Python regular expression for phone numbers

Question

I am very new to regular expression & seeking help to parse out phone numbers from HTML text

At source site, the html tags are very distorted & does not have any unique selectors that i can use . Below if the list of possibilities i am looking to parse.

raw = """+49 39291 55-217
02102 7007064
0152 01680970
+49 39291 55-216
02102 3802 22
0800 333004 451-100
+49 221 9937 26950
02151-47974510
+49(0)6105 937 -539
0211/409 2268
+49(0)6105 937 -539
+49211/584-623
0211 58422 2012
+49 (9131) 7-35335
+49 521 9488 2470
+ 49-40-70 70 84 - 0
0211 17 95 99 04
02151-47974327
+49 203 28900 1121
0211 9449-2555
+49 (5 41) 9 98 -2268"""

I tried this pattern but could not make out more from it

import re, requests

Phones = re.findall(re.compile(r'.*?(\(?\d{3}\D{0,3}\d{3}\D{0,3}\d{4}).*?'),raw)

phones
['102 7007064', '152 0168097', '151-4797451', '937 -539\n0211', '937 -539\n+4921', '584-623\n0211', '151-4797432']

Any advise or help is highly appreciated. Thank you

`\D` matches line break chars , too. You should replace it with something like `[-./]?` — Wiktor Stribiżew, Aug 30 '18 at 09:30
@WiktorStribiżew, i tired that but doesn't outputs well all, I observed a 2 points. either the phone numbers starts with +49 or a 0 — Shekhar Samanta, Aug 30 '18 at 09:32
The first related question on the right side: https://stackoverflow.com/q/123559/8881141 — Mr. T, Aug 30 '18 at 09:33
Guys, please do not suggest phone validation threads, it is about extraction of phone numbers from a longer text. — Wiktor Stribiżew, Aug 30 '18 at 09:34
Yes true, its about extracting phone number from HTML source. — Shekhar Samanta, Aug 30 '18 at 09:35
Try [`\+? {0,2}\d+ {0,2}[(-]?\d(?:[ \d]*\d)?[)-]? {0,2}\d+[/ -]?\d+[/ -]?\d+(?: *- *\d+)?`](https://regex101.com/r/opzBnV/1) — Wiktor Stribiżew, Aug 30 '18 at 09:51
@WiktorStribiżew , its a very broader one, it returns lots of unwanted matches — Shekhar Samanta, Aug 30 '18 at 09:59
try the pattern on this HTML : https://pastebin.com/WNxjLfhR — Shekhar Samanta, Aug 30 '18 at 10:01
That is too long. Please provide exact specs for your pattern. — Wiktor Stribiżew, Aug 30 '18 at 10:11
Probably, https://regex101.com/r/opzBnV/2 will be more precise. — Wiktor Stribiżew, Aug 30 '18 at 10:16
what about using `sub` instead? ie `re.sub('[^+0-9\n]','',raw).split()` — Onyambu, Aug 30 '18 at 10:17
@WiktorStribiżew , i actually trying to get the phone numbers from job pages like this : https://de.indeed.com/viewjob?jk=1d06971a8e322ba2&tk=1cm4o74d7958gfs7&from=serp&vjs=3 .those above phone numbers are possibilities that i get from this kind of job pages — Shekhar Samanta, Aug 30 '18 at 10:32
If you are not going to provide the specs, the question is off-topic/too broad, I suggest closing until you clear it all up. — Wiktor Stribiżew, Aug 30 '18 at 10:33
I am testing using this pattern : regex101.com/r/opzBnV/2 @WiktorStribiżew could you please post it as answer & I will accept soon — Shekhar Samanta, Aug 30 '18 at 10:52
@WiktorStribiżew , yes i upvoted. I trying to learn in a more detailed way , could you please guide me to some good online tutorials that explains steps in detail — Shekhar Samanta, Sep 01 '18 at 15:34
I do not know your level of regex knowledge so that I can only suggest doing all lessons at [regexone.com](http://regexone.com/), reading through [regular-expressions.info](http://www.regular-expressions.info), [regex SO tag description](http://stackoverflow.com/tags/regex/info) (with many other links to great online resources), and the community SO post called [What does the regex mean](http://stackoverflow.com/questions/22937618/reference-what-does-this-regex-mean). Also, [rexegg.com](http://rexegg.com) is worth having a look at. — Wiktor Stribiżew, Sep 01 '18 at 20:12

score 5 · Accepted Answer · answered Aug 30 '18 at 10:54

I suggest using this pattern:

(?:\B\+ ?49|\b0)(?: *[(-]? *\d(?:[ \d]*\d)?)? *(?:[)-] *)?\d+ *(?:[/)-] *)?\d+ *(?:[/)-] *)?\d+(?: *- *\d+)?

See the regex demo. Note it is written based on your comment saying the phone numbers starts with +49 or a 0 and on the list of examples you provided. It may be considered "work in progress" since you have not provided more specific rules for phone number extraction.

Pattern details

(?:\B\+ ?49|\b0) - a +, optional space, 49 or a 0, both substrings cannot be preceded with a word char
(?: *[(-]? *\d(?:[ \d]*\d)?)? - an optional substring matching 0+ spaces, then an optional ( or -, 0+ spaces, a digit and then an optional sequence of digits/spaces followed with a digit
*(?:[)-] *)? - 0+ spaces and then an optional sequence of ) or - followed with 0+ spaces
\d+ - 1+ digits
* - 0+ spaces
(?:[/)-] *)? - an optional sequence of /, ) or - followed with 0+ spaces
\d+ - 1+ digits
*(?:[/)-] *)? - 0+ spaces and then an optional sequence of /, ) or - followed with 0+ spaces
\d+ - 1+ digits
(?: *- *\d+)? - an optional sequence: 0+ spaces, -, 0+ spaces, 1+ digits.

Python regular expression for phone numbers

1 Answers1