Confusion about UEFI/BIOS and GPT/MBR compatibility issues
So a user said in another post that they were NOT able to boot in UEFI mode and install Fedora even though it is UEFI compatible. But this person was able to useLegacy mode and install Fedora, and furthermore, was able to "keep the Windows partition." I'm guessing that means that Win 8 that came with it, which would be installed in UEFI/GPT mode, correct? I'm specificly referring to the Y510P but from what I understand *every* laptop that comes preinstalled with Win 8 must be UEFI/GPT.
So the way I understand it is that the installed Fedora is in "BIOS"/GPT mode rather than "BIOS"/MBR mode because you can't have both GPT and MBR on the same disc.
I have just started learning about this UEFI/BIOS and MBR/GPT nonsense, and it's going to drive me crazy until I finally understand it. So I guess what I'm asking is:
1) When I get my y510p I assume it will be in UEFI/GPT mode. Can I install non-UEFI distros alongside it as I take it was done with Fedora?
2) If I install a linux distro alongside Win 8, do I have to worry about compatibility issues with a drive that is in GPT format? Or does the MBR/GPT issue have nothing to do with it, so I don't have to worry about ever changing my drive to MBR?
For example, I read that Win 7 must be installed either as BIOS/MBR or UEFI/GPT. This can not be mixed and matched. This means that if I could not get the Windows 7 installer to boot in UEFI I would have to install as MBR. This also means I would have to format the drive and reinstall Win 8 on the MBR.
So my question is do other OS's like Linux have these restrictions? (For example, if a particular distro will not boot in UEFI and therefor MUST install on MBR)
3) I have a pendrive with YUMI installed with a ton of distros/tools/Win installs/etc. (It is a USB boot tool like unetbootin that allow you to add multiple bootable images.) When I tried it recently on my dad's laptop. I have used it many, many times with my older computers, none of which were UEFI. It works great. Now that I've had a recent encounter with my dad's ASUS Windows 8 computer (not with the y510p yet) I found out that UEFI seems to be complicating the crap out of things (for me, at least.)
So when I used this computer, I noticed that when I boot (with legacy mode enabled) and enter the "boot selection screen" in order to boot with USB, I have two options a) UEFI:"name of usb" and b) "name of USB". The UEFI option would NOT boot, but it would boot without the UEFI: option.
So does this mean that I am booting in non-UEFI mode and once I have booted this way and choose a distro to install that it CANNOT install in UEFI mode? I recently saw a tool called Rufus that I have yet to try that has an option to set the bootable USB to UEFI, so that would possibly work if I wanted to install a UEFI compatible distro (Arch linux is what I'm wanting.)
4) If installing a UEFI compatible distro (such as Arch) requires that the USB device be able to boot in UEFI mode has anyone been able to do this? Has anyone even been able to boot a device in UEFI mode to do *anything* such as run a live linux?
I'm 99% sure I would be able to boot in legacy mode and run a live linux (because I did so on my dad's computer) but the problems arise when I consider how to INSTALL.
I would really like to know the answers to these questions (as scattered as they are.) Any help would be appreciated!
Unnecessary info:
(I started learning about BIOS/UEFI and MBR/GPT the hard way a few days ago by trying for hours to install Windows 7 on my dad's Windows 8 laptop because I could NOT get Win 7 installation to work...it kept asking for drivers before I could install until I finally used the Windows USB install tool, put the stick in a different USB, AND formatted the drive as MBR because Windows 7 would NOT install on the existing GPT drive until I used diskpart.exe -clean. And I have read that Win 7 64 bit will work fine on a UEFI/GPT setup. I used the Windows 7 USB boot tool which did NOT give me a UEFI: and regular option. It showed up simply as "name of usb" without a UEFI in front. Since I read that Windows 7 must either be in BIOS/MBR mode or UEFI/GPT mode that this drive would not boot in UEFI mode, and I don't know why...Although I believe I read that Win 7 cannot be installed from a USB in UEFI/GPT mode, only BIOS/MBR. UEFI/GPT mode requires a DVD install but I did not have a drive to test this.)
I have a Y510p which is running dual boot Windows 8.1 and Arch Linux. I think that it is strongly advised to do plenty of reading ahead of any install if you will be using UEFI and Linux so that you understand all the issued before making critical changes to the existing system.
Yes, if the machine comes with Windows 8 (as mine did) then the disk will be formatted with a GPT partition table (instead of the old MBR partitioning scheme), and will boot using UEFI. If you are going to try to keep the existing Windows 8 system and add Linux then you will need to keep the disk with its GPT partition table and partition structure, but you can shrink the Windows 8 C: drive to make space for the Linux partitions that are needed ( a root partition and at least a /home and/or /opt partition and possibly a linux swap partition also ). If you want to boot the Linux install via UEFI then you can simply add the required boot directory to the EFI System Partition (ESP).
However it is very important that before trying to do any linux install that you switch off Fastboot from within WIndows 8 (or 8.1). Also most Linux distributions are in some difficulty booting using Secure Boot, though a few such a Ubuntu and Fedora are supposed to be able to do so. Hence it is much easier to work with Linux if Secure Boot is first switched off from the BIOS settings menu.
The order of operations that I used was;
1) Switch off Secure Boot from the BIOS - and boot back into the Windows 8 system to check that it boots OK.
2) With Windows 8 running go into the settings and switch off Fastboot (which does a hybrid suspend when it shuts down instead of a full normal shutdown - if you don't do this then the memory gets overwritten when booting Linux in the future which means booting back into Windows will fail).
3) Reboot back into WIndows and check all is well, and if so then use the disk management facility within Windows 8 to shrink the C: drive to make room for the Linux partitions.
4) Reboot to check Windows 8 still boots OK.
5) If you are going to update to Windows 8.1 then do so, and then update everything once it is booted (it is a huge update and takes ages!). Once done then you will likely have to update drivers for the graphics cards, the clickpad and possibly the wireless chip and ethernet chip. I found that I needed to get drivers that were newer than were available on the Lenovo website, by going to the relevant hardware manufacturer website (eg for synaptics for the clickpad). Then spent a week or so in the evenings getting Windows 8.1 configured the way I like it.
6) Then I did a lot of reading about the various options for the boot manager that would suit a UEFI boot for a dual boot system for Windows 8.1 and Arch Linux and there was a choice of Grub, Gummiboot, rEFInd, and others - and after reading the details I decided on rEFInd as my boot manager which can boot not only any new Arch Linux install but automatically finds the Windows UEFI boot files and presents the options in a nice graphical window once the system gets past POST at bootup.
7) It was important to check which partition was the ESP and to know what partitions I needed to create for the Arch Linux system. Then I went ahead and booted from a usbkey to a uefi install system, and very carefully proceeded with a standard Arch Linux install, being particularly careful to know where to put the rEFInd boot manager files and the kernel and initrd files. Also I used efibootmgr to write the appropriate NVRAM boot entry in the motherboard memory so that the uefi boot system knows where to find the rEFInd uefi boot files in the ESP.
8) Once complete the system boots to Arch Linux as the default, with a nice Windows icon which you can select with the arrow keys within the boot timeout period (default 20 seconds).
I noted also that it is possible to create boot stanzas in the rEFInd boot manager config files which allow rEFInd to chain load other Linux systems or even other bootloaders if you wish - so it is very flexible. So if you want to you could install a grub standalone set of diretories/files so that if the normal linux boot fails then you can select the grub icon from rEFInd and chainload grub to boot either the same Archlinux install, or point to a third linux distribution if you have more partitions containing that third install which might be Ubuntu or Mint or ....
Either way although getting to understand how uefi boot works is a learning curve it is actually generally simpler than the old legacy BIOS boot. With uefi you no longer need an MBR on the drive, and only a suitable EFI System Partition which has to be VFAT formatted. However if you want to have one of the linux distributions booting from legacy MBR then you need to create an MBR at the start of the drive - so you would need to move the start of the first partition and create a suitable sized Master Boot Record otherwise MBR boot can't work. If you do that then of course you have to be careful if the Windows partition is the one being re-sized that it doesn't mess up the Windows boot! However since using uefi to boot rEFInd allows a chainload to grub/gummiboot or other bootloaders then there should be no need to mess with MBR booting if you go down that route.
If you are interested in rEFInd then the author Rod Smith has a good set of documentation that describe the details at http://www.rodsbooks.com/refind/
He is also the author of a really excellent disk partitioner for GPT disks - http://www.rodsbooks.com/gdisk/
Clearly it is necessary to read up on the boot facilities available for any linux distribution that you plan to put on the system.
One nice thing is that uefi boot with an efistub supported kernel build is really fast on the Y510p. My system boots Arch linux in about 7 seconds to the KDE login prompt once the POST is complete and that only takes a couple of seconds. Of course Windows is much slower once it is selected at the rEFInd screen and takes somewherearound 40 seconds or so to boot, but at least Linux is super fast!
Anyway I hope that this helps.
Similar Messages
-
Just bought a Nikon d750 and confused about adobe LR4 and PS6 support for the RAW files. I have DNG 8.7 but wondering if LR and PS will import direct soon Thanks for any advice
Support for the Nikon D750 was introduced in the latest version of LR 5.7 and ACR 8.7 on Novemder 18th 2014.
Further updates to LR 4 were stopped when LR 5 was released on June 9th 2013. No further updates for bug fixes and new camera support.
Nada, LR 4 will never support Nikon D750. The Nikon D750 was introduced into the market in September 2014 some 15 months after further development of LR 4 was discontinued.
You can use the Adobe DNG program (free download for the package) to convert the Nef (raw) files from your Nikon D750 to the Adobe DNG format which will permit you to import those into LR 4. This is the crutch provided by Adobe to allow for the processing of raw files with outdated versions of LR and ACR.
You can also update the ACR plugin for PS CS6 to version 8.7 which can also work with the raw files from the D750. For direct support in Lightroom you will need to upgrade (paid) to version 5.7. -
I'm horribly confused about student licensing and commercial use
As the title says I'm horribly confused about student licensing and using it for commercial use.
I currently have a Student Licensing version of Adobe Creative Suite 4 that I purchased through my school's journeyEd portal.
Seeing how CS5 is now out I was browsing looking at prices (why not upgrade while I'm still a student, right?) and while browsing I bumped into one source that says that Student Licensing can not be used for commercial purposes, and this is when the confusion started. I remember reading before that we are able to use student licensing for commercial purposes, okay time to google search. I found one Adobe FAQ that says I can. .
http://www.adobe.com/education/students/studentteacheredition/faq.html
" Can I use my Adobe Student and Teacher Edition software for commercial use?
Yes. You may purchase a Student and Teacher Edition for personal as well as commercial use. "
and I found this old thread;
http://forums.adobe.com/thread/314304
Where an poster listed as an employee of Adobe states
"There is no upgrade from the CS3 Educational Edition to the comparable CS3 editions sold in non-academic environments. If you have an educational version of for CS3 obtained legitimately (i.e., you qualified for the educational version when you obtained it), you may continue to use that software for the indefinite future, even for commercial use! You cannot sell or otherwise transfer that license, though! When the next version of the Creative Suite is released, you will have two choices: (1) If you still qualify for the educational version, you can buy a copy of that next version (there is no special upgrade pricing from one educational version to another; the price is already very low) or (2) you can upgrade from the educational version of CS3 to the full version of the next version of the Creative Suite as an upgrade from CS3 at the prices published at that time. "
Okay cool, hmm what this? Adobe is asking me if I want to IM with live costumer service agent, sure why not? Then the conversation started and I asked her my question about using my CS4 license for commercial use, she asks for my product code and email to verify my product, then informs me I can purchase the upgrade version of CS5 and use that for commercial, okay great, but not really answering my question. I reword it and give her a link to that FAQ page it goes like this. ..
"[CS Rep] : [My name], I would like to inform you that Adobe Student and Teacher Editions are not allowed for
commercial use.
[CS Rep] : However, you can upgrade your current software to a normal upgrade version, and you can continue
using it for commercial purpose.
[Me] : Then is the FAQ page mistaken? Because it is very misleading if it is. But thank you for the information.
[CS Rep] : You are welcome.
[CS Rep] : I apologize for the misleading information in the FAQ."
. .And after that, I went back to being confused.
SO my questions are. . . Can I or can't I use my Adobe Creative Suite 4 student licensing for commercial purposes? and If I purchase a Student Licensing of CS5 can I use that for commercial purposes as well?
Sorry for the long post, I just want to be perfectly clear on what I can and can not do with my purchase.The rules differ in various parts of the world. In North America you can use it for commercial work.
There are no student/academic upgrades. The pricing is so low that in many cases you're better off buying another full student license but you are eligible for upgrade pricing for commercial versions once you're out of school.
You may not transfer the student license in any way.
Bob -
ILife 08 and iMovie 06 compatibility issues?
Sorry if any of this has already been covered in other threads...
Has anyone had a chance to thoroughly test the 6.04 version of iMovie (download version if you install iLife 08) and know what compatibility issues there are?
I.E. How well does iPhoto 08 play with iMovie 06, or iMovie 06 with iDVD 08. Are the third party plugins problematic with v6.04?
From what I saw at the Apple store, iPhoto08 and iDVD08 would be the only apps that I would really want since iMovie08 seems to be a step backwards and iWeb is still lacking in everything that I wanted improved. But I don't want to waste the money to discover all sorts of "gotchas" with trying to get iMovie06 to play nice with the rest of iLife08...
Also, has anyone done an install excluding iMovie08 entirely? Do you still need to download 6.04 if installed this way? Are there any pitfalls?
This is like Deja vu or something...When iLife06 first came out, I installed everything but iMovie and stuck with iMovie05 for a long time cause of all the bugs and how long it took to get the third party plugins working...Of all the things for people to moan about with iMovie '08, the real unfortunate thing is it's inability to import effects, etc. (essentially your completed movie) from iMovie HD. One could somewhat forgive Apple for this debacle, if only it played nice with iMovie HD. Then it could be everything it's supporters say it is, i.e. "a different app," "a simple app for simpletons in a hurry" (although I find it's lack of tools and compatibility issues more difficult to manage).
...and a lot of us feel like this is the thing we're going to be stuck with going forward.
I keep thinking that something else is going to happen, because this can't be right. I think this diver/video engineer who came up with this concept is suffering from the bends.
Apple doesn't need the money, (as in making us buy Final Cut Pro for decent video editing), trust me, so I don't understand why it would take the PR hit.
iMovie '08 spoofs are already on YouTube, etc.
Message was edited by: rmig68 -
Final Cut Pro 4.5 and Quicktime 7 compatibility issues. Error 39!
I see some people saying Quicktime 7 is not compatible with Final Cut 4.5 and some saying it is.
I upgraded to 10.3.9 and then Quicktime 7 and I started getting error 39 at this one spot in my timeline and then crashing if I tried to play just a few seconds ahead of that spot. I tried trashing the prefs but that didn't do any good. The individual files play fine outside the timeline so it isn't corrupt media. I tried rerendering the effects that are at that point but that didn't help. I tried copying the entire timeline to a new sequence and that didn't help.
The strange thing was that when I took the part of the movie that was having trouble and deleted it from the timeline it still got error 39 when just playing over the empty timeline where the media used to be! This led me to realize that the problem lay in some Quicktime audio effects I had about a second in front ot the error 39 spot. I took off the effects from that audio but it didn't help. Then I deleted that audio from the timeline and just pasted it back in from the original clip and then it finally got to working.
Now I know nothing about the software, but I thought there might be some connection between the fact that I had effects that FCP labels as "quicktime effects" on there from my days in Quicktime 6.5, then upgraded to Q7, and perhaps those old effects don't work in Q7? I don't know.
So, my immediate problem is solved but I thought this experience may be of use to someone else. And perhaps some compatibility issues can be brought to light.I would assume the Software Update program has sufficient logic within it to detect that my friend is running Final Cut Pro 4.0 and it would therefore be disastrous to install QuickTime 7.3, thus removing QuickTime 7.3 from her list of suggested updates. Would someone at Apple please fix this?
Sorry, but Quicktime is a component for SO MANY things, not just FCP. Sorry, but the OS doesn't detect that older versions of software are on there and won't be compatible...this is where you need to have a little knowledge. And sorry, but you need to have a little knowledge. If things are working, DO NOT UPGRADE unless it has been tested and cleared. In the case of FCP 4...it has not.
Rule of thumb...if you use older software, you have to stick to older versions of the OS and components that that software relies on.
How many things were broken with Vista? So many that the HUGE community let out a cry and now XP is available again configured on machines or as a stand alone buy. And no, Vista didn't warn that certain applications wouldn't work either.
You gotta research and find out. This place here...this forum...is a FANTASTIC place to find stuff like this out.
BTW...Apple doesn't read these forums, only us users. Gotta submit feedback here:
http://www.apple.com/feedback/finalcutpro.html
Shane -
Experiment and conclusion on notorious Bios and turbo boost related issue.
Hi everyone,
Like everyone here, I suffered from the turbo boost issue and I decided to find a practical way to solve this problem,so .I conducted some experiments trying to find out the relationship between power strategy, graphic type, booting power source and the CPU frequency.. Here are my results:
Variable: Power strategy for AC/BAT: Max/Balanced, Max/Max(didn't try for balanced AC power strategy)
Graphic type: Integrated, discrete and Nvidia Optimus
power source when the machine booted(BAT,AC)
Testing model: W520 42762QU i7-2720QM Quodro2000M 16G Crucial M4 128G, 500G HDD, BIOS1.30
Testing programs: Intel Turbo Boost programs and CPU-Z
power strategy
Graphics
booting:AC
AC-BAT
AC-BAT-AC
Booting: BAT
BAT-AC
BAT-AC-BAT
Max/Balance
Discrete
Turbo
1.5Ghz
Turbo
1.5Ghz
800Mhz
1.5Ghz
Max/Max
Turbo
1.5Ghz
Turbo
1.5Ghz
800Mhz
1.5Ghz
Max/Balance
Integrated
Turbo
Turbo
Turbo
Turbo
800Mhz
Turbo
Max/Max
Turbo
Turbo
Turbo
Turbo
800Mhz
Turbo
Max/Balance
Optimus
Turbo
800-2400
Turbo
800-2400
800Mhz
800-2400
Max/Max
Turbo
800-2400
Turbo
800-2400
800Mhz
800-2400
Notice the '800-2400' state, it is a special situation when CPU-Z occationally gives me a speed of approx 2400 when I tried some head-load programs, which lasted about few seconds and the frequencies went back to 800 or 1500(in most cases, 800) when I closed the program.
Conclution:
1. power stratagy will not influnce the cpu frequency.
2. booting power source and graphic types will do.
3. I'm sick and tired of pluging and unpluggin my W520.
4. I'm not a fan of Nvidia Optimus.
5. Lenovo did a terrible job on this issue.
6. I expect a same result for Bios 1.26
Practical solutions:
1.When you need a powerful W520, use Discrete graphic and plug your AC while booting.
2, For a longer battery life, use integrated graphic, and if you wan't to plug the AC, please turn off your machine and boot again.(not rebooting!)
Thank you for reading my post and sorry for my bad english
Robin LiuSomeOtherGuy:
Thanks very much for keeping us all informed about the "work in progress" on the BIOS and the Turbo-thing. That is thoughtful of you and will (hopefully) cut down on the amount of carping and complaining about this issue.
Michael
W520 (4270 CTO), which replaced a W500 (4062-27U), which replaced a T42P, which replaced an A21P... -
E2100L Cisco Connect and Network Magic compatibility issues with Mac OS X 10.7 Lion
When you will release the latest version for Wireless router Linksys E2100L?, because I bought a MAC with Lion and I can´t use my Router.
Regards,
DanielOlder versions of Cisco Connect for Mac computers have compatibility issues with the latest release of Mac OS X (10.7 “Lion”). Only Cisco Connect versions 1.4.11222 and newer are compatible with the Lion operating system.
Here is the link to ensure that Cisco Connect can be installed on your Mac: http://www6.nohold.net/Cisco2/ukp.aspx?vw=1&docid=a4e763be954243cfa0ab960c0fac7105_KB_23xxx_EN_v1.xm...
There is no such information about the release of Cisco Connect for E2100L for the Mac OS X (10.7 “Lion”) compatibility, so you need to wait for further release of the firmrware for E2100L. -
Premiere CS6 and CC: any compatibility issues working back and forth between both versions?
Hello,
What are the compatibility issues with Premiere CS6 and CC versions? Can the user go back and forth? I run a computer lab at a college and we will be upgrading to CC but many students will still be using the older CS6 version of Premiere until we update all the computers.... Will students run into problems?
Thanks,
dmichaelMy concern is going back and forth between the versions... I read somewhere there are compatibility issues with Premiere....Can the user working in CC go back to premiere CS6? and vice versa?
Thanks,
dmichael -
Confusion about BPM suite and BPA suite
Hi,
I'm very very new to Oracle BPM.
I just checked the website and found Oracle provides two suites - BPM and BPA for business process management. I'm quite confused what's the difference of their usage? It looks to me that they are very similar to each other in functional.
Is there any article that has already explained their difference?
btw: if i want to choose a suite just for modeling, implementation and monitoring purpose, does that mean a BPM suite is more suitable?
Thank you
LesleyHi Lesley,
BPA is an robust process modeling tool used by business analysts to model both their processes and the entire enterprise. BPA has many different types of diagrams that let you see the different levels of decomposition and abstraction. As you'd exect, BPA has a robust process simulation capability. If you hit the BPA forum (Business Process Analysis Suite you'll get more information about it from experts in BPA.
Oracle BPM was built and architected from the beginning as full life cycle Business Process Management (BPM) tool. It is similar to BPA in that it supports process modeling, documentation and simulation. I use the same diagram to explain the process to executives, managers, IT, SME and business analysts. As a business analyst, the tool is not complicated and can be learned in just a couple hours.
Here's where the two differ. Oracle BPM was architected from the beginning as a full featured Business Process Management tool that implements and monitors business processes. Its capabilities include:
<li> Logic - BPA is not a BPM tool so it is not intended to support the process logic needed for runtime. To support this, Oracle BPEL is used in conjunction with BPA. As a developer using Oracle BPM, I use templates and drag and drop to create much of my logic. Once I catalog an object it can be reused across multiple projects either by using the Project Dependency option or by importing the artifacts. I test my logic either by using the method editor debugger or at the process level. As I create objects used in my logic using Oracle BPM, I inherit attributes provided by introspected components. This means that if you have an ERP system object with 138 attributes, you do not have to rebuild this object from scratch in Oracle BPM.
<li> Runtime / Execute the Business Processes - End users interact with the processes at runtime using an OOTB Workspace. End users are given various roles. When they log into the Workspace, the end users only see the work item instances that are in the roles that they have been assigned. There is an Engine that stores the work item instance information as it flows through the processes at runtime. The Engine is like a traffic cop, ensuring that the right work item instance goes to the right person in the right activity at the right time.
<li> UI - The End user interface screens and complex end user interaction with a variety of screens can be built inside the Oracle BPM toolset's editor. The forms you build are automatically presented to end users in the Workspace.
<li> Integration to Existing IT Assets - Oracle BPM can expose and consume IT components directly or through a service bus. Once consumed, the components can be used by any process needing to invoke them at runtime.
Hope this helps,
Dan -
Confused about nonGUI beans and events
The crux of this is the question "When will a NON GUI bean use events to communicate?"
Articles I have read advocate that Beans are not just GUI components, and I agree. However I've only ever seen examples of GUI Beans!
I am writing a non GUI Bean to represent a credit card.
Before I got concerned with beans I would have just written the class with some accessor methods and some functional methods to do what I want, something like;
void setCardNumber(String n)
boolean isNumberValid() //check the card number using MOD10
and then used that class in my application.
But now I want to make it conform closer to JavaBeans. I know that beans have methods, but I am confused somewhat about how events figure into this. Do nonGUI beans fire events/listen for events? Very often?
What I am asking is, what events should a Bean like CreditCard fire? The main function of this Bean is to determine if the number conforms to MOD10 checks using isNumberValid() method.
Consider that I want to write another Bean, a GUI component that takes card details and updates the nonGUI CreditCard Bean. Of course I want loose coupling, not every user will use my CreditCard Bean (especially if I dont get these events straight!). If I use property change events to fill in the card number etc in the CreditCard Bean, should I make it (the CreditCard bean) listen for focusLost events from the interface to determine when all the details have been filled in and fire an event about the validity of the card number?? Doesn't this break the rules a bit, by providing prior knowledge to the CreditCard nonGUI Bean??
Should the CreditCard nonGUI bean fire an event to say that the number is valid/invalid?
You will probably say, "WHY use events, when you only need methods?" and that is my question! When would a non GUI bean fire or listen to events? I want my non GUI bean to be completely ignorant of GUI stuff, so no focusLost listener.
Should the credit card interface fire a custom event (CardDetailsEnteredEvent) and the non GUI bean listen for these? But that's tighter coupling than just implementing the methods isn't it?
I think that the nonGUI bean will have property change listeners for the card number etc, but this begs the question, what is the point of setter methods if I am bound the properties anyway??
I've read everything pertinent at http://java.sun.com/products/javabeans/training.html
Thanks to anyone who can stop my head hurting.
JimAh, I think I may have figured it out from the spec.
"Conceptually, events are a mechanism for propagating state change notifications between a
source object and one or more target listener objects."
key phrase being "state change" - my non GUI bean CreditCard only really has two states; 1. incomplete card details and 2. complete card details. So I guess I could/should only fire an event (DetailsEntered) when the details are completely 'filled in'.
Which leaves me with the GU interface bean, it has two states also, 1. details being edited (focus gained) and 2. details finished being edited (focus lost) - should then this bean fire its own event (DetailsEdited) or should it just rely on other beans listening for the focus lost event? It's my functional interpretation that a focus lost event signifies that the details have been edited, perhaps I should therefore implement my own event that has this contextual meaning (DetailsEdited).
I'm aware that the names of these events are quite poor, I always find good naming to be another very important head scratcher!!
Jim -
Confusion about backing beans and multiple pages
Hello:
I'm new to JSF and am confused by several high level and/or design issues surrounding backing beans. The
application I am working on (not unlike many apps out there, I suspect), is comprised of several forms which
collect various query parameters and other pages that display the query results via dataTables. Additionally, the
displayed results have hyper-text enabled content sprinkled throughout in order to allow further "drill-down" to
greater details about a given result.
Obviously, the initial forms are backed by a managed bean that has the getters/setters for each of the various
fields. From a modularization perspective, I'd prefer to have the query execution in a separate class entirely,
which I can accomplish via delegation or actionListeners. Where I get confused is in the case that one of the
hyper-text enabled fields needs to invoke yet another query requiring specific values from the original result.
In this case, how should I design the beans to store the pertinent selection information AND still be able to
have the resulting 'details' query in a separate class.
For example, let's assume that I have a form with a simple backing bean to collect initial query params. The user
fills out the form and clicks on submit. The action field of the <h:commandButton> tag inside the <h:form>
delegates to a method which performs the query and returns a navigation string to the results display page. The
results page then uses a <h:dataTable> tag to display the results. However, through the process of displaying the
results, it will create the hyper-text links as follows:
<h:form>
<h:commandLink action="viewDetails">
<h:inputHidden id="P_ID" value="#{results.id}"/>
<h:outputText value="Click here to see details"/>
</h:commandLink>
</h:form>Notice that the value="#{results.id}" is the key field (in some case I may need more than one) that the subsequent
details page will need to perform the query. However, this value is stored in the results attribute of the original
query backing bean. I do NOT want to add yet another level of queries into the same backing bean as I would like
to keep the original and details query beans separated for maintenance and style reasons. How then, can I get this
specific value ("#{results.id}") into another backing bean that could be accessed by the subsequent details page
and query? In general, how does one map a single display element to multiple backing beans (or accomplish the
effect of having one 'source' bean's value end up in another 'destination' bean's attr? If this is not possible,
then isn't the details query/page forced to use the same backing bean in order to get the data it needs? If so,
how can it determine which id was selected in this case?
I'd appreciate any thoughts, ideas and suggestions here! Is there a better way to design the above in order to
maintain the separation of logic/control? What is the 'best practice' approach for this sort of stuff?
Thanks much in advance!This is what I have done in the pass
<h:commandLink action="#{AppAction.getRecord}" >
<h:outputText value="#{bbr.id}"/>
<f:param name = "recordid" value ="#{bbr.id}" />
</h:commandLink>
within the getRecord method
String key = "";
FacesContext facesContext = FacesContext.getCurrentInstance();
ServletContext servletContext = ServletContext)facesContext.getExternalContext().getContext();
key = (String)facesContext.getExternalContext().getRequestParameterMap().get("recordid");
then call you dbmethod passing in the key of the record the user wants to edit
and then simply return a string to what page you want the user to go to
I think this is what your looking for -
Confused about wiki pages and groups (i'm new to wiki services).
we set up wiki services on our newly upgraded 10.5 server, but there is something that i can't seem to understand.
it seems that the only way to create a new wiki is to create a new group. i don't quite get how you can have some members of one group and some members of another, access the same wiki page.
this will get very messy for our company because we want to set up wikis for the multiple projects we have going. this means that the amount of groups will escalate rapidly and surely will cause problems and confusion.
we currently have 7 groups created on the WGM and my company is about 50 employees.
each group currently has their own wiki which leads me to my original problem/question.
if we create a new wiki, that creates a new group, and we don't want that.
i could totally be missing something here as i mentioned i am new to the wiki services and 10.5 server in general.
if you need more information from me, let me know.
thanks in advance.anyone?? bueller...? bueller...?
i can't seem to find any information about this. i even looked here http://www.apple.com/server/macosx/resources/ in the wiki documentation and didn't find anything.
perhaps i wasn't entirely clear in my initial post.
say i want people in the accounting and marketing group to share a wiki page. the only way i know how to do this, is to create a new group (let's call it wiki_group) and to have members of both accounting and marketing be members of wiki_group.
as i mentioned in my initial post, this is going to be a huge problem.
the only work-around that i can think of is to have the accounting group and marketing group be members of both. this is not acceptable due to security and resource sharing.
how can i create wikis without having to create a new group, and to have members of other groups view and edit other wikis? -
Confused about polkit-gnome and lxpolkit [solved]
I recently updated and new version of goffice and gnumeric came down with this warning:
==> The agent is no longer autostarted by default except in GNOME Flashback.
For Xfce, LXDE etc., "lxpolkit" is the suggested lightweight alternative.
See https://wiki.archlinux.org/index.php/Polkit#Authentication_agents for
more details.
I use xfce4 and do not know gnome-polkit installed at all. I removed it and pacman did not complain about dependencies. What exactly do I need lxpolkit for? I am able to use sudo without it.
Last edited by maggie (2013-09-16 15:40:49)I'am confused too:
Using lxde I was not even aware that there is an alternative to polkit-gnome, as it is always
mentioned as quasi dependend on gvfs:
The gvfs package needs to be installed, along with polkit-gnome for the polkit rules.
This seems to be wrong as neither gvfs nor the polkit rules do depend on polkit-gnome authentication-agent in any way.
Informations through the wiki(s) are very inconsistent!
From LXDE-Page:
polkit-gnome provides an authentication and will need to be started on login:
$ mkdir -p ~/.config/autostart
$ cp /etc/xdg/autostart/polkit-gnome-authentication-agent-1.desktop ~/.config/autostart
What is it good for? Is there any reason to start it twice? The file mentioned is installed at '/usr/share/applications/polkit-gnome-authentication-agent-1.desktop' by default.
Polkit-Page:
To autostart it everywhere except in GNOME and KDE, copy the file /usr/share/applications/polkit-gnome-authentication-agent-1.desktop to /etc/xdg/autostart/
LXDE-Wiki:
gvfs and its dependencies (optional, but highly recommended):
policykit-gnome (required for authentication for volume management)
And last but not least the awareness that the LX alternative exists since March 28th, 2010 where PCMan himself told us polkit-gnome is not needed anymore:
LXPOLKIT – SIMPLE POLICYKIT AUTHENTICATION AGENT:
Generally when one needs to use policykit, he or she needs to install policykit-gnome. Now we have our own.
A new component LXPolkit was added. It’s minimal policykit authentication agent.
As far as I get it, polkit-gnome only provides the 'query of password window' and can be replaced by lxpolkit which is more lightweight.
And as it was never 'auto'started (see above) except in GNOME, the pacman.log message is IMHO highly questionable and confusing. -
Confused about C99 _Bool and stdbool.h
C99 defines the _Bool type, which seems like a good language feature, but it does not seem to be designed very well.
The main point of confusion I have is that boolean expressions apparently do not give a _Bool type. These expressions both give a size of 4: sizeof(1==1), sizeof(true). What is the point of having a boolean type if the languages doesn't even use it for boolean values? C++ is consistent is boolean sizes, so maybe the C99 boolean types are meant more for C++ function compatibility.
Also, why do the true and false values in stdbool.h not include casts to actually be defined as Bool types? Even if boolean expressions are not the Bool type, shouldn't sizeof(true) match sizeof(_Bool)?
My guess is that Bool is not implemented very well because it would break a lot of C code that assumes 'int' for boolean values. Maybe the next standard should also define types like bool8t and bool32_t, so that the established int-booleans can work along with a 1-byte _Bool type.I can answer some of your questions. Sizeof(1==1) is 4 because, per the C standard, 6.5.9(2), the result of an equality operator is type int. Sizeof(true) is 4 because "true" is defined (in stdbool.h) to be an integer constant (the value 1), which of course has a size of 4. One could make a case that these may not be expected, or logical, but they are following the current standard.
C++ is "consistent" (versus C) with these cases because C++ defines the result type of relational operators as boolean. Thus, sizeof(1==1) and sizeof(bool) are equal.
As for the inequality of sizeof(true) and sizeof(_Bool), it's not any different than something like sizeof('a') versus sizeof(char). Sizeof applied to a character constant returns 4 and sizeof(char) returns 1. -
Confused about SATA RAID and SATA DVD
I'm planning to build a system based on the P965 Platinum board and I'm considering implementing a RAID 0 array. However, I have read of many problems with RAID and SATA DVD writers. Are most problems just due to BIOS problems or are there more inherent problems? Are people successfully running this board with a RAID array and SATA DVD? Can you boot from the SATA DVD drive?
If I have a pair of drives in a RAID array, can I add further hard drives that will not be in a RAID array? I won't trust RAID for all my data and I don't want to use mirroring.
I'm confused and I can't seem to find any definitive answers on this.
Thanks
TeresaQuote
Thanks Maesus. Will the drive from an AHCI machine only work in another machine set up with AHCI?
Teresa, yes but not always, this AHCI is driver bound. And because this AHCI is driver bound, it makes it hard to change from IDE to AHCI for a IDE driver installed OS, as the OS doesn't have AHCI driver installed. So when you change to AHCI setting in the BIOS, the OS refuse to load further. AHCI is driver bound, and it is also controller/chipset bound. If you installed an OS with AHCI settings enabled in an AMD system, you may not run that in an AHCI Intel system. Vice versa.
Quote from: Teresa on 20-May-07, 19:51:36
I ask this because there's a thread here that gives instructions on how to turn on AHCI after Win XP has been installed. That tends to suggest that AHCI makes a difference to an operating system install.
This is only because the OS installer doesn't have AHCI driver built-in. If you slip-streamed your OS installer with AHCI drivers, your OS installation will go smoothly.
Quote
Or is it a case that the disk's data can still be read on another machine without AHCI but it can't be used to run Win XP? I just need to know that I can pull my data hard disk out of my system and access it on another older machine that doesn't have ACHI.
Yes, the disk's data still can be read anywhere else. Just that it cannot be used to boot up to WinXP, if that WinXP doesn't have AHCI driver installed.
Quote
Hopefully one day soon I'll reach a critical threshold of knowledge and actually start ordering the parts for my build.
Teresa
Let's make it simple. Treat IDE, AHCI and RAID-0 as different HDD drivers. WinXP has IDE driver built-in. Vista should have all drivers built-in. IDE and AHCI data structure is the same. However, RAID-0 data structure stored in the HDD is different and is controller-bound.
For AHCI mode installed HDD, you can read in IDE mode without single issue.
For IDE mode installed HDD, you can also read in AHCI mode without issue, provided AHCI drivers are installed upfront.
For RAID-0 installed HDD, you cannot read in IDE or AHCI mode, and you may not be able to read in other different controller/chipset systems even though you set them as RAID-0 exactly.
Maybe you are looking for
-
imported a slide on my old mac book pro to my iTune i had on it, but it got a new one how can i get my old iTune back with that slide i imported
-
Using GP Runtime Manager in CAF
Hi, I would like to create a CAF operation to access our Guided Procedures processes via the GP Runtime Manager. In my CAF project I have added the dependencies to both caf/eu/gp/api and caf/eu/gp/api/wd and checked all the 3 'Design Time', 'Deploy T
-
BareCode reader and insert String into actual selected JTextField
Hi everyone, I can't invent anything appropriate about my concept. I would like to write a program for BareCode reading. I have working code witch gets a text string from reader which is connected over RS232. But I have to send this String to actual
-
IOS 7, iPad 4: keyboard problems with Shift key
Since I have upgraded my iPad 4 to iOS 7 a few days ago the keyboard drives me mad: I did not experience the overall slowness of the keyboard as reported by so many others. Anyway, I reset my settings just to be safe. But the reset had absolutely no
-
How can I open a document with applescript without the doc. window opening.
How can I open a document with applescript without opening the document window and speak the text of the document? I can get it to open the file but the doc window always opens up.