I was given a task to do which requires a long time to do.
The image say it all :
This is what I have : (x100 times):
And I need to extract this value only
How can I capture the value ?
I have made it with this regex :
DbCommand.*?\("(.*?)"\);
As you can see it does work :
And after the replace function (replace to $1) I do get the pure value :
but the problem is that I need only the pure values and not the rest of the unmatched group :
Question : In other words :
How can I get the purified result like :
Eservices_Claims_Get_Pending_Claims_List
Eservices_Claims_Get_Pending_Claims_Step1
Here is my code at Online regexer
Is there any way of replacing "all besides the matched group" ?
p.s. I know there are other ways of doing it but I prefer a regex solution ( which will also help me to understand regex better)
Unfortunately, JavaScript doesn't understand lookbehind. If it did, you could change your regular expression to match .*? preceded (lookbehind) by DbCommand.*?\(" and followed (lookahead) by "\);.
With that solution denied, i believe the cleanest solution is to perform two matches:
// you probably want to build the regexps dynamically
var regexG = /DbCommand.*?\("(.*?)"\);/g;
var regex = /DbCommand.*?\("(.*?)"\);/;
var matches = str.match(regexG).map(function(item){ return item.match(regex)[1] });
console.log(matches);
// ["Eservices_Claims_Get_Pending_Claims_List", "Eservices_Claims_Get_Pending_Claims_Step1"]
DEMO: http://jsbin.com/aqaBobOP/2/edit
You should be able to do a global replace of:
public static DataTable.*?{.*?DbCommand.*?\("(.*?)"\);.*?}
All I've done is changed it to match the whole block including the function definition using a bunch of .*?s.
Note: Make sure your regex settings are such that the dot (.) matches all characters, including newlines.
In fact if you want to close up all whitespace, you can slap a \s* on the front and replace with $1\n:
\s*public static DataTable.*?{.*?DbCommand.*?\("(.*?)"\);.*?}
Using your test case: http://regexr.com?37ibi
You can use this (without the ignore case and multiline option, with a global search):
pattern: (?:[^D]+|\BD|D(?!bCommand ))+|DbCommand [^"]+"([^"]+)
replace: $1\n
Try simply replacing the whole document replacing using this expression:
^(?: |\t)*(?:(?!DbCommand).)*$
You will then only be left with the lines that begin with the string DbCommand
You can then remove the spaces in between by replacing:
\r?\n\s* with \n globally.
Here is an example of this working: http://regexr.com?37ic4
Related
I have a string in which I need to get the value between either "[ValueToBeFetched]" or "{ValueToBeFetched}".
var test = "I am \"{now}\" doing \"[well]\"";
test.match(/"\[.*?]\"/g)
the above regex serves the purpose and gets the value between square brackets and I can use the same for curly brackets also.
test.match(/"\{.*?}\"/g)
Is there a way to keep only one regex and do this, something like an or {|[ operator in regex.
I tried some scenarios but they don't seem to work.
Thanks in advance.
You could try following regex:
(?:{|\[).*?(?:}|\])
Details:
(?:{|\[): Non-capturing group, gets character { or [
.*?: gets as few as possible
(?:}|\]): Non-capturing group, gets character } or ]
Demo
Code in JavaScript:
var test = "I am \"{now}\" doing \"[well]\"";
var result = test.match(/"(?:{|\[).*?(?:}|\])"/g);
console.log(result);
Result:
["{now}", "[well]"]
As you said, there is an or operator which is |:
[Edited as suggested] Let's catch all sentences that begins with an "a" or a "b" :
/^((a|b).*)/gm
In this example, if the line parsed begins with a or b, the entire sentence will be catched in the first result group.
You may test your regex with an online regex tester
For your special case, try something like that, and use the online regex tester i mentionned before to understand how it works:
((\[|\{)\w*(\]|\}))
I am passing a URL to a block of code in which I need to insert a new element into the regex. Pretty sure the regex is valid and the code seems right but no matter what I can't seem to execute the match for regex!
//** Incoming url's
//** url e.g. api/223344
//** api/11aa/page/2017
//** Need to match to the following
//** dir/api/12ab/page/1999
//** Hence the need to add dir at the front
var url = req.url;
//** pass in: /^\/api\/([a-zA-Z0-9-_~ %]+)(?:\/page\/([a-zA-Z0-9-_~ %]+))?$/
var re = myregex.toString();
//** Insert dir into regex: /^dir\/api\/([a-zA-Z0-9-_~ %]+)(?:\/page\/([a-zA-Z0-9-_~ %]+))?$/
var regVar = re.substr(0, 2) + 'dir' + re.substr(2);
var matchedData = url.match(regVar);
matchedData === null ? console.log('NO') : console.log('Yay');
I hope I am just missing the obvious but can anyone see why I can't match and always returns NO?
Thanks
Let's break down your regex
^\/api\/ this matches the beginning of a string, and it looks to match exactly the string "/api"
([a-zA-Z0-9-_~ %]+) this is a capturing group: this one specifically will capture anything inside those brackets, with the + indicating to capture 1 or more, so for example, this section will match abAB25-_ %
(?:\/page\/([a-zA-Z0-9-_~ %]+)) this groups multiple tokens together as well, but does not create a capturing group like above (the ?: makes it non-captuing). You are first matching a string exactly like "/page/" followed by a group exactly like mentioned in the paragraph above (that matches a-z, A-Z, 0-9, etc.
?$ is at the end, and the ? means capture 0 or more of the precending group, and the $ matches the end of the string
This regex will match this string, for example: /api/abAB25-_ %/page/abAB25-_ %
You may be able to take advantage of capturing groups, however, and use something like this instead to get similar results: ^\/api\/([a-zA-Z0-9-_~ %]+)\/page\/\1?$. Here, we are using \1 to reference that first capturing group and match exactly the same tokens it is matching. EDIT: actually, this probably won't work, since the text after /api/ and the text after /page/ will most likely be different, carrying on...
Afterwards, you are are adding "dir" to the beginning of your search, so you can now match someting like this: dir/api/abAB25-_ %/page/abAB25-_ %
You have also now converted the regex to a string, so like Crayon Violent pointed out in their comment, this will break your expected funtionality. You can fix this by using .source on your regex: var matchedData = url.match(regVar.source); https://developer.mozilla.org/en-US/docs/Web/JavaScript/Reference/Global_Objects/RegExp/source
Now you can properly match a string like this: dir/api/11aa/page/2017 see this example: https://repl.it/Mj8h
As mentioned by Crayon Violent in the comments, it seems you're passing a String rather than a regular expression in the .match() function. maybe try the following:
url.match(new RegExp(regVar, "i"));
to convert the string to a regular expression. The "i" is for ignore case; don't know that's what you want. Learn more here:
https://developer.mozilla.org/en-US/docs/Web/JavaScript/Reference/Global_Objects/RegExp
I've been hoving around by some answers here, and I can't find a solution to my problem:
I have this regexp which matches everyting inside an HTML span tag, including contents:
<span\b[^>]*>(.*?)</span>
and I want to find a way to make a search in all the text, except for what is matched with that regexp.
For example, if my text is:
var text = "...for there is a class of <span class="highlight">guinea</span> pigs which..."
... then the regexp would match:
<span class="highlight">guinea</span>
and I want to be able to make a regexp such that if I search for "class", regexp will match "...for there is a class of..."
and will not match inside the tag, like in
"... class="highlight"..."
The word to be matched ("class") might be anywhere within the text. I've tried
(?!<span\b[^>]*>(.*?)</span>)class
but it keeps searching inside tags as well.
I want to find a solution using only regexp, not dealing with DOM nor JQuery. Thanks in advance :).
Although I wouldn't recommend this, I would do something like below
(class)(?:(?=.*<span\b[^>]*>))|(?:(?<=<\/span>).*)(class)
You can see this in action here
Rubular Link for this regex
You can capture your matches from the groups and work with them as needed. If you can, use a HTML parser and then find matches from the text element.
It's not pretty, but if I get you right, this should do what you wan't. It's done with a single RegEx but js can't (to my knowledge) extract the result without joining the results in a loop.
The RegEx: /(?:<span\b[^>]*>.*?<\/span>)|(.)/g
Example js code:
var str = '...for there is a class of <span class="highlight">guinea</span> pigs which...',
pattern = /(?:<span\b[^>]*>.*?<\/span>)|(.)/g,
match,
res = '';
match = pattern.exec(str)
while( match != null )
{
res += match[1];
match = pattern.exec(str)
}
document.writeln('Result:' + res);
In English: Do a non capturing test against your tag-expression or capture any character. Do this globally to get the entire string. The result is a capture group for each character in your string, except the tag. As pointed out, this is ugly - can result in a serious number of capture groups - but gets the job done.
If you need to send it in and retrieve the result in one call, I'd have to agree with previous contributors - It can't be done!
I have this regular expression
// Look for /en/ or /en-US/ or /en_US/ on the URL
var matches = req.url.match( /^\/([a-zA-Z]{2,3}([-_][a-zA-Z]{2})?)(\/|$)/ );
Now with the above regular express it will cause the problem with the URL such as:
http://mydomain.com/css/bootstrap.css
or
http://mydomain.com/js/jquery.js
because my regular expression is to strip off 2-3 characters from A-Z or a-z
My question is how would I add in to this regular expression to not strip off anything with
js or img or css or ext
Without impacting the original one.
I'm not so expert on regular expression :(
Negative lookahead?
var matches = req.url.match(/^\/(?!(js|css))([a-zA-Z]{2,3}([-_][a-zA-Z]{2})?)(\/|$)/ );
\ not followed by js or css
First of all you have not defined what exactly you are searching for.
Define an array with lowercased common language codes (Common language codes)
This way you'll know what to look for.
After that, convert your url to lowercase and replace all '_' with '-' and search for every member of the array in the resulting string using indexOf().
Since you said you're using the regex to replace text, I changed it to a replace function. Also, you forced the regex to match the start of the string; I don't see how it would match anything with that. Anyway, here's my approach:
var result = req.url.replace(/\/([a-z]{2,3}([-_][a-z]{2})?)(?=\/|$)/i,
function(s,t){
switch(t){case"js":case"img":case"css":case"ext":return s;}
return "";
}
);
Given this string:
var str = 'A1=B2;C3,D0*E9+F6-';
I would like to retrieve the substring that goes from the beginning of the string up to 'D0*' (excluding), in this case:
'A1=B2;C3,'
I know how to achieve this using the combination of the substr and indexOf methods:
str.substr(0, str.indexOf('D0*'))
Live demo: http://jsfiddle.net/simevidas/XSu22/
However, this is obviously not the best solution since it contains a redundancy (the str name has to be written twice). This redundancy can be avoided by using the match method together with a regular expression that captures the substring:
str.match(/???/)[1]
Which regular expression literal do we have to pass into match to ensure that the correct substring is returned?
My guess is this: /(.*)D0\*/ (and that works), but my experience with regular expressions is rather limited, so I'm going to need a confirmation...
Try this:
/(.*?)D0\*/.exec(str)[1]
Or:
str.match(/(.*?)D0\*/)[1]
DEMO HERE
? directly following a quantifier makes the quantifier non-greedy (makes it match minimum instead of maximum of the interval defined).
Here's where that's from
/^(.+?)D0\*/
Try it here: http://rubular.com/r/TNTizJLSn9
/^.*(?=D0\*)/
more text to hit character limit...
You can do a number-group, like your example.
/^(.*?)foo/
It mean somethink like:
Store all in group, from start (the 0)
Stop, but don't store on found foo (the indexOf)
After that, you need match and get
'hello foo bar foo bar'.match(/^(.*?)foo/)[1]; // will return "hello "
It mean that will work on str variable and get the first (and unique) number-group existent. The [0] instead [1] mean that will get all matched code.
Bye :)