Slashdot Log In
Windows PowerShell in Action
Posted by
samzenpus
on Wed May 02, 2007 01:51 PM
from the read-all-about-it dept.
from the read-all-about-it dept.
jlcopeland writes "For two decades I've hated the command prompt in DOS and Windows.
Inconsistencies abound and everything is a special case. The
fallback on a Microsoft box has been running a Unix shell under Cygwin or
installing Microsoft's
own Services for Unix (or its predecessor, Softway's Interix),
or by scripting in Perl, but those only get
you so far. Having co-written nine years worth of trade rag columns
using mostly Perl as the implementation language for the samples,
and thinking of every problem that comes across
my desk as an excuse to write a little bit of scripting code,
I've got some well-formed views about scripting languages
and what works and what doesn't. That means
I've been eagerly watching the development of PowerShell since it
was called Monad. It's got the advantage of being a unified command-line
interface and scripting language for Windows, even if it does have
a dorky name." Read the rest of Jeffrey's review.
Bruce Payette's Windows PowerShell in Action is a great overview of PowerShell, aimed at an audience that's got some experience with other scripting languages. Bruce's book is a big improvement over Andy Oakley's earlier book, Monad, which I had been using: it's more complete and it's up-to-date for the first release of PowerShell. It's got great (and sometimes amusing) examples, and feels like the Perl Camel book in flow. When I was reading it in the gym or someplace else away from the keyboard, I kept wanting to run back to the office to try something out. There are also useful "why it works this way" digressions, which provide a lot of context. Since Bruce was on the original development team, wrote most of the commandlets, and was responsible for much of the language design, those digressions are more authoratitive than the directors' commentary tracks on most DVDs.
In outline, the nine chapters in the first part of the book build up as you'd expect: overview and concepts, to data types, to operators, to regular expressions, to syntax, to functions, to interpreting errors. It covers that ground better than many language books that now litter my shelves. The explanations are clear, and the examples are almost all exactly on point. It took me a second reading to realize that my complaints about the regular expression sub-chapter wasn't about the chapter itself, but about some of the implementation decisions; that's an argument about style more than substance, and an observation about me, not about Bruce's writing or PowerShell. The first part of the book is the "mandatory reading," if you will, to get the language down and begin exploring on your own.
The second part is where the real applications are covered. That's the part that you especially want to read sitting next to the keyboard. As you'd expect, the example code is available from the publisher's web site to start you off — look for "Example Code" under "Resources." There's a very good discussion of text processing and how-to-handle XML, complete with some not-obvious warnings about traps to avoid. I've been working very carefully through the really good chapter on using GUIs with PowerShell, "Getting Fancy — .NET and WinForms," and my own proof of concept for that has been rebuilding an old C++ data entry application into a much simpler PowerShell script. As a nice side effect, Bruce's book (and the WinForms chapter in particular) provide a gentle overview to some concepts in the .NET framework, which I hadn't had an opportunity to delve into. The appendix on using PowerShell as a management application will be especially useful to system managers; that was one of the original PoweShell target audiences, and the language achieved that goal very well. The appendix on the language's grammar is really useful, and I keep flipping back to it to check on things.
After Oakley's Monad appeared, there was a long gap before the next PowerShell book appeared. Bruce's book looks to be the first of the post-release wave. If all it had going for it was the authoratative pedigree of the writer, it might be worth it, but it's also well-written, well-organized, and thorough, which I think makes it invaluable as both a learning tool and a reference.
You can purchase Windows PowerShell in Action from amazon.com. Slashdot welcomes readers' book reviews -- to see your own review here, read the book review guidelines, then visit the submission page.
This discussion has been archived.
No new comments can be posted.
The Fine Print: The following comments are owned by whoever posted them. We are not responsible for them in any way.
Full
Abbreviated
Hidden
Loading... please wait.
Monad (Score:5, Funny)
Re:Monad (Score:5, Informative)
Parent
At this rate... (Score:5, Insightful)
Re:At this rate... (Score:4, Informative)
"Those who do not understand Unix are condemned to reinvent it, poorly"
Parent
Or, Greenspun's 10th Rule: (Score:5, Funny)
Parent
Re:At this rate... (Score:4, Informative)
Parent
Re:At this rate... (Score:4, Informative)
Wake me up when *nix gets an object-oriented (rather than text-oriented) shell. Because that is what makes Powershell so unique. Yes, it has plenty of builtin functions to make tasks easier, but the real advantage is that everything you pass between commands is an object.
You don't have to worry about interpreting text output - you just access whatever data you want directly. Many of the commands are easily chainable into something like "ls | select fullname,length | sort name | format-list | out-printer".
Parent
Re:At this rate... (Score:4, Funny)
"Far out in the uncharted backwaters of the unfashionable end of the Western Spiral arm of the Galaxy lies a small unregarded yellow sun. Orbiting this at a distance of roughly ninety-eight million miles is an utterly insignificant little blue-green planet whose ape-descended life forms are so amazingly primitive that they still think digital watches are a pretty neat idea." (Douglas Adams)
Parent
Re: (Score:3, Insightful)
Sure, the pipe can pass anything, but as there is no standard to pass anything but ASCII encoded text, you're more or less restricted to use this on Unix otherwise you'll have to modify all the tools: watch how slow the transition to Unicode was..
In theory, one could define a 'common' binary encoding of objects easily on Linux, but then you have to modify each tool, each scripting language to make it work, a huge task..
In practice, espec
Re: (Score:3, Interesting)
Why not just use "rename"?
I'm actually thinking of something just a bit more complex: Not that I have a use for something like that, mind you...Re:At this rate... (Score:4, Informative)
The FOR command [microsoft.com] in the "legacy" Windows shell is pretty powerful, too. It even has horrible syntax, just like its UNIX fathers.
Yes, the legacy Windows shell sucks, but not as badly as most people assume. The NT shell can do a lot of stuff that most people don't even think to try. Great gobs of functionality have been added over the years, starting with Windows NT 3.5. And contrary to what many slashdotters think, the legacy shell on Windows NT-derived systems is not DOS, nor is it 16-bit. CMD.EXE is just another 32-bit or 64-bit process running on the NT kernel.
Parent
Re:PowerShell isn't a Shell It's a scripting langu (Score:5, Insightful)
The idea benind the Unix toolset and shell is that everything is reduced to a common lingo -- a character stream. Each tool can then be "used as designed", or "misused". The classic example is the original Unix spell implementation. The tool designer promises to accept as wide an input range as possible, to output consistent streams, and not be verbose.
The actual use (misuse) to the tool is left to the shell and user.
The Object philosophy means that input to a tool MUST have certain methods available. If the correct methods are not implemented over the object, an adaptor tool must be used. Microsoft ensures that all PowerShell tools work together IN NORMAL USE. Obvious "misuse" is not (necessarily) supported.
This makes common usage cases easy, but makes "outrageous" cases almost impossible (unless you reach for VC++ and write your own adaptors).
As an example -- I do a lot of "performance analysis", which entails examination of log files, conversion to normalized scales, and running the results through GNU Plot to get images to paste into reports. There is almost always a need for custom shell scripting to do the log examination and reduction.
Now, this IS possible in PowerShell, but only by treating it as a "Unix (gasp, how horrible!)" type shell.
Since the exploration phase (and creative "misuse") is my primary area, PowerShell doesn't have much to offer me. But, for a developer living in the straight and narrow land of "how it should work", it is probably the next best thing to sliced bread.
As to the "PowerShell" equivalent of the non-Microsoft world: I find that I still (occasionally) cook up SNOBOL scripts. When writing compilers for an old course, it was the ONLY programming language specifically not allowed for assignments (it made lexing, parsing and generation much too simple).
Parent
Re:At this rate... (Score:5, Interesting)
I was skeptical when I first heard about Monad. I mean it seemed obvious to me that Microsoft just didn't get the point of a "shell" which is supposed to be simple. With a pending install of Exchange 2007, the power shell is required so I figured I'd start to dig into it. I have to say I'm rather impressed.
First of all, it is actually simple. Not only that, but the syntax is EXTREMELY CONSISTENT. And honestly I cannot stress that enough, because if you think you know part of a command you can usually figure out the verb/noun syntax to use. It also allows shortcut versions of commands so you don't have to type the entire "wordy" version of the command. Aliases are supported too. Another cool feature? You can navigate the registry like the rest of the file heirarchy.
I'm a big fan of bash, but I must admit that at times it gets old shuffling stuff with awk and cut and so forth. By getting objects you can take what you want out of it, and not worry about the biggest Unix terror - the text output changing. If whatever you're trying to do doesn't support
Overall it's pretty awesome technology and I must give MS credit where it's due. Not that I'll be switching any of my FreeBSD servers to Windows because of it, but it makes windows administration orders of magnitude better. Too bad it got dumped in Vista. I've heard it will be included in service pack 1 though.
Parent
Re: (Score:3, Informative)
I can offer a link to the download.microsoft.com page that pulls it for Vista. (all links are English x86, if you run English x64, you'll be able to navigate the download center by now).
Vista x86: http://www.microsoft.com/downloads/details.aspx?Fa milyID=c6ef4735-c7de-46a2-997a-ea58fdfcba63&Displa yLang=en [microsoft.com]
XP x86: http://www.microsoft.com/downloads/info.aspx?na=22 &p=1&SrcDisplayLang=en&SrcCategoryId=&SrcFamilyId= [microsoft.com]
Re: (Score:3, Informative)
http://geophile.com/osh/index.html [geophile.com]
http://dispatch.sourceforge.net/ [sourceforge.net]
Don't knock it until you try it (Score:4, Interesting)
Powershell is very powerful. Much more so than cmd.exe. I don't have significant enough experience with bash to compare the two but I would not be surprised to learn Powershell equals if not beats bash at the shell game. I wouldn't say it is ready to replace any of the scripting languages just yet.
I have been using it for a while now and the single (semi-major) problem I can find is memory usage. It is a hog at best, and at worst when you are using it semi-heavily it can easily chew up 1GB of memory. That's even with giving the GC something to work with, ie unsetting $vars when you are done with their data.
Re:Don't knock it until you try it (Score:5, Informative)
Unless MS rewrites all of their other commands to accept STDIN/OUT, Monad will never surpass the shells. The power of the shells isnt' their programming flexibility, it's their ability to incorporate all the other UNIX tools and commands via pipes to do what you want.
Another reason it will never surpass the shells. They're lightweight, and flexible, and I don't need a Garbage Collector running in the back end to clean up my object allocation.
Parent
Re:Don't knock it until you try it (Score:5, Informative)
It really fails in two places. Firstly, it's slow. You wouldn't have thought it possible for a command shell to be that slow, but it is. It's so slow it was actually quicker for me to use explorer. It is god-awfully, mind-bogglingly slow.
The second problem is that it had no easy way of being accessed over a network link, last time I looked at it. So there's no chance of SSHing into a Windows box and administrating it from there, at least not without fiddling with a lot of hacks and workarounds I couldn't get to work.
The other place where unix shells have an advantage over Powershell is in there interface, as Powershell is currently quite basic in that department. There's limited tab completion and a prompt that can be altered (like PS1 under sh derivitives), but not much over that. Certainly nowhere near my personal favourite, Fish [wikipedia.org].
Parent
Re: (Score:3)
I'm stupified that anybody would find value in that. The shells ferry data around. They themselves generally don't need to or have any business actually understanding that. They funnel it to other commands to deal with that. Wrapping of common utilities output in XML is a ridiculous exercise in buzzword compliance that could only possibly help people who are dealing with unstable utiliti
Monads are windowless, get it? (Score:5, Interesting)
I wish they'd kept "monad" as the name. It was a deft tip of the hat to Leibniz's Monadologie, which held that monads were the windowless metaphysical atoms of perception itself.
Well, *that's* unique (Score:5, Funny)
You should switch to bash and the GNU tools.
Oh, wait.
Everyone should follow their lead... (Score:5, Funny)
I sure hope they charge extra for it, make it a resource hog, lock out third-pary extensions, and then discontinue it as soon as I'm dependant on it. I really liked the 1980s and look forward to reliving them.
Great Review (Score:5, Interesting)
Some clever spacing (Score:5, Funny)
I guess I always suspected it was true, but the beastie mascot of BSD made me wonder if there wasn't room for a little UNIX in Hades too.
I'd love Powershell, if it weren't for one thing: (Score:5, Interesting)
It takes a few seconds for the prompt to appear, and if I run a "dir" operation with both cmd.exe and PS in a directory with hundreds of files, cmd.exe will beat it in seconds.
I'm not running a slow machine(core duo 2, 1GB of RAM). Is there something that needs to be configured to make it suck less?
Re: (Score:3, Informative)
So basically, what makes it suck less, is to use it more.
Re:I'd love Powershell, if it weren't for one thin (Score:3, Interesting)
In cmd.exe, "dir" is "dir." It gives you a text listing of the files and subdirectories of your current directory. In PowerShell, "dir" is an alias for "get-childitem," which returns an object array that can either be parsed for display in the console, or passed down the pipeline to another commandlet.
Re:I'd love Powershell, if it weren't for one thin (Score:3, Insightful)
What about MKS? (Score:3, Informative)
A quick intro to Monad (Score:4, Interesting)
The whole point of Monad is that it's not just text, it's objects. So, unlike Unix, where you work with a command and then filter its output (which is just text), the output of Monad, while looking like text, is actually kind of like pointers back to the real thing, so instead of just doing a Unix-style command | filter | filter, you can say "OK, run this command, now of the things it output, go back and tell me this and this about them." Like, "Of these things that are running, tell me which five are using the most CPU time, then tell me the version of each, and how much memory they're using."
PS: "...even if it does have a dorky name"--which name were you referring to: the one that sounds like 'testicle' or the one that makes me think of the Lottery?
You can knock it but it IS VERY USEFUL (Score:3, Interesting)
The only other choice to do something similar on Windows is VB script which I find painful at best. It may not be the best Shell ever but at the moment its the best Windows integrated shell with access to
The book is great by the way, and his blog has loads of useful tips too.
Anybody who is actually going to try it, will find the powershell community extension very useful. G
FYI - PowerShell is not just for Vista (Score:3, Interesting)
...PowerShell is avlaibel for MS OS's older than Vista too:
http://www.microsoft.com/windowsserver2003/techno
Bean Shell (Score:4, Interesting)
Same problems too as well - memory consumption up the wazoo and slow as hell. Every time you've got to do a "pipe" you need to look at miles and miles of API.
Re:It's amazing people still use windows. (Score:4, Funny)
2017, the year of the linux desktop!
Parent
Re:It's amazing people still use windows. (Score:5, Insightful)
I'd say that power users who think they don't need a shell don't know what they're missing ; they could have a lot more power.
The gentleman sitting next to me has recently discovered and raves about WinGrep, a GUI file search/replace utility with RegExp support. It's not bad, but it can't compare with a shell - you can't, for instance, search for a bunch of files containing your desired pattern match and invoke an external utility to process each one. And anything that the original application designer didn't visualise as a feature is excluded. He's easily capable of comprehending grep and sed, which would do the same job for free, but he's more comfortable with the GUI.
In a *nix style shell, the ability to pipeline STDOUT through a whole bunch of little utils is the tool of a real power user - and it has a nice easy learning curve, you can pick up new commands as and when you like, and combine them with old favourites. The downside to the *nix shell is that very often you have to perform some esoteric text processing to get what you want, which means learning tools like awk and sed. Powershell works by passing objects through the pipeline - objects that have useful properties. It's even an improvement with old-style executables that emit pure text - the
The GUI equivalent of a shell for a power user would be a pipeline composer where you can take various widgets representing actions and plug them together. Perhaps something like the DTS transform designer in SQL Enterprise Manager. Or maybe not
Parent
Re: (Score:3, Insightful)
Sure you can live without a vacuum cleaner, but your carpet and overall cleanliness is going to suffer for it. I know people who live in apartments without vacuum cleaners and they are very similar to computers who's users don't use or understand shells.
The philosophy behind textual data (Score:4, Interesting)
Then they have absolutely no idea about what they are doing. The one big advantage in using pipes is that any application can handle text data.
Let me give one example: I use the sort command all the time, it sorts data by lines of text, lines are compared according to criteria passed in command options.
Now, imagine if it depended on binary objects. For every sort operation one would have to write a comparison function to decide which object should come before the other. Writing a special function would mean declaring some form of callback, or maybe some people would call it a closure, whatever. And so on.
Here's one simple command I use when a disk starts getting full to see which directory is the data hog:
du -x / | sort -nr > mem.txt &
What this command does is check the disk usage (du command) in the root directory (/) without looking at symbolic links to other disks (-x option). The result is piped to the sort function, where it's sorted by the numeric value of the first column in reverse order (-nr option). The sorted result is sent to a file named mem.txt. Since checking the whole disk may take some time, it's done in the background (& command). After it finishes, I have a file with the size of each directory in the disk, one line per directory, ordered by size, larger directories first.
See how powerful it is, having data represented as text? Try writing this line as a Powershell script.
Another advantage of having data in text format is that you can test it using the keyboard and screen very easily, no need to run a special debugger.
Instead of trying to reinvent Unix poorly, Microsoft would do a favor to its customers if they accepted Unix is a mighty fine OS and adopted without shame its best features.
Parent
Re: (Score:3, Insightful)
Sort by name:
du -x / | sort --method 'toString'
Sort by size:
du -x / | sort --method 'getSize'
Also you could do things like make your own output formats without having to use
Re: (Score:3)
There is already a comparison function written for each type of object (especially ints and strings, the only objects sort can really compare).
Piping objects is a super-set of the power of piping textual strings, as a sibling poster mentioned: You can always convert objects to strings, but not vice versa.
A powershell equivalent (with my own invented syntax):
du -x / | sort
Ofcourse since
Re:The philosophy behind textual data (Score:5, Insightful)
it treats piped arguments as objects instead of strings, Psh lets programs access the data directly, instead of having to manipulate large amounts of textual data with tools such as grep or PERL.
Then they have absolutely no idea about what they are doing. The one big advantage in using pipes is that any application can handle text data.
Let me give one example: I use the sort command all the time, it sorts data by lines of text, lines are compared according to criteria passed in command options.
Now, imagine if it depended on binary objects. For every sort operation one would have to write a comparison function to decide which object should come before the other. Writing a special function would mean declaring some form of callback, or maybe some people would call it a closure, whatever. And so on.
Here's one simple command I use when a disk starts getting full to see which directory is the data hog:
du -x / | sort -nr > mem.txt &
What this command does is check the disk usage (du command) in the root directory (/) without looking at symbolic links to other disks (-x option). The result is piped to the sort function, where it's sorted by the numeric value of the first column in reverse order (-nr option). The sorted result is sent to a file named mem.txt. Since checking the whole disk may take some time, it's done in the background (& command). After it finishes, I have a file with the size of each directory in the disk, one line per directory, ordered by size, larger directories first.
See how powerful it is, having data represented as text? Try writing this line as a Powershell script.
Another advantage of having data in text format is that you can test it using the keyboard and screen very easily, no need to run a special debugger.
Instead of trying to reinvent Unix poorly, Microsoft would do a favor to its customers if they accepted Unix is a mighty fine OS and adopted without shame its best features.
I strongly disagree with most of what you've said. Here's why.
.NET provides this, of course. Similar technologies could provide the same thing on Linux. It's a tall order, mainly because of all the programs in the world that are written for naive assumptions about the shell environment...)
The supposed "ease" of dealing with text is an illusion - it's a fallacy that's built out of
1: the fact that textual formats are usually organized such that we humans can read them if we send the data out to a console
2: the fact that Unix types have built up a formidable array of text-wrangling utilities to deal with these problems
3: a general assumption that the reading and writing of formats passed between processes won't pose any real challenge to process.
The relative "ease" of text is negated if three corresponding conditions are met in a shell dealing in structured data:
1: Data is structured (or information about the structure of the data is communicated) using mechanisms that are respected by all tools involved. (In other words, there's some kind of Lengua Franca for the structured data -
2: The structured data that is used in inter-process pipelines is given suitable (preferably interactive) display methods
3: An appropriate set of data handling tools are introduced, generic enough to work for most problems and powerful enough to be effective.
The problem with textual formats for structured data is that there will always appear ways that you can do it wrong. For instance, what if a field contains the character (or set of characters) you're expecting to use to delimit the fields? Well, that's why find has a -print0 option, isn't it? Now, what if the field could contain null bytes, too? Then maybe you use escape characters - and the process of reading in the output from the previous command starts to become a more complicated parsing problem.
You cite how easy it is to sort based on a field (and, to extend the exam
Parent
Re: (Score:3, Informative)
Actually, 3D acceleration in Linux is technologically ahead of Windows. What's behind is driver support, although that's coming around.
People use Windows because most people are not looking for the same things in an OS that you are.
Well, nobody in my family uses Windows anymore: they have all switched to either Mac OS or Ubuntu, both of which are considerably less hassle and overall cheaper.
Re: (Score:3, Informative)
$> diff --brief a/ b/
It just seemed worth pointing out.
Re: (Score:3, Informative)
% mkfifo
% mkfifo
% ls a >
% ls b >
% diff
That's just disgusting.
Yes that is disgusting. Fortunately, you're not the first person to notice this. Try this instead:
diff <(ls a) <(ls b)
There is a simila
Re:Windows "power shell"? (Score:5, Interesting)
I really, really want the Linux CLI to be modernized. (Lots of time is spent working on the Linux GUI, but it seems like when it comes to the CLI people are content to let it rot.) I've spent a lot of time trying to figure out how I would go about doing that. I've read a bit about PowerShell - it seems interesting, at least, and promising. For instance:
- It can wrangle live
- It encourages a unified interface (conventions for command options, etc.) for CLI programs and utilities
- It applies these new techniques in conjunction with existing, traditional shell mechanisms, like pipes.
- It endeavors to make documentation and general information about commands easier to access
Now, there's also things I don't like so much - for instance, the distinction between "commandlets" and normal commands. (To be fair, this is largely due to the fact that most existing code in the world is written either for a traditional CLI or a GUI - so most code isn't going to know how to deal with a smart CLI anyway. But I think there are better solutions.)
I think it's kind of a drag that Microsoft may now have a better CLI than Linux - but I think that's a situation that can be changed.
Parent
Re:Windows "power shell"? (Score:5, Interesting)
If you look at Powershell, they've got the "read immediately" process down - commands like "Format-Table" come to mind. Yeah, big deal, right? It's what PERL was born to do. But nevertheless - if you run a command and it generates an object, a meaningful printed representation of that data appears on the console. If you want it to look nicer, there are commands to format the output. I think the shell would be better if the user didn't need to handle that step explicitly - I have some ideas for how that could be done. (I am very interested in writing a Linux shell with these kinds of capabilities.)
As long as Monad has that, it's probably decent. But that's going to be application-specific.
In what sense is Monad's pipeline communication format "application-specific"? It can deal with any
Parent
Re: Windows "power shell"? (Score:5, Insightful)
There is, however, no universally agreed syntax for "objects". Sure, there have been attempts, but I doubt any of them will succeed, maybe ever. Different systems have so vastly different opinions of what an object is, and I believe that is how it should be, because if all systems would have to have the same idea of an object, you would be locking them into a predefined design pattern, and innovation might decrease. I don't know if maybe people said the same thing about bytes in the 50s and 60s, so I wouldn't bet my prediction will turn out to be correct, though.
Of course, this is perfect for Microsoft. They don't want other systems, anyway. As long as anyone can agree on the .NET definition of an "object", Microsoft will be happy. However, even then, the fact remains that not every .NET object is serializable -- you can't just take an arbitrary object and squirt (pun intended) it over the network or store it on disk. As long as you wish to communicate with anything outside your own VM, text (or at least a byte stream) is necessary.
Heh, that's one of the weirder statements I've seen as of late. Kind of like saying that you can't "speak" in a telephone, it's just a PCM stream anyway. Call me weird, but I'd argue that text is human readable by definition. I do (kind of) see your point, though, but I don't agree. Text is always human readable, because it has such an internal structure that makes it human readable with an extremely simple and universally standardized (except for charset) algorithm. If you just have an "object", though, there's no universal algorithm for turning it into a visual structure. Usually, each object class even has its own such algorithm, which isn't usually reversible (unlike text), and not every class even does. To begin with, there is, as I wrote above, no guarantee of any sort that an object is even slightly serializable.Not that I think that you're wrong in every possible way. I definitely think that an object-oriented shell may have its virtues, but it's never going to work outside its own VM. Text is universal, since you can send it anywhere and receive it from anywhere. That "anywhere" includes a human, too.
Parent
Re: (Score:3, Funny)
Re:Come one, sell me this shit. (Score:5, Insightful)
Now imagine the same thing, but instead of passing in strings you could pass in/out native data types, full objects, other methods, etc. That's PowerShell.
Parent
Re: (Score:3, Informative)
Re: (Score:3, Interesting)