正则表达式匹配一行中的所有模式

Question

I am trying to scrape through some log files to find a line, such as this:

'A-Topeka-Firesale\:\s\*132\*\d{2,5}\*[23]\d{9}\#'

and once that line is matched, to go backward in the file and find a preceding line, like this:

2016-12-30 11:02:12 DEBUG[ispatcher-18269] ab.talk.this.api.Api - http://hostname:19991/trapeze?session_id=176764&manager_event=old&apostrophe=2341231231234&_operation=doc 3da48a90-0f4f-4eb3-a241-94a1f05b891b requesting:

and I need to match "http://hostname:19991/trapeze?" , "manager_event=old" and requesting: for the second line (which is usually between 3-5 lines above the first match, to be a match.

So far I have tried variations of this:

for each, line in enumerate(f):
first_match = re.search(b'A-Topeka-Firesale\:\s\*132\*\d{2,5}\*[23]\d{9}\#', line)     
    if first_match:
for i in range(each, -1, -1)
    if re.match("|".join(['http://hostname:19991/trapeze', 'manager_event=old', 'requesting:']), str(f[i])):
        break

and this:

for each, line in enumerate(f):
    first_match = re.search(b'A-Topeka-Firesale\:\s\*132\*\d{2,5}\*[23]\d{9}\#', line)     
        if first_match:
            for i in range(each, -1, -1)
                if all(re.match(regex_str, str(f[i])) for regex_str in ['http://hostname:19991/trapeze', 'manager_event=old', 'requesting: ']):
                break

And the call matches wrong lines (eg lines starting with blank spaces and with an instance of one of the matches (trapeze)). Please what am I doing wrong and how can I do it better?

Sample input:

 2016-01-30 00:00:27 DEBUG[-dispatcher-411] ab.talk.this.api.Api - http://hostname:19991/trapeze?manager_id=40178&manager_event=old&apostrophe=2341231231234&_operation=doc dgfgdffb-8123-4f05-ac15-7ac841afad14 requesting:
  HEADERS:
  this-is-a-header: 200*01231231234
  A-Topeka-Firesale: *132*200*01231231234#
  Host: hostname:19991
  Accept: */*
  User-Agent: AHC/2.0
  Timeout-Access: <function1>
 CONTENT:

2015-03-12 00:00:28 DEBUG[-dispatcher-747] ab.talk.this.api.Api - http://hostname:19991/trapeze?manager_id=84942&manager_event=old&apostrophe=2341231231235&_operation=ogle abcdf8237-393f-4c4b-bc46-e184cbf08d9a requesting:
  HEADERS:
  this-is-a-header: 100
  A-Topeka-Firesale: *132*100#
  Host: hostname:19991
  Accept: */*
  User-Agent: AHC/2.0
  Timeout-Access: <function1>
 CONTENT:

Answer 1

Very unclear what it is you really want, but after some guessing - could this be what you want?

2016-12-30 11:02:12 DEBUG[ispatcher-18269] ab.talk.this.api.Api - http://hostname:19991/trapeze?session_id=176764&manager_event=old&apostrophe=2341231231234&_operation=doc 3da48a90-0f4f-4eb3-a241-94a1f05b891b requesting:

bla bla bla

bla bla bla

bla bla bla

A-Topeka-Firesale: *132*12345*2123456789#

In the text above you want to match the last line. (You've only given a regex, so I made one up matching the criteria.) Finding that line will lead you to the first line, matching http://hostname:19991/trapeze? , manager_event=old and requesting: in that order, but not directly after one another.

If I'm guessing correctly, this regex

(http://hostname:19991/trapeze.*?manager_event=old.*?requesting:).*?A-Topeka-Firesale\:\s\*132\*\d{2,5}\*[23]\d{9}\#

should (could) do it for you. It captures the first (complete) line, which is what I understand is what you're after. (You haven't specified if it's something spcific you're after in it, like session_id , or whatever, but that could be "targeted" directly of course.)

Check it out here at regex101 .

Note that the e x ample uses the extended flag to allow splitting the regex up to (somewhat) improve readability, and the s ingle line flag to have . match line feeds.