HTML tag closure detection and Repair

HTML tag closure detection and repair, said a bit large, and did not consider a very complete, no use of regular expressions, applicable to the HTML file only the start tag does not have the end tag, there is an end tag is not the beginning of the tag case. The position of the label closure needs to be adjusted according to demand $str = ' "> Content Content full This is content This is content This is cont This is content This is content '; $str _len = strlen ($STR); Record start tag $pre _data = Array (); Record Start label position $pre _pos = Array (); $last _data = Array (); $error _data = Array (); $error _pos = Array (); $i = 0; Mark as < start $start _flag = false; while ($i < $str _len) { if ($str [$i]== < && $str [$i +1]!= '/' && $str [$i +1]!= '! ') { $i + +; $_tmp_str = "; Mark as < start $start _flag = true; Mark Blank $space _flag = false; while ($str [$i]!= > && $str [$i]!= "'" && $str [$i]!= ' "' && $str [$i]! = '/' && $i <$ Str_len) { if ($str [$i]== ') { $space _flag = true; } if (! $space _flag) { $_tmp_str. = $str [$i]; } $i + +; } $pre _data[] = $_tmp_str; $pre _pos[] = $i; } else if ($str [$i]== < && $str [$i +1]== '/') { $i + = 2; $_tmp_str = "; while ($str [$i]!= > && $i < $str _len) { $_tmp_str. = $str [$i]; $i + +; } $last _data[] = $_tmp_str; View the previous value of the start tag if (count ($pre _data) >0) { $last _pre_node = Getlastnode ($pre _data, 1); if ($last _pre_node = = $_tmp_str) { Pairing, delete the value of the corresponding position Array_pop ($pre _data); Array_pop ($pre _pos); Array_pop ($last _data); } else { There are two cases of no pairing. Case one: Only closed tags, no start tag Case two: Only start tag, no closed tag Array_pop ($last _data); $error _data[] = $_tmp_str; $error _pos[] = $i; } } else { Array_pop ($last _data); $error _data[] = $_tmp_str; $error _pos[] = $i; } }else if ($str [$i]== < && $str [$i +1]== "!") { $i + +; while ($i < $str _len) { if ($str [$i]== "-" && $str [$i +1]== "-" && $str [$i +2]== ">") { $i + +; Break } else { $i + +; } } $i + +; }else if ($str [$i]== '/' && $str [$i +1]== ' > ') { Skip Automatic Single closed labels if ($start _flag) { Array_pop ($pre _data); Array_pop ($pre _pos); $i +=2; } }else if ($str [$i]== "/" && $str [$i +1]== "*") { $i + +; while ($i < $str _len) { if ($str [$i]== "*" && $str [$i +1]== "/") { $i + +; Break } else { $i + +; } $i + +; } }else if ($str [$i]== "'") { $i + +; while ($str [$i]!= "'" && $i < $str _len) { $i + +; } $i + +; } else if ($str [$i]== ' ") { $i + +; while ($str [$i]!= ' "' && $i < $str _len) { $i + +; } $i + +; } else { $i + +; } } Determine the location of the start tag function Confirm_pre_pos ($str, $pre _pos) { $str _len = strlen ($STR); $j = $pre _pos; while ($j < $str _len) { if ($str [$j] = = ' "') { $j + +; while ($j < $str _len) { if ($str [$j]== ' "') { $j + +; Break } $j + +; } } else if ($str [$j] = = "'") { $j + +; while ($j < $str _len) { if ($str [$j]== "'") { $j + +; Break } $j + +; } } else if ($str [$j]== ">") { $j + +; while ($j < $str _len) { if ($str [$j]== "<") { Back to original content location $j--; Break } $j + +; } Break } else { $j + +; } } return $j; } Determine the location of the start tag function Confirm_err_pos ($str, $err _pos) { $j = $err _pos; $j--; while ($j > 0) { if ($str [$j] = = ' "') { $j--; while ($j < $str _len) { if ($str [$j]== ' "') { $j--; Break } $j--; } } else if ($str [$j] = = "'") { $j--; while ($j < $str _len) { if ($str [$j]== "'") { $j--; Break } $j--; } } else if ($str [$j]== ">") { $j + +; Break } else { $j--; } } return $j; } Gets the value of the penultimate number of the array function Getlastnode (array $arr, $num) { $len = count ($arr); if ($len > $num) { return $arr [$len-$num]; } else { return $arr [0]; } } Collate the data, primarily looking backwards, for further inspection Function Sort_data (& $pre _data, & $pre _pos, & $error _data, & $error _pos) { $rem _key_array = Array (); $rem _i_array = Array (); Get the value you want to delete foreach ($error _data as $key = + $value) { $count = count ($pre _data); for ($i = ($count-1); $i >=0; $i-) { if ($pre _data[$i] = = $value &&!in_array ($i, $rem _i_array)) { $rem _key_array[] = $key; $rem _i_array[] = $i; Break } } } Remove the corresponding value from the start tag foreach ($rem _key_array as $_item) { Unset ($error _pos[$_item]); Unset ($error _data[$_item]); } Delete the corresponding value of the end tag foreach ($rem _i_array as $_item) { Unset ($pre _data[$_item]); Unset ($pre _pos[$_item]); } } Organize data, close labels function Modify_data ($str, $pre _data, $pre _pos, $error _data, $error _pos) { $move _log = Array (); Only closed-label data foreach ($error _data as $key = + $value) { # code ... $_tmp_move_count = 0; foreach ($move _log as $pos _key = $move _value) { # code ... if ($error _pos[$key]>= $pos _key) { $_tmp_move_count + = $move _value; } } $data = Insert_data ($str, $value, $error _pos[$key]+$_tmp_move_count, false); $STR = $data [' str ']; $move _log[$data [' pos ']] = $data [' Move_count ']; } Data only from the start tag foreach ($pre _data as $key = + $value) { # code ... $_tmp_move_count = 0; foreach ($move _log as $pos _key = $move _value) { # code ... if ($pre _pos[$key]>= $pos _key) { $_tmp_move_count + = $move _value; } } $data = Insert_data ($str, $value, $pre _pos[$key]+$_tmp_move_count, True); $STR = $data [' str ']; $move _log[$data [' pos ']] = $data [' Move_count ']; } return $str; } Insert data, $type indicate how data is inserted function Insert_data ($str, $insert _data, $pos, $type) { $len = strlen ($STR); Start tag Type if ($type ==true) { $move _count = strlen ($insert _data) +3; $pos = Confirm_pre_pos ($str, $pos); $pre _str = substr ($str, 0, $pos); $end _str = substr ($str, $pos); $mid _str = " "; Closed label Type } else { $pos = Confirm_err_pos ($str, $pos); $move _count = strlen ($insert _data) + 2; $pre _str = substr ($str, 0, $pos); $end _str = substr ($str, $pos); $mid _str = "<". $insert _data. ">"; } $str = $pre _str. $mid _str. $end _str; return Array (' str ' = $str, ' pos ' = = $pos, ' move_count ' + $move _count); } Sort_data ($pre _data, $pre _pos, $error _data, $error _pos); $new _str = Modify_data ($str, $pre _data, $pre _pos, $error _data, $error _pos); echo $new _str; Print_r ($pre _data); Print_r ($pre _pos); Print_r ($error _data); Print_r ($error _pos); echo strlen ($STR); foreach ($pre _pos as $value) { $value = Confirm_pre_pos ($str, $value); for ($i = $value-5; $i <= $value; $i + +) { echo $str [$i]; // } echo "\ n"; // } foreach ($error _pos as $value) { for ($i = $value-5; $i <= $value; $i + +) { echo $str [$i]; // } echo "\ n"; // } ?> Copy Code

