Diff for /parser3/src/include/pa_common.h between versions 1.144 and 1.192

version 1.144, 2011/02/18 06:01:21 version 1.192, 2023/09/26 20:49:07
Line 1 Line 1
 /** @file  /** @file
         Parser: commonly used functions.          Parser: commonly used functions.
   
         Copyright (c) 2001-2009 ArtLebedev Group (http://www.artlebedev.com)          Copyright (c) 2001-2023 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_COMMON_H  #ifndef PA_COMMON_H
 #define PA_COMMON_H  #define PA_COMMON_H
   
 static const char * const IDENT_COMMON_H="$Date$";  #define IDENT_PA_COMMON_H "$Id$"
   
 #include "pa_string.h"  #include "pa_string.h"
 #include "pa_hash.h"  #include "pa_hash.h"
Line 16  static const char * const IDENT_COMMON_H Line 16  static const char * const IDENT_COMMON_H
 class Request;  class Request;
   
 // defines  // defines
 #define HTTP_USER_AGENT "user-agent"  #define HTTP_STATUS             "status"
   
 #define HTTP_STATUS     "status"  
 #define HTTP_STATUS_CAPITALIZED "Status"  #define HTTP_STATUS_CAPITALIZED "Status"
   
 #define HTTP_CONTENT_LENGTH     "content-length"  #define HTTP_CONTENT_LENGTH             "content-length"
 #define HTTP_CONTENT_LENGTH_CAPITALIZED "Content-Length"  
   
 #define HTTP_CONTENT_TYPE       "content-type"  #define HTTP_CONTENT_TYPE               "content-type"
 #define HTTP_CONTENT_TYPE_UPPER "CONTENT-TYPE"  #define HTTP_CONTENT_TYPE_UPPER         "CONTENT-TYPE"
 #define HTTP_CONTENT_TYPE_CAPITALIZED   "Content-Type"  #define HTTP_CONTENT_TYPE_CAPITALIZED   "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 CONTENT_TRANSFER_ENCODING_NAME "content-transfer-encoding"  
 #define CONTENT_TRANSFER_ENCODING_CAPITALIZED "Content-Transfer-Encoding"  
   
 #define CONTENT_DISPOSITION     "content-disposition"  #define CONTENT_DISPOSITION             "content-disposition"
   #define CONTENT_DISPOSITION_UPPER       "CONTENT-DISPOSITION"
 #define CONTENT_DISPOSITION_CAPITALIZED "Content-Disposition"  #define CONTENT_DISPOSITION_CAPITALIZED "Content-Disposition"
 #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 CONTENT_DISPOSITION_ATTACHMENT  "attachment"
   #define CONTENT_DISPOSITION_INLINE      "inline"
   #define CONTENT_DISPOSITION_FILENAME_NAME "filename"
   
 const String content_transfer_encoding_name(CONTENT_TRANSFER_ENCODING_NAME);  #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_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);  const String content_disposition_filename_name(CONTENT_DISPOSITION_FILENAME_NAME);
   
   
 #define HASH_ORDER  #define HASH_ORDER
   
 #ifdef HASH_ORDER  #ifdef HASH_ORDER
Line 63  typedef HASH_STRING<Value*> HashStringVa Line 52  typedef HASH_STRING<Value*> HashStringVa
   
 // replace system s*nprintf with our versions  // replace system s*nprintf with our versions
 #undef vsnprintf   #undef vsnprintf 
 int __vsnprintf(char *, size_t, const char* , va_list);  int pa_vsnprintf(char *, size_t, const char* , va_list);
 #define vsnprintf __vsnprintf   #define vsnprintf pa_vsnprintf 
 #undef snprintf  #undef snprintf
 int __snprintf(char *, size_t, const char* , ...);  int pa_snprintf(char *, size_t, const char* , ...);
 #define snprintf __snprintf  #define snprintf pa_snprintf
   
 #if _MSC_VER  #ifdef _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); }  
 */  
   
 //access  //access
 #define F_OK 0  #define F_OK 0
Line 88  inline long lseek( int handle, long offs Line 69  inline long lseek( int handle, long offs
 #ifndef strcasecmp  #ifndef strcasecmp
 #       define strcasecmp _stricmp  #       define strcasecmp _stricmp
 #endif  #endif
   
 #ifndef strncasecmp  #ifndef strncasecmp
 #       define strncasecmp _strnicmp  #       define strncasecmp _strnicmp
 #endif  #endif
 #ifndef mkdir  
 #       define mkdir(path, mode) _mkdir(path)  
 #endif  
   
 #ifndef putenv  #define stat __stat64
 #       define putenv _putenv  #define pa_fstat _fstat64
 #endif  
   
 #endif  int pa_stat(const char *pathname, struct stat *buffer);
   int pa_open(const char *pathname, int flags, int mode=0);
   FILE *pa_fopen(const char *pathname, const char *mode);
   
 const char* capitalize(const char* s);  #define pa_lseek _lseeki64
   
 /** under WIN32 "t" mode fixes DOS chars OK,   #else
         can't say that about other systems/ line break styles  
   #define pa_stat stat
   #define pa_fstat fstat
   
   #define pa_open open
   #define pa_fopen fopen
   
   #define pa_lseek lseek
   
   #endif
   
   /** 
           file related functions
 */  */
 void fix_line_breaks(  
                         char *str,  #define FILE_BUFFER_SIZE        4096
                         size_t& length///< may change! used to speedup next actions  
                         );  
   
 int pa_lock_shared_blocking(int fd);  int pa_lock_shared_blocking(int fd);
 int pa_lock_exclusive_blocking(int fd);  int pa_lock_exclusive_blocking(int fd);
Line 118  int pa_unlock(int fd); Line 108  int pa_unlock(int fd);
   
 void create_dir_for_file(const String& file_spec);  void create_dir_for_file(const String& file_spec);
   
 typedef void (*File_read_action)(  int pa_get_valid_file_options_count(HashStringValue& options);
                                 struct stat& finfo,  
                                 int f,   typedef void (*File_read_action)(struct stat& finfo, int f, const String& file_spec, void *context);
                                 const String& file_spec, const char* fname, bool as_text,  
                                 void *context);  
   
 /**  /**
         shared-lock specified file,           shared-lock specified file, 
Line 137  bool file_read_action_under_lock(const S Line 125  bool file_read_action_under_lock(const S
                                 bool fail_on_read_problem=true);                                  bool fail_on_read_problem=true);
   
 /**  /**
         read specified text file using           read specified text file
         if fail_on_read_problem is true[default] throws an exception          if fail_on_read_problem is true[default] throws an exception
   
         WARNING: charset is used for http header case conversion, it's not a charset of input file!          WARNING: charset is used for http header case conversion, it's not a charset of input file!
 */  */
 char *file_read_text(Request_charsets& charsets,   char *file_read_text(Request_charsets& charsets, const String& file_spec, bool fail_on_read_problem = true);
                                         const String& file_spec,   char *file_load_text(Request& r, const String& file_spec, bool fail_on_read_problem = true, HashStringValue* options = 0, bool transcode_result = true);
                                         bool fail_on_read_problem=true,  
                                         HashStringValue* options=0,  
                                         bool transcode_result=true);  
   
 char *file_load_text(Request& r,   
                                         const String& file_spec,   
                                         bool fail_on_read_problem=true,  
                                         HashStringValue* options=0,  
                                         bool transcode_result=true);  
   
 struct File_read_result {  struct File_read_result {
         bool success;          bool success;
Line 161  struct File_read_result { Line 140  struct File_read_result {
 };  };
   
 /**  /**
         read specified file using           read specified binary file
         if fail_on_read_problem is true[default] throws an exception          if fail_on_read_problem is true[default] throws an exception
   */
   
         WARNING: charset is used for http header case conversion, it's not a charset of input file!  File_read_result file_read_binary(const String& file_spec, bool fail_on_read_problem = true, char* buf = 0, uint64_t offset = 0, size_t limit = 0);
   
   /**
           load specified file
           if fail_on_read_problem is true[default] throws an exception
 */  */
 File_read_result file_read(Request_charsets& charsets,  
                                 const String& file_spec,  
                                 bool as_text,  
                                 HashStringValue* options=0,  
                                 bool fail_on_read_problem=true,  
                                 char* buf=0, size_t offset=0, size_t size=0, bool transcode_text_result=true);  
   
 File_read_result file_load(Request& r,  File_read_result file_load(Request& r, const String& file_spec, bool as_text, HashStringValue* options=0, bool fail_on_read_problem=true, bool transcode_text_result=true);
                                 const String& file_spec,  
                                 bool as_text,  
                                 HashStringValue* options=0,  
                                 bool fail_on_read_problem=true,  
                                 char* buf=0, size_t offset=0, size_t size=0, bool transcode_text_result=true);  
   
 typedef void (*File_write_action)(int f, void *context);  typedef void (*File_write_action)(int f, void *context);
   
Line 217  void file_write( Line 190  void file_write(
         delete specified file           delete specified file 
         throws an exception in case of problems          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           move specified file 
         throws an exception in case of problems          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 char* fname, struct stat *afinfo=0);
 bool entry_exists(const String& file_spec);  bool entry_exists(const String& file_spec);
Line 231  bool dir_exists(const String& file_spec) Line 204  bool dir_exists(const String& file_spec)
 const String* file_exist(const String& path, const String& name);  const String* file_exist(const String& path, const String& name);
 bool file_executable(const String& file_spec);  bool file_executable(const String& file_spec);
   
 bool file_stat(const String& file_spec,   bool file_stat(const String& file_spec, uint64_t& rsize, time_t& ratime, time_t& rmtime, time_t& rctime, bool fail_on_read_problem=true);
                                 size_t& rsize,   size_t check_file_size(uint64_t size, const String* file_spec);
                                 time_t& ratime,  
                                 time_t& rmtime,  
                                 time_t& rctime,  
                                 bool fail_on_read_problem=true);  
   
 /**  size_t stdout_write(const void *buf, size_t size);
         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  void check_safe_mode(struct stat finfo, const String& file_spec, const char* fname); 
         assigns @a *row_ref to point right after delimiter if there were one  
         or to zero if no @a delim were found.  ssize_t file_block_read(const int f, void* 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 *getrow(char **row_ref,char delim='\n');
 char *lsplit(char *string, char delim);  char *lsplit(char *string, char delim);
 char *lsplit(char **string_ref,char delim);  char *lsplit(char **string_ref,char delim);
 char *rsplit(char *string, 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 js=false/*true==decode \uXXXX and don't convert '+' to space*/);
   
   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  #ifdef WIN32
 void back_slashes_to_slashes(char *s);  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  #endif
   
 bool StrStartFromNC(const char* str, const char* substr, bool equal=false);  
 size_t strpos(const char *str, const char *substr);  size_t strpos(const char *str, const char *substr);
   
 Charset* detect_charset(const char* content_type);  size_t remove_crlf(char *start, char *end);
   
 #define SECS_PER_DAY (60*60*24)  
 int getMonthDays(int year, int month);  
   
 int remove_crlf(char *start, char *end);  
   
 inline bool pa_isalpha(unsigned char c) {  inline bool pa_isalpha(unsigned char c) { return (((c>='A') && (c<='Z')) || ((c>='a') && (c<='z'))); }
         return (((c>='A') && (c<='Z')) || ((c>='a') && (c<='z')));  inline bool pa_isalnum(unsigned char c) { return (((c>='0') && (c<='9')) || pa_isalpha(c)); }
 }  
   
 inline bool pa_isalnum(unsigned char c) {  
         return (((c>='0') && (c<='9')) || pa_isalpha(c));  
 }  
   
 void check_safe_mode(struct stat finfo, const String& file_spec, const char* fname);   
   
 void pa_base64_decode(const char *in, size_t in_size, char*& result, size_t& result_size);  void pa_strncpy(char *dst, const char *src, size_t n);
 char* pa_base64_encode(const char *in, size_t in_size);  char *pa_strcat(const char *a, const char *b, const char *c = 0);
 struct File_base64_action_info {  const char *pa_filename(const char *path);
         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  const char* capitalize(const char* s);
 static unsigned long crc32Table[256];  char *str_lower(const char *s, size_t length);
 static void InitCrc32Table()  char *str_upper(const char *s, size_t length);
 {  inline char *str_lower(const char *s) { return str_lower(s, strlen(s)); }
         if(crc32Table[1] == 0){  inline char *str_upper(const char *s) { return str_upper(s, strlen(s)); }
                 // 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 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(; src<end; src++) {  
                 *dest++=hex[*src/0x10];  
                 *dest++=hex[*src%0x10];  
         }  
         *dest=0;  
   
         return bytes_hex;  const char* hex_string(unsigned char* bytes, size_t size, bool upcase);
 }  extern const char* hex_digits;
   
 int pa_get_valid_file_options_count(HashStringValue& options);  const char *pa_idna_encode(const char *in, Charset &source);
   const char *pa_idna_decode(const char *in, Charset &source);
   
 // some stuff for use with .for_each  unsigned long pa_crc32(const char *in, size_t in_size);
 static void copy_all_overwrite_to(  unsigned long pa_crc32(const String& file_spec);
                                                                 HashStringValue::key_type key,   
                                                                 HashStringValue::value_type value,   
                                                                 HashStringValue* dest) {  
         dest->put(key, value);  
 }  
   
 static void remove_key_from(  /** 
                                                         HashStringValue::key_type key,           Mix functions
                                                         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));  
 }  
   
 static int lastposafter(const String& s, size_t after, const char* substr, size_t substr_size, bool beforelast=false) {  #define PA_DEFAULT(A,B) ((A) ? (A):(B) )
         size_t size=0; // just to calm down compiler  
         if(beforelast)  
                 size=s.length();  
         size_t at;  
         while((at=s.pos(String::Body(substr), after))!=STRING_NOT_FOUND) {  
                 size_t newafter=at+substr_size/*skip substr*/;  
                 if(beforelast && newafter==size)  
                         break;  
                 after=newafter;  
         }  
   
         return after;  Charset* detect_charset(const char* content_type);
 }  
   
 // globals  // globals
   
Line 405  extern const String file_status_name; Line 281  extern const String file_status_name;
   
 // global defines for file options which are handled but not checked elsewhere, we check them  // global defines for file options which are handled but not checked elsewhere, we check them
   
 #define PA_SQL_LIMIT_NAME "limit"  
 #define PA_SQL_OFFSET_NAME "offset"  
 #define PA_COLUMN_SEPARATOR_NAME "separator"  #define PA_COLUMN_SEPARATOR_NAME "separator"
 #define PA_COLUMN_ENCLOSER_NAME "encloser"  #define PA_COLUMN_ENCLOSER_NAME "encloser"
 #define PA_CHARSET_NAME "charset"  #define PA_CHARSET_NAME "charset"
   #define PA_RESPONSE_CHARSET_NAME "response-charset"
   
 // globals defines for sql options  // globals defines for sql options
   
 #define SQL_BIND_NAME "bind"  #define SQL_BIND_NAME "bind"
   #define PA_SQL_LIMIT_NAME "limit"
   #define PA_SQL_OFFSET_NAME "offset"
 #define SQL_DEFAULT_NAME "default"  #define SQL_DEFAULT_NAME "default"
 #define SQL_DISTINCT_NAME "distinct"  #define SQL_DISTINCT_NAME "distinct"
 #define SQL_VALUE_TYPE_NAME "type"  #define SQL_VALUE_TYPE_NAME "type"
   
   extern String sql_bind_name;
   extern String sql_limit_name;
   extern String sql_offset_name;
   extern String sql_default_name;
   extern String sql_distinct_name;
   extern String sql_value_type_name;
   
 #ifndef DOXYGEN  #ifndef DOXYGEN
 enum Table2hash_distint { D_ILLEGAL, D_FIRST };  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
   
 #endif  #endif

Removed from v.1.144  
changed lines
  Added in v.1.192


E-mail: