[SOLVED] Need a regex to find a string using python .

call_krushna · 12-19-2015, 06:23 PM

Hi Team,

I have a file with below contents .

test.txt
------------------------------------------------------
obId>TX-4440</jobId><user>sysadmin</user><started>2014-10-02T19:18:41.000Z</started><status>FINISHED</status><type>EXPORT</type><priority>MEDIUM</priority></job><job><jobId>TX-4445</jobId><user>sysadmin</user><started>2014-10-02T19:17:42.000Z</started><status>FINISHED</status><type>EXPORT</type><priority>MEDIUM</priority></job><job><jobId>TX-4455</jobId><user>sysadmin</user><started>2014-10-03T06:37:49.000Z</started><status>FINISHED</status><type>EXPORT</type><priority>MEDIUM</priority></job><job><jobId>TX-4456</jobId><user>sysadmin</user><started>2014-10-03T06:38:06.000Z</started><status>FINISHED</status><type>EXPORT</type><priority>MEDIUM</priority></job><job><jobId>TX-4458</jobId><user>sysadmin</user><started>2014-10-03T06:55:41.000Z</started><status>FINISHED</status><type>EXPORT</type><priority>MEDIUM</priority></job><job><jobId>TX-4922</jobId><user>sysadmin</user><started>2014-10-05T00:39:40.000Z</started><status>FINISHED</status><type>EXPORT</type><priority>MEDIUM</priority></job><job><jobId>TX-5020</jobId>

----------------------- file ends here ------------------------

I need a regex to find all the job ID that start with TX ( e.g. TX-4440, TX-4456 )

I tried in below way .

reg = re.compile('[TX]+-+[0-9]+')

reg.search(mystring).group()

I am able to get very first value i.e (TX-4440) , but I need all the values as list .

Any help is highly appreciable .

Thanks & Regards

norobro · 12-19-2015, 07:32 PM

This example in the python docs should get you started.

grail · 12-19-2015, 11:43 PM

You might also wish to check your current regex as it will also match the following :- TXTXTXTX-----0

I would be guessing you would not wish to match such a combination.

dugan · 12-20-2015, 12:45 AM

Code:

import re

with open('test.txt') as f:
    text = f.read()

print re.findall('>(TX-[0-9]{4})<', text)

That said, did you consider fixing the XML file so that it's proper XML, and then using an XML parser? The Python standard library includes one. Most people would use lxml.

call_krushna · 12-23-2015, 07:43 PM

Thanks Dugan,

That re code helped quickly to resolve the issue . Can u explain what it it doing really the below code .

re.findall('>(TX-[0-9]{4})<', text)) .

Thanks and regards

dugan · 12-23-2015, 07:51 PM

Quote:

Originally Posted by call_krushna

Thanks Dugan,

That re code helped quickly to resolve the issue . Can u explain what it it doing really the below code .

re.findall('>(TX-[0-9]{4})<', text)) .

Thanks and regards

Well, I assume you can look up what "re.findall" does. Don't dispel that assumption.

I saw that all of the instances you wanted were ">TX-####<". Tag-closing angle bracket, "TX-" constant, 4 numbers, tag-opening angle bracket. So we're searching for strings between ">" and "<". Hence: ">(...)<". The parenetheses specify a group, and what's inside the group will be found. So inside the group, I have "TX-". That's a constant. Followed by "[0-9]", a digit, followed by "{4}", repeated 4 times.

See how unwise it is to ask someone to "explain" a regular expression?

EDIT: it's unwise because it's really difficult to explain them in a way that's easy to follow and that doesn't come off as word salad.

call_krushna · 12-23-2015, 08:23 PM

Hi Dugan,

Many thank for the explanation . We are using {4} which will search numbers repeated 4 times only. It wont work if number repeated 5 times e.g.(TX-12345) . Is there any way we can search

i) any time repeated numbers , something like {*)

ii.) 4 or more than 4 time repeated numbers {4+}

Thanks and regards

grail · 12-24-2015, 02:32 AM

{4,} will allow for 4 or more assuming you do not want less than 4 which then you would just use * or +

call_krushna · 12-24-2015, 02:34 AM

Many thanks Grail .

Thanks and Regards