159:                 @param partial 
                    160:                        returns partial match status. 
1.51      paf       161:                        - -1: strings too different
                    162:                        -  0: full match
                    163:                        -  1: means @c this starts @c src
                    164:                        -  2: means @src starts @this
1.50      paf       165:        */
1.61      paf       166:        int cmp(int& partial, const char* src_ptr, size_t src_size=0, 
1.72      paf       167:                size_t this_offset=0, Untaint_lang lang=UL_UNSPECIFIED) const;
1.139     paf       168:        /// this starts with src
                    169:        bool starts_with(const char* src_ptr, size_t src_size=0) const {
                    170:                int p; cmp(p, src_ptr, src_size);
                    171:                return p==0 || p==2;
                    172:        }
1.50      paf       173:        bool operator == (const char* src_ptr) const { 
                    174:                size_t src_size=src_ptr?strlen(src_ptr):0;
                    175:                if(size() != src_size)
                    176:                        return false;
                    177:                int partial; // unused
1.56      paf       178:                return cmp(partial, src_ptr, src_size)==0; 
1.50      paf       179:        }
1.80      paf       180:        bool operator != (const char* src_ptr) const { 
                    181:                int partial; // unused
                    182:                return cmp(partial, src_ptr, 0)!=0; 
                    183:        }
                    184: 
1.119     paf       185:        String& append(const String& src, uchar lang, bool forced=false);
1.76      paf       186:        String& operator << (const String& src) { return append(src, UL_PASS_APPENDED); }
                    187:        String& operator << (const char *src) { return APPEND_CONST(src); }
1.8       paf       188: 
1.41      paf       189:        /// simple hash code of string. used by Hash
1.14      paf       190:        uint hash_code() const;
1.100     parser    191: 
1.142     paf       192:        /// extracts first char of a string, if any
                    193:        char first_char() const {
1.144.2.2  paf       194:                return is_empty()?0:*felements[0].ptr;
1.142     paf       195:        }
1.54      paf       196: 
                    197:        /// extracts [start, finish) piece of string
1.144.2.4  paf       198:        StringPtr mid(size_t start, size_t finish) const;
1.55      paf       199: 
1.59      paf       200:        /// @return position of substr in string, -1 means "not found" [String version]
1.62      paf       201:        int pos(const String& substr, 
1.112     paf       202:                int this_offset=0, Untaint_lang lang=UL_UNSPECIFIED) const;
1.59      paf       203:        /// @return position of substr in string, -1 means "not found" [const char* version]
1.97      parser    204:        int pos(const char *substr, size_t substr_size=0, 
1.112     paf       205:                int this_offset=0, Untaint_lang lang=UL_UNSPECIFIED) const;
1.62      paf       206: 
1.144.2.5! paf       207:        void split(Array<ConstStringPtr>& result, 
1.62      paf       208:                size_t *pos_after_ref, 
                    209:                const char *delim, size_t delim_size, 
1.86      paf       210:                Untaint_lang lang=UL_UNSPECIFIED, int limit=-1) const;
1.144.2.5! paf       211:        void split(Array<ConstStringPtr>& result, 
1.62      paf       212:                size_t *pos_after_ref, 
                    213:                const String& delim, 
1.86      paf       214:                Untaint_lang lang=UL_UNSPECIFIED, int limit=-1) const;
1.62      paf       215: 
1.144.2.5! paf       216:        typedef void (*Row_action)(Table& table, Array<ConstStringPtr> *row, 
1.136     paf       217:                int prestart, int prefinish, 
                    218:                int poststart, int postfinish,
1.68      paf       219:                void *info);
1.87      parser    220:        /**
1.144.2.5! paf       221:                @return table of found items, if any.
1.87      parser    222:                table format is defined and fixed[can be used by others]: 
                    223:                @verbatim
                    224:                        prematch/match/postmatch/1/2/3/...
                    225:                @endverbatim
                    226:        */
1.144.2.5! paf       227:        TablePtr match(Charset& source_charset,
        !           228:                ConstStringPtr aorigin,         
1.64      paf       229:                const String& regexp, 
1.144.2.5! paf       230:                ConstStringPtr options,
1.99      parser    231:                Row_action row_action, void *info,
                    232:                bool *was_global=0) const;
1.87      parser    233:        enum Change_case_kind {
                    234:                CC_UPPER,
                    235:                CC_LOWER
                    236:        };
1.144.2.5! paf       237:        StringPtr change_case(Pool& pool, Charset& source_charset,
1.87      parser    238:                Change_case_kind kind) const;
1.144.2.5! paf       239:        StringPtr replace(Pool& pool, Dictionary& dict) const;
1.96      parser    240:        double as_double() const;
                    241:        int as_int() const;
1.137     paf       242: 
1.144.2.5! paf       243:        StringPtr join_chains(Pool& pool) const;
1.13      paf       244: 
1.41      paf       245: #ifndef NO_STRING_ORIGIN
                    246:        /// origin of string. calculated by first row
1.144.2.2  paf       247:        const String_fragment::Origin& origin() const;
1.41      paf       248: #endif
1.8       paf       249: 
1.129     paf       250: private:
1.10      paf       251: 
1.125     paf       252:        Untaint_lang lang_of(size_t offset) const;
1.79      paf       253: 
1.144.2.5! paf       254:        size_t cstr_bufsize(Untaint_lang lang=UL_UNSPECIFIED,
        !           255:                SQL_Connection *connection=0,
        !           256:                Charset *buf_charset=0) const;
1.79      paf       257:        /// convert to C string, store to 'dest' which must be big enough for proper untaint
1.82      paf       258:        char *store_to(char *dest, Untaint_lang lang=UL_UNSPECIFIED, 
                    259:                SQL_Connection *connection=0,
1.133     paf       260:                Charset *store_to_charset=0,
                    261:                const char *store_to_charset_name=0) const;
1.144.2.2  paf       262: /*
1.144.2.1  paf       263:        void join_chain(
1.130     paf       264:                                           const Chunk*& achunk, const Chunk::Row*& arow, uint& acountdown, 
1.119     paf       265:                                           uchar& joined_lang, const char *& joined_ptr, size_t& joined_size) const;
1.144.2.2  paf       266: */
1.7       paf       267: private: //disabled
                    268: 
1.12      paf       269:        String& operator = (const String&) { return *this; }
1.7       paf       270: 
1.1       paf       271: };
1.144.2.5! paf       272: 
        !           273: // statics
        !           274: 
        !           275: const ConstStringPtr String::undefined_origin=ConstStringPtr(0);
        !           276: 
        !           277: // helpers
        !           278: 
        !           279: #define STRING_PREPARED_FOREACH_FRAGMENT(code) \
        !           280:        for(; fragment<last; fragment++) { \
        !           281:                code; \
        !           282:        }
        !           283: 
        !           284: #define STRING_FOREACH_FRAGMENT(code) \
        !           285:        { \
        !           286:                element_type *last = felements+fused; \
        !           287:                element_type *fragment = felements; \
        !           288:                STRING_PREPARED_FOREACH_FRAGMENT(code) \
        !           289:        }
1.1       paf       290: 
                    291: #endif

E-mail: