|
|
| version 1.95, 2002/02/20 12:40:24 | version 1.115.4.1, 2003/03/17 14:48:06 |
|---|---|
| Line 1 | Line 1 |
| /** @file | /** @file |
| Parser: String class part: untaint mechanizm. | Parser: String class part: untaint mechanizm. |
| Copyright(c) 2001, 2002 ArtLebedev Group (http://www.artlebedev.com) | Copyright(c) 2001, 2003 ArtLebedev Group (http://www.artlebedev.com) |
| Author: Alexandr Petrosian <paf@design.ru> (http://paf.design.ru) | Author: Alexandr Petrosian <paf@design.ru> (http://paf.design.ru) |
| $Id$ | |
| */ | */ |
| static const char* IDENT_UNTAINT_C="$Date$"; | |
| #include "pa_pool.h" | #include "pa_pool.h" |
| #include "pa_string.h" | #include "pa_string.h" |
| #include "pa_hash.h" | #include "pa_hash.h" |
| Line 24 | Line 24 |
| ulong string_string_shortcut_economy=0; | ulong string_string_shortcut_economy=0; |
| #endif | #endif |
| #define escape(action) \ | #define escape(action) \ |
| { \ | { \ |
| const char *src=row->item.ptr; \ | const char *src=row->item.ptr; \ |
| Line 48 ulong string_string_shortcut_economy=0; | Line 47 ulong string_string_shortcut_economy=0; |
| dest+=bsize; \ | dest+=bsize; \ |
| inline bool need_file_encode(unsigned char c){ | inline bool need_file_encode(unsigned char c){ |
| // theoretical problem with, for instance, "_2B" and "." fragments, | // russian letters and space ENABLED |
| // they would yield the same | // encoding only these... |
| // because need_file_encode('_')=false | return strchr( |
| // but we need to delete such files somehow, getting names from ^index | "*?'\"<>|" |
| #ifndef WIN32 | |
| if((c>='0') &&(c<='9') ||(c>='A') &&(c<='Z') ||(c>='a') &&(c<='z')) | ":\\~" |
| return false; | |
| return !strchr( | |
| " _./()-" | |
| #ifdef WIN32 | |
| ":\\~" | |
| #endif | #endif |
| , c); | , c)!=0; |
| } | } |
| inline bool need_uri_encode(unsigned char c){ | inline bool need_uri_encode(unsigned char c){ |
| if((c>='0') &&(c<='9') ||(c>='A') &&(c<='Z') ||(c>='a') &&(c<='z')) | if((c>='0') &&(c<='9') ||(c>='A') &&(c<='Z') ||(c>='a') &&(c<='z')) |
| Line 157 inline bool need_quote_http_header(const | Line 150 inline bool need_quote_http_header(const |
| if no language-change specified and src not yet appended to some other string[last_chunk!=0] | if no language-change specified and src not yet appended to some other string[last_chunk!=0] |
| shrinking dest last_chunk[preparing it for linking], | shrinking dest last_chunk[preparing it for linking], |
| ///shrinking src last_chunk[preparing it to be linked, consequent dest.appends would go there], | ///shrinking src last_chunk[preparing it to be linked, consequent dest.appends would go there], |
| linking[dest.last_chunk = src.head] | linking[dest.last_chunk = src.head.chunk] |
| if some language-change specified or src already appended to some other string[last_chunk==0] | if some language-change specified or src already appended to some other string[last_chunk==0] |
| cloning pieces. | cloning pieces. |
| */ | */ |
| String& String::append(const String& src, uchar lang, bool forced) { | String& String::append(const String& src, uchar lang, bool forced) { |
| if(!last_chunk) // growth stopped [we're appended as string to somebody] | if(!last_chunk) // growth stopped [we're appended as string to somebody] |
| throw Exception(0, 0, | throw Exception(0, |
| this, | this, |
| "string growth stopped (append string)"); | "string growth stopped (append string)"); |
| Line 179 String& String::append(const String& src | Line 172 String& String::append(const String& src |
| #endif | #endif |
| // using fact: | // using fact: |
| // src.head.count initally equeals this.head.count and shrinks-only, | // src.head.chunk.count initally equeals this.head.chunk.count and shrinks-only, |
| // so can't be more than this.head.count, | // so can't be more than this.head.chunk.count, |
| // which means that we know that | // which means that we know that |
| // src.head would fit into this.head | // src.head.chunk would fit into this.head.chunk |
| if(is_empty()) { // our head is empty | if(is_empty()) { // our head.chunk is empty |
| // they have more than head? we need all head : we need only filled-part of head | // they have more than head.chunk? we need all head.chunk : we need only filled-part of head.chunk |
| Chunk *src_head_link=src.head.rows[src.head.count].link; | Chunk *src_head_link=src.head.chunk.rows[src.head.chunk.count].link; |
| size_t head_count=src_head_link?src.head.count:(src.append_here-src.head.rows); | size_t head_count=src_head_link?src.head.chunk.count:(src.append_here-src.head.chunk.rows); |
| // "your head is my head" | // "your head.chunk is my head.chunk" |
| memcpy(head.rows, src.head.rows, sizeof(Chunk::Row)*(head_count)); | memcpy(head.chunk.rows, src.head.chunk.rows, sizeof(Chunk::Row)*(head_count)); |
| if(src_head_link) { | if(src_head_link) { |
| // "your body is my body" | // "your body is my body" |
| head.rows[head.count=head_count].link=src_head_link; | head.chunk.rows[head.chunk.count=head_count].link=src_head_link; |
| // "your last_chunk is mine now" | // "your last_chunk is mine now" |
| last_chunk=src.last_chunk; | last_chunk=src.last_chunk; |
| // "your append_here is mine now" | // "your append_here is mine now" |
| append_here=src.append_here; | append_here=src.append_here; |
| } else { | } else { |
| // "your last_chunk is mine now" | // "your last_chunk is mine now" |
| last_chunk=&head; | last_chunk=&head.chunk; |
| // "your append_here is recalc-mine now" | // "your append_here is recalc-mine now" |
| append_here=head.rows+head_count; | append_here=head.chunk.rows+head_count; |
| } | } |
| } else { // our head contains something | } else { // our head.chunk contains something |
| // "chopping off my tail-reserve" | // "chopping off my tail-reserve" |
| last_chunk->count=append_here-last_chunk->rows; | |
| // "you is my tail" | // "you is my tail" |
| last_chunk->rows[last_chunk->count=append_here-last_chunk->rows].link=&src.head; | append_here->link=&src.head.chunk; |
| // "your last_chunk is mine now" | // "your last_chunk is mine now" |
| last_chunk=src.last_chunk; | last_chunk=src.last_chunk; |
| // "your append_here is mine now" | // "your append_here is mine now" |
| Line 248 String& String::append(const String& src | Line 242 String& String::append(const String& src |
| row->item.lang==UL_TAINTED?lang:row->item.lang, | row->item.lang==UL_TAINTED?lang:row->item.lang, |
| row->item.origin.file, row->item.origin.line); | row->item.origin.file, row->item.origin.line); |
| ); | ); |
| break2: | /* |
| for(Chunk::Row *row=last_chunk->rows; row<append_here; row++) | |
| if(row->link==(void*)0xcdcdcdcd) | |
| _asm int 3;*/ | |
| return *this; | return *this; |
| } | } |
| Line 287 size_t String::cstr_bufsize(Untaint_lang | Line 284 size_t String::cstr_bufsize(Untaint_lang |
| case UL_MAIL_HEADER: | case UL_MAIL_HEADER: |
| // tainted, untaint language: mail-header | // tainted, untaint language: mail-header |
| if(buf_charset) { | if(buf_charset) { |
| int parts_count=1; | |
| const char *prev=row->item.ptr; | |
| size_t size=row->item.size; | |
| while(const char *comma_at=(const char*)memchr(prev, ',', size)) { | |
| parts_count++; | |
| size-=comma_at-prev; | |
| prev=comma_at+1; | |
| } | |
| // Subject: Re: parser3: =?koi8-r?Q?=D3=C5=CD=C9=CE=C1=D2?= | // Subject: Re: parser3: =?koi8-r?Q?=D3=C5=CD=C9=CE=C1=D2?= |
| dest+= | dest+= |
| row->item.size*3+ | row->item.size*3+ |
| buf_charset->name().size()+MAX_STRING/* worst: =?charset?Q?=%XX?= */; | parts_count*(buf_charset->name().size()+MAX_STRING/* worst: =?charset?Q?=%XX?= */); |
| } else | } else |
| dest+=row->item.size; | dest+=row->item.size; |
| break; | break; |
| Line 332 size_t String::cstr_bufsize(Untaint_lang | Line 337 size_t String::cstr_bufsize(Untaint_lang |
| break; | break; |
| } | } |
| ); | ); |
| break2: | |
| return dest; | return dest; |
| } | } |
| /** http://www.ietf.org/rfc/rfc2047.txt | |
| RFC | |
| (3) As a replacement for a 'word' entity within a 'phrase', for example, | |
| one that precedes an address in a From, To, or Cc header. The ABNF | |
| definition for 'phrase' from RFC 822 thus becomes: | |
| phrase = 1*( encoded-word / word ) | |
| In this case the set of characters that may be used in a "Q"-encoded | |
| 'encoded-word' is restricted to: <upper and lower case ASCII | |
| letters, decimal digits, "!", "*", "+", "-", "/", "=", and "_" | |
| (underscore, ASCII 95.)>. An 'encoded-word' that appears within a | |
| 'phrase' MUST be separated from any adjacent 'word', 'text' or | |
| 'special' by 'linear-white-space'. | |
| ... | |
| (2) The 8-bit hexadecimal value 20 (e.g., ISO-8859-1 SPACE) may be | |
| represented as "_" (underscore, ASCII 95.). (This character may | |
| not pass through some internetwork mail gateways, but its use | |
| will greatly enhance readability of "Q" encoded data with mail | |
| readers that do not support this encoding.) Note that the "_" | |
| always represents hexadecimal 20, even if the SPACE character | |
| occupies a different code position in the character set in use. | |
| paf: obviously, | |
| without "=", or one could not differ "=E0" and "russian letter a" | |
| and without "_", or in would mean 0x20 | |
| */ | |
| static bool mail_header_char_valid_within_Qencoded(char c) { | |
| return c>='A' && c<='Z' | |
| || c>='a' && c<='Z' | |
| || c>='0' && c<='9' | |
| || strchr("!*+-/", c); | |
| } | |
| char *String::store_to(char *dest, Untaint_lang lang, | char *String::store_to(char *dest, Untaint_lang lang, |
| SQL_Connection *connection, | SQL_Connection *connection, |
| Charset *store_to_charset) const { | Charset *store_to_charset, |
| const char *store_to_charset_name) const { | |
| // WARNING: | // WARNING: |
| // before any changes check cstr_bufsize first!!! | // before any changes check cstr_bufsize first!!! |
| bool whitespace=true; | bool whitespace=true; |
| // expanded STRING_FOREACH_ROW here for debugging purposes | //debug |
| const Chunk *chunk=&head; \ | const Chunk *chunk=&(*this).head.chunk; \ |
| do { \ | const Chunk::Row *row=chunk->rows; \ |
| const Chunk::Row *row=chunk->rows; \ | uint countdown=chunk->count; \ |
| for(uint i=0; i<chunk->count; i++, row++) { \ | while(row!=(*this).append_here) { \ |
| if(row==append_here) \ | if(countdown==0) { \ |
| goto break2; \ | chunk=row->link; \ |
| \ | row=chunk->rows; \ |
| countdown=chunk->count; \ | |
| }; | |
| //*/ | |
| // STRING_FOREACH_ROW( | |
| uchar to_lang=lang==UL_UNSPECIFIED?row->item.lang:lang; | uchar to_lang=lang==UL_UNSPECIFIED?row->item.lang:lang; |
| char *start=dest; | char *start=dest; |
| Line 371 char *String::store_to(char *dest, Untai | Line 414 char *String::store_to(char *dest, Untai |
| case UL_FILE_SPEC: | case UL_FILE_SPEC: |
| // tainted, untaint language: file [name] | // tainted, untaint language: file [name] |
| escape( | escape( |
| encode(need_file_encode, '_'); | // Macintosh has problems with small Russian letter 'r' |
| if( *src=='\xF0' && pool().get_source_charset().name()=="windows-1251" ) { | |
| // fixing that letter for most common charset | |
| to_char('p'); | |
| } else // fallback to default | |
| encode(need_file_encode, '_'); | |
| ); | ); |
| break; | break; |
| case UL_URI: | case UL_URI: |
| Line 384 char *String::store_to(char *dest, Untai | Line 432 char *String::store_to(char *dest, Untai |
| { | { |
| const char *src=(const char *)client_ptr; | const char *src=(const char *)client_ptr; |
| for(int size=client_size; size--; src++) | for(int size=client_size; size--; src++) |
| switch(*src) { | encode(need_uri_encode, '%'); |
| case ' ': to_char('+'); break; | |
| default: encode(need_uri_encode, '%'); | |
| }; | |
| } | } |
| break; | break; |
| case UL_HTTP_HEADER: | case UL_HTTP_HEADER: |
| // tainted, untaint language: http-field-content-text | // tainted, untaint language: http-field-content-text |
| escape(switch(*src) { | escape( |
| case ' ': to_char('+'); break; | encode(need_uri_encode, '%'); |
| default: encode(need_uri_encode, '%'); | ); |
| }); | |
| break; | break; |
| case UL_MAIL_HEADER: | case UL_MAIL_HEADER: |
| // tainted, untaint language: mail-header | // tainted, untaint language: mail-header |
| if(store_to_charset) { | // http://www.ietf.org/rfc/rfc2047.txt |
| if(store_to_charset && store_to_charset_name) { | |
| const void *mail_ptr; | const void *mail_ptr; |
| size_t mail_size; | size_t mail_size; |
| Charset::transcode(pool(), | Charset::transcode(pool(), |
| Line 409 char *String::store_to(char *dest, Untai | Line 454 char *String::store_to(char *dest, Untai |
| // Subject: Re: parser3: =?koi8-r?Q?=D3=C5=CD=C9=CE=C1=D2?= | // Subject: Re: parser3: =?koi8-r?Q?=D3=C5=CD=C9=CE=C1=D2?= |
| const char *src=(const char *)mail_ptr; | const char *src=(const char *)mail_ptr; |
| bool to_quoted_printable=false; | bool to_quoted_printable=false; |
| for(int size=mail_size; size--; src++) { | |
| if(*src & 0x80) { | bool email=false; |
| for(const char *end=src+mail_size; src<end; src++) { | |
| //RFC + An 'encoded-word' MUST NOT appear in any portion of an 'addr-spec'. | |
| if(to_quoted_printable && (*src==',' || *src=='<')) { | |
| email=*src=='<'; | |
| dest+=sprintf(dest, "?="); | |
| to_quoted_printable=false; | |
| } | |
| if(!email && ( | |
| !to_quoted_printable && (*src & 0x80) // starting quote-printable-encoding on first 8bit char | |
| || to_quoted_printable && !mail_header_char_valid_within_Qencoded(*src) | |
| )) { | |
| if(!to_quoted_printable) { | if(!to_quoted_printable) { |
| dest+=sprintf(dest, "=?%s?Q?", store_to_charset->name().cstr()); | dest+=sprintf(dest, "=?%s?Q?", store_to_charset_name); |
| to_quoted_printable=true; | to_quoted_printable=true; |
| } | } |
| dest+=sprintf(dest, "=%02X", *src & 0xFF); | //RFC Upper case should be used for hexadecimal digits "A" through "F" |
| } else { | if(*src == 0x20) // RFC The 8-bit hexadecimal value 20 (e.g., ISO-8859-1 SPACE) |
| *dest++=*src; | *dest++='_'; // RFC may be represented as "_" |
| } | else |
| dest+=sprintf(dest, "=%02X", *src & 0xFF); | |
| } else | |
| *dest++=*src; | |
| if(*src=='>') | |
| email=false; | |
| } | } |
| if(to_quoted_printable) // close | if(to_quoted_printable) // close |
| dest+=sprintf(dest, "?="); | dest+=sprintf(dest, "?="); |
| Line 441 char *String::store_to(char *dest, Untai | Line 502 char *String::store_to(char *dest, Untai |
| if(connection) | if(connection) |
| dest+=connection->quote(dest, row->item.ptr, row->item.size); | dest+=connection->quote(dest, row->item.ptr, row->item.size); |
| else | else |
| throw Exception(0, 0, | throw Exception(0, |
| this, | this, |
| "untaint in SQL language failed - no connection specified"); | "untaint in SQL language failed - no connection specified"); |
| break; | break; |
| Line 475 char *String::store_to(char *dest, Untai | Line 536 char *String::store_to(char *dest, Untai |
| }); | }); |
| break; | break; |
| default: | default: |
| throw Exception(0, 0, | throw Exception(0, |
| this, | this, |
| "unknown untaint language #%d", | "unknown untaint language #%d", |
| static_cast<int>(row->item.lang)); // sould never | static_cast<int>(to_lang)); // should never |
| break; // never | break; // never |
| } | } |
| Line 500 char *String::store_to(char *dest, Untai | Line 561 char *String::store_to(char *dest, Untai |
| break; | break; |
| }; | }; |
| } else // piece without optimization | } else // piece without optimization |
| //debug | |
| whitespace=false; | whitespace=false; |
| row++; countdown--; \ | |
| } \ | } |
| chunk=row->link; \ | /* |
| } while(chunk); \ | ); |
| */ | |
| break2: | |
| return dest; | return dest; |
| } | } |
| char *String::cstr_debug_origins() const { | char *String::cstr_debug_origins() const { |
| //_asm int 3; | //_asm int 3; |
| char *result=(char *)malloc(size()+used_rows()*MAX_STRING*2); | char *result=(char *)malloc_atomic(size()+used_rows()*MAX_STRING*2); |
| char *dest=result; | char *dest=result; |
| const Chunk *chunk=&head; | STRING_FOREACH_ROW( |
| do { | IFNDEF_NO_STRING_ORIGIN( |
| const Chunk::Row *row=chunk->rows; | if(row->item.origin.file) |
| for(uint i=0; i<chunk->count; i++, row++) { | dest+=sprintf(dest, ORIGIN_FILE_LINE_FORMAT, |
| if(row==append_here) | row->item.origin.file, |
| goto break2; | 1+row->item.origin.line); |
| else | |
| #ifndef NO_STRING_ORIGIN | dest+=sprintf(dest, "<unknown>"); |
| if(row->item.origin.file) | ); |
| dest+=sprintf(dest, ORIGIN_FILE_LINE_FORMAT, | uchar show_lang=row->item.lang & ~UL_OPTIMIZE_BIT; |
| row->item.origin.file, | if(show_lang>=sizeof(String_Untaint_lang_name)/sizeof(String_Untaint_lang_name[0])) |
| 1+row->item.origin.line); | throw Exception(0, |
| else | this, |
| dest+=sprintf(dest, "<unknown>"); | "unknown untaint language #%d", |
| #endif | static_cast<int>(show_lang)); // sould never |
| uchar show_lang=row->item.lang & ~UL_OPTIMIZE_BIT; | |
| if(show_lang>=sizeof(String_Untaint_lang_name)/sizeof(String_Untaint_lang_name[0])) | |
| throw Exception(0, 0, | |
| this, | |
| "unknown untaint language #%d", | |
| static_cast<int>(show_lang)); // sould never | |
| dest+=sprintf(dest, "#%s%s: ", | |
| String_Untaint_lang_name[show_lang], | |
| row->item.lang & UL_OPTIMIZE_BIT?".O":""); | |
| char *dest_after_origins=dest; | |
| memcpy(dest, row->item.ptr, row->item.size); | dest+=sprintf(dest, "#%s%s: ", |
| dest+=row->item.size; | String_Untaint_lang_name[show_lang], |
| row->item.lang & UL_OPTIMIZE_BIT?".O":""); | |
| char *dest_after_origins=dest; | |
| remove_crlf(dest_after_origins, dest); | memcpy(dest, row->item.ptr, row->item.size); |
| to_char('\n'); | dest+=row->item.size; |
| } | |
| chunk=row->link; | remove_crlf(dest_after_origins, dest); |
| } while(chunk); | to_char('\n'); |
| ); | |
| break2: | |
| *dest=0; | *dest=0; |
| return result; | return result; |
| } | } |