Annotation of parser3/src/include/pa_string.h, revision 1.144.2.14

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: 
1.144.2.14! paf        11: static const char* IDENT_STRING_H="$Date: 2003/01/29 09:16:15 $";
1.144.2.10  paf        12: 
                     13: // includes
1.1       paf        14: 
1.15      paf        15: #include "pa_pool.h"
1.144.2.12  paf        16: #include "pa_types.h"
                     17: #include "pa_array.h"
1.4       paf        18: 
1.144.2.12  paf        19: // forwards
1.144.2.10  paf        20: 
1.144.2.12  paf        21: class String; DECLARE_OBJECT_PTR(String);
                     22: typedef object_ptr<const String> ConstStringPtr; const ConstStringPtr ConstStringPtrZero(0);
1.144.2.10  paf        23: 
                     24: // forwards
                     25: 
1.144.2.9  paf        26: class Table;
                     27: typedef object_ptr<Table> TablePtr;
1.144.2.5  paf        28: class SQL_Connection;
                     29: class Dictionary;
1.144.2.14! paf        30: struct Request_charsets;
1.144.2.5  paf        31: 
                     32: // helpers
                     33: 
1.9       paf        34: #ifndef NO_STRING_ORIGIN
1.33      paf        35: #      define STRING_APPEND_PARAMS \
                     36:                const char *src, size_t size,  \
1.144.2.7  paf        37:                String_UL lang, \
1.33      paf        38:                const char *file, uint line
1.54      paf        39: /// appends piece to String  @see String::real_append
                     40: #      define APPEND(src, size, lang, file, line) \
                     41:                real_append(src, size, lang, file, line)
1.9       paf        42: #else
1.33      paf        43: #      define STRING_APPEND_PARAMS \
                     44:                const char *src, \
                     45:                size_t size, \
1.144.2.7  paf        46:                String_UL lang
1.54      paf        47: /// appends piece to String  @see String::real_append
                     48: #      define APPEND(src, size, lang, file, line) \
                     49:                real_append(src, size, lang)
                     50: #endif
1.83      paf        51: /// appends clean piece to String @see String::real_append
1.54      paf        52: #define APPEND_CLEAN(src, size, file, line) \
1.62      paf        53:        APPEND(src, size, String::UL_CLEAN, file, line)
1.83      paf        54: /// appends piece to String as-is @see String::real_append
                     55: #define APPEND_AS_IS(src, size, file, line) \
                     56:        APPEND(src, size, String::UL_AS_IS, file, line)
1.48      paf        57: /// appends tainted piece to String  @see String::real_append
1.54      paf        58: #define APPEND_TAINTED(src, size, file, line) \
1.62      paf        59:        APPEND(src, size, String::UL_TAINTED, file, line)
1.48      paf        60: /// handy: appends const char* piece to String  @see String::real_append
1.83      paf        61: #define        APPEND_CONST(src) APPEND_AS_IS(src, 0, 0, 0)
1.9       paf        62: 
1.144.2.7  paf        63: typedef uint String_UL;
                     64: 
1.144.2.2  paf        65: struct String_fragment { 
1.144.2.7  paf        66:        String_UL/*Untaint_lang*/ lang; ///< untaint flag, later untaint language
1.144.2.2  paf        67:        const char *ptr;  ///< pointer to the start
                     68:        size_t size;  ///< length
1.41      paf        69: 
1.144.2.2  paf        70: #ifndef NO_STRING_ORIGIN
                     71:        /// all String pieces hold information of where they come from
                     72:        struct Origin {
                     73:                const char *file;  ///< macros file name | load file name | sql query text
                     74:                uint line; ///< file line no | record no
                     75:        } origin;  ///< origin
                     76: #endif
                     77: };
                     78: 
                     79: /** 
                     80:        String which knows the language of all it's fragments.
1.41      paf        81: 
                     82:        All pieces remember 
                     83:        - the file and its line they are from [can be turned off by NO_STRING_ORIGIN]
                     84:        - whether they are tainted or not, 
                     85:          and the language which should be used to detaint them
                     86: */
1.48      paf        87: 
1.144.2.2  paf        88: class String: public Array<String_fragment> {
                     89:        size_t fsize;
1.144.2.5  paf        90:        int count() const; // hiding fro accidental use instead of size()
1.144.2.2  paf        91: public:
1.1       paf        92: 
1.106     parser     93:        /** piece is tainted or not. the language to use when detaint
                     94:                remember to change String_Untaint_lang_name @ untaint.C along
                     95:        */
1.27      paf        96:        enum Untaint_lang {
1.72      paf        97:                UL_UNSPECIFIED=0, ///< zero value handy for hash lookup @see untaint_lang_name2enum
1.122     paf        98:                // these two must go before others, there are checks for >UL_AS_IS
1.62      paf        99:                UL_CLEAN, ///< clean
1.122     paf       100:                UL_AS_IS,     ///< leave all characters intact
                    101: 
1.48      paf       102:                UL_PASS_APPENDED,
1.41      paf       103:                        /**<
                    104:                                leave language built into string being appended.
                    105:                                just a flag, that value not stored
                    106:                        */
1.122     paf       107:                UL_TAINTED,  ///< tainted, untaint language as assigned later 
                    108:                // untaint languages. assigned by ^untaint[lang]{...}
1.103     parser    109:                UL_FILE_SPEC, ///< file specification
1.74      paf       110:                UL_HTTP_HEADER,    ///< text in HTTP response header
                    111:                UL_MAIL_HEADER,    ///< text in mail header
1.48      paf       112:                UL_URI,       ///< text in uri
                    113:                UL_TABLE,     ///< ^table:set body
                    114:                UL_SQL,       ///< ^table:sql body
                    115:                UL_JS,        ///< JavaScript code
1.102     parser    116:                UL_XML,         ///< ^dom:set xml
1.48      paf       117:                UL_HTML,      ///< HTML code (for editing)
1.144.2.7  paf       118:                UL_OPTIMIZE_BIT = 0x8000  ///< flag, requiring cstr whitespace optimization
1.27      paf       119:        };
                    120: 
1.8       paf       121: public:
                    122: 
1.144.2.6  paf       123:        explicit String(const char *src=0, size_t src_size=0, bool tainted=false);
1.14      paf       124:        String(const String& src);
1.144.2.2  paf       125:        bool is_empty() const { return !size(); }
                    126:        size_t size() const { return fsize; }
1.69      paf       127:        /// convert to C string. if 'lang' known, forcing 'lang' to it
1.144.2.5  paf       128:        CharPtr cstr(Untaint_lang lang=UL_AS_IS, 
1.82      paf       129:                SQL_Connection *connection=0,
1.144.2.14! paf       130:                const Request_charsets *charsets=0) const {
1.88      parser    131: 
1.144.2.12  paf       132:                CharPtr result(new char[cstr_bufsize(lang, connection, charsets)]);
                    133:                char *eol=store_to(result.get(), lang, connection, charsets);
1.50      paf       134:                *eol=0;
                    135:                return result;
                    136:        }
1.144.2.9  paf       137:        char *cstr(Pool& pool, Untaint_lang lang=UL_AS_IS, 
                    138:                SQL_Connection *connection=0,
1.144.2.14! paf       139:                const Request_charsets *charsets=0) const {
1.144.2.9  paf       140: 
1.144.2.12  paf       141:                char *result=new(pool) char[cstr_bufsize(lang, connection, charsets)];
                    142:                char *eol=store_to(result, lang, connection, charsets);
1.144.2.9  paf       143:                *eol=0;
                    144:                return result;
                    145:        }
1.144.2.12  paf       146:        char *cstr_debug_origins(Pool& pool) const;
1.108     parser    147:        /// puts pieces to buf
1.144.2.5  paf       148:        void serialize(Pool& pool, size_t prolog_size,  char *& buf, size_t& buf_size) const;
1.108     parser    149:        /// appends pieces from buf to self
1.132     paf       150:        bool deserialize(size_t prolog_size, void *buf, size_t buf_size, const char *file);
1.46      paf       151:        /** append fragment
1.83      paf       152:                @see APPEND_AS_IS, APPEND_CLEAN, APPEND_TAINTED, APPEND_CONST
1.46      paf       153:        */
1.9       paf       154:        String& real_append(STRING_APPEND_PARAMS);
1.44      paf       155:        /// @return <0 ==0 or >0 depending on comparison result
1.62      paf       156:        int cmp (int& partial, const String& src, 
1.72      paf       157:                size_t this_offset=0, Untaint_lang lang=UL_UNSPECIFIED) const;
1.56      paf       158:        bool operator < (const String& src) const {     int p; return cmp(p, src)<0; }
                    159:        bool operator > (const String& src) const {     int p; return cmp(p, src)>0; }
                    160:        bool operator <= (const String& src) const { int p; return cmp(p, src)<=0; }
                    161:        bool operator >= (const String& src) const { int p; return cmp(p, src)>=0; }
1.26      paf       162:        bool operator == (const String& src) const {
                    163:                if(size()!=src.size()) // can speed up in trivial case
                    164:                        return false;
1.56      paf       165:                int p; return cmp(p, src)==0;
1.26      paf       166:        }
1.56      paf       167:        bool operator != (const String& src) const { int p; return cmp(p, src)!=0; }
1.26      paf       168: 
1.50      paf       169:        /**
                    170:                 @param partial 
                    171:                        returns partial match status. 
1.51      paf       172:                        - -1: strings too different
                    173:                        -  0: full match
                    174:                        -  1: means @c this starts @c src
                    175:                        -  2: means @src starts @this
1.50      paf       176:        */
1.61      paf       177:        int cmp(int& partial, const char* src_ptr, size_t src_size=0, 
1.72      paf       178:                size_t this_offset=0, Untaint_lang lang=UL_UNSPECIFIED) const;
1.139     paf       179:        /// this starts with src
                    180:        bool starts_with(const char* src_ptr, size_t src_size=0) const {
                    181:                int p; cmp(p, src_ptr, src_size);
                    182:                return p==0 || p==2;
                    183:        }
1.50      paf       184:        bool operator == (const char* src_ptr) const { 
                    185:                size_t src_size=src_ptr?strlen(src_ptr):0;
                    186:                if(size() != src_size)
                    187:                        return false;
                    188:                int partial; // unused
1.56      paf       189:                return cmp(partial, src_ptr, src_size)==0; 
1.50      paf       190:        }
1.80      paf       191:        bool operator != (const char* src_ptr) const { 
                    192:                int partial; // unused
                    193:                return cmp(partial, src_ptr, 0)!=0; 
                    194:        }
                    195: 
1.144.2.11  paf       196:        String& append_to(String& dest, uchar lang, bool forced) const;
                    197:        String& append(const String& src, uchar lang, bool forced=false) { 
                    198:                return src.append_to(*this, lang, forced);
                    199:        }
1.76      paf       200:        String& operator << (const String& src) { return append(src, UL_PASS_APPENDED); }
1.144.2.8  paf       201:        String& operator << (const StringPtr src) { return append(*src, UL_PASS_APPENDED); }
1.76      paf       202:        String& operator << (const char *src) { return APPEND_CONST(src); }
1.8       paf       203: 
1.144.2.7  paf       204:        /// simple hash code of string
1.14      paf       205:        uint hash_code() const;
1.100     parser    206: 
1.142     paf       207:        /// extracts first char of a string, if any
                    208:        char first_char() const {
1.144.2.2  paf       209:                return is_empty()?0:*felements[0].ptr;
1.142     paf       210:        }
1.54      paf       211: 
                    212:        /// extracts [start, finish) piece of string
1.144.2.4  paf       213:        StringPtr mid(size_t start, size_t finish) const;
1.55      paf       214: 
1.59      paf       215:        /// @return position of substr in string, -1 means "not found" [String version]
1.62      paf       216:        int pos(const String& substr, 
1.112     paf       217:                int this_offset=0, Untaint_lang lang=UL_UNSPECIFIED) const;
1.59      paf       218:        /// @return position of substr in string, -1 means "not found" [const char* version]
1.97      parser    219:        int pos(const char *substr, size_t substr_size=0, 
1.112     paf       220:                int this_offset=0, Untaint_lang lang=UL_UNSPECIFIED) const;
1.62      paf       221: 
1.144.2.5  paf       222:        void split(Array<ConstStringPtr>& result, 
1.62      paf       223:                size_t *pos_after_ref, 
                    224:                const char *delim, size_t delim_size, 
1.86      paf       225:                Untaint_lang lang=UL_UNSPECIFIED, int limit=-1) const;
1.144.2.5  paf       226:        void split(Array<ConstStringPtr>& result, 
1.62      paf       227:                size_t *pos_after_ref, 
                    228:                const String& delim, 
1.86      paf       229:                Untaint_lang lang=UL_UNSPECIFIED, int limit=-1) const;
1.62      paf       230: 
1.144.2.9  paf       231:        typedef void (*Row_action)(Table& table, Array<ConstStringPtr> *row, 
1.136     paf       232:                int prestart, int prefinish, 
                    233:                int poststart, int postfinish,
1.144.2.9  paf       234:                void *info);
1.87      parser    235:        /**
1.144.2.5  paf       236:                @return table of found items, if any.
1.87      parser    237:                table format is defined and fixed[can be used by others]: 
                    238:                @verbatim
                    239:                        prematch/match/postmatch/1/2/3/...
                    240:                @endverbatim
                    241:        */
1.144.2.9  paf       242:        TablePtr match(Charset& source_charset,
1.144.2.5  paf       243:                ConstStringPtr aorigin,         
1.64      paf       244:                const String& regexp, 
1.144.2.5  paf       245:                ConstStringPtr options,
1.99      parser    246:                Row_action row_action, void *info,
1.144.2.9  paf       247:                bool *was_global=0) const;
1.87      parser    248:        enum Change_case_kind {
                    249:                CC_UPPER,
                    250:                CC_LOWER
                    251:        };
1.144.2.5  paf       252:        StringPtr change_case(Pool& pool, Charset& source_charset,
1.87      parser    253:                Change_case_kind kind) const;
1.144.2.5  paf       254:        StringPtr replace(Pool& pool, Dictionary& dict) const;
1.96      parser    255:        double as_double() const;
                    256:        int as_int() const;
1.137     paf       257: 
1.144.2.5  paf       258:        StringPtr join_chains(Pool& pool) const;
1.13      paf       259: 
1.41      paf       260: #ifndef NO_STRING_ORIGIN
                    261:        /// origin of string. calculated by first row
1.144.2.2  paf       262:        const String_fragment::Origin& origin() const;
1.41      paf       263: #endif
1.8       paf       264: 
1.144.2.5  paf       265:        size_t cstr_bufsize(Untaint_lang lang=UL_UNSPECIFIED,
                    266:                SQL_Connection *connection=0,
1.144.2.14! paf       267:                const Request_charsets *charsets=0) const;
1.79      paf       268:        /// convert to C string, store to 'dest' which must be big enough for proper untaint
1.82      paf       269:        char *store_to(char *dest, Untaint_lang lang=UL_UNSPECIFIED, 
                    270:                SQL_Connection *connection=0,
1.144.2.14! paf       271:                const Request_charsets *charsets=0) const;
1.144.2.2  paf       272: /*
1.144.2.1  paf       273:        void join_chain(
1.130     paf       274:                                           const Chunk*& achunk, const Chunk::Row*& arow, uint& acountdown, 
1.119     paf       275:                                           uchar& joined_lang, const char *& joined_ptr, size_t& joined_size) const;
1.144.2.2  paf       276: */
1.7       paf       277: private: //disabled
                    278: 
1.12      paf       279:        String& operator = (const String&) { return *this; }
1.7       paf       280: 
1.1       paf       281: };
1.144.2.7  paf       282: 
                    283: /// simple hash code of string. used by Hash
                    284: inline uint hash_code(const String& self) {
                    285:        return self.hash_code();
                    286: }
                    287: /// simple hash code of string. used by Hash
                    288: inline uint hash_code(ConstStringPtr self) {
                    289:        return self->hash_code();
                    290: }
1.144.2.5  paf       291: 
                    292: // helpers
                    293: 
                    294: #define STRING_PREPARED_FOREACH_FRAGMENT(code) \
                    295:        for(; fragment<last; fragment++) { \
                    296:                code; \
                    297:        }
                    298: 
                    299: #define STRING_FOREACH_FRAGMENT(code) \
                    300:        { \
                    301:                element_type *last = felements+fused; \
                    302:                element_type *fragment = felements; \
                    303:                STRING_PREPARED_FOREACH_FRAGMENT(code) \
                    304:        }
1.1       paf       305: 
                    306: #endif

E-mail: