perl learning : xpath search
学习Perl过程中，昨天自行实现了XML文件的读取，将其内容放入到一散列中。
但至于其内容是否正确与否却不得而知，于是需要添加一些测试代码来验证。
以前写C/C++代码时，常用XPATH来搜索XML文件中的节点，这需要XML引擎来处理。
但现在读取XML文件采用的是“土方法”（自行实现的），所以这引擎也得自己来写喽了。
XPATH引擎，我这其实就一函数，给定XPATH字串和存有XML文件内容的散列（当然是用MyXMLSimple方法读取进来的），
就可以搜索到你想找的XML节点，当然返回值也是个节点。
XPATH是很强大的，哥哥这里只是实现了一个基本的功能：查找节点。
我这里支持的XPATH路径只有两种：全局搜索的路径，即以双斜杠开头的；或是从根开始搜索的，即以间斜杠或直接以节点TAG开头的。
节点可以指定属性的值。支持的XPATH样板：
//TAG[@name="value" or @name="value"]/TAG[@name="value" and @name="value"] #全局递归搜索
/TAG/TAG/TAG[@name=”value"] #从根开始搜索
TAG/TAG[@name="value"] #从根开始
其中的TAG表示节点的节点名字，name为节点的某一属性名字，value为某一属性的值，这些字串暂时不支持通配符查找。
程序中用来一些正则表达式，那个也刚学不太懂，所以一些正则表达式可能写得比较乱或是根本就是错误的，请原谅小弟的无知。
下面是Perl源程序。
####perl pragram###################################################
###############################my xml simple protected sub##############
#get node tag of a nodestring like <%TAG% ... >
#in param: the nodestring
#return value: the tag : %TAG%
sub GetNodeTag
{
my ($NodeString) = @_;
$NodeString =~ m/</s*//(/w+)/s*>/ or $NodeString =~ m/</s*(/w+).*?>/;
return $1;
}
#get the attributes of a node that description by a string like
#in param1: the nodestring
#in param2: a array to store the names and values of the attributes
#in the array: name1, value1, name2, value2, name3, value3,...
sub GetAtribList
{
my ($nodestring, $atribarray) = @_;
@$atribarray = $nodestring =~ m/(/w+)/s*=/s*/"(/w*?)/"/g;
}
#push the attributes of a node into a hash that reprent the node
#in param1: the nodestring
#in param2: the hash that will reprent the node
sub pushattributes
{
my ($nodestring, $hash) = @_;
my @tsarray = ();
GetAtribList($nodestring, /@tsarray);
if (@tsarray)
{
for (my $j = 0; $j < @tsarray;)
{
my $key = @tsarray[$j];
my $value = @tsarray[$j+1];
$$hash{$key} = $value;
$j += 2;
}
}
}
#push a couple of key and value to a hash
#in param1: the key
#in param2: the value
#in param3: the hash
sub pushhash
{
my ($key, $value, $hash) = @_;
if ($hash =~ /^HASH/)
{
if (exists
if ($ekvalue =~ /^ARRAY/)
{
my @array=@{$ekvalue};
my $index = @array;
}
}
else
{
$$hash{$key} = $value;
}
}
} #generate a xml dom hash by a nodearray that contained the nodestring of the xml file
#in param1: the nodearray that contained all the nodestring in the xml file <...>
#in param2: the begin pos to generate the xml dom hash in the nodearray
#in param3: the end pos to generate the xml dom hash in the nodearray
#in param4: the hash
sub GenerateXMLDom
{
my ($nodearray, $begin, $length, $root)=@_;
for (my $i=$begin; $i<$begin + $length; )
{
$temp = $$nodearray[$i];
my ($nodetag) = GetNodeTag($temp);
if ($temp =~ m////s*>/) #single node
{
my %tshash1 = ();
pushattributes($temp, /%tshash1);
pushhash($nodetag, /%tshash1, $root);
$i++;
next;
}
elsif ($temp =~ m/</s*///) #end of complex node
{
$i++;
next;
}
else #complex node
{
my %tshash2 = ();
pushattributes($temp, /%tshash2);
my $headnum = 1;
my $endnum = 0;
my $start = $i;
while ($endnum != $headnum && $i < $begin + $length-1)
{
$i++;
my $nextnode = $$nodearray[$i];
if ($nextnode =~ m////s*>/) #single node
{
$headnum++;
$endnum++;
}
elsif ($nextnode =~ m/</s*///) #end of complex node
{
$endnum++;
}
else
{
$headnum++;
}
}
$i++;
my $len = $i - $start;
GenerateXMLDom(/@nodearray, $start+1, $len-2, /%tshash2);
pushhash($nodetag, /%tshash2, $root);
next;
}
}
}
###############################end of my xml simple protected sub##############
################################my xml simple public sub#####################
#my xml simple, to read a xml file into a hash
#in param: the xml file to read
#return value: the hash that contained the xml dom content
sub MyXMLSimple
{
my ($xmlfile) = @_;
open (INFILE, "$xmlfile") or die "can't open $xmlfile for reading $!";
while ($line=)
{
chomp $line;
$filecontent.=$line;
}
close INFILE;
#delete xml comment
$filecontent =~ s///g;
#cut file context by <...>
@nodearray = $filecontent =~ m/<.*?>/g;
$length = @nodearray;
%rootnode = (); #the root node of the xml file
#construct the xml dom struct
GenerateXMLDom(/@nodearray, 0, $length, /%rootnode);
return %rootnode;
}
########################end of my xml simple#############################
###########################xpath protected sub###########################
#use to search the node by attributes, the node tag has already matched
#in param1:search xpath suchas TAGNAME[@attributename="attributevalue" and @attributename="attributevalue" or @attributename="attributevalue" ...]
#in param2:a hash with the key of TAGNAME
#in param3:a array to store the find node which was reprented by a hash
#return value: 1 means has found the node match the attributes or 0 means not found
sub SearchNodeByAttribute
{
my ($TagString, $hash, $result) = @_;
#get the attribute name list
#the element in the array is such as : name, =|!=, value, name, =|!=, value,...
my @attributes = $TagString =~ m//@(/w+)/s*(!?=)/s*[/"|'](.*?)[/"|']/g;
#get the logical relations between attributes if existed more then one attributes
my $index = index($TagString, "[");
my $endex = index($TagString, "]");
my $attrstr;
if ($index != -1 && $endex != -1)
{
my $len = $endex - $index - 1;
$attrstr = substr($TagString, $index+1, $len);
}
#the logical relations list
#a sample : and,or,and,or
my (@atrirelation) = $attrstr =~ m//@/w+/s*!?=/s*[/"|'].*?[/"|']/s*(and|or)/s*/g;
#if do not give the attributes, then found the node
if (@attributes <= 0)
{
push (@$result, $hash);
return 1;
}
#compare the existed attributes with the given attributes and store the result to a array
my @tparray = ();
for (my $i=0; $i < @attributes;)
{
my $attname = $attributes[$i];
my $oper = $attributes[$i+1];
my $attvalue = $attributes[$i+2];
my $exvalue = $$hash{$attname};
my $b = 0;
if ($oper =~ //s*=/s*/)
{
if ($exvalue =~ m/$attvalue/)
{
$b = 1;
}
}
else
{
if ($exvalue =~ m/$attvalue/)
{
$b = 0;
}
else
{
$b = 1;
}
}
push(@tparray, $b);
$i += 3;
}
#use the logical relations beweent attributes to combine the compare results
my $tpret = $tparray[0];
for (my $j = 0; $j < @atrirelation; $j++)
{
if ($atrirelation[$j] =~ //s*and/s*/)
{
$tpret = $tpret && $tparray[$i+1];
}
else
{
$tpret = $tpret || $tparray[$i+1];
}
}
#if the combined result is 1, then found it
if ($tpret)
{
push(@$result, $hash);
}
return $tpret;
}
#use to search a node by the give tag and resultes from its parent node hash
#in param1:search xpath suchas TAGNAME[@attributename="attributevalue" and @attributename="attributevalue" or @attributename="attributevalue" ...]
#in param2:a hash that is the parent node
#in param3:a array to store the find node which was reprented by a hash
#return value: 1 means has found the node match the attributes or 0 means not found
sub SearchNode
{
my ($TagString, $hash, $result) = @_;
if (length($TagString) == 0)
{
return 0;
}
#get the node tag
$TagString =~ m/(/w+)/;
my $Tag = $1;
if ($hash =~ /^HASH/) # must be a hash
{
if (exists
if ($value =~ /^ARRAY/) #there is existed more then on hash node that has the node tag of $Tag
{
my @values = @{$value};
my $ret = 0;
for (my $k=0; $k < @values; $k++)
{
#match each node in the array to find the node
my ($b) = SearchNodeByAttribute($TagString, $values[$i], $result);
if ($b)
{
$ret = 1;
}
} return $ret;
}
else
{
#compare the given attributes to find the node
return SearchNodeByAttribute($TagString, $value, $result);
}
}
else
{
return 0;
}
}
else
{
return 0;
}
}
#the $hash as the root hash, and the $xpath was givened from the root, then to search the node by given xpath
#in param2:a hash that is the parent node
#in param3:a array to store the find node which was reprented by a hash
#return value: 1 means has found the node match the attributes or 0 means not found
sub SearchFromRoot
{
my ($xpath, $hash, $result) = @_;
if ($xpath =~ m/^///)
{
$xpath = substr($xpath, 1);
}
if (length($xpath) == 0)
{
return 0;
}
#get the first node tag and attributes such as TAGNAME[@attributename="attributevalue" and ... or ...]
my ($tag) = $xpath =~ m/(/s*/w+/s*(/[(/s*/@/w+/s*=/s*[/"|']/w+[/"|']/s*(and|or)/s*)*/s*/@/w+/s*=/s*[/"|']/w+[/"|']/s*/])?/s*)/;
#remove the first node tag then for the second recure search
$xpath =~ s//s*/w+/s*(/[(/s*/@/w+/s*=/s*[/"|']/w+[/"|']/s*(and|or)/s*)*/s*/@/w+/s*=/s*[/"|']/w+[/"|']/s*/])?/s*//;
my @nodelist = ();
my $ret = 0;
#first find the hash node that match the first node tag and attributes that givened
$ret = SearchNode($tag, $hash, /@nodelist);
#if do not found, then return 0
if ($ret == 0)
{
return 0;
}
else
{
#found the first node tag and attributes node
if (length($xpath) == 0)
{
#recure search end, then store the results
for (my $i=0; $i < @nodelist; $i++)
{
push(@$result, $nodelist[$i]);
}
return 1;
}
else
{
my $ret = 0;
for (my $i=0; $i < @nodelist; $i++)
{
#then search the second node tag and attributes
my ($b) = GetNodeByPath($xpath, $nodelist[$i], $result);
if ($b)
{
$ret = 1;
}
}
return $ret;
}
}
}
#used in recure search xpath sub.
#call the recure search sub in each element in the array
#in param1:a xpath
#in param2:a hash that is the root node
#in param3:a array to store the find node which was reprented by a hash
#return value: 1 means has found the node match the attributes or 0 means not found
sub RecureSearchArrayNode
{
my ($xpath, $value, $result) = @_;
if ($value =~ /^HASH/)
{
return RecureSearchNode($xpath, $value, $result);
}
elsif ($value =~ /^ARRAY/)
{
my @tparray = @{$value};
my $ret = 0;
for (my $k=0; $k < @tparray; $k++)
{
my ($b) = RecureSearchArrayNode($xpath, $tparray[$k], $result);
if ($b)
{
$ret = 1;
}
}
return $ret;
}
else
{
return 0;
}
}
#used for the recure search when the xpath was givened like : //TAG/TAG[...]/...
#thought that the xpath is givened from the root node hash, and each hash node in the xml hash dom can be the root node, then do the recure search
#in param1:a xpath
#in param2:a hash that is the root node
#in param3:a array to store the find node which was reprented by a hash
#return value: 1 means has found the node match the attributes or 0 means not found
sub RecureSearchNode
{
my ($xpath, $hash, $result) = @_;
my $ret = 0;
#first search the node from the given hash node
$ret = SearchFromRoot($xpath, $hash, $result);
#then recure the each element in the current hash node
while (($key, $value) = each(%$hash))
{
if ($value =~ /^HASH/)
{
my ($b) = RecureSearchNode($xpath, $value, $result);
if ($b)
{
$ret = 1;
}
}
elsif ($value =~ /^ARRAY/)
{
my @tparray = @{$value};
my $tpret = 0;
for (my $k=0; $k < @tparray; $k++)
{
if ($tparray[$k] =~ /^HASH/)
{
my ($b) = RecureSearchNode($xpath, $tparray[$k], $result);
if ($b)
{
$tpret = 1;
}
}
elsif ($tparray[$k] =~ /^ARRAY/)
{
my ($b) = RecureSearchArrayNode($xpath, $tparray[$k], $result);
if ($b)
{
$tpret = 1;
}
}
else
{
next;
}
}
if ($tpret)
{
$ret = 1;
}
}
else
{
next;
}
}
return $ret;
}
#used search hash node by a string that flow xpath syntax, the xpath is a single path
#in param1:a xpath
#in param2:a hash that is the root node of a xml dom struct
#in param3:a array to store the find node which was reprented by a hash
#return value: 1 means has found the node match the attributes or 0 means not found
sub SearchPathNode
{
my ($xpath, $hash, $result) = @_;
if (length($xpath) != 0)
{
if ($xpath =~ m/^/////) #recure search
{
$xpath = substr($xpath, 2);
return RecureSearchNode($xpath, $hash, $result);
}
else #search from root
{
return SearchFromRoot($xpath, $hash, $result);
}
}
else
{
return 0;
}
}
###########################end of xpath protected sub###########################
########################################public xpath sub########################
#used search hash node by a string that flow xpath syntax, the xpath is a complex path that combine by more then one xpath
#in param1:a xpath
#in param2:a hash that is the root node of a xml dom struct
#in param3:a array to store the find node which was reprented by a hash
#return value: 1 means has found the node match the attributes or 0 means not found
sub GetNodeByPath
{
my ($xpathlist, $hash, $result) = @_;
if ($xpathlist =~ m/(((////|//)?/s*/w+/s*(/[(/s*/@/w+/s*=/s*[/"|']/w+[/"|']/s*(and|or)/s*)*/s*/@/w+/s*=/s*[/"|']/w+[/"|']/s*/])?/s*///s*)*(////|//)?/s*/w+/s*(/[(/s*/@/w+/s*=/s*[/"|']/w+[/"|']/s*(and|or)/s*)*/s*/@/w+/s*=/s*[/"|']/w+[/"|']/s*/])?/s*/|/s*)*((////|//)?/s*/w+/s*(/[(/s*/@/w+/s*=/s*[/"|']/w+[/"|']/s*(and|or)/s*)*/s*/@/w+/s*=/s*[/"|']/w+[/"|']/s*/])?/s*///s*)*(////|//)?/s*/w+/s*(/[(/s*/@/w+/s*=/s*[/"|']/w+[/"|']/s*(and|or)/s*)*/s*/@/w+/s*=/s*[/"|']/w+[/"|']/s*/])?/s*/)
{
my @xpatharray = split(//|/,$xpathlist);
$ret = 0;
foreach my $xpath (@xpatharray)
{
if (length($xpath) == 0)
{
next;
}
my ($b) = SearchPathNode($xpath, $hash, $result);
if ($b)
{
$ret = 1;
}
}
return $ret;
}
else
{
print "the xpath $xpathlist does not match the syntax of xpath!/n";
return;
}
}
###########################end of my xpath#########################################
#test code
(%xmldoc) = MyXMLSimple("test.xml");
#$xpath = "////CTRL[/@name=/"cs_edit/" or /@value=/"ce_edit/"]//C//EN//EVT[/@name='EVT_CON_FE_OK' or /@value=/"0x1013/"]";
$xpath = "////EN//EVT";
@nodelist = ();
$ret = GetNodeByPath($xpath, /%xmldoc, /@nodelist);
if ($ret)
{
foreach my $value (@nodelist)
{
$a = $$value{"name"};
print "$a /n";
}
}
#end test
但至于其内容是否正确与否却不得而知，于是需要添加一些测试代码来验证。
以前写C/C++代码时，常用XPATH来搜索XML文件中的节点，这需要XML引擎来处理。
但现在读取XML文件采用的是“土方法”（自行实现的），所以这引擎也得自己来写喽了。
XPATH引擎，我这其实就一函数，给定XPATH字串和存有XML文件内容的散列（当然是用MyXMLSimple方法读取进来的），
就可以搜索到你想找的XML节点，当然返回值也是个节点。
XPATH是很强大的，哥哥这里只是实现了一个基本的功能：查找节点。
我这里支持的XPATH路径只有两种：全局搜索的路径，即以双斜杠开头的；或是从根开始搜索的，即以间斜杠或直接以节点TAG开头的。
节点可以指定属性的值。支持的XPATH样板：
//TAG[@name="value" or @name="value"]/TAG[@name="value" and @name="value"] #全局递归搜索
/TAG/TAG/TAG[@name=”value"] #从根开始搜索
TAG/TAG[@name="value"] #从根开始
其中的TAG表示节点的节点名字，name为节点的某一属性名字，value为某一属性的值，这些字串暂时不支持通配符查找。
程序中用来一些正则表达式，那个也刚学不太懂，所以一些正则表达式可能写得比较乱或是根本就是错误的，请原谅小弟的无知。
下面是Perl源程序。
####perl pragram###################################################
###############################my xml simple protected sub##############
#get node tag of a nodestring like <%TAG% ... >
#in param: the nodestring
#return value: the tag : %TAG%
sub GetNodeTag
{
my ($NodeString) = @_;
$NodeString =~ m/</s*//(/w+)/s*>/ or $NodeString =~ m/</s*(/w+).*?>/;
return $1;
}
#get the attributes of a node that description by a string like
#in param1: the nodestring
#in param2: a array to store the names and values of the attributes
#in the array: name1, value1, name2, value2, name3, value3,...
sub GetAtribList
{
my ($nodestring, $atribarray) = @_;
@$atribarray = $nodestring =~ m/(/w+)/s*=/s*/"(/w*?)/"/g;
}
#push the attributes of a node into a hash that reprent the node
#in param1: the nodestring
#in param2: the hash that will reprent the node
sub pushattributes
{
my ($nodestring, $hash) = @_;
my @tsarray = ();
GetAtribList($nodestring, /@tsarray);
if (@tsarray)
{
for (my $j = 0; $j < @tsarray;)
{
my $key = @tsarray[$j];
my $value = @tsarray[$j+1];
$$hash{$key} = $value;
$j += 2;
}
}
}
#push a couple of key and value to a hash
#in param1: the key
#in param2: the value
#in param3: the hash
sub pushhash
{
my ($key, $value, $hash) = @_;
if ($hash =~ /^HASH/)
{
if (exists
hash{$key}) { $ekvalue =
hash{$key};if ($ekvalue =~ /^ARRAY/)
{
my @array=@{$ekvalue};
my $index = @array;
hash{$key}[$index] = $value; } else { my @tparray = (); push (@tparray, $ekvalue); push (@tparray, $value);
hash{$key}=[@tparray];}
}
else
{
$$hash{$key} = $value;
}
}
} #generate a xml dom hash by a nodearray that contained the nodestring of the xml file
#in param1: the nodearray that contained all the nodestring in the xml file <...>
#in param2: the begin pos to generate the xml dom hash in the nodearray
#in param3: the end pos to generate the xml dom hash in the nodearray
#in param4: the hash
sub GenerateXMLDom
{
my ($nodearray, $begin, $length, $root)=@_;
for (my $i=$begin; $i<$begin + $length; )
{
$temp = $$nodearray[$i];
my ($nodetag) = GetNodeTag($temp);
if ($temp =~ m////s*>/) #single node
{
my %tshash1 = ();
pushattributes($temp, /%tshash1);
pushhash($nodetag, /%tshash1, $root);
$i++;
next;
}
elsif ($temp =~ m/</s*///) #end of complex node
{
$i++;
next;
}
else #complex node
{
my %tshash2 = ();
pushattributes($temp, /%tshash2);
my $headnum = 1;
my $endnum = 0;
my $start = $i;
while ($endnum != $headnum && $i < $begin + $length-1)
{
$i++;
my $nextnode = $$nodearray[$i];
if ($nextnode =~ m////s*>/) #single node
{
$headnum++;
$endnum++;
}
elsif ($nextnode =~ m/</s*///) #end of complex node
{
$endnum++;
}
else
{
$headnum++;
}
}
$i++;
my $len = $i - $start;
GenerateXMLDom(/@nodearray, $start+1, $len-2, /%tshash2);
pushhash($nodetag, /%tshash2, $root);
next;
}
}
}
###############################end of my xml simple protected sub##############
################################my xml simple public sub#####################
#my xml simple, to read a xml file into a hash
#in param: the xml file to read
#return value: the hash that contained the xml dom content
sub MyXMLSimple
{
my ($xmlfile) = @_;
open (INFILE, "$xmlfile") or die "can't open $xmlfile for reading $!";
while ($line=)
{
chomp $line;
$filecontent.=$line;
}
close INFILE;
#delete xml comment
$filecontent =~ s///g;
#cut file context by <...>
@nodearray = $filecontent =~ m/<.*?>/g;
$length = @nodearray;
%rootnode = (); #the root node of the xml file
#construct the xml dom struct
GenerateXMLDom(/@nodearray, 0, $length, /%rootnode);
return %rootnode;
}
########################end of my xml simple#############################
###########################xpath protected sub###########################
#use to search the node by attributes, the node tag has already matched
#in param1:search xpath suchas TAGNAME[@attributename="attributevalue" and @attributename="attributevalue" or @attributename="attributevalue" ...]
#in param2:a hash with the key of TAGNAME
#in param3:a array to store the find node which was reprented by a hash
#return value: 1 means has found the node match the attributes or 0 means not found
sub SearchNodeByAttribute
{
my ($TagString, $hash, $result) = @_;
#get the attribute name list
#the element in the array is such as : name, =|!=, value, name, =|!=, value,...
my @attributes = $TagString =~ m//@(/w+)/s*(!?=)/s*[/"|'](.*?)[/"|']/g;
#get the logical relations between attributes if existed more then one attributes
my $index = index($TagString, "[");
my $endex = index($TagString, "]");
my $attrstr;
if ($index != -1 && $endex != -1)
{
my $len = $endex - $index - 1;
$attrstr = substr($TagString, $index+1, $len);
}
#the logical relations list
#a sample : and,or,and,or
my (@atrirelation) = $attrstr =~ m//@/w+/s*!?=/s*[/"|'].*?[/"|']/s*(and|or)/s*/g;
#if do not give the attributes, then found the node
if (@attributes <= 0)
{
push (@$result, $hash);
return 1;
}
#compare the existed attributes with the given attributes and store the result to a array
my @tparray = ();
for (my $i=0; $i < @attributes;)
{
my $attname = $attributes[$i];
my $oper = $attributes[$i+1];
my $attvalue = $attributes[$i+2];
my $exvalue = $$hash{$attname};
my $b = 0;
if ($oper =~ //s*=/s*/)
{
if ($exvalue =~ m/$attvalue/)
{
$b = 1;
}
}
else
{
if ($exvalue =~ m/$attvalue/)
{
$b = 0;
}
else
{
$b = 1;
}
}
push(@tparray, $b);
$i += 3;
}
#use the logical relations beweent attributes to combine the compare results
my $tpret = $tparray[0];
for (my $j = 0; $j < @atrirelation; $j++)
{
if ($atrirelation[$j] =~ //s*and/s*/)
{
$tpret = $tpret && $tparray[$i+1];
}
else
{
$tpret = $tpret || $tparray[$i+1];
}
}
#if the combined result is 1, then found it
if ($tpret)
{
push(@$result, $hash);
}
return $tpret;
}
#use to search a node by the give tag and resultes from its parent node hash
#in param1:search xpath suchas TAGNAME[@attributename="attributevalue" and @attributename="attributevalue" or @attributename="attributevalue" ...]
#in param2:a hash that is the parent node
#in param3:a array to store the find node which was reprented by a hash
#return value: 1 means has found the node match the attributes or 0 means not found
sub SearchNode
{
my ($TagString, $hash, $result) = @_;
if (length($TagString) == 0)
{
return 0;
}
#get the node tag
$TagString =~ m/(/w+)/;
my $Tag = $1;
if ($hash =~ /^HASH/) # must be a hash
{
if (exists
hash{$Tag}) # in the hash existed a key value match the node tag, means there is existed a child node in the hash that has the tag of $Tag { my $value =
hash{$Tag}; #get the existed child node hash or arrayif ($value =~ /^ARRAY/) #there is existed more then on hash node that has the node tag of $Tag
{
my @values = @{$value};
my $ret = 0;
for (my $k=0; $k < @values; $k++)
{
#match each node in the array to find the node
my ($b) = SearchNodeByAttribute($TagString, $values[$i], $result);
if ($b)
{
$ret = 1;
}
} return $ret;
}
else
{
#compare the given attributes to find the node
return SearchNodeByAttribute($TagString, $value, $result);
}
}
else
{
return 0;
}
}
else
{
return 0;
}
}
#the $hash as the root hash, and the $xpath was givened from the root, then to search the node by given xpath
#in param2:a hash that is the parent node
#in param3:a array to store the find node which was reprented by a hash
#return value: 1 means has found the node match the attributes or 0 means not found
sub SearchFromRoot
{
my ($xpath, $hash, $result) = @_;
if ($xpath =~ m/^///)
{
$xpath = substr($xpath, 1);
}
if (length($xpath) == 0)
{
return 0;
}
#get the first node tag and attributes such as TAGNAME[@attributename="attributevalue" and ... or ...]
my ($tag) = $xpath =~ m/(/s*/w+/s*(/[(/s*/@/w+/s*=/s*[/"|']/w+[/"|']/s*(and|or)/s*)*/s*/@/w+/s*=/s*[/"|']/w+[/"|']/s*/])?/s*)/;
#remove the first node tag then for the second recure search
$xpath =~ s//s*/w+/s*(/[(/s*/@/w+/s*=/s*[/"|']/w+[/"|']/s*(and|or)/s*)*/s*/@/w+/s*=/s*[/"|']/w+[/"|']/s*/])?/s*//;
my @nodelist = ();
my $ret = 0;
#first find the hash node that match the first node tag and attributes that givened
$ret = SearchNode($tag, $hash, /@nodelist);
#if do not found, then return 0
if ($ret == 0)
{
return 0;
}
else
{
#found the first node tag and attributes node
if (length($xpath) == 0)
{
#recure search end, then store the results
for (my $i=0; $i < @nodelist; $i++)
{
push(@$result, $nodelist[$i]);
}
return 1;
}
else
{
my $ret = 0;
for (my $i=0; $i < @nodelist; $i++)
{
#then search the second node tag and attributes
my ($b) = GetNodeByPath($xpath, $nodelist[$i], $result);
if ($b)
{
$ret = 1;
}
}
return $ret;
}
}
}
#used in recure search xpath sub.
#call the recure search sub in each element in the array
#in param1:a xpath
#in param2:a hash that is the root node
#in param3:a array to store the find node which was reprented by a hash
#return value: 1 means has found the node match the attributes or 0 means not found
sub RecureSearchArrayNode
{
my ($xpath, $value, $result) = @_;
if ($value =~ /^HASH/)
{
return RecureSearchNode($xpath, $value, $result);
}
elsif ($value =~ /^ARRAY/)
{
my @tparray = @{$value};
my $ret = 0;
for (my $k=0; $k < @tparray; $k++)
{
my ($b) = RecureSearchArrayNode($xpath, $tparray[$k], $result);
if ($b)
{
$ret = 1;
}
}
return $ret;
}
else
{
return 0;
}
}
#used for the recure search when the xpath was givened like : //TAG/TAG[...]/...
#thought that the xpath is givened from the root node hash, and each hash node in the xml hash dom can be the root node, then do the recure search
#in param1:a xpath
#in param2:a hash that is the root node
#in param3:a array to store the find node which was reprented by a hash
#return value: 1 means has found the node match the attributes or 0 means not found
sub RecureSearchNode
{
my ($xpath, $hash, $result) = @_;
my $ret = 0;
#first search the node from the given hash node
$ret = SearchFromRoot($xpath, $hash, $result);
#then recure the each element in the current hash node
while (($key, $value) = each(%$hash))
{
if ($value =~ /^HASH/)
{
my ($b) = RecureSearchNode($xpath, $value, $result);
if ($b)
{
$ret = 1;
}
}
elsif ($value =~ /^ARRAY/)
{
my @tparray = @{$value};
my $tpret = 0;
for (my $k=0; $k < @tparray; $k++)
{
if ($tparray[$k] =~ /^HASH/)
{
my ($b) = RecureSearchNode($xpath, $tparray[$k], $result);
if ($b)
{
$tpret = 1;
}
}
elsif ($tparray[$k] =~ /^ARRAY/)
{
my ($b) = RecureSearchArrayNode($xpath, $tparray[$k], $result);
if ($b)
{
$tpret = 1;
}
}
else
{
next;
}
}
if ($tpret)
{
$ret = 1;
}
}
else
{
next;
}
}
return $ret;
}
#used search hash node by a string that flow xpath syntax, the xpath is a single path
#in param1:a xpath
#in param2:a hash that is the root node of a xml dom struct
#in param3:a array to store the find node which was reprented by a hash
#return value: 1 means has found the node match the attributes or 0 means not found
sub SearchPathNode
{
my ($xpath, $hash, $result) = @_;
if (length($xpath) != 0)
{
if ($xpath =~ m/^/////) #recure search
{
$xpath = substr($xpath, 2);
return RecureSearchNode($xpath, $hash, $result);
}
else #search from root
{
return SearchFromRoot($xpath, $hash, $result);
}
}
else
{
return 0;
}
}
###########################end of xpath protected sub###########################
########################################public xpath sub########################
#used search hash node by a string that flow xpath syntax, the xpath is a complex path that combine by more then one xpath
#in param1:a xpath
#in param2:a hash that is the root node of a xml dom struct
#in param3:a array to store the find node which was reprented by a hash
#return value: 1 means has found the node match the attributes or 0 means not found
sub GetNodeByPath
{
my ($xpathlist, $hash, $result) = @_;
if ($xpathlist =~ m/(((////|//)?/s*/w+/s*(/[(/s*/@/w+/s*=/s*[/"|']/w+[/"|']/s*(and|or)/s*)*/s*/@/w+/s*=/s*[/"|']/w+[/"|']/s*/])?/s*///s*)*(////|//)?/s*/w+/s*(/[(/s*/@/w+/s*=/s*[/"|']/w+[/"|']/s*(and|or)/s*)*/s*/@/w+/s*=/s*[/"|']/w+[/"|']/s*/])?/s*/|/s*)*((////|//)?/s*/w+/s*(/[(/s*/@/w+/s*=/s*[/"|']/w+[/"|']/s*(and|or)/s*)*/s*/@/w+/s*=/s*[/"|']/w+[/"|']/s*/])?/s*///s*)*(////|//)?/s*/w+/s*(/[(/s*/@/w+/s*=/s*[/"|']/w+[/"|']/s*(and|or)/s*)*/s*/@/w+/s*=/s*[/"|']/w+[/"|']/s*/])?/s*/)
{
my @xpatharray = split(//|/,$xpathlist);
$ret = 0;
foreach my $xpath (@xpatharray)
{
if (length($xpath) == 0)
{
next;
}
my ($b) = SearchPathNode($xpath, $hash, $result);
if ($b)
{
$ret = 1;
}
}
return $ret;
}
else
{
print "the xpath $xpathlist does not match the syntax of xpath!/n";
return;
}
}
###########################end of my xpath#########################################
#test code
(%xmldoc) = MyXMLSimple("test.xml");
#$xpath = "////CTRL[/@name=/"cs_edit/" or /@value=/"ce_edit/"]//C//EN//EVT[/@name='EVT_CON_FE_OK' or /@value=/"0x1013/"]";
$xpath = "////EN//EVT";
@nodelist = ();
$ret = GetNodeByPath($xpath, /%xmldoc, /@nodelist);
if ($ret)
{
foreach my $value (@nodelist)
{
$a = $$value{"name"};
print "$a /n";
}
}
#end test
Comments
Post a Comment