Wikipedia:Reference desk/Archives/Computing/2007 April 27

From Wikipedia, the free encyclopedia

Computing desk
< April 26 << Mar | April | May >> April 28 >
Welcome to the Wikipedia Computing Reference Desk Archives
The page you are currently viewing is an archive page. While you can leave answers for any questions shown below, please ask new questions on one of the current reference desk pages.


Contents


[edit] April 27

[edit] Viewing deleted page content in a PDF file

Acrobat Professional tells me (through the Examine Document function) that a PDF file contains "deleted hidden page or image content." How can I undelete or view this deleted content? NeonMerlin 00:11, 27 April 2007 (UTC)

[edit] Creating an exception to the Windows Firewall for Mac/PC networking

I'm currently using a desktop Windows XP Home Edition, SP2, and an Apple MacBook at home. They're connected with a basic wireless router. I was easily able to configure the PC to connect to the Macintosh and access its files, but I have a harder time getting the Mac to find the PC. The only way I've been able to achieve it thus far is by disabling the Windows Firewall. Is there any way to configure an exception to the firewall to allow my Mac to easily access the PC's files? Thanks! --McMillin24 contribstalk 03:15, 27 April 2007 (UTC)

-Try clicking Start -> run and enter "wscui.cpl" and press enter. A Security Centre page will load, Click Windows Firewall and then the Exceptions tab at the top. In here tick the File and Printer sharing box and click OK. Once done windows will happy share your files to the network, but make sure you have a router which doesn't share your netbios port (port 139) or anyone on the internet who knows your IP address will be able to browse your hard disk. A good way of checking this is by visiting http://www.grc.com and use their Shields Up firewall tester for free. Hope this helps, Phill Upson (support@expertrepair.co.uk)

[edit] Sticky Excel/Access 2000 problem regarding importing DB with URLs hidden with anchor text

If anyone can either answer my question or direct me to where to get more info, I'd be most appreciative.

I have a large flat file database in Excel2000 which includes a bunch (2000 or so) hyperlinks with anchor text. I'd like to upload the whole thing into Access2000 but it turns the hyperlinks into plain text (showing the anchor text, not the web addresses), and changing the field type from text to hyperlink isn't available in the import.

So... I'd like to figure out a way to do this. The way that seems most obvious is to remove the anchor text in the excel file so that I can upload as html, but I'm open to other solutions. So, how do I remove the anchor text to reveal the URLs? Or how do I do this another way? Thanks in advance. Anchoress 03:20, 27 April 2007 (UTC)

My untested guess is that it would be easiest to write a little VBScript that will go through and separate the links into two columns, one of links and the other of anchor text. Now unfortunately I don't know anything about using hyperlinks in VBScript and imagine that will take a little hunting around to get right — potentially in the Hyperlinks collection but I don't really know. I see a lot on a quick Google search on adding hyperlinks but imagine there must also be ways of accessing their properties as well. Then it's just a matter of iterating through the recordset and putting the links into a new column or something along those lines, which will take a minute or two but should work fine. --24.147.86.187 03:45, 27 April 2007 (UTC)
If your input file is a comma-separated values file like this:
 thing1,thing2,<a href="http://some/url">some anchor text</a>,thing3
then a bit of Perl like this:
 perl -pe 's#<a href="([^"]+)">([^<]+)</a>#$1,$2#' < input.csv > output.csv
will turn your rows into this:
 thing1,thing2,http://some/url,some anchor text,thing3
--TotoBaggins 04:02, 27 April 2007 (UTC)
Thanks for the answers you guys, but it's not as simple as that. The data was copied from web pages as anchor-texted hyperlinks, so there's not two sets of data to separate in the .xls file; the actual typed contents of the cells is the anchor text. It's going to have to be something to do with a dynamic paste or something, I have no idea. I've fiddled with a bunch of paste specials, and looked through the installed formulas, but I don't know. Anchoress 05:15, 27 April 2007 (UTC)
Saving (or exporting) the spreadsheet as a .csv file will presumably turn the hyperlinks into their natural <a href="..."> state, and you could then do the text manipulation above. I don't have Excel, so I'm somewhat out of my element, but it can't be so daft as to just discard the URLs when exporting. One hopes. --TotoBaggins 05:23, 27 April 2007 (UTC)
No, it didn't, unfortunately. It just exported the anchor text. I was able to paste the plain URL as a hyperlink to another cell after copying, with a particular paste function, but it only worked with one cell, not a range. Anchoress 05:36, 27 April 2007 (UTC)
Here is some VBScript that will cycle through all hyperlinks in the file and display their name and link. I wouldn't run it on a large file, but it should give you (or others) and idea of how to go about extracting the link information.
For Each h In Worksheets(1).Hyperlinks
 MsgBox h.Name
 MsgBox h.Address
Next h
I'd do more but I'm on the run... --24.147.86.187 12:23, 27 April 2007 (UTC)
Here's a link to a similar request. --TotoBaggins 12:37, 27 April 2007 (UTC)
Thanks Toto and 24, checking the solutions out. Anchoress 22:18, 27 April 2007 (UTC)

Have you tried the "Replace all" function - but I understand that it won't help you with the part of the URL code that you want to replace that changes with each one - instead you'd need something more dynamic to do that - where it could say "Delete everything after the '>'". Check the link above from TotoBaggins which seems to take care of what you're looking for. I'd be able to help you otherwise by writing up some code, but I battle a bit with VB - so worst case come back here and I'll write it for you. Rfwoolf 13:10, 27 April 2007 (UTC)

That won't work, it doesn't store them as HTML code it stores them as Hyperlink data objects. --140.247.240.159 16:45, 27 April 2007 (UTC)
Here's another solution: I think that Access allows a hyperlink data type, no? It might only be in later versions than 2000.... --140.247.240.159 16:45, 27 April 2007 (UTC)
Yes, it does, but as I said in my question, it isn't available as a field type in this import, and by the time the data is imported, it's already been delinked. I'm going to try a lazy solution suggested on the Access Developers yahoogroup, and if that doesn't work I'm going to try the VB solutions proposed here. Anchoress 22:18, 27 April 2007 (UTC)
XML alternative: One approach may be to save the file in "XML Spreadsheet format" (if it is supported by your version of Excel). What that will do for you is convert the "hyperlink objects" into readily-modified text that you can then extract with an XML parser or script or whatever. If I remember correctly, Access also has an XML format that it understands for import, so you might not even have to mess with the XML before importing to Access. (See e.g., Microsoft_Office_Excel#File_formats apparently, XML output is not supported in excel 2000? :( .dr.ef.tymac 04:23, 28 April 2007 (UTC)
Yeah, that sounds good, I'm going to try it. Anchoress 09:10, 28 April 2007 (UTC)
Thanks dr.ef.tymac, your many suggestions and the work you've put into helping me has been very helpful. Unfortunately I just checked and my version of Excel doesn't support the XML format, as you feared. I'll check some of the other solutions, I'm just feeling a bit discouraged for now and I think I'll take a break from this whole thing. ;-) Anchoress 04:48, 30 April 2007 (UTC)
UDL alternative: Another approach may be to specify an OLE DB connection using a UDL file that uses the OLEDB data provider for msft excel. Then use a script to pipe that from excel into access using regular SQL queries. dr.ef.tymac 18:08, 28 April 2007 (UTC)

Request for clarification:
If you want me to come up with the VB code, please clearly explain what you want. Is this correct:
The cells in Excel contain hyperlinks and the raw code for each one reads

<a href="http://www.example1.com">Example1</a>

and you want to extract it to say

www.example1.com

Is that correct?
Rfwoolf 08:15, 28 April 2007 (UTC)

You certainly don't need to do anything for me, but no, that's not at all what it's like. As I've said a couple of times, it's an embedded object, so the actual contents of the cells is just the anchor text. There's no <a href> stuff at all. So if for instance the text is copied and pasted somewhere, or if the file is saved as .txt or .csv, or if it's imported into access, the URL field shows just the plain text representation of the anchor text. Anchoress 09:10, 28 April 2007 (UTC)

[edit] Running a website anonymously

I want to make an adult website, but don't want anyone to be able to trace it back to me. Is there any way to get a website without giving out my name, credit card numbers, or phone numbers? I wouldn't want to necessarily make a profit, and I wouldn't want one of those "payment required in order to view the site". Banner ad support is okay because I need to no go into debt because visitors. I guess I should also say- I only want to be anonymous because I'm very shy-- i'm certainly not trying to be criminal or anything, so I'm don't care if the police or FBI or legitimate authorities have the ability to trace the site back to me, I want to make sure that nobody else can find out who I am. Monoceros —The preceding unsigned comment was added by Monoceros144 (talk • contribs) 04:46, 27 April 2007 (UTC).

Try Zoomshare my dad made a great art web site to host pics of his paintings and it's free (Note: There is a paid version of zoomshare that has bigger storage and removes the adds). --Lwarf 10:07, 27 April 2007 (UTC)
Well if you want your own domain you basically can't do it. You have to somehow pay them and you will need to provide some personal information for the record. --antilivedT | C | G 10:26, 27 April 2007 (UTC)
Just to let you know that some "hosts" do not allow such content on their servers. Typically you might be able to get away with it for a short period of time, but ultimately if it's not allowed then your account may be suspended - so just a tip, you might want to read over the terms and conditions of any given host -- just in case you didn't know.
Also, when you register a Domain name, you provide a variety of information about who owns the domain, their billing address, their contact details, the webmaster name, the server person's name, etc. but I've seen quite a few hosts offering to "hide" this information from the general public - but charging you to do so - that might be an option. Of course all you need to do is look at a way to obfuscate this information when you register the domain - that's not to say you must do anything illegal, but you might look into whether you can get your host to use their details and not yours - although that would mean that they own the domain name and not you. My advice is to look for some "Adult hosts" or just for hosts - and talk to them about your request and see what they can do for you.
Good luck Rfwoolf 12:42, 27 April 2007 (UTC)
Sorry if this sounds a bit like advertising, but ComputingHost allow you to register domains and keep your information completely private in WHOIS searches. They also allow you to host adult content. --saxsux 18:09, 27 April 2007 (UTC)
Are you sure they're real? A lot of these deals that say they're webhosting are scams—that's why I like to go by word of mouth. Most hosts do not allow adult content and will end your service if they find you with it up. [Mαc Δαvιs] ❖ 11:08, 28 April 2007 (UTC)
Yup. They host several of my websites. --saxsux 12:48, 28 April 2007 (UTC)
I got given a "take-away" website domain, wherein you enter the name of your website i.e. www.adult.com and then you can edit it how you want, all you have to do is buy it again every year, for about $3. not sure how you do it though i haven't even opened the box.

Nebuchandezzar 14:49, 3 May 2007 (UTC)

[edit] what is the architecture of a core 2 duo processor?

what is the architecture of a core 2 duo processor? —The preceding unsigned comment was added by 203.177.158.106 (talk) 06:33, 27 April 2007 (UTC).

The Intel Core 2 Duos are X86 Architecture CPUs. The Core 2 Article has information on this, as well as the instruction sets the processor uses. These processors also get called "X86-64" sometimes (in reference to them having a 64bit instruction set). gorffy 06:42, 27 April 2007 (UTC)

[edit] RDBMS

Where the relational data base management system really works...?

See relational databases and relational database management systems. --h2g2bob 11:02, 27 April 2007 (UTC)

[edit] Hard Disk Partitioning Query

Hi All, Could anyone please explain me,In the partitions, which one is better (state the reasons):

  • Create one drive.
  • Create two drives.
  • Create three or more drives.

Thank you in advance for any help. —The preceding unsigned comment was added by 193.188.105.228 (talk) 12:18, 27 April 2007 (UTC).

That sounds awfully like a homework question. The answer is, of course, "it depends". Without more information about the intended usage, it's like asking how long a piece of string is. --Robert Merkel 13:20, 27 April 2007 (UTC)

hi, well,it is a homework question,and I googled for the answer too but couldn't get any useful information. to be more specific,this partioning is for Windows XP's FAT and NTFS file systems. so plzz provide me with the answer.

He did; the answer is "it depends". --TotoBaggins 14:07, 27 April 2007 (UTC)
42 of course. Seriously, what are you trying to do? Splintercellguy 15:05, 27 April 2007 (UTC)
Tell your teacher, from me, that that is a dumb question. But anyway, take a look at the disk partitioning article. Partitions are most used to allow more than one operating system to live on a single computer disk drive. The general rule is one OS per partition. Linux normally uses an extra swap partition, so a Windows-Linux dual-boot would be 3 partitions total. Partitions can also be used for encryption (an encrypted disk) or to use a different disk format for part of the disk. --h2g2bob 15:26, 27 April 2007 (UTC)
An answer from a practical point of view - I would say 2. I normally create a small partition for XP (say between 10 and 25 gig for XP depending on how many programs you are going to install). The remaining space of your drive is your other partition and could be used for data/binaries/source/documents/etc and installing large programs that store data. The reasons for this are: 1. XP's partition gets very fragmented as your swap and temp files are here, so you can defragment often. 2. If XP goes kaput, which happens often, you can reinstall without bothering to backup and restore your data on the other partition. Sandman30s 18:27, 29 April 2007 (UTC)
Unless you need to keep your data separate as said above, partitioning a single drive doesn't offer any huge benefits. If you suffer a physical disk problem, all your data is on a bad drive and there aren't really any performance gains. Separate disks in a RAID configuration however offer all kinds of benefits. For reference my linux box uses a boot partition, root, data, swap and a separate disk for backups, so 5 partitions in all, i only do this to ease kernel and other software updates. —The preceding unsigned comment was added by Phillip.upson (talk • contribs) 16:03, 3 May 2007 (UTC).

[edit] Rogue television streaming

a somewhat-recent edition of the wall street journal featured an article that talked about rogue sites posting television broadcasts and the like ... in order to bypass all the advertising one has to sit through in order to watch the same content on the "official" network sites. Two men were named. Does anyone remember the names of the men, also is there a link to the article somewhere on the net for free? I dont remember the publication date and i dont have the paper anymore. NoClutter 14:40, 27 April 2007 (UTC)

[edit] del.icio.us for windows XP (tagging local files and folders just like you can tag links)

Is there anything such as del.icio.us but one that is usable on the desktop instead of just for web pages? NoClutter 14:42, 27 April 2007 (UTC)

I'm not exactly sure what you mean. del.icio.us is a Social Bookmarking website for sharing bookmarks to other web sites. What would you share on your desktop between different users? Are you wanting to share files or internet shortcuts to other websites on your WinXP desktop? —Mitaphane ?|! 23:08, 28 April 2007 (UTC)
No. I mean the ability to "tag" local files and folders on my own machine the same way one can "tag bookmarks". (be able to manage tags, tag groups, and descriptions on individual files locally, exactly like the social bookmarking site, without having to use the internet at all). NoClutter 05:06, 29 April 2007 (UTC)
Gotcha. I know XP you can put metadata on files by right clicking them and hitting properties. However you can't put your own user-defined fields in these properties (as I presume you'd like to do) or manage them in a easy way in Windows Explorer. Had Microsoft, went through using WinFS for Vista it might have been able to accommodate what you'd like to do. The only thing else I could suggest is searching the internet for something like windows file metadata manager. Good luck. —Mitaphane ?|! 20:42, 29 April 2007 (UTC)

[edit] "blind group" email

Is there a general way to send an email to a group, with the group members thinking that they are receiving it individually? In other words, the recipient's address will be the only one on the email, but it will go out to a group of people. Thanks. 150.174.104.137 14:56, 27 April 2007 (UTC)

You could try using blind carbon copy. Other than that, it's whatever your email software or email service provides for you. --h2g2bob 15:02, 27 April 2007 (UTC)

Thanks. 150.174.104.137 15:21, 27 April 2007 (UTC)

As said above, sending using the bcc field is easiest. Other options include using a mailing list manager (mailman if you are a linux user), creating a script for your email client which processes one email per address book entry in a loop (e.g. Outlook would support this approach). Finally if you wanted a mailing list your users could manage, try reading up on PHP, you could let users subscribe and unsubscribe via a webpage, their details get stored in a database (could also be a text file) and then you can use a php script to loop through the address book sending mail on your behalf. Mail me if you need more help. Phillip.upson 16:09, 3 May 2007 (UTC)

[edit] Installing a Windows XP MBR

I have two hard drives: we'll call them Drive 1 and Drive 2. My installation of Windows is on Drive 2. For historical reasons, the boot drive is Drive 1. Presumably (bootloaders confuse me, so I might have something wrong) when the BIOS starts it looks for a bootloader in the MBR of Drive 1, which then boots the NTLDR on Drive 2, which in turn boots Windows.

Setting the BIOS to boot Drive 1 works fine. Setting it to boot Drive 2 leads to an error message, presumably because there is no Windows bootloader in the MBR of Drive 2, so naturally nothing can boot.

Now, I want to format Drive 1 so I can install Linux on it (yes, yes, hold the cheering), but presumably formatting Drive 1 would remove the MBR and stop me being able to run Windows. I know GRUB should be able to boot Windows, but in case of any problems I'd rather that I could just boot Windows off Drive 2: but I have no idea how to install a Windows bootloader on the Drive 2 MBR.

Any ideas? Sum0 16:35, 27 April 2007 (UTC)

If you have your WinXP CD this can be done from the recovery console. Boot from the CD and enter the console at the appropriate time. Then, use the fixboot and fixmbr commands as shown: fixboot and fixmbr gorffy 16:55, 27 April 2007 (UTC)
I have tried FIXBOOT, which seemed to work fine, but I'm not sure that that was the problem. I ran FIXMBR too: it said I had a "non-standard or invalid master boot record". After some consideration, I went ahead anyway, and it said it completed successfully - but it didn't solve anything. Which I don't understand, unless possibly it wrote to the Drive 1 MBR by default (I didn't specify a drive). I'll try specifying Drive 2. Sum0 17:03, 27 April 2007 (UTC)
No luck with that... Sum0 17:40, 27 April 2007 (UTC)
I believe Hiren's Boot CD has MBR tools that can fix your problem. Splintercellguy 22:03, 27 April 2007 (UTC)
I think all XP MBRs are created equal, so you can use dd to copy the MBR from the working drive to the other one:
  • boot into knoppix or another liveCD (personally I found Ubuntu's install/liveCD to be very satisfactory)
  • dd if=/dev/hda of=/dev/hdb bs=512 count=1
that's assuming /dev/hda is drive1 and /dev/hdb is drive2 : you should change these as appropriate - if you get the above wrong you'll copy drive2's bad MBR over drive1's good one, and thus be doublescrewed - so test mount them both to make superultramegasure. -- Finlay McWalter | Talk 22:12, 27 April 2007 (UTC)
OK, I'll consider that. That just copies the boot code, right? It won't screw around with the other MBR stuff (I'm pretty hazy, but I believe the stuff there is Important)? Also, one drive is FAT32 and the other is NTFS. That won't matter, will it? Hiren's Boot CD looks worth a try too. Sum0 22:23, 27 April 2007 (UTC)
No tool can get you fired faster than dd, so you're wise to be cautious. That 512 will copy all of the MBR, which Master boot record notes does include a partition table - copying 446 bytes instead should copy just the code (but it's not clear that this will be sufficient for you, as we don't really seem to know what the condition of your drive2 MBR is. If you're nervous (wisely) you can save drive2's MBR to a file on a USB flash drive with dd if=/dev/hdb of=/media/usbdisk/drive2_old_mbr bs=512 count=1, so you can always restore it from there (by exchanging if and of) later if things go bad. -- Finlay McWalter | Talk 22:34, 27 April 2007 (UTC)
Progress! Before trying anything drastic with dd, I decided to open up XP Disk Management. Right-clicked on Drive 2, clicked "Set as active partition" and now when I boot up, I get "NTLDR not found". Obviously not quite fixed but I think I can solve it from here (installing NTLDR shouldn't be too hard, I hope!). Thanks everyone. Sum0 19:07, 28 April 2007 (UTC)

[edit] compilers and interpreters.

hi ... i have doubt abt the difference between compilers and interpreters and assemblers..pls help me thanx in advance chakradhar 17:43, 27 April 2007 (UTC).

A compiler compiles source code into machine code that the computer can run directly. An interpreter interprets code as it's run, changing it into 'directions' for the computer as it goes along. That's the best I can explain it, someone else can probably do better. The articles on both topics cover differences between the two and advantages to each. -- Consumed Crustacean (talk) 17:52, 27 April 2007 (UTC)
Though a lot of "interpreters" these days use a form of compilation to increase performance. As was stated, read the articles. -- mattb 17:56, 27 April 2007 (UTC)
Bytecode, Just-in-time compilation. -- Consumed Crustacean (talk) 17:59, 27 April 2007 (UTC)
A compiler turns a portable high-level language like this:
int square_it(int x) {
    int squared = x * x;
    return squared;
}
into machine-specific assembly language like this:
       .file   "foo.c"
       .text
.globl square_it
       .type   square_it, @function
square_it:
       pushl   %ebp
       movl    %esp, %ebp
       subl    $16, %esp
       movl    8(%ebp), %eax
       imull   8(%ebp), %eax
       movl    %eax, -4(%ebp)
       movl    -4(%ebp), %eax
       leave
       ret
       .size   square_it, .-square_it
       .ident  "GCC: (GNU) 4.0.2 20050808 (prerelease) (Ubuntu 4.0.1-4ubuntu9)"
       .section        .note.GNU-stack,"",@progbits
An assembler then turns that output into a series of bits that you can feed directly to a CPU with no further changes. It's not readable by (normal) humans, but looks something like this for the file above:
457f 464c 0101 0001 0000 0000 0000 0000
0001 0003 0001 0000 0000 0000 0000 0000
00c4 0000 0000 0000 0034 0000 0000 0028
0007 0006 8955 83e5 10ec 458b 0f08 45af
8908 fc45 458b c9fc 00c3 0000 4700 4343
203a 4728 554e 2029 2e34 2e30 2032 3032
3530 3830 3830 2820 7270 7265 6c65 6165
6573 2029 5528 7562 746e 2075 2e34 2e30
2d31 7534 7562 746e 3975 0029 2e00 6873
7473 7472 6261 2e00 6574 7478 2e00 6164
6174 2e00 7362 0073 632e 6d6f 656d 746e
2e00 6f6e 6574 472e 554e 732d 6174 6b63
0000 0000 0000 0000 0000 0000 0000 0000
...
Traditional interpreters are programs that just read the high-level language and follow its instructions as they go. A modern example is bash. Many modern languages (like Perl and Java) do something more like the compiler example above, except in the last step they interpret the binary file and do what *it* says, rather than feeding it to a CPU. --TotoBaggins 18:54, 27 April 2007 (UTC)
It's incumbent on me (as a fatally flawed nitpicker) to observe that assemblers rarely if ever produce something "you can feed directly to a CPU with no further changes", and indeed the binary example you show is such a case. An assembler generates a binary object file (in this case ELF - the first four bytes, in Intel byte order, are [DEL]ELF). That has to be sent to the linker to have any external references resolved. And even then the file is still ELF (or whatever), and that's still not executable. Finally the code has to be loaded (or "located"), where fixups are resolved, .bss sections wiped, and .data sections initialised from .text. In the case of a "real OS" this is done by the ELF loader, or its equivalent, when you exec the code, but in the case of embedded systems this is done explicitly by the locator, the last element of the tool chain. As a long-time systems developer it vexes me some (and this isn't a dig at anyone posting here, honest) that CS departments spend ages in compiler class on lexical analysis and syntax handling, wave their hands a bit at code generation, and make no mention at all of the latter half of the compilation chain. -- Finlay McWalter | Talk 00:50, 28 April 2007 (UTC)
Yes, good followup! I thought about mentioning loading, but decided it would get too far afield for the OP, so stuck with a lie to children. --TotoBaggins 04:10, 28 April 2007 (UTC)
See also Cross-platform, and Write once, compile anywhere for additional insights on some of the points mentioned above. dr.ef.tymac 00:49, 28 April 2007 (UTC)

[edit] Installing Software on XP

Is it possible to install, say a web browser, onto a single user account on a Win XP PC with multiple users, so that only that one user will be able to use it?

The reason I am asking is because certain software become defaults and open themselves for certain file extensions. Is it possible to make it so that it will only 'exist' on one account?

Thanks in advance. Skumbag - 酢薫バッグ 18:43, 27 April 2007 (UTC)

In some installers it will ask if you want to install the software the current user or all users. --Lwarf 10:34, 28 April 2007 (UTC)
You can just as well bind the file extensions differently after the offending software was installed. —The preceding unsigned comment was added by 84.187.18.197 (talk) 16:38, 28 April 2007 (UTC).