当前页面位置: » 丰搜网 » 文档中心 » 详细内容
正则表达式结合数组提取文章中的文件名
今天编zblog上传模块的时候，需要用到一个提取文章中文件名的子程，开始我把问题想复杂了，匹配了所有可能的文件名，不仅正则表达式写了一大串，而且下面还split了半天，后来我突然发现z-blog的上传文件都存在upload下，白写了那么多复杂的匹配。。。哎，不能浪费掉啊，还是贴上来万一哪个兄弟以后需要也好拿去用~~
下面这段子程基本上可以算是比较不错的通用匹配了。(ps:我突然发现code_lite把我的ubb转义了！！！晕,我只好自己转义了。。。)
--------------------------------------------------------------------------------
dim objregexp,matches,i,dc9_dot_cn_match
dim arymatch()
redim preserve arymatch(0)
set objregexp=new regexp
objregexp.ignorecase =true
objregexp.global=true
objregexp.pattern="(?:\[[^\]]+\]([^\[]+)\[\/[^\]]+\])(?:(?:hrefsrc)=([^\s^>]+)["">\s\'])"
set matches = objregexp.execute("[img]2312.jpg[/img][img]2312.jpg[/img]hh[img]http://www.dc9.cn/cm/upload/2ed312.jpg[/img]<a href=""http://www.dc9.cn/ddd.jpg""></a> href=""[img_right=400,300,title]upload/2assas.jpg[/img_right]http://www.dc9.cn/dccdd.jpg"">[img_right=400,300,title]upload/2assas.jpg[/img_right]<sctipt src=""http://www.dc9.cn/upload/sss.jpg""")
dim tmpmatch
for i=0 to matches.count-1
call insertdatatoarray(matches(i).submatches(0),arymatch)
call insertdatatoarray(matches(i).submatches(1),arymatch)
next
for i=0 to ubound(arymatch)
if not isnull(arymatch(i)) and trim(arymatch(i))<>"" then response.write arymatch(i)&"<br>"
next
function insertdatatoarray(data,byref arymatch)
if trim(data)<>"" then
data=replace(data,"'","")
data=replace(data,"""","")
data=replace(data,"\","/")
data=split(data,"/")(ubound(split(data,"/")))
redim preserve arymatch(ubound(arymatch)+1)
dc9_dot_cn_match=false
for j=0 to ubound(arymatch)
if arymatch(j)=data then dc9_dot_cn_match=true
next
if not dc9_dot_cn_match then arymatch(ubound(arymatch))=data
end if
end function
--------------------------------------------------------------------------------
实际上，z-blog的话就匹配upload就可以了（但是我为了省事，也为了更精确些，就干脆在上面的那个匹配上面加上了upload，所以看上去好像下面这个更复杂些，其实理论上应该下面这个简单些，毕竟有upload管着呢。。），也不用split那个/.这么写就可以了：
--------------------------------------------------------------------------------
dim objregexp,matches,i,zc_upload_match
dim arymatch()
redim preserve arymatch(0)
set objregexp=new regexp
objregexp.ignorecase =true
objregexp.global=true
objregexp.pattern="(?:\[[^\]]+\][^\[]*upload\/([^\[^\\^\/]+)\[\/[^\]]+\])(?:(?:hrefsrc)=""{0,1}[^\s^""^>^']*upload\/([^\s^>]+)["">\s\'])"
set matches = objregexp.execute("[img]2312.jpg[/img][img]2312.jpg[/img]hh[img]http://www.dc9.cn/cm/upload/2ed312.jpg[/img]<a href=""http://www.dc9.cn/ddd.jpg""></a> href=""[img_right=400,300,title]upload/2assas.jpg[/img_right]http://www.dc9.cn/dccdd.jpg"">[img_right=400,300,title]upload/2assas.jpg[/img_right]<sctipt src=""http://www.dc9.cn/upload/sss.jpg"" src=http://www.dc9.cn/upload/sasds.jpg>")
dim tmpmatch
for i=0 to matches.count-1
call insertdatatoarray(matches(i).submatches(0),arymatch)
call insertdatatoarray(matches(i).submatches(1),arymatch)
next
for i=0 to ubound(arymatch)
if not isnull(arymatch(i)) and trim(arymatch(i))<>"" then response.write arymatch(i)&"<br>"
next
function insertdatatoarray(data,byref arymatch)
if trim(data)<>"" then
data=replace(data,"'","")
data=replace(data,"""","")
redim preserve arymatch(ubound(arymatch)+1)
zc_upload_match=false
for j=0 to ubound(arymatch)
if arymatch(j)=data then zc_upload_match=true
next
if not zc_upload_match then arymatch(ubound(arymatch))=data
end if
end function
出处：http://www.dc9.cn/post/279.html
