--- parser3/src/include/pa_common.h 2009/09/03 11:08:28 1.136 +++ parser3/src/include/pa_common.h 2016/07/29 13:30:41 1.166 @@ -1,14 +1,14 @@ /** @file Parser: commonly used functions. - Copyright (c) 2001-2009 ArtLebedev Group (http://www.artlebedev.com) + Copyright (c) 2001-2015 Art. Lebedev Studio (http://www.artlebedev.com) Author: Alexandr Petrosian (http://paf.design.ru) */ #ifndef PA_COMMON_H #define PA_COMMON_H -static const char * const IDENT_COMMON_H="$Date: 2009/09/03 11:08:28 $"; +#define IDENT_PA_COMMON_H "$Id: pa_common.h,v 1.166 2016/07/29 13:30:41 moko Exp $" #include "pa_string.h" #include "pa_hash.h" @@ -16,34 +16,33 @@ static const char * const IDENT_COMMON_H class Request; // defines -#define HTTP_USER_AGENT "user-agent" -#define HTTP_STATUS "status" -#define HTTP_CONTENT_LENGTH "content-length" +#define HTTP_STATUS "status" +#define HTTP_STATUS_CAPITALIZED "Status" -#define HTTP_CONTENT_TYPE "content-type" -#define HTTP_CONTENT_TYPE_UPPER "CONTENT-TYPE" -#define HTTP_CONTENT_TYPE_FORM_URLENCODED "application/x-www-form-urlencoded" -#define HTTP_CONTENT_TYPE_MULTIPART_FORMDATA "multipart/form-data" -#define HTTP_CONTENT_TYPE_MULTIPART_RELATED "multipart/related" -#define HTTP_CONTENT_TYPE_MULTIPART_MIXED "multipart/mixed" +#define HTTP_CONTENT_LENGTH "content-length" +#define HTTP_CONTENT_LENGTH_CAPITALIZED "Content-Length" -#define CONTENT_TRANSFER_ENCODING_NAME "content-transfer-tncoding" +#define HTTP_CONTENT_TYPE "content-type" +#define HTTP_CONTENT_TYPE_UPPER "CONTENT-TYPE" +#define HTTP_CONTENT_TYPE_CAPITALIZED "Content-Type" + +#define CONTENT_DISPOSITION "content-disposition" +#define CONTENT_DISPOSITION_UPPER "CONTENT-DISPOSITION" +#define CONTENT_DISPOSITION_CAPITALIZED "Content-Disposition" -#define CONTENT_DISPOSITION "content-disposition" -#define CONTENT_DISPOSITION_ATTACHMENT "attachment" -#define CONTENT_DISPOSITION_INLINE "inline" +#define CONTENT_DISPOSITION_ATTACHMENT "attachment" +#define CONTENT_DISPOSITION_INLINE "inline" #define CONTENT_DISPOSITION_FILENAME_NAME "filename" -const String http_content_type(HTTP_CONTENT_TYPE); +#define HTTP_CONTENT_TYPE_FORM_URLENCODED "application/x-www-form-urlencoded" +#define HTTP_CONTENT_TYPE_MULTIPART_FORMDATA "multipart/form-data" +#define HTTP_CONTENT_TYPE_MULTIPART_RELATED "multipart/related" +#define HTTP_CONTENT_TYPE_MULTIPART_MIXED "multipart/mixed" -const String content_transfer_encoding_name(CONTENT_TRANSFER_ENCODING_NAME); +#define BASE64_STRICT_OPTION_NAME "strict" -const String content_disposition(CONTENT_DISPOSITION); -const String content_disposition_inline(CONTENT_DISPOSITION_INLINE); -const String content_disposition_attachment(CONTENT_DISPOSITION_ATTACHMENT); const String content_disposition_filename_name(CONTENT_DISPOSITION_FILENAME_NAME); - #define HASH_ORDER #ifdef HASH_ORDER @@ -62,15 +61,7 @@ int __vsnprintf(char *, size_t, const ch int __snprintf(char *, size_t, const char* , ...); #define snprintf __snprintf -#if _MSC_VER -/* -inline int open( const char* filename, int oflag ) { return _open(filename, oflag); } -inline int close( int handle ) { return _close(handle); } -inline int read( int handle, void *buffer, unsigned int count ) { return _read(handle,buffer,count); } -inline int write( int handle, const void *buffer, unsigned int count ) { return _write(handle,buffer,count); } -inline int stat( const char* path, struct _stat *buffer ) { return _stat(path, buffer); } -inline long lseek( int handle, long offset, int origin ) { return _lseek(handle, offset, origin); } -*/ +#ifdef _MSC_VER //access #define F_OK 0 @@ -81,28 +72,18 @@ inline long lseek( int handle, long offs #ifndef strcasecmp # define strcasecmp _stricmp #endif + #ifndef strncasecmp # define strncasecmp _strnicmp #endif -#ifndef mkdir -# define mkdir(path, mode) _mkdir(path) -#endif -#ifndef putenv -# define putenv _putenv #endif -#endif - -const char* capitalize(const char* s); - -/** under WIN32 "t" mode fixes DOS chars OK, - can't say that about other systems/ line break styles +/** + file related functions */ -void fix_line_breaks( - char *str, - size_t& length///< may change! used to speedup next actions - ); + +#define FILE_BUFFER_SIZE 4096 int pa_lock_shared_blocking(int fd); int pa_lock_exclusive_blocking(int fd); @@ -111,11 +92,9 @@ int pa_unlock(int fd); void create_dir_for_file(const String& file_spec); -typedef void (*File_read_action)( - struct stat& finfo, - int f, - const String& file_spec, const char* fname, bool as_text, - void *context); +int pa_get_valid_file_options_count(HashStringValue& options); + +typedef void (*File_read_action)(struct stat& finfo, int f, const String& file_spec, void *context); /** shared-lock specified file, @@ -198,21 +177,24 @@ bool file_write_action_under_lock( throws an exception in case of problems */ void file_write( + Request_charsets& charsets, const String& file_spec, - const char* data, size_t size, + const char* data, + size_t size, bool as_text, - bool do_append=false); + bool do_append=false, + Charset* asked_charset=0); /** delete specified file throws an exception in case of problems */ -bool file_delete(const String& file_spec, bool fail_on_problem=true); +bool file_delete(const String& file_spec, bool fail_on_problem=true, bool keep_empty_dirs=false); /** move specified file throws an exception in case of problems */ -void file_move(const String& old_spec, const String& new_spec); +void file_move(const String& old_spec, const String& new_spec, bool keep_empty_dirs=false); bool entry_exists(const char* fname, struct stat *afinfo=0); bool entry_exists(const String& file_spec); @@ -221,153 +203,82 @@ bool dir_exists(const String& file_spec) const String* file_exist(const String& path, const String& name); bool file_executable(const String& file_spec); -bool file_stat(const String& file_spec, - size_t& rsize, - time_t& ratime, - time_t& rmtime, - time_t& rctime, - bool fail_on_read_problem=true); +bool file_stat(const String& file_spec, size_t& rsize, time_t& ratime, time_t& rmtime, time_t& rctime, bool fail_on_read_problem=true); -/** - scans for @a delim[default \n] in @a *row_ref, - @return piece of line before it or end of string, if no @a delim found - assigns @a *row_ref to point right after delimiter if there were one - or to zero if no @a delim were found. +size_t stdout_write(const void *buf, size_t size); + +void check_safe_mode(struct stat finfo, const String& file_spec, const char* fname); + +int file_block_read(const int f, unsigned char* buffer, const size_t size); + +/** + String related functions +*/ + +/** under WIN32 "t" mode fixes DOS chars OK, + can't say that about other systems/ line break styles */ +void fix_line_breaks(char *str, size_t& length /* < may change! used to speedup next actions */); + char *getrow(char **row_ref,char delim='\n'); char *lsplit(char *string, char delim); char *lsplit(char **string_ref,char delim); char *rsplit(char *string, char delim); -const char* format(double value, char *fmt); -size_t stdout_write(const void *buf, size_t size); +const char* format(double value, const char *fmt); + +char* unescape_chars(const char* cp, int len, Charset* client_charset=0, bool js=false/*true==decode \uXXXX and don't convert '+' to space*/); -char* unescape_chars(const char* cp, int len, Charset* client_charset=0, bool ignore_plus=false); +char *search_stop(char*& current, char cstop_at); + +inline int pa_strncasecmp(const char* str, const char* substr, size_t count=0) { + return strncasecmp(str, substr, count ? count : strlen(substr)); +} #ifdef WIN32 void back_slashes_to_slashes(char *s); -//void slashes_to_back_slashes(char *s); -#endif - -#ifndef _qsort -# define _qsort(names,cnt,sizeof_names,func_addr) \ - qsort(names,cnt,sizeof_names,func_addr) #endif -bool StrStartFromNC(const char* str, const char* substr, bool equal=false); size_t strpos(const char *str, const char *substr); -Charset* detect_charset(const char* content_type); +int remove_crlf(char *start, char *end); -#define SECS_PER_DAY (60*60*24) -int getMonthDays(int year, int month); +inline bool pa_isalpha(unsigned char c) { return (((c>='A') && (c<='Z')) || ((c>='a') && (c<='z'))); } +inline bool pa_isalnum(unsigned char c) { return (((c>='0') && (c<='9')) || pa_isalpha(c)); } -int remove_crlf(char *start, char *end); +const char* capitalize(const char* s); +char *str_lower(const char *s, size_t helper_length=0); +char *str_upper(const char *s, size_t helper_length=0); -#ifdef PA_SAFE_MODE -void check_safe_mode(struct stat finfo, const String& file_spec, const char* fname); -#endif +const char* hex_string(unsigned char* bytes, size_t size, bool upcase); +extern const char* hex_digits; -void pa_base64_decode(const char *in, size_t in_size, char*& result, size_t& result_size); +void pa_base64_decode(const char *in, size_t in_size, char*& result, size_t& result_size, bool strict=false); char* pa_base64_encode(const char *in, size_t in_size); -struct File_base64_action_info { - unsigned char** base64; -}; char* pa_base64_encode(const String& file_spec); -static void file_base64_file_action( - struct stat& finfo, - int f, - const String&, const char* /*fname*/, bool, - void *context); - -#define FILE_BUFFER_SIZE 4096 -static unsigned long crc32Table[256]; -static void InitCrc32Table() -{ - if(crc32Table[1] == 0){ - // This is the official polynomial used by CRC32 in PKZip. - // Often times the polynomial shown reversed as 0x04C11DB7. - static const unsigned long dwPolynomial = 0xEDB88320; - - for(int i = 0; i < 256; i++) - { - unsigned long dwCrc = i; - for(int j = 8; j > 0; j--) - { - if(dwCrc & 1) - dwCrc = (dwCrc >> 1) ^ dwPolynomial; - else - dwCrc >>= 1; - } - crc32Table[i] = dwCrc; - } - } -} - -int file_block_read(const int f, unsigned char* buffer, const size_t size); -inline void CalcCrc32(const unsigned char byte, unsigned long &crc32) -{ - crc32 = ((crc32) >> 8) ^ crc32Table[(byte) ^ ((crc32) & 0x000000FF)]; -} +const char *pa_idna_encode(const char *in, Charset &source); +const char *pa_idna_decode(const char *in, Charset &source); const unsigned long pa_crc32(const char *in, size_t in_size); const unsigned long pa_crc32(const String& file_spec); -static void file_crc32_file_action( - struct stat& finfo, - int f, - const String&, const char* /*fname*/, bool, - void *context); - -static const char* hex_string(unsigned char* bytes, size_t size, bool upcase) { - char *bytes_hex=new(PointerFreeGC) char [size*2/*byte->hh*/+1/*for zero-teminator*/]; - unsigned char *src=bytes; - unsigned char *end=bytes+size; - char *dest=bytes_hex; - - const char *hex=upcase?"0123456789ABCDEF":"0123456789abcdef"; - - for(; srcput(key, value); } -static void remove_key_from( - HashStringValue::key_type key, - HashStringValue::value_type /*value*/, - HashStringValue* dest) { +static void remove_key_from(HashStringValue::key_type key, HashStringValue::value_type /*value*/, HashStringValue* dest) { dest->remove(key); } -static String::C date_gmt_string(tm* tms) { - /// http://www.w3.org/Protocols/rfc2616/rfc2616-sec3.html#sec3.3 - static const char month_names[12][4]={ - "Jan","Feb","Mar","Apr","May","Jun","Jul","Aug","Sep","Oct","Nov","Dec"}; - static const char days[7][4]={ - "Sun","Mon","Tue","Wed","Thu","Fri","Sat"}; - - char *buf=new(PointerFreeGC) char[MAX_STRING]; - return String::C(buf, - snprintf(buf, MAX_STRING, "%s, %.2d %s %.4d %.2d:%.2d:%.2d GMT", - days[tms->tm_wday], - tms->tm_mday,month_names[tms->tm_mon],tms->tm_year+1900, - tms->tm_hour,tms->tm_min,tms->tm_sec)); -} - +Charset* detect_charset(const char* content_type); // globals @@ -380,6 +291,7 @@ extern const String file_status_name; #define PA_COLUMN_SEPARATOR_NAME "separator" #define PA_COLUMN_ENCLOSER_NAME "encloser" #define PA_CHARSET_NAME "charset" +#define PA_RESPONSE_CHARSET_NAME "response-charset" // globals defines for sql options @@ -390,7 +302,7 @@ extern const String file_status_name; #ifndef DOXYGEN enum Table2hash_distint { D_ILLEGAL, D_FIRST }; -enum Table2hash_value_type { C_HASH, C_STRING, C_TABLE }; +enum Table2hash_value_type { C_HASH, C_STRING, C_TABLE, C_CODE }; #endif #endif