ColdFusion regex to detect first URL without image extension - regex

Question

I've been digging around stackoverflow this evening but nothing worked so far. What I'm trying to achieve is to extract URL from a string (mainly HTML) that doesn't have any image extension at the end. So if the given HTML string has URL ending with .jpg and then somewhere few lines down another URL without any image extension, regex would get that second one and stop. Alternatively it can return all the 'good' URLs, just leave out the images.
So far I've got:
<cfset c= reMatch('(http|ftp|https):\/\/[\w\-_]+(\.[\w\-_]+)+([\w\-\.,#?^=%&:/~\+##]*[\w\-\#?^=%&/~\+##])?',htmlString)>
I know image detecting part should be somewhere towards the end, but had only managed to freeze the server with my attempts so far.
Sample string to match:
'<tr> <td style="vertical-align: top; padding-right: 12px;"><img src="http://static01.nyt.com/images/2016/01/31/us/why-iowaALT/why-iowaALT-thumbStandard.jpg" /></td> <td> <h6 style="font-size: 10px; font-weight: normal; text-transform: uppercase; color: ##000000; margin: 0; margin-bottom: 2px"></h6> <h1 style="font-weight: normal; font-family: georgia,"times new roman",times,serif; font-size: 23px; margin: 0; margin-bottom: 4px"><a href="http://p.nytimes.com/email/re?location=InCMR7g4BCJTYuyKqXu41s2MxgEX9Okc&user_id=7b8478da99b24f28abb9c2f1be86c807&email_type=eta&task_id=1454290534529254&regi_id=0" style="color: ##004276; text-decoration: none !important;">'
Note: it should be ColdFusion regex version, which is at times a bit limited
Thanks!

Answer 1

Consider that your code works perfectly fine about extracting valid html links and you have them stored in array. All you have to do is go through this array and find does any of urls stored in this array doesn't contain extension -if not, return this value.
list_of_extensions = '(bmp|jpg|png|gif)'; //you can make this list longer
for(my_url in urls){
if(not reFind(list_of_extensions, my_url)){ // you can be more specific in this reFind call
return my_url; // it will return first invalid url
}
}

Answer 2

You can switch to Java for what you are trying to achieve like this:
<!--- Java Regular Expression Pattern Object --->
<cfset local.objPattern = createObject(
"java",
"java.util.regex.Pattern"
).compile(
javaCast( "string", '(?:https?:\/\/)[^"]*(?<!\.jpg)(?=\")' )
)
/>
<!--- Get Pattern Matcher for your html content --->
<cfset local.objMatcher = local.objPattern.matcher(
JavaCast( "string", local.htmlContent )
) />
<!--- Find Matching URLs --->
<cfloop condition="local.objMatcher.find()">
<cfdump var="#local.objMatcher.group()# <br>">
</cfloop>