Extract value from string in file powershell

Question

I am trying to search through the newest log file in directory for a particular string. The string reads Number of days since last service= and after the equals sign is a number. I want to extract that number for use later in my powershell program. This is what i have so far

$logDir= "C:logs\"
$latest = Get-ChildItem -Path $logdir | Sort-Object LastAccessTime -Descending | Select-Object -First 1
$findStr = Select-String -Path $latest.name -Pattern "Number of days since last service="

but now i am stuck. This is what it actually looks like in the log

<Msg>Number of days since last service =18</Msg>

so what i don't know how to do is extract the number between the equals sign and the < for this particular string

It's great that they're using an XML structured log format, so you don't have to use regular expressions and text slicing to get the data. :facepalm: — TessellatingHeckler
– TessellatingHeckler, Commented Jul 28, 2017 at 18:39

colsw · Accepted Answer · 2017-07-28 19:11:52Z

5

You can use Select-String with regex capture groups.

 $Matches = Select-String -Path $latest.name -Pattern "last service =(\d+)" | % {$_.Matches.Groups[1].Value}

answered Jul 28, 2017 at 19:11

colsw

3,3661 gold badge18 silver badges29 bronze badges

Sign up to request clarification or add additional context in comments.

Comments

Zach Alexander · Accepted Answer · 2017-07-28 20:57:14Z

1

You've probably got your answer by now, but "Imma giv eya more!" Assuming we have these defined ahead of time:

$Log = "<Stuff><Msg>Number of days since last service =18</Msg></Stuff>"
$StringHint = "Number of days since last service ="

1. Your typical Select-String method of doing things with Regexes. yep.

Select-String -InputObject $Log -Pattern "(?<=$StringHint)\d{1,}(?=<)" -AllMatches | % {$_.Matches.Groups[0].Value}

2. Doing the splits:

((($Log -split $StringHint)[1]) -split "<")[0]

3. The XML way. This only works for XML log files and if there is no other MSG tag (which is unlikely in this case). I chose to include this because I think it's cool how a simple type cast can dramatically alter how we can work with the data. Plus, it's a really easy way of stripping the nasty <'s and >'s away.

$xml = [xml]$Log
$EqualsLoc = $xml.Stuff.Msg.IndexOf("=")
$xml.Stuff.Msg.Substring($EqualsLoc+1)

The above would have been even better if you had a log file like this (yes, I know this is a configuration file and not a log file):

$XML = [xml]"<Application>
    <version>1.3.53</version>
    <AdvancedSettings>
        <Network>
            <InternetForWeebs>False</InternetForWeebs>
            <HackAllTheSystems>True</HackAllTheSystems>
            <L33tHandle>Gingervitis</L33tHandle>
        </Network>
    </AdvancedSettings>
    <BasicSettings>
        <Controls>
            <TheAnyKeyDesignation>0x42</TheAnyKeyDesignation>
            <CanHazCheeseburgerMode>True</CanHazCheeseburgerMode>
        </Controls>
    </BasicSettings>
    <Diagnostics>
        <DaysSinceLastService>18</DaysSinceLastService>
    </Diagnostics>
</Application>"

Then, getting the entry you needed would be something as easy as this:

$XML.Application.Diagnostics.DaysSinceLastService

edited Jul 28, 2017 at 20:57

answered Jul 28, 2017 at 20:05

Zach Alexander

4034 silver badges9 bronze badges

2 Comments

TheMadTechnician Over a year ago

Or if you don't know the path, and just know it's a Msg node, you can use Select-Xml as such: $XML | Select-Xml -XPath './/Msg' | Where{$_.Node.InnerXML -match "(?<=$StringHint)(\d{1,})(?=<)"} | %{$Matches[1]}

colsw Over a year ago

\d{1,} is the same as \d+ in that it matches one or more occurances.

BenH · Accepted Answer · 2017-07-28 18:18:39Z

0

Since you know the string before and after the data you want, you could just the -Replace operator to remove the unwanted text:

$Days = $findStr -replace '<Msg>Number\ of\ days\ since\ last\ service\ =','' -replace '</Msg>',''

Alternatively, since the information is the only number in the string, you could use -Match with \d+ (one or more numbers) to get the value.

$findStr -match '\d+'
$Days = $Matches.Values

answered Jul 28, 2017 at 18:18

BenH

10.1k1 gold badge26 silver badges38 bronze badges

1 Comment

TheMadTechnician Over a year ago

For your first answer you could more simply do $Days = $findstr -replace '\<Msg\>Number of days since last service =\s*(\d+).*','$1'

user1984300 · Accepted Answer · 2017-07-28 19:02:49Z

0

I was able to do it like this

$result =[regex]::Match($findStr, '=([^/)]+)<').Groups[1].Value

answered Jul 28, 2017 at 19:02

user1984300

994 silver badges18 bronze badges

1 Comment

TessellatingHeckler Over a year ago

Select-String outputs regex MatchInfo objects, so $findStr is already a thing you can use Groups and Value on, if you put the grouping in your select-string regex. (Instead of forcing it back into strings, and matching them as regexes again)

Collectives™ on Stack Overflow

Extract value from string in file powershell

4 Answers 4

Comments

2 Comments

1 Comment

1 Comment

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

4 Answers 4

Comments

2 Comments

1 Comment

1 Comment

Your Answer

Sign up or log in

Post as a guest

Related