I am really used to doing grep -iIr on the Unix shell but I haven't been able to get a PowerShell equivalent yet.
Basically, the above command searches the target folders recursively and ignores binary files because of the "-I" option. This option is also equivalent to the --binary-files=without-match option, which says "treat binary files as not matching the search string"
So far I have been using Get-ChildItems -r | Select-String as my PowerShell grep replacement with the occasional Where-Object added. But I haven't figured out a way to ignore all binary files like the grep -I command does.
How can binary files be filtered or ignored with Powershell?
So for a given path, I only want Select-String to search text files.
EDIT: A few more hours on Google produced this question How to identify the contents of a file is ASCII or Binary. The question says "ASCII" but I believe the writer meant "Text Encoded", like myself.
EDIT: It seems that an isBinary() needs to be written to solve this issue. Probably a C# commandline utility to make it more useful.
EDIT: It seems that what grep is doing is checking for ASCII NUL Byte or UTF-8 Overlong. If those exists, it considers the file binary. This is a single memchr() call.
On Windows, file extensions are usually good enough:
# all C# and related files (projects, source control metadata, etc)
dir -r -fil *.cs* | ss foo
# exclude the binary types most likely to pollute your development workspace
dir -r -exclude *exe, *dll, *pdb | ss foo
# stick the first three lines in your $profile (refining them over time)
$bins = new-list string
$bins.AddRange( [string[]]#("exe", "dll", "pdb", "png", "mdf", "docx") )
function IsBin([System.IO.FileInfo]$item) { !$bins.Contains($item.extension.ToLower()) }
dir -r | ? { !IsBin($_) } | ss foo
But of course, file extensions are not perfect. Nobody likes typing long lists, and plenty of files are misnamed anyway.
I don't think Unix has any special binary vs text indicators in the filesystem. (Well, VMS did, but I doubt that's the source of your grep habits.) I looked at the implementation of Grep -I, and apparently it's just a quick-n-dirty heuristic based on the first chunk of the file. Turns out that's a strategy I have a bit of experience with. So here's my advice on choosing a heuristic function that is appropriate for Windows text files:
Examine at least 1KB of the file. Lots of file formats begin with a header that looks like text but will bust your parser shortly afterward. The way modern hardware works, reading 50 bytes has roughly the same I/O overhead as reading 4KB.
If you only care about straight ASCII, exit as soon you see something outside the character range [31-127 plus CR and LF]. You might accidentally exclude some clever ASCII art, but trying to separate those cases from binary junk is nontrivial.
If you want to handle Unicode text, let MS libraries handle the dirty work. It's harder than you think. From Powershell you can easily access the IMultiLang2 interface (COM) or Encoding.GetEncoding static method (.NET). Of course, they are still just guessing. Raymond's comments on the Notepad detection algorithm (and the link within to Michael Kaplan) are worth reviewing before deciding exactly how you want to mix & match the platform-provided libraries.
If the outcome is important -- ie a flaw will do something worse than just clutter up your grep console -- then don't be afraid to hard-code some file extensions for the sake of accuracy. For example, *.PDF files occasionally have several KB of text at the front despite being a binary format, leading to the notorious bugs linked above. Similarly, if you have a file extension that is likely to contain XML or XML-like data, you might try a detection scheme similar to Visual Studio's HTML editor. (SourceSafe 2005 actually borrows this algorithm for some cases)
Whatever else happens, have a reasonable backup plan.
As an example, here's the quick ASCII detector:
function IsAscii([System.IO.FileInfo]$item)
{
begin
{
$validList = new-list byte
$validList.AddRange([byte[]] (10,13) )
$validList.AddRange([byte[]] (31..127) )
}
process
{
try
{
$reader = $item.Open([System.IO.FileMode]::Open)
$bytes = new-object byte[] 1024
$numRead = $reader.Read($bytes, 0, $bytes.Count)
for($i=0; $i -lt $numRead; ++$i)
{
if (!$validList.Contains($bytes[$i]))
{ return $false }
}
$true
}
finally
{
if ($reader)
{ $reader.Dispose() }
}
}
}
The usage pattern I'm targeting is a where-object clause inserted in the pipeline between "dir" and "ss". There are other ways, depending on your scripting style.
Improving the detection algorithm along one of the suggested paths is left to the reader.
edit: I started replying to your comment in a comment of my own, but it got too long...
Above, I looked at the problem from the POV of whitelisting known-good sequences. In the application I maintained, incorrectly storing a binary as text had far worse consequences than vice versa. The same is true for scenarios where you are choosing which FTP transfer mode to use, or what kind of MIME encoding to send to an email server, etc.
In other scenarios, blacklisting the obviously bogus and allowing everything else to be called text is an equally valid technique. While U+0000 is a valid code point, it's pretty much never found in real world text. Meanwhile, \00 is quite common in structured binary files (namely, whenever a fixed-byte-length field needs padding), so it makes a great simple blacklist. VSS 6.0 used this check alone and did ok.
Aside: *.zip files are a case where checking for \0 is riskier. Unlike most binaries, their structured "header" (footer?) block is at the end, not the beginning. Assuming ideal entropy compression, the chance of no \0 in the first 1KB is (1-1/256)^1024 or about 2%. Luckily, simply scanning the rest of the 4KB cluster NTFS read will drive the risk down to 0.00001% without having to change the algorithm or write another special case.
To exclude invalid UTF-8, add \C0-C1 and \F8-FD and \FE-FF (once you've seeked past the possible BOM) to the blacklist. Very incomplete since you're not actually validating the sequences, but close enough for your purposes. If you want to get any fancier than this, it's time to call one of the platform libraries like IMultiLang2::DetectInputCodepage.
Not sure why \C8 (200 decimal) is on Grep's list. It's not an overlong encoding. For example, the sequence \C8 \80 represents Ȁ (U+0200). Maybe something specific to Unix.
Ok, after a few more hours of research I believe I've found my solution. I won't mark this as the answer though.
Pro Windows Powershell had a very similar example. I had completely forgot that I had this excellent reference. Please buy it if you are interested in Powershell. It went into detail on Get-Content and Unicode BOMs.
This Answer to a similar questions was also very helpful with the Unicode identification.
Here is the script. Please let me know if you know of any issues it may have.
# The file to be tested
param ($currFile)
# encoding variable
$encoding = ""
# Get the first 1024 bytes from the file
$byteArray = Get-Content -Path $currFile -Encoding Byte -TotalCount 1024
if( ("{0:X}{1:X}{2:X}" -f $byteArray) -eq "EFBBBF" )
{
# Test for UTF-8 BOM
$encoding = "UTF-8"
}
elseif( ("{0:X}{1:X}" -f $byteArray) -eq "FFFE" )
{
# Test for the UTF-16
$encoding = "UTF-16"
}
elseif( ("{0:X}{1:X}" -f $byteArray) -eq "FEFF" )
{
# Test for the UTF-16 Big Endian
$encoding = "UTF-16 BE"
}
elseif( ("{0:X}{1:X}{2:X}{3:X}" -f $byteArray) -eq "FFFE0000" )
{
# Test for the UTF-32
$encoding = "UTF-32"
}
elseif( ("{0:X}{1:X}{2:X}{3:X}" -f $byteArray) -eq "0000FEFF" )
{
# Test for the UTF-32 Big Endian
$encoding = "UTF-32 BE"
}
if($encoding)
{
# File is text encoded
return $false
}
# So now we're done with Text encodings that commonly have '0's
# in their byte steams. ASCII may have the NUL or '0' code in
# their streams but that's rare apparently.
# Both GNU Grep and Diff use variations of this heuristic
if( $byteArray -contains 0 )
{
# Test for binary
return $true
}
# This should be ASCII encoded
$encoding = "ASCII"
return $false
Save this script as isBinary.ps1
This script got every text or binary file I tried correct.
i agree that the other answers are more 'complete' but - because i do not know what file extensions i will encounter within a folder and i want to look thru them all, this is the easiest solution for me.
how about instead of avoiding searching thru binary files you just ignore the errors that you get from searching thru binary files?
it doesn't take long to run a search even if there are binary files within the folder being searched.
in the end, all that you care about is the strings that match the pattern (which there is next to no chance of it would find a string that matches the pattern inside of a binary file).
GCI -Recurse -Force -ErrorAction SilentlyContinue | ForEach-Object { GC $_ -ErrorAction SilentlyContinue | Select-String -Pattern "Pattern" } | Out-File -FilePath C:\temp\grep.txt -Width 999999
Related
I need to delete the first couple of lines of a .txt-file in powershell. There are plenty of questions and answers already on SA how to do it. Most of them copy the whole filecontent into memory, cut out the first x lines and then save the content into the textfile again.
However, in my case the textfiles are huge (500MB+), so loading them completly into memory, just to delete the first couple of lines, takes very long and feels like a huge waste of resources.
Is there a more elegant approach? If you only want to read the first x lines, you can use
Get-Content in.csv -Head 10
, which only reads the first 10 lines. Is there something similar for deletion?
Here is a another way to do it using StreamReader and StreamWriter, as noted in comments, it's important to know the encoding of your file for this use case.
See Remarks from the Official Documentation:
The StreamReader object attempts to detect the encoding by looking at the first four bytes of the stream. It automatically recognizes UTF-8, little-endian Unicode, big-endian Unicode, little-endian UTF-32, and big-endian UTF-32 text if the file starts with the appropriate byte order marks. Otherwise, the user-provided encoding is used. See the Encoding.GetPreamble method for more information.
If you need to specify an Encoding you can target the StreamReader(String, Encoding) Constructor. For example:
$reader = [System.IO.StreamReader]::new('path\to\input.csv', [System.Text.Encoding]::UTF8)
As noted previously in Remarks, this might not be needed for common encodings.
An alternative to below code, could be the use of $reader.ReadToEnd() as Brice points out in his comment, after skipping the first 10 lines, this would read the entire contents of the file in memory before writing to the new file. I haven't used this method for this answer since, mklement0's helpful answer provides a very similar solution to the problem and this answer was intended to be a memory friendly solution.
try {
$reader = [System.IO.StreamReader]::new('absolute\path\to\input.csv')
$writer = [System.IO.StreamWriter]::new('absolute\path\to\output.csv')
# skip 10 lines
foreach($i in 1..10) {
$null = $reader.ReadLine()
}
while(-not $reader.EndOfStream) {
$writer.WriteLine($reader.ReadLine())
}
}
finally {
($reader, $writer).foreach('Dispose')
}
It's very also worth noting zett42's helpful comment using $reader.ReadBlock(Char[], Int32, Int32) method and $writer.Write(..) instead of $write.WriteLine(..) could be an even faster and still memory friendly alternative to read and write in chunks.
You're essentially attempting to remove the starting bytes of the file without modifying the remaining bytes, Raymond C has a good read posted here about why that can't be done.
The underlying abstract model for storage of file contents is in the form of a chunk of bytes, each indexed by the file offset. The reason appending bytes and truncating bytes is so easy is that doing so doesn’t alter the file offsets of any other bytes in the file. If a file has ten bytes and you append one more, the offsets of the first ten bytes stay the same. On the other hand, deleting bytes from the front or middle of a file means that all the bytes that came after the deleted bytes need to “slide down” to close up the space. And there is no “slide down” file system function.
As Mike Anthony's helpful answer explains, there is no system-level function that efficiently implements what you're trying to do, so you have no choice but to rewrite your file.
While memory-intensive, the following solution is reasonably fast:
Read the file as a whole into memory, as a single string, using Get-Content's -Raw switch...
This is orders of magnitude faster than the line-by-line streaming that Get-Content performs by default.
... then use regex processing to strip the first 10 lines ...
... and save the trimmed content back to disk.
Important:
Since this rewrites the file in place, be sure to have a backup copy of your file.
Use -Encoding with Get-Content / Set-Content to correctly interpret the input / control the output character encoding (PowerShell fundamentally doesn't preserve the information about the character encoding of a file that was read with Get-Content). Without -Encoding, the default encoding is the system's active ANSI code page in Windows PowerShell, and, more sensibly, BOM-less UTF-8 in PowerShell (Core) 7+.
# Use -Encoding as needed.
(Get-Content -Raw in.csv) -replace '^(?:.*\r?\n){10}' |
Set-Content -NoNewLine in.csv
If the file is too large to fit into memory:
If you happen to have WSL installed, an efficient, streaming tail solution is possible:
Note:
Your input file must use a character encoding in which a LF character is represented as a single 0xA byte - which is true of most single-byte encodings and also of the variable-width UTF-8 encoding, but not of, say, UTF-16.
You must output to a different file (which you can later replace the input file with).
bash.exe -c 'tail +11 in.csv > out.csv'
Otherwise, line-by-line processing is required.
Note: I'm leaving aside other viable approaches, namely those that either read and write the file in large blocks, as zett42 recommends, or an approach that collects (large) groups of output lines before writing them to the output file in a single operation, as shown in Theo's helpful answer.
Caveat:
All line-by-line processing approaches risk inadvertently changing the newline format of the original file: on writing the lines back to a file, it is invariably the platform-native newline format that is used (CLRF on Windows, LF on Unix-like platforms).
Also, the information as to whether the input file had a trailing newline or not is lost.
Santiago's helpful answer shows a solution based on .NET APIs, which performs well by PowerShell standards.
Brice came up with an elegant and significant optimization that lets a .NET method perform the (lazy) iteration over the file's lines, which is much faster than looping in PowerShell code:
[System.IO.File]::WriteAllLines(
"$pwd/out.csv",
[Linq.Enumerable]::Skip(
[System.IO.File]::ReadLines("$pwd/in.csv"),
10
)
)
For the sake of completeness, here's a comparatively slower, PowerShell-native solution using a switch statement with the -File parameter for fast line-by-line reading (much faster than Get-Content):
& {
$i = 0
switch -File in.csv {
default { if (++$i -ge 11) { $_ } }
}
} | Set-Content out.csv # use -Encoding as needed
Note:
Since switch doesn't allow specifying a character encoding for the input file, this approach only works if the character encoding is correctly detected / assumed by default. While BOM-based files will be read correctly, note that switch makes different assumptions about BOM-less files based on the PowerShell edition: in Windows PowerShell, the system's active ANSI code page is assumed; in PowerShell (Core) 7+, it is UTF-8.
Because language statements cannot directly serve as pipeline input, the switch statement must be called via a script block (& { ... })
Streaming the resulting lines to Set-Content via the pipeline is what slows the solution down. Passing the new file content as an argument, to Set-Content's -Value parameter would drastically speed up the operation - but that would again require that the file fit into memory as a whole:
# Faster reformulation, but *input file must fit into memory as whole*.
# `switch` offers a lot of flexibility. If that isn't needed
# and reading the file in full is acceptable, the
# the Get-Content -Raw solution at the top is the fastest Powershell solution.
Set-Content out.csv $(
$i = 0
switch -File in.csv {
default { if (++$i -ge 11) { $_ } }
}
)
There may be another alternative by using switch to read the files line-by line and buffering a certain maximum amount of lines in a List.
This would be lean on memory consumtion and at the same time limit the number of disk writes to speed up the process.
Something like this perhaps
$maxBuffer = 10000 # the maximum number of lines to buffer
$linesBuffer = [System.Collections.Generic.List[string]]::new()
# get an array of the files you need to process
$files = Get-ChildItem -Path 'X:\path\to\the\input\files' -Filter '*.txt' -File
foreach ($file in $files) {
# initialize a counter for omitting the first 10 lines lines and clear the buffer
$omitCounter = 0
$linesBuffer.Clear()
# create a new file path by appending '_New' to the input file's basename
$outFile = '{0}\{1}_New{2}' -f $file.DirectoryName, $file.BaseName, $file.Extension
switch -File $file.FullName {
default {
if ($omitCounter -ge 10) {
if ($linesBuffer.Count -eq $maxBuffer) {
# write out the buffer to the new file and clear it for the next batch
Add-Content -Path $outFile -Value $linesBuffer
$linesBuffer.Clear()
}
$linesBuffer.Add($_)
}
else { $omitCounter++ } # no output, just increment the counter
}
}
# here, check if there is still some data left in the buffer
if ($linesBuffer.Count) { Add-Content -Path $outFile -Value $linesBuffer }
}
Working with files named in Japanese, and having trouble getting the encoding to process right. After running
chcp 50222
$OutputEncoding = [console]::outputencoding
[Console]::OutputEncoding = [Text.Encoding]::GetEncoding(50222)
I can view Japanese properly on the console, and see things like "【お題箱】琴浦さん", it shows up fine in the dir listing as it should; and when redirecting it to a file, it stores properly.
HOWEVER, when I try to pipe things through the tee commandlet, to see it on the console and feed it to a file at the same time, I get "・・・・・｡・・・ｴ・ｵｦ・ｹ・・ｾ・・ｭ・" instead.
Best I can tell it's being re-encoded to something else between being output to the console, and being fed into tee.... so what can I do to fix that? Or is there something that'd do it better than tee?
(I've also noticed that things fed into tee from a 3rd-party download manager I have, have a significant delay until it shows up on the screen. It will pause a while, show a few screens in a burst, pause for a while, show another few screens, etc)
Based on Get-Help Tee-Object -Full, the command always uses Unicode (meaning UTF-16 LE or code page 1200) encoding. Code page 50222 (iso-2022-jp/Japanese (JIS-Allow 1 byte Kana - SO/SI)) isn't a standard encoding that Add-Content or Out-File supports, either, so the common workaround of Add-Content -Passthru won't work. I suspect you'll have to use a StreamWriter to even be able to write this encoding to a file.
I also have no idea if the console host that PowerShell uses actually respects chcp.exe or if it supports code page 50222.
Keep in mind, too, that internally all strings in .Net are Unicode (code page 1200). If there are glyphs that cannot be represented with code page 1200 that can be represented with code page 50222, you may have problems.
Try duplicating Tee-Object with ForEach-Object and a StreamWriter:
$Encoding = [System.Text.Encoding]::GetEncoding(50222)
$Append = $true
$StreamWriter = New-Object System.IO.StreamWriter -ArgumentList $OutputFile, $Append, $Encoding
previousCommand.exe | ForEach-Object {
$StreamWriter.WriteLine($_);
$_;
}
$StreamWriter.Close();
My real suspicion, though, is that you may end up having to work very hard to get the system to take input in this encoding and to treat it correctly.
Using Powershell to Strip Content from PDF While Keeping PDF Format.
My Task:
I have been attempting to perform what would be a simple task if the documents were not in PDF format. I have a bunch of PDFs that have unwanted data before the bulk of usable data starts, this is anything that comes before ‘%PDF’ in the documents. A script that pulls all the desired data and exports it to a new file was needed. That part was super easy.
The Problem:
The data that is exported appears to be formatted correctly, except it doesn’t open as a PDF anymore. I can open it in Notepad++ and it looks identical to one that was clean manually and works. Examining the raw code of the Powershell altered PDF it appears that the ‘lines’ are much shorter than they should be.
$Path = 'C:\FileLocation'
$Output = '.\MyFile.pdf'
$LineArr = #()
$Target = Get-ChildItem -Path $Path -Filter *.pdf -Recurse -ErrorAction SilentlyContinue | Get-Content -Encoding default | Out-String -stream
$Target.Where({ $_ -like '*%PDF*' }, 'SkipUntil') | ForEach-Object{
If ($_.contains('%PDF')){
$LineArr += "%" + $_.Split('%')[1]
}
else{
$LineArr += $_
}
}
$LineArr | Out-File -Encoding Default -FilePath $Output
I understand the PDF format doesn't really use lines, so that might be where the problem is being created. Either when the data is being initially put into an array, or when it’s being written the PDF format is probably being broken. Is there a way to retain the format of the PDF while it is modified and then saved? It’s probably the case that I’m missing something simple.
So I was about to start looking at iTextSharp and decided to give an older language a try first, Winbatch. (bleh!) I almost made a screen scraper to do the work but the shame of taking that route got the better of me. So, the function library was the next stop.
This is just a little blurb I spit out with no error checking or logging going on at this point. All that will be added in along with file searches later. All in all it manages to clear all the unwanted extras in the PDF but keeping the exact format that is required by PDFs.
strPDFdoco = "C:\TestPDFs\Test.pdf"
strPDFString = "%%PDF"
strPDFendString = "%%%%END"
If FileExist(strPDFdoco)
strPDFName = ItemExtract(-1, strPDFdoco, "\")
strFixedPDFFullPath = ("C:\TestPDF\Fixed\": strPDFName)
strCurrentPDFFileSize = FileSize(strPDFdoco) ; Get size of PDF file
hndOldPDFFile = BinaryAlloc(strCurrentPDFFileSize) ; Allocate memory for reading PDF file
BinaryRead(hndOldPDFFile, strPDFdoco) ; Read PDF file
strStartIndex = BinaryIndexEx(hndOldPDFFile, 0, strPDFString, #FWDSCAN, #FALSE) ; Find start point for copy
strEndIndex = BinaryEodGet(hndOldPDFFile) ; find eof
strCount = strEndIndex - strStartIndex
strWritePDF = BinaryWriteEx( hndOldPDFFile, strStartIndex, strFixedPDFFullPath, 0, strCount)
BinaryFree(hndOldPDFFile)
ENDIF
Now that I have an idea how this works, making a tool to do this in PS sounds more doable. There's a PS function out there in the wild called Get-HexDump that might be a good base to educate myself on bits and hex in PS. Since this works in Winbatch I assume there is some sort of equivalent in AutoIt and it could be reproduced in most basic languages.
There appears to be a lot of people out there trying to clear crud from before the header and after the end of their PDF docos, Hopefully this helps, I've got a half mill to hit with whatever script I morph this into. I might update with a PS version if I decide to go that route again, and if I remember.
I discovered that I could force a Unicode file to ASCII using the script below, which is really great. I assume it's based on my environment or Windows default, but it's adding a CR and LF at the end of each line. Is there a way to force just a LF character rather than both without loading the entire file into memory? I have seen some solutions that load the entire file into memory and basically do a string replace, which won't work because some of my files are multiple GB.
Thanks!
get-content -encoding utf8 $inputFile | Out-file -force -encoding ASCII $outputFile
I suggest you use .NET System.File.IO classes from within your script. In particular the System.File.IO.StreamWriter class has a property, NewLine which you can set to whatever characters you want the line terminator characters to be. (Although to be readable by StreamReader the line terminator chars must be \n or \r\n (in C/C++ notation because of conflict with SO and PS on backtick)).
Secondary benefit of using IO.StreamWriter, according to this blog is much better perf.
Basic code flow is something like this (not tested):
# Note that IO.StreamWriter will use process's current working directory,
# not PS's. So safer to specify full paths
$inStream = [System.IO.StreamReader] "c:\temp\orig.txt"
$outStream = new-object System.IO.StreamWriter "c:\temp\copy.txt",
[text.encoding]::ASCII
$outStream.NewLine = '`n'
while (-not $inStream.endofstream) {
$outStream.WriteLine( $instream.Readline())
}
$inStream.close()
$outStream.close()
This script should have constant memory requirements, but hard to know what .NET might do under the covers.
I want to find a piece of text in a large xml file and want to replace with some other text. The size of the file is around (50GB). I want to do this in command line. I am looking at PowerShell and want to know if it can handle the large size.
Currently I am trying something like this but it does not like it
Get-Content C:\File1.xml | Foreach-Object {$_ -replace "xmlns:xsi=\"http:\/\/www\.w3\.org\/2001\/XMLSchema-instance\"", ""} | Set-Content C:\File1.xml
The text I want to replace is xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" with an empty string "".
Questions
Can PowerShell handle large
files
I don't want the replace to happen in
memory and prefer streaming assuming
that will not bring the server to
its knees.
Are there any other approaches I can take (different
tools/strategy?)
Thanks
I had a similar need (and similar lack of powershell experience) but cobbled together a complete answer from the other answers on this page plus a bit more research.
I also wanted to avoid the regex processing, since I didn't need it either -- just a simple string replace -- but on a large file, so I didn't want it loaded into memory.
Here's the command I used (adding linebreaks for readability):
Get-Content sourcefile.txt
| Foreach-Object {$_.Replace('http://example.com', 'http://another.example.com')}
| Set-Content result.txt
Worked perfectly! Never sucked up much memory (it very obviously didn't load the whole file into memory), and just chugged along for a few minutes then finished.
Aside from worrying about reading the file in chunks to avoid loading it into memory, you need to dump to disk often enough that you aren't storing the entire contents of the resulting file in memory.
Get-Content sourcefile.txt -ReadCount 10000 |
Foreach-Object {
$line = $_.Replace('http://example.com', 'http://another.example.com')
Add-Content -Path result.txt -Value $line
}
The -ReadCount <number> sets the number of lines to read at a time. Then the ForEach-Object writes each line as it is read. For a 30GB file filled with SQL Inserts, I topped out around 200MB of memory and 8% CPU. While, piping it all into Set-Content at hit 3GB of memory before I killed it.
It does not like it because you can't read from a file and write back to it at the same time using Get-Content/Set-Content. I recommend using a temp file and then at the end, rename file1.xml to file1.xml.bak and rename the temp file to file1.xml.
Yes as long as you don't try to load the whole file at once. Line-by-line will work but is going to be a bit slow. Use the -ReadCount parameter and set it to 1000 to improve performance.
Which command line? PowerShell? If so then you can invoke your script like so .\myscript.ps1 and if it takes parameters then c:\users\joe\myscript.ps1 c:\temp\file1.xml.
In general for regexes I would use single quotes if you don't need to reference PowerShell variables. Then you only need to worry about regex escaping and not PowerShell escaping as well. If you need to use double-quotes then the back-tick character is the escape char in double-quotes e.g. "`$p1 is set to $ps1". In your example single quoting simplifies your regex to (note: forward slashes aren't metacharacters in regex):
'xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"'
Absolutely you want to stream this since 50GB won't fit into memory. However, this poses an issue if you process line-by-line. What if the text you want to replace is split across multiple lines?
If you don't have the split line issue then I think PowerShell can handle this.
This is my take on it, building on some of the other answers here:
Function ReplaceTextIn-File{
Param(
$infile,
$outfile,
$find,
$replace
)
if( -Not $outfile)
{
$outfile = $infile
}
$temp_out_file = "$outfile.temp"
Get-Content $infile | Foreach-Object {$_.Replace($find, $replace)} | Set-Content $temp_out_file
if( Test-Path $outfile)
{
Remove-Item $outfile
}
Move-Item $temp_out_file $outfile
}
And called like so:
ReplaceTextIn-File -infile "c:\input.txt" -find 'http://example.com' -replace 'http://another.example.com'
The escape character in powershell strings is the backtick ( ` ), not backslash ( \ ). I'd give an example, but the backtick is also used by the wiki markup. :(
The only thing you should have to escape is the quotes - the periods and such should be fine without.