|
|
| version 1.217, 2016/09/07 14:40:07 | version 1.236, 2025/05/25 17:44:27 |
|---|---|
| Line 1 | Line 1 |
| /** @file | /** @file |
| Parser: string class decl. | Parser: string class decl. |
| Copyright (c) 2001-2015 Art. Lebedev Studio (http://www.artlebedev.com) | Copyright (c) 2001-2024 Art. Lebedev Studio (http://www.artlebedev.com) |
| Author: Alexandr Petrosian <paf@design.ru> (http://paf.design.ru) | Authors: Konstantin Morshnev <moko@design.ru>, Alexandr Petrosian <paf@design.ru> |
| */ | */ |
| #ifndef PA_STRING_H | #ifndef PA_STRING_H |
| Line 16 | Line 16 |
| extern "C" { // cord's author forgot to do that | extern "C" { // cord's author forgot to do that |
| #define CORD_NO_IO | #define CORD_NO_IO |
| #include "cord.h" | #include "../lib/cord/include/cord.h" |
| #ifdef CORD_CAT_OPTIMIZATION | #ifdef CORD_CAT_OPTIMIZATION |
| #define CORD_cat(x, y) CORD_cat_optimized(x, y) | #define CORD_cat(x, y) CORD_cat_optimized(x, y) |
| Line 47 class SQL_Connection; | Line 47 class SQL_Connection; |
| class Dictionary; | class Dictionary; |
| class Request_charsets; | class Request_charsets; |
| class String; | class String; |
| typedef Array<const String*> ArrayString; | |
| class VRegex; | class VRegex; |
| #ifdef NDEBUG | |
| typedef Array<const String*> ArrayString; | |
| #else | |
| class ArrayString : public Array<const String*> { | |
| public: | |
| inline ArrayString(size_t initial=0) : Array(initial){ | |
| } | |
| inline Array& operator+=(element_type src) { | |
| assert(src != NULL); | |
| return Array::operator+=(src); | |
| } | |
| inline element_type get(size_t index) const { | |
| element_type result=Array::get(index); | |
| assert(result != NULL); | |
| return result; | |
| } | |
| }; | |
| #endif | |
| // generally useful | // generally useful |
| int pa_atoi(const char* str, const String* problem_source=0); | double pa_atod(const char* str, const String* problem_source); |
| double pa_atod(const char* str, const String* problem_source=0); | int pa_atoi(const char* str, int base=10, const String* problem_source=0); |
| unsigned int pa_atoui(const char *str, int base, const String* problem_source=0); | unsigned int pa_atoui(const char *str, int base=10, const String* problem_source=0); |
| unsigned long long int pa_atoul(const char *str, int base, const String* problem_source=0); | uint64_t pa_atoul(const char *str, int base=10, const String* problem_source=0); |
| /// this is result of pos functions which mean that substr were not found | /// this is result of pos functions which mean that substr were not found |
| #define STRING_NOT_FOUND ((size_t)-1) | #define STRING_NOT_FOUND ((size_t)-1) |
| Line 73 unsigned long long int pa_atoul(const ch | Line 92 unsigned long long int pa_atoul(const ch |
| class String: public PA_Object { | class String: public PA_Object { |
| friend class StringSplitHelper; | |
| public: | public: |
| /** piece is tainted or not. the lang to use when detaint | /** piece is tainted or not. the lang to use when detaint |
| Line 89 public: | Line 109 public: |
| WARNING WARNING WARNING WARNING WARNING WARNING | WARNING WARNING WARNING WARNING WARNING WARNING |
| */ | */ |
| #if _MSC_VER >= 1900 | |
| /// required for VS2015+ to make sizeof(Languages::opt) == sizeof(CORD), will be 16 byte under x64 without it | |
| enum Language : size_t { | |
| #else | |
| enum Language { | enum Language { |
| L_UNSPECIFIED=0, ///< no real string has parts of this lange: it's just convinient to check when string's empty | #endif |
| L_UNSPECIFIED=0, ///< no real string has parts of this lange: it's just convinient to check when string's empty | |
| // these two must go before others, there are checks for >L_AS_IS | // these two must go before others, there are checks for >L_AS_IS |
| L_CLEAN='0', ///< clean WARNING: read above warning before changing | L_CLEAN='0', ///< clean WARNING: read above warning before changing |
| L_AS_IS='A', ///< leave all characters intact WARNING: read above warning before changing | L_AS_IS='A', ///< leave all characters intact WARNING: read above warning before changing |
| L_PASS_APPENDED='P', | L_TAINTED='T', ///< tainted, untaint lang as assigned later |
| /**< | |
| leave lang built into string being appended. | |
| just a flag, that value not stored | |
| */ | |
| L_TAINTED='T', ///< tainted, untaint lang as assigned later | |
| // untaint langs. assigned by ^untaint[lang]{...} | // untaint langs. assigned by ^untaint[lang]{...} |
| L_FILE_SPEC='F', ///< file specification | L_FILE_SPEC='F', ///< file specification |
| L_HTTP_HEADER='h', ///< text in HTTP response header | L_HTTP_HEADER='h', ///< text in HTTP response header |
| L_MAIL_HEADER='m', ///< text in mail header | L_MAIL_HEADER='m', ///< text in mail header |
| L_URI='U', ///< text in uri | L_URI='U', ///< text in uri |
| L_SQL='Q', ///< ^table:sql body | L_SQL='Q', ///< ^table:sql body |
| L_JS='J', ///< JavaScript code | L_JS='J', ///< JavaScript code |
| L_XML='X', ///< ^xdoc:create xml | L_XML='X', ///< ^xdoc:create xml |
| L_HTML='H', ///< HTML code | L_HTML='H', ///< HTML code |
| L_REGEX='R', ///< RegExp | L_REGEX='R', ///< RegExp |
| L_JSON='S', ///< JSON code | L_JSON='S', ///< JSON code |
| L_HTTP_COOKIE='C', ///< cookies encoded as %uXXXX for compartibility with js functions encode/decode | L_HTTP_COOKIE='C', ///< cookies encoded as %uXXXX for compartibility with js functions encode/decode |
| L_PARSER_CODE='p', ///< ^process body | L_PARSER_CODE='p', ///< ^process body |
| // READ WARNING ABOVE BEFORE ADDING ANYTHING | // READ WARNING ABOVE BEFORE ADDING ANYTHING |
| Line 175 public: | Line 196 public: |
| public: | public: |
| const char* v() const; | const char* visualize() const; |
| void dump() const; | void dump() const; |
| Languages(): langs(0) {} | Languages(): langs(0) {} |
| Line 316 public: | Line 337 public: |
| size_t length; | size_t length; |
| C(): str(0), length(0) {} | C(): str(0), length(0) {} |
| C(const char *astr, size_t asize): str(astr), length(asize) {} | C(const char *astr, size_t asize): str(astr), length(asize) {} |
| explicit C(Body abody): str(abody.cstr()), length(abody.length()) {} | |
| }; | }; |
| struct Cm { | struct Cm { |
| Line 323 public: | Line 345 public: |
| size_t length; | size_t length; |
| Cm(): str(0), length(0) {} | Cm(): str(0), length(0) {} |
| Cm(char *astr, size_t asize): str(astr), length(asize) {} | Cm(char *astr, size_t asize): str(astr), length(asize) {} |
| explicit Cm(Body abody): str(abody.cstrm()), length(abody.length()) {} | |
| }; | }; |
| class Body { | class Body { |
| Line 352 public: | Line 375 public: |
| public: | public: |
| const char* v() const; | |
| void dump() const; | void dump() const; |
| Body(): body(CORD_EMPTY) INIT_HASH_CODE(0) INIT_LENGTH(0) {} | Body(): body(CORD_EMPTY) INIT_HASH_CODE(0) INIT_LENGTH(0) {} |
| Line 378 public: | Line 400 public: |
| #endif | #endif |
| } | } |
| static Body uitoa(size_t aindex); | |
| static Body Format(int value); | |
| void clear() { ZERO_LENGTH ZERO_HASH_CODE body=CORD_EMPTY; } | void clear() { ZERO_LENGTH ZERO_HASH_CODE body=CORD_EMPTY; } |
| Line 388 public: | Line 409 public: |
| inline CORD get_cord() const { return body; } | inline CORD get_cord() const { return body; } |
| uint get_hash_code() const; | uint get_hash_code() const; |
| // never null | |
| const char* cstr() const { | const char* cstr() const { |
| #ifdef STRING_LENGTH_CACHING | #ifdef STRING_LENGTH_CACHING |
| string_length = length(); | string_length = length(); |
| Line 400 public: | Line 422 public: |
| return CORD_to_const_char_star(body, length()); | return CORD_to_const_char_star(body, length()); |
| } | } |
| // never null | |
| char* cstrm() const { return CORD_to_char_star(body, length()); } | char* cstrm() const { return CORD_to_char_star(body, length()); } |
| #ifdef STRING_LENGTH_CACHING | #ifdef STRING_LENGTH_CACHING |
| Line 486 public: | Line 509 public: |
| size_t* out_start=0, size_t* out_length=0, Charset* source_charset=0) const; | size_t* out_start=0, size_t* out_length=0, Charset* source_charset=0) const; |
| }; | }; |
| protected: | private: |
| Body body; ///< all characters of string | Body body; ///< all characters of string |
| Languages langs; ///< string characters lang | Languages langs; ///< string characters lang |
| const char* v() const; | |
| void dump() const; | void dump() const; |
| #define ASSERT_STRING_INVARIANT(string) \ | #define ASSERT_STRING_INVARIANT(string) \ |
| assert((string).langs.invariant((string).body.length())) | assert((string).langs.invariant((string).body.length())) |
| Line 565 public: | Line 587 public: |
| return langs.opt.lang; | return langs.opt.lang; |
| } | } |
| char* visualize_langs() const; | |
| /// puts pieces to buf | /// puts pieces to buf |
| Cm serialize(size_t prolog_size) const; | Cm serialize(size_t prolog_size) const; |
| /// appends pieces from buf to self | /// appends pieces from buf to self |
| Line 587 public: | Line 611 public: |
| return Body(x).ncmp(0/*x_begin*/, body, 0/*y_begin*/, length())==0; | return Body(x).ncmp(0/*x_begin*/, body, 0/*y_begin*/, length())==0; |
| } | } |
| String& append_to(String& dest) const; | |
| String& append_to(String& dest, Language lang, bool forced=false) const; | String& append_to(String& dest, Language lang, bool forced=false) const; |
| String& append(const String& src, Language lang, bool forced=false) { | String& append(const String& src, Language lang, bool forced=false) { |
| return src.append_to(*this, lang, forced); | return src.append_to(*this, lang, forced); |
| Line 599 public: | Line 624 public: |
| return *this; | return *this; |
| } | } |
| String& operator << (const String& src) { return append(src, L_PASS_APPENDED); } | String& operator << (const String& src) { return src.append_to(*this); } |
| String& operator << (const char* src) { return append_help_length(src, 0, L_AS_IS); } | String& operator << (const char* src) { return append_help_length(src, 0, L_AS_IS); } |
| String& operator << (const Body& src){ | String& operator << (const Body& src){ |
| langs.appendHelper(body, L_AS_IS, src); | langs.appendHelper(body, L_AS_IS, src); |
| Line 626 public: | Line 651 public: |
| String& mid(size_t substr_begin, size_t substr_end) const; | String& mid(size_t substr_begin, size_t substr_end) const; |
| String& mid(Charset& charset, size_t from, size_t to, size_t helper_length=0) const; | String& mid(Charset& charset, size_t from, size_t to, size_t helper_length=0) const; |
| /** | /// return position of substr in string, -1 means "not found" [const char* version] |
| ignore lang if it's L_UNSPECIFIED | size_t pos(const char* substr, size_t this_offset=0) const { return body.pos(substr, this_offset); } |
| but when specified: look for substring that lies in ONE fragment in THAT lang | size_t pos(const Body substr, size_t this_offset=0) const { return body.pos(substr, this_offset); } |
| @return position of substr in string, -1 means "not found" [const char* version] | size_t pos(const String& substr, size_t this_offset=0) const { return body.pos(substr.body, this_offset); } |
| */ | size_t pos(char c, size_t this_offset=0) const { return body.pos(c, this_offset); } |
| size_t pos(const Body substr, size_t this_offset=0, Language lang=L_UNSPECIFIED) const; | /// ignore lang if it's L_UNSPECIFIED, otherwise look for substring that lies in ONE fragment in THAT lang |
| /// String version of @see pos(const char*, int, Language) | size_t pos(const Body substr, size_t this_offset, Language lang) const; |
| size_t pos(const String& substr, size_t this_offset=0, Language lang=L_UNSPECIFIED) const; | size_t pos(const String& substr, size_t this_offset, Language lang) const; |
| size_t pos(char c, size_t this_offset=0) const { | |
| return body.pos(c, this_offset); | |
| } | |
| size_t pos(Charset& charset, const String& substr, size_t this_offset=0, Language lang=L_UNSPECIFIED) const; | size_t pos(Charset& charset, const String& substr, size_t this_offset=0, Language lang=L_UNSPECIFIED) const; |
| size_t strrpbrk(const char* chars, size_t left=0) const { | size_t strrpbrk(const char* chars, size_t left=0) const { |
| Line 653 public: | Line 675 public: |
| return body.rskipchars(chars, left, right); | return body.rskipchars(chars, left, right); |
| } | } |
| void split(ArrayString& result, size_t& pos_after, const char* delim, Language lang=L_UNSPECIFIED, int limit=-1) const; | void split(ArrayString& result, size_t pos_after, const char* delim, Language lang=L_UNSPECIFIED) const; |
| void split(ArrayString& result, size_t& pos_after, const String& delim, Language lang=L_UNSPECIFIED, int limit=-1) const; | void split(ArrayString& result, size_t pos_after, const String& delim, Language lang=L_UNSPECIFIED) const; |
| typedef void (*Row_action)(Table& table, ArrayString* row, int prestart, int prefinish, int poststart, int postfinish, void *info); | typedef void (*Row_action)(Table& table, ArrayString* row, int prestart, int prefinish, int poststart, int postfinish, void *info); |
| Line 676 public: | Line 698 public: |
| const String& replace(const Dictionary& dict) const; | const String& replace(const Dictionary& dict) const; |
| const String& trim(Trim_kind kind=TRIM_BOTH, const char* chars=0, Charset* source_charset=0) const; | const String& trim(Trim_kind kind=TRIM_BOTH, const char* chars=0, Charset* source_charset=0) const; |
| double as_double() const { return pa_atod(cstr(), this); } | double as_double() const { return pa_atod(cstr(), this); } |
| int as_int() const { return pa_atoi(cstr(), this); } | int as_int() const { return pa_atoi(cstr(), 0, this); } |
| bool as_bool() const { return as_int()!=0; } | bool as_bool() const { return as_int()!=0; } |
| const String& escape(Charset& source_charset) const; | const String& escape(Charset& source_charset) const; |