php 实现文本文件差异比较及自动合并文件
大家都用过git和svn，都知道本地代码与代码服务器的代码不一样的时候会自动合并，非常方便，我想开发一个版本管理系统，所以用到了代码比较及合并的功能，代码差异比较用的是diff，但是这个库不提供合并功能，我就在他的基础上写了一个合并的功能
<?php
class Diff{
// define the constants
const UNMODIFIED = 0;
const DELETED = 1;
const INSERTED = 2;
/* Returns the diff for two strings. The return value is an array, each of
* whose values is an array containing two values: a line (or character, if
* $compareCharacters is true), and one of the constants DIFF::UNMODIFIED (the
* line or character is in both strings), DIFF::DELETED (the line or character
* is only in the first string), and DIFF::INSERTED (the line or character is
* only in the second string). The parameters are:
*
* $string1 - the first string
* $string2 - the second string
* $compareCharacters - true to compare characters, and false to compare
* lines; this optional parameter defaults to false
*/
public static function compare(
$string1, $string2, $compareCharacters = false){
// initialise the sequences and comparison start and end positions
$start = 0;
if ($compareCharacters){
$sequence1 = $string1;
$sequence2 = $string2;
$end1 = strlen($string1) - 1;
$end2 = strlen($string2) - 1;
}else{
$sequence1 = preg_split('/\R/', $string1);
$sequence2 = preg_split('/\R/', $string2);
$end1 = count($sequence1) - 1;
$end2 = count($sequence2) - 1;
}
// skip any common prefix
while ($start <= $end1 && $start <= $end2
&& $sequence1[$start] == $sequence2[$start]){
$start ++;
}
// skip any common suffix
while ($end1 >= $start && $end2 >= $start
&& $sequence1[$end1] == $sequence2[$end2]){
$end1 --;
$end2 --;
}
// compute the table of longest common subsequence lengths
$table = self::computeTable($sequence1, $sequence2, $start, $end1, $end2);
// generate the partial diff
$partialDiff =
self::generatePartialDiff($table, $sequence1, $sequence2, $start);
// generate the full diff
$diff = array();
for ($index = 0; $index < $start; $index ++){
$diff[] = array($sequence1[$index], self::UNMODIFIED);
}
while (count($partialDiff) > 0) $diff[] = array_pop($partialDiff);
for ($index = $end1 + 1;
$index < ($compareCharacters ? strlen($sequence1) : count($sequence1));
$index ++){
$diff[] = array($sequence1[$index], self::UNMODIFIED);
}
// return the diff
return $diff;
}
/* Returns the diff for two files. The parameters are:
*
* $file1 - the path to the first file
* $file2 - the path to the second file
* $compareCharacters - true to compare characters, and false to compare
* lines; this optional parameter defaults to false
*/
public static function compareFiles(
$file1, $file2, $compareCharacters = false){
// return the diff of the files
return self::compare(
file_get_contents($file1),
file_get_contents($file2),
$compareCharacters);
}
/* Returns the table of longest common subsequence lengths for the specified
* sequences. The parameters are:
*
* $sequence1 - the first sequence
* $sequence2 - the second sequence
* $start - the starting index
* $end1 - the ending index for the first sequence
* $end2 - the ending index for the second sequence
*/
private static function computeTable(
$sequence1, $sequence2, $start, $end1, $end2){
// determine the lengths to be compared
$length1 = $end1 - $start + 1;
$length2 = $end2 - $start + 1;
// initialise the table
$table = array(array_fill(0, $length2 + 1, 0));
// loop over the rows
for ($index1 = 1; $index1 <= $length1; $index1 ++){
// create the new row
$table[$index1] = array(0);
// loop over the columns
for ($index2 = 1; $index2 <= $length2; $index2 ++){
// store the longest common subsequence length
if ($sequence1[$index1 + $start - 1]
== $sequence2[$index2 + $start - 1]){
$table[$index1][$index2] = $table[$index1 - 1][$index2 - 1] + 1;
}else{
$table[$index1][$index2] =
max($table[$index1 - 1][$index2], $table[$index1][$index2 - 1]);
}
}
}
// return the table
return $table;
}
/* Returns the partial diff for the specificed sequences, in reverse order.
* The parameters are:
*
* $table - the table returned by the computeTable function
* $sequence1 - the first sequence
* $sequence2 - the second sequence
* $start - the starting index
*/
private static function generatePartialDiff(
$table, $sequence1, $sequence2, $start){
// initialise the diff
$diff = array();
// initialise the indices
$index1 = count($table) - 1;
$index2 = count($table[0]) - 1;
// loop until there are no items remaining in either sequence
while ($index1 > 0 || $index2 > 0){
// check what has happened to the items at these indices
if ($index1 > 0 && $index2 > 0
&& $sequence1[$index1 + $start - 1]
== $sequence2[$index2 + $start - 1]){
// update the diff and the indices
$diff[] = array($sequence1[$index1 + $start - 1], self::UNMODIFIED);
$index1 --;
$index2 --;
}elseif ($index2 > 0
&& $table[$index1][$index2] == $table[$index1][$index2 - 1]){
// update the diff and the indices
$diff[] = array($sequence2[$index2 + $start - 1], self::INSERTED);
$index2 --;
}else{
// update the diff and the indices
$diff[] = array($sequence1[$index1 + $start - 1], self::DELETED);
$index1 --;
}
}
// return the diff
return $diff;
}
/* Returns a diff as a string, where unmodified lines are prefixed by ' ',
* deletions are prefixed by '- ', and insertions are prefixed by '+ '. The
* parameters are:
*
* $diff - the diff array
* $separator - the separator between lines; this optional parameter defaults
* to "\n"
*/
public static function toString($diff, $separator = "\n"){
// initialise the string
$string = '';
// loop over the lines in the diff
foreach ($diff as $line){
// extend the string with the line
switch ($line[1]){
case self::UNMODIFIED : $string .= ' ' . $line[0];break;
case self::DELETED : $string .= '- ' . $line[0];break;
case self::INSERTED : $string .= '+ ' . $line[0];break;
}
// extend the string with the separator
$string .= $separator;
}
// return the string
return $string;
}
/**
* 合并代码生成新的文本
* @param unknown $diff
* @return string
*/
public static function merge($diff){
// initialise the HTML
$_code ="";
// loop over the lines in the diff
$index = 0;
$indentation="";
$separator="\n";
$_i=1;
while ($index < count($diff)){
// determine the line type
switch ($diff[$index][1]){
// display the content on the left and right
case self::UNMODIFIED:
$leftCell =
self::getCellcode(
$diff, $indentation, $separator, $index, self::UNMODIFIED);
$rightCell = $leftCell;
break;
// display the deleted on the left and inserted content on the right
case self::DELETED:
$leftCell =
self::getCellcode(
$diff, $indentation, $separator, $index, self::DELETED);
$rightCell =
self::getCellcode(
$diff, $indentation, $separator, $index, self::INSERTED);
break;
// display the inserted content on the right
case self::INSERTED:
$leftCell = '';
$rightCell =
self::getCellcode(
$diff, $indentation, $separator, $index, self::INSERTED);
break;
}
// extend the HTML with the new row
if($leftCell == $rightCell){
$_code.=$rightCell;
}else{
if($leftCell == ''){
$_code.="++\n".$rightCell."++\n";
}else{
$_code.="--\n".$leftCell."--\n";
if($rightCell!=""){
$_code.="++\n".$rightCell."++\n";
}
}
}
$_i++;
}
return $_code;
}
/* Returns a diff as an HTML string, where unmodified lines are contained
* within 'span' elements, deletions are contained within 'del' elements, and
* insertions are contained within 'ins' elements. The parameters are:
*
* $diff - the diff array
* $separator - the separator between lines; this optional parameter defaults
* to '<br>'
*/
public static function toHTML($diff, $separator = '<br>'){
// initialise the HTML
$html = '';
// loop over the lines in the diff
foreach ($diff as $line){
// extend the HTML with the line
switch ($line[1]){
case self::UNMODIFIED : $element = 'span'; break;
case self::DELETED : $element = 'del'; break;
case self::INSERTED : $element = 'ins'; break;
}
$html .=
'<' . $element . '>'
. htmlspecialchars($line[0])
. '</' . $element . '>';
// extend the HTML with the separator
$html .= $separator;
}
// return the HTML
return $html;
}
/* Returns a diff as an HTML table. The parameters are:
*
* $diff - the diff array
* $indentation - indentation to add to every line of the generated HTML; this
* optional parameter defaults to ''
* $separator - the separator between lines; this optional parameter
* defaults to '<br>'
*/
public static function toTable($diff, $indentation = '', $separator = '<br>'){
// initialise the HTML
$html = $indentation . "<table class=\"diff\">\n";
// loop over the lines in the diff
$index = 0;
while ($index < count($diff)){
// determine the line type
switch ($diff[$index][1]){
// display the content on the left and right
case self::UNMODIFIED:
$leftCell =
self::getCellContent(
$diff, $indentation, $separator, $index, self::UNMODIFIED);
$rightCell = $leftCell;
break;
// display the deleted on the left and inserted content on the right
case self::DELETED:
$leftCell =
self::getCellContent(
$diff, $indentation, $separator, $index, self::DELETED);
$rightCell =
self::getCellContent(
$diff, $indentation, $separator, $index, self::INSERTED);
break;
// display the inserted content on the right
case self::INSERTED:
$leftCell = '';
$rightCell =
self::getCellContent(
$diff, $indentation, $separator, $index, self::INSERTED);
break;
}
// extend the HTML with the new row
$html .=
$indentation
. " <tr>\n"
. $indentation
. ' <td class="diff'
. ($leftCell == $rightCell
? 'Unmodified'
: ($leftCell == '' ? 'Blank' : 'Deleted'))
. '">'
. $leftCell
. "</td>\n"
. $indentation
. ' <td class="diff'
. ($leftCell == $rightCell
? 'Unmodified'
: ($rightCell == '' ? 'Blank' : 'Inserted'))
. '">'
. $rightCell
. "</td>\n"
. $indentation
. " </tr>\n";
}
// return the HTML
return $html . $indentation . "</table>\n";
}
private static function getCellcode(
$diff, $indentation, $separator, &$index, $type){
$_code = '';
while ($index < count($diff) && $diff[$index][1] == $type){
$_code .=
$diff[$index][0] . $separator;
$index ++;
}
return $_code;
}
/* Returns the content of the cell, for use in the toTable function. The
* parameters are:
*
* $diff - the diff array
* $indentation - indentation to add to every line of the generated HTML
* $separator - the separator between lines
* $index - the current index, passes by reference
* $type - the type of line
*/
private static function getCellContent(
$diff, $indentation, $separator, &$index, $type){
// initialise the HTML
$html = '';
// loop over the matching lines, adding them to the HTML
while ($index < count($diff) && $diff[$index][1] == $type){
$html .=
'<span>'
. htmlspecialchars($diff[$index][0])
. '</span>'
. $separator;
$index ++;
}
// return the HTML
return $html;
}
}
?>
好了，怎么调用呢<?php $_mergestr=Diff::merge(Diff::compareFiles("a.txt", "b.txt")) ;
//将合并后的文件保存
file_put_contents("c.txt", $$_mergestr);
?>
{{collectdata}}
网友评论0