EOQualifier bad pattern for qualifier
Hi there, If I have: searchString = '*' + input + '*'; new EOKeyValueQualifier( anAttribute, EOQualifier.QualifierOperatorCaseInsensitiveLike, searchString ); And someone puts in some strange characters (e.g., "find[me"), I get the following exception. java.lang.IllegalArgumentException: Bad pattern *photogra[hy*for qualifier at com .webobjects .foundation ._NSStringUtilities ._bracketedStringFromPatternAnalyzer(_NSStringUtilities.java:740) at com .webobjects .foundation ._NSStringUtilities.stringMatchesPattern(_NSStringUtilities.java:785) at com .webobjects .eocontrol .EOKeyValueQualifier.evaluateWithObject(EOKeyValueQualifier.java:141) at com .webobjects .eocontrol.EOOrQualifier.evaluateWithObject(EOOrQualifier.java:103) at com .webobjects .eocontrol.EOAndQualifier.evaluateWithObject(EOAndQualifier.java:96) at com .webobjects .eocontrol.EOQualifier.filteredArrayWithQualifier(EOQualifier.java:635) <...> I thought EOF would escape things automatically, but perhaps there's a bug there(?). How else might I escape the chars (WO5.3.3)? Thanks. with regards, -- Lachlan Deck ___ Do not post admin requests to the list. They will be ignored. Webobjects-dev mailing list (Webobjects-dev@lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/webobjects-dev/archive%40mail-archive.com This email sent to [EMAIL PROTECTED]
Re: Accessing Session
On 02.04.2008, at 21:25, Eleni Manou wrote: I am just copying some lines from my project in WO 5.3. There in Session.java I define public String userName; and then in a component I set this variable using ((Session)session()).userName = . This is not working in WOLips project, wher I get the error message Unresolved compilation problem: username cannot be resolved or is not a field Add public accessor methods to your session class for the username and call ((Session)session()).setUserName("aUserName"); I can not understand what you mean by "Do you have some things in packages and others not? Are you importing the correct Session class?". Sorry, I am a newcomer in Eclipse/WOLips. Can you help me a little bit more? Background is, that sometimes people *think* they are asking the right Session class, but they import a Session class from a completely different package (JavaMail for example). cug -- http://www.event-s.net ___ Do not post admin requests to the list. They will be ignored. Webobjects-dev mailing list (Webobjects-dev@lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/webobjects-dev/archive%40mail-archive.com This email sent to [EMAIL PROTECTED]
Re: WO 5.4 Linux Deployment
wonder sources. Works great for Apache 2 and 2.2 On Tue, Apr 1, 2008 at 11:58 PM, Aleksey Novicov <[EMAIL PROTECTED]> wrote: > Which source do you recommend using? The source that is distributed with > Xcode and is in /Developer/Examples/WebObjects/Source/Adaptors? Or the > source in Project Wonder? > > > > On Mar 30, 2008, at 10:22 PM, Joe Little wrote: > > > > On Tue, Mar 25, 2008 at 6:41 PM, Aleksey Novicov <[EMAIL PROTECTED]> wrote: > > > > > I am looking for input and instruction on deploying WO 5.4 apps on Red > > > Hat Linux. Are there any updated instructions for WO 5.4? Is there a > > > pre-compiled Apache 2.2 WebObjects adapter for Red Hat available? How > > > about deploying on Tomcat? Is it preferred over JavaMonitor/wotaskd? I > > > have experience with JavaMonitor on OS X but have never tried > > > deploying on Tomcat. > > > > > > > > > > I'll just add that I never deploy as a servlet, but use native > > adaptors. Nothing precompiled, but the instructions out there for 5.3 > > still work for 5.4, albeit using the newer 5.4 frameworks/jars. The > > FreeBSD install script that has been posted might be more enlightening > > as I think its already updated to 5.4. > > > > > > > By the way, this mail list is invaluable. Thanks to all of the > > > contributors. > > > > > > Aleksey > > > ___ > > > Do not post admin requests to the list. They will be ignored. > > > Webobjects-dev mailing list (Webobjects-dev@lists.apple.com) > > > Help/Unsubscribe/Update your Subscription: > > > > http://lists.apple.com/mailman/options/webobjects-dev/jmlittle%40gmail.com > > > > > > This email sent to [EMAIL PROTECTED] > > > > > > > > > > ___ Do not post admin requests to the list. They will be ignored. Webobjects-dev mailing list (Webobjects-dev@lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/webobjects-dev/archive%40mail-archive.com This email sent to [EMAIL PROTECTED]
Re: Accessing Session
Please keep questions on the list. On 2-Apr-08, at 11:19 PM, Eleni Manou wrote: Hi David I am just copying some lines from my project in WO 5.3. There in Session.java I define public String userName; and then in a component I set this variable using ((Session)session()).userName = . This should work, though personally I'd wrap userName with accessor methods. This is not working in WOLips project, wher I get the error message Unresolved compilation problem: username cannot be resolved or is not a field Why is username not here not camelCased like above? Can you help more? Many thanks Stavros On 02 Απρ 2008, at 6:23 ΠΜ, David LeBer wrote: On 1-Apr-08, at 11:14 PM, Eleni Manou wrote: Hi all, Can anyone help e on how to access session variables? Before (WO 5.3) I had simply to declare a variable in to the session.java class and then I cold read and and set this variable using (Session)session(). This seems that does not work in my Eclipse/ WOLips project. Am I doing something wrong (probably)? Many thanks in advance Are your variables private or protected? They need to be public or (better yet) have public accessors. ;david -- David LeBer Codeferous Software 'co-def-er-ous' adj. Literally 'code-bearing' site: http://codeferous.com blog: http://davidleber.net profile: http://www.linkedin.com/in/davidleber -- Toronto Area Cocoa / WebObjects developers group: http://tacow.org ___ Do not post admin requests to the list. They will be ignored. Webobjects-dev mailing list (Webobjects-dev@lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/webobjects-dev/archive%40mail-archive.com This email sent to [EMAIL PROTECTED]
Re: [OT] Limiting File Upload Sizes
i've successfully implemented the SWFUpload with WO. unfortunately i didn't know how to play nice with all the javascript SWFUpload and WO stuff so i placed the PHP upload feature (which is quite nice with smooth ajax interface) in an inline frame and send the results to a DA for processing / inserting file information into my WOApp. i pass variables into the PHP code to limit upload size, the # of files to upload, path to write, etc... i'll try to "clean" this up and remove it from my project for sharing. ./james { primaxSTUDIO : http://primaxSTUDIO.com : 513.772.1223 } On Apr 1, 2008, at 4:49 AM, Owen McKerrow wrote: Hi Guys, Thanks for the suggestions. I was after something specifically that stopped the user even trying to upload something so they don't waste time sitting around waiting for something to upload only have it to fail half way through. I will check out Joshua's implementation. Owen McKerrow WebMaster, emlab Ph : +61 02 4221 5517 http://emlab.uow.edu.au - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - "I like the way this project has somehow, against all common sense, got itself made." - Peter Jackson, "The Lord of The Rings" On 01/04/2008, at 3:01 PM, Chuck Hill wrote: On Mar 31, 2008, at 6:46 PM, Miguel Arroz wrote: Hi! The KISS solution: configure the maximum size of a request in Apache itself, using the LimitRequestBody option in the configuration. You should find a way to handle nicely the error pages the user will see when submitting a file too large. This of course depends on your setup, if you are doing the upload using some Ajax trickery or not, etc. If you don't want to use the Apache thing, you should be able to do a streaming upload (stream it to disk) and cut off the copying after a certain point. It should be straightforward to return an error message. Chuck On 2008/04/01, at 02:04, Owen McKerrow wrote: Hi All, Just wondering if anyone has any recommendations on what to use to limit the size of files uses upload. Im currently looking at http://swfupload.org/. has anyone had any experience with it, in particular getting it to play nice with a WOApp ? Or is there something better out there ? Owen McKerrow WebMaster, emlab Ph : +61 02 4221 5517 http://emlab.uow.edu.au - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - "As of tomorrow, employees will only be able to access the building using individual security cards. Pictures will be taken next Wednesday employees will receive their cards in two weeks." - "Dilbert Quotes" Winner, Fred Dales, Microsoft Corp ___ Do not post admin requests to the list. They will be ignored. Webobjects-dev mailing list (Webobjects-dev@lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/webobjects-dev/arroz%40guiamac.com This email sent to [EMAIL PROTECTED] Miguel Arroz http://www.terminalapp.net http://www.ipragma.com ___ Do not post admin requests to the list. They will be ignored. Webobjects-dev mailing list (Webobjects-dev@lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/webobjects-dev/chill%40global-village.net This email sent to [EMAIL PROTECTED] -- Practical WebObjects - for developers who want to increase their overall knowledge of WebObjects or who are trying to solve specific problems. http://www.global-village.net/products/practical_webobjects ___ Do not post admin requests to the list. They will be ignored. Webobjects-dev mailing list (Webobjects-dev@lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/webobjects-dev/jameslist%40primax.com This email sent to [EMAIL PROTECTED] ___ Do not post admin requests to the list. They will be ignored. Webobjects-dev mailing list (Webobjects-dev@lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/webobjects-dev/archive%40mail-archive.com This email sent to [EMAIL PROTECTED]
Re: The sky is falling! The world is ending! WebObjects is Dead!
They can't do WO sessions, almost all the great speakers of the community are speaking at WOWODC :-) WWDC sessions are out and there are no WO sessions!It is the end! WebObjects is dead! Dead, dead I say! http://developer.apple.com/wwdc/sessions/ Was I first this year? Was I? Someone was going to start the panic. It was my turn. For anyone new to WO: the WO sessions get announced later. WebObjects is definitely not dead. Be patient. But if you want even more real cool WO stuff, you want to come to WOWODC: http://www.wocommunity.org/wowodc08/ See you there! Chuck -- Practical WebObjects - for developers who want to increase their overall knowledge of WebObjects or who are trying to solve specific problems. http://www.global-village.net/products/practical_webobjects ___ Do not post admin requests to the list. They will be ignored. Webobjects-dev mailing list (Webobjects-dev@lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/webobjects-dev/probert%40os.ca This email sent to [EMAIL PROTECTED] ___ Do not post admin requests to the list. They will be ignored. Webobjects-dev mailing list (Webobjects-dev@lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/webobjects-dev/archive%40mail-archive.com This email sent to [EMAIL PROTECTED]
Re: The sky is falling! The world is ending! WebObjects is Dead!
I can't wait for the annual noob fest where they all jump in with... "I'm not gonna spend my time learning this if Apple is EOL'ing it". I look forward to that *every* year :-) Alan On Apr 2, 2008, at 5:03 PM, David Holt wrote: Darn, you beat me to it! David :-) On 2-Apr-08, at 3:57 PM, Chuck Hill wrote: WWDC sessions are out and there are no WO sessions!It is the end! WebObjects is dead! Dead, dead I say! http://developer.apple.com/wwdc/sessions/ Was I first this year? Was I? Someone was going to start the panic. It was my turn. For anyone new to WO: the WO sessions get announced later. WebObjects is definitely not dead. Be patient. But if you want even more real cool WO stuff, you want to come to WOWODC: http://www.wocommunity.org/wowodc08/ See you there! Chuck -- Practical WebObjects - for developers who want to increase their overall knowledge of WebObjects or who are trying to solve specific problems. http://www.global-village.net/products/practical_webobjects ___ Do not post admin requests to the list. They will be ignored. Webobjects-dev mailing list (Webobjects-dev@lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/webobjects-dev/ programmingosx%40mac.com This email sent to [EMAIL PROTECTED] ___ Do not post admin requests to the list. They will be ignored. Webobjects-dev mailing list (Webobjects-dev@lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/webobjects-dev/award% 40apple.com This email sent to [EMAIL PROTECTED] ___ Do not post admin requests to the list. They will be ignored. Webobjects-dev mailing list (Webobjects-dev@lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/webobjects-dev/archive%40mail-archive.com This email sent to [EMAIL PROTECTED]
Re: The sky is falling! The world is ending! WebObjects is Dead!
On 03/04/2008, at 10:03 AM, David Holt wrote: Darn, you beat me to it! David :-) +1 ___ Do not post admin requests to the list. They will be ignored. Webobjects-dev mailing list (Webobjects-dev@lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/webobjects-dev/archive%40mail-archive.com This email sent to [EMAIL PROTECTED]
Re: The sky is falling! The world is ending! WebObjects is Dead!
Darn, you beat me to it! David :-) On 2-Apr-08, at 3:57 PM, Chuck Hill wrote: WWDC sessions are out and there are no WO sessions!It is the end! WebObjects is dead! Dead, dead I say! http://developer.apple.com/wwdc/sessions/ Was I first this year? Was I? Someone was going to start the panic. It was my turn. For anyone new to WO: the WO sessions get announced later. WebObjects is definitely not dead. Be patient. But if you want even more real cool WO stuff, you want to come to WOWODC: http://www.wocommunity.org/wowodc08/ See you there! Chuck -- Practical WebObjects - for developers who want to increase their overall knowledge of WebObjects or who are trying to solve specific problems. http://www.global-village.net/products/practical_webobjects ___ Do not post admin requests to the list. They will be ignored. Webobjects-dev mailing list (Webobjects-dev@lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/webobjects-dev/programmingosx %40mac.com This email sent to [EMAIL PROTECTED] ___ Do not post admin requests to the list. They will be ignored. Webobjects-dev mailing list (Webobjects-dev@lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/webobjects-dev/archive%40mail-archive.com This email sent to [EMAIL PROTECTED]
The sky is falling! The world is ending! WebObjects is Dead!
WWDC sessions are out and there are no WO sessions!It is the end! WebObjects is dead! Dead, dead I say! http://developer.apple.com/wwdc/sessions/ Was I first this year? Was I? Someone was going to start the panic. It was my turn. For anyone new to WO: the WO sessions get announced later. WebObjects is definitely not dead. Be patient. But if you want even more real cool WO stuff, you want to come to WOWODC: http://www.wocommunity.org/wowodc08/ See you there! Chuck -- Practical WebObjects - for developers who want to increase their overall knowledge of WebObjects or who are trying to solve specific problems. http://www.global-village.net/products/practical_webobjects ___ Do not post admin requests to the list. They will be ignored. Webobjects-dev mailing list (Webobjects-dev@lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/webobjects-dev/archive%40mail-archive.com This email sent to [EMAIL PROTECTED]
Re: an interesting lockRowComparingAttributes:
On Apr 2, 2008, at 3:27 PM, Lachlan Deck wrote: On 03/04/2008, at 9:11 AM, Chuck Hill wrote: On Apr 2, 2008, at 2:57 PM, Lachlan Deck wrote: On 03/04/2008, at 4:55 AM, Chuck Hill wrote: On Apr 2, 2008, at 10:38 AM, David Avendasora wrote: If the SELECT ... FOR LOCK can't find any rows that have fields with values that match exactly with what the DB has, then it wouldn't get any rows either. No? Yes, but... SELECT ... FOR LOCK is _pessimistic_ locking and EOF should be using optimistic locking. If EOF is issuing pessimistic locking SQL, then something has gone wrong. Maybe log the SQL and then compare what WO is looking for with what is actually in the DB? That is not the issue. The important symptom is that EOF is doing a SELECT ... FOR LOCK. Didn't you see this when you were using Vertical Inheritance? Rings a bell for me... (Pierre, see radar://5632577 for suggested fix) Was it you that I was discussing this with? We did discuss this at one point a while ago. Do you recall what was happening? So many obscure bugs, so little time... Yes, see thread "Snapshot weirdness" Dec 07 http://sourceforge.net/mailarchive/message.php?msg_id=866FD117-33B9-488F-BD23-77FBA7636F88%40gmail.com Essentially... that thread will give you the background of the problem... but the real solution I posted to yourself/Mike offlist due to certain code mods. Yes, that is what I was remembering. It wasn't David this time. This time... But unless Johann is using VI, that does not shed much light on his problem. Thanks! Chuck -- Practical WebObjects - for developers who want to increase their overall knowledge of WebObjects or who are trying to solve specific problems. http://www.global-village.net/products/practical_webobjects ___ Do not post admin requests to the list. They will be ignored. Webobjects-dev mailing list (Webobjects-dev@lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/webobjects-dev/archive%40mail-archive.com This email sent to [EMAIL PROTECTED]
Re: an interesting lockRowComparingAttributes:
On 03/04/2008, at 9:11 AM, Chuck Hill wrote: On Apr 2, 2008, at 2:57 PM, Lachlan Deck wrote: On 03/04/2008, at 4:55 AM, Chuck Hill wrote: On Apr 2, 2008, at 10:38 AM, David Avendasora wrote: If the SELECT ... FOR LOCK can't find any rows that have fields with values that match exactly with what the DB has, then it wouldn't get any rows either. No? Yes, but... SELECT ... FOR LOCK is _pessimistic_ locking and EOF should be using optimistic locking. If EOF is issuing pessimistic locking SQL, then something has gone wrong. Maybe log the SQL and then compare what WO is looking for with what is actually in the DB? That is not the issue. The important symptom is that EOF is doing a SELECT ... FOR LOCK. Didn't you see this when you were using Vertical Inheritance? Rings a bell for me... (Pierre, see radar://5632577 for suggested fix) Was it you that I was discussing this with? We did discuss this at one point a while ago. Do you recall what was happening? So many obscure bugs, so little time... Yes, see thread "Snapshot weirdness" Dec 07 http://sourceforge.net/mailarchive/message.php?msg_id=866FD117-33B9-488F-BD23-77FBA7636F88%40gmail.com Essentially... that thread will give you the background of the problem... but the real solution I posted to yourself/Mike offlist due to certain code mods. with regards, -- Lachlan Deck ___ Do not post admin requests to the list. They will be ignored. Webobjects-dev mailing list (Webobjects-dev@lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/webobjects-dev/archive%40mail-archive.com This email sent to [EMAIL PROTECTED]
Re: Why no 1to1 relationships?
On Apr 2, 2008, at 4:19 PM, Chuck Hill wrote: On Apr 2, 2008, at 1:09 PM, netBrackets wrote: Thanks for the great comments. And 1:1, owned/mandatory data etc. is what I want. I perhaps have an unusual case in that for a small minority of days I experience bursts very heavy activity, with changes occurring on only a small subset of my data during this period And exactly what problem does this result in? Just the performance, and being able to keep up with the rate of transactions without having to buy more hardware. I guess I'm worried about the speed of database transactions and all that checking being very slow, maybe more than I should? and furthermore, changes to any one record can be guaranteed (through disabling a few features of the site temporarily) to be made by only a single session. I have an idea that I'd like to segregate this changing data into its own table, and then on the "heavy days", turning off all locking on the table (along with disabling the required site features to assure single session changes) to speed things up. Which leads to my next question, might there be a way to easily disable lock checking on a single table, and then turn it back on without having to create a whole new build and or re-create my database schema? I think it would be fair to characterize that method of handling this situation as "insane", "crazy", "wacked out", and "totally nuts". Unless your goal is data inconsistency. Maybe we should look at the real problem? Now don't hold back, tell us what you really think. :-) There's no real problem other than improving throughput for a short period of time. It's really how it's handled now, except using flat files and not a database to save the data (basically each session updates its own file), and it's not been a problem. Any specifics as to what's "totally nuts" as long as the guarantee holds up? Or is it the guarantee (no two existing sessions will ever update the same record) itself that you question? Chuck On Wednesday, April 02, 2008, at 02:12PM, "Art Isbell" <[EMAIL PROTECTED] > wrote: On Apr 2, 2008, at 8:45 AM, Robert Walker wrote: Yes, it is possible to create one-to-one relationships, joined on the primary keys. One side of the relationship will need "Propagates primary key," as has been mentioned. I just wanted to add one caveat to this configuration: The relationship will always be 1-to-1 and will never be 1-to-0. In other words if you created an instance of the entity on the side propagating the key, you will get an instance of the other side of the relationship. This will happen even if no fields are set on the other side. This means that if any attributes (other than the PK) on the other side of the relationship are set to required, neither object will save. Also if you attempt to create a new instance on other other side (not propagating the key) you will also have problems saving. As far as I recall this is the experience that I've had when attempting to do this. Because of these complications I now opt for the more flexible one-to-one style by setting it up in the model as a one-to-many relationship. Then force the many-side to allow only one or zero related rows though validation logic. Another way of stating this is that a one-to-one relationship must be mandatory (i.e., it cannot be optional). This makes sense because the foreign key in a one-to-one relationship is the primary key which cannot be null and is guaranteed to be unique. If a foreign key is not null, the destination object must exist. If a foreign key is not the primary key, it is not guaranteed to be unique, so an attribute that is not the primary key cannot be the foreign key of a one-to- one relationship. I think that this agrees with relational database theory; i.e., all one-to-one relationships must be 1:1, not 1:0. A 1:0 relationship is just a special case of a 1:many relationship. Aloha, Art ___ Do not post admin requests to the list. They will be ignored. Webobjects-dev mailing list (Webobjects-dev@lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/webobjects-dev/jeffandmonica%40mac.com This email sent to [EMAIL PROTECTED] ___ Do not post admin requests to the list. They will be ignored. Webobjects-dev mailing list (Webobjects-dev@lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/webobjects-dev/chill%40global-village.net This email sent to [EMAIL PROTECTED] -- Practical WebObjects - for developers who want to increase their overall knowledge of WebObjects or who are trying to solve specific problems. http://www.global-village.net/products/practical_webobjects ___ Do not post admin requests to th
Re: an interesting lockRowComparingAttributes:
On Apr 2, 2008, at 2:57 PM, Lachlan Deck wrote: On 03/04/2008, at 4:55 AM, Chuck Hill wrote: On Apr 2, 2008, at 10:38 AM, David Avendasora wrote: If the SELECT ... FOR LOCK can't find any rows that have fields with values that match exactly with what the DB has, then it wouldn't get any rows either. No? Yes, but... SELECT ... FOR LOCK is _pessimistic_ locking and EOF should be using optimistic locking. If EOF is issuing pessimistic locking SQL, then something has gone wrong. Maybe log the SQL and then compare what WO is looking for with what is actually in the DB? That is not the issue. The important symptom is that EOF is doing a SELECT ... FOR LOCK. Didn't you see this when you were using Vertical Inheritance? Rings a bell for me... (Pierre, see radar://5632577 for suggested fix) Was it you that I was discussing this with? Do you recall what was happening? So many obscure bugs, so little time... Chuck -- Practical WebObjects - for developers who want to increase their overall knowledge of WebObjects or who are trying to solve specific problems. http://www.global-village.net/products/practical_webobjects ___ Do not post admin requests to the list. They will be ignored. Webobjects-dev mailing list (Webobjects-dev@lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/webobjects-dev/archive%40mail-archive.com This email sent to [EMAIL PROTECTED]
Re: an interesting lockRowComparingAttributes:
On 03/04/2008, at 4:55 AM, Chuck Hill wrote: On Apr 2, 2008, at 10:38 AM, David Avendasora wrote: If the SELECT ... FOR LOCK can't find any rows that have fields with values that match exactly with what the DB has, then it wouldn't get any rows either. No? Yes, but... SELECT ... FOR LOCK is _pessimistic_ locking and EOF should be using optimistic locking. If EOF is issuing pessimistic locking SQL, then something has gone wrong. Maybe log the SQL and then compare what WO is looking for with what is actually in the DB? That is not the issue. The important symptom is that EOF is doing a SELECT ... FOR LOCK. Didn't you see this when you were using Vertical Inheritance? Rings a bell for me... (Pierre, see radar://5632577 for suggested fix) with regards, -- Lachlan Deck ___ Do not post admin requests to the list. They will be ignored. Webobjects-dev mailing list (Webobjects-dev@lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/webobjects-dev/archive%40mail-archive.com This email sent to [EMAIL PROTECTED]
Re: stateless component
On Apr 2, 2008, at 2:37 PM, Archibald Singleton wrote: Hello, From the WO API docs: "For extra efficiency, you can create stateless components: components that can be shared between sessions. Stateless components aren't replicated each time they're needed; rather, a single shared instance is referenced each time the component is used." But logging calls to my stateless component constructor (subclass of ERXStatelessComponent), I see that's it's being called for every request and that new instances are created for each request. Maybe worth noting is the fact that the component is instantiated by a direct action (hence there are no sessions). I basically replicated what David showed in his lastest screencast re: the first hit to the app being a default action instead of a component action. I would have thought that that "shared instance" business means it's created once and reused all along? Could someone please be kind enough to enlighten me as to what's happening? Are you creating this using pageWithName()? These are stateless _components_ not _pages_. That is all that comes immediately to mind. If you use more than one in a thread, new ones will be created. Chuck -- Practical WebObjects - for developers who want to increase their overall knowledge of WebObjects or who are trying to solve specific problems. http://www.global-village.net/products/practical_webobjects ___ Do not post admin requests to the list. They will be ignored. Webobjects-dev mailing list (Webobjects-dev@lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/webobjects-dev/archive%40mail-archive.com This email sent to [EMAIL PROTECTED]
stateless component
Hello, From the WO API docs: "For extra efficiency, you can create stateless components: components that can be shared between sessions. Stateless components aren't replicated each time they're needed; rather, a single shared instance is referenced each time the component is used." But logging calls to my stateless component constructor (subclass of ERXStatelessComponent), I see that's it's being called for every request and that new instances are created for each request. Maybe worth noting is the fact that the component is instantiated by a direct action (hence there are no sessions). I basically replicated what David showed in his lastest screencast re: the first hit to the app being a default action instead of a component action. I would have thought that that "shared instance" business means it's created once and reused all along? Could someone please be kind enough to enlighten me as to what's happening? TIA = tmk = ___ Do not post admin requests to the list. They will be ignored. Webobjects-dev mailing list (Webobjects-dev@lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/webobjects-dev/archive%40mail-archive.com This email sent to [EMAIL PROTECTED]
Re: WO 5.4 Linux Deployment
Hi there, On 02/04/2008, at 5:37 PM, Aleksey Novicov wrote: I tried this and found that the latest Eclipse/WOLips tools are very complete in packaging up woapps for servlet deployment. However, there were a couple of caveats worth mentioning: 1. Starting with a WOnder Application template there is no need (and no option!) to create a project with servlet support. All that is needed after the project is created is to check Servlet Deployment and check Autogenerate web.xml in WOLips Build within the project properties. This will create the proper WAR file as a fully self- contained servlet single directory deployment (SSDD) with all of the necessary WO frameworks. It's very slick. 2. Project dependencies (such as other frameworks) however will not be included in the SSDD. They must be included as framework dependencies as well. See http://wiki.objectstyle.org/confluence/display/WOL/Add+a+Framework+Dependency . 3. The generated web.xml file in WEB-INF has a parameter named WOAppMode which is set to "Development". I found that with this setting, Tomcat and the WebObjects Resource Manager would not serve up any static WebServerResources correctly. I kept getting links like "/MyApp/WebObjects/MyApp.woa/_wr_/wodata=/Library/stuff". I don't know if I had a particular property value set incorrectly. But I did find that if I changed the WOAppMode value to "Deployment", the problem went away since the static resources get served up directly by the web server. This of course requires a split install (Tomcat + Apache). Because I'm using Ajax, I had to remember to include Ajax.framework/WebServerResources in the split install since there are critical javascript files that are required. <...> You might like to adjust the project template for, adding the optional tickboxes for all of these, and provide them back to WOLips. with regards, -- Lachlan Deck ___ Do not post admin requests to the list. They will be ignored. Webobjects-dev mailing list (Webobjects-dev@lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/webobjects-dev/archive%40mail-archive.com This email sent to [EMAIL PROTECTED]
Re: an interesting lockRowComparingAttributes:
On Apr 2, 2008, at 2:18 PM, Art Isbell wrote: On Apr 2, 2008, at 7:38 AM, David Avendasora wrote: Couldn't this be the same problem though? If the SELECT ... FOR LOCK can't find any rows that have fields with values that match exactly with what the DB has, then it wouldn't get any rows either. No? Maybe, but I think Chuck's point is that SELECT ... FOR LOCK shouldn't be generated unless the fetch specification was configured to lock objects (i.e., fetchSpec.locksObjects() returns true). This is not recommended. Yes. And there is at least one place in EOF where an unexpected exception or null or something, causes it to assume that it is a lock operation. I _think_ it has something to do with adaptor operations. I just can't quite recall what it was. Chuck -- Practical WebObjects - for developers who want to increase their overall knowledge of WebObjects or who are trying to solve specific problems. http://www.global-village.net/products/practical_webobjects ___ Do not post admin requests to the list. They will be ignored. Webobjects-dev mailing list (Webobjects-dev@lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/webobjects-dev/archive%40mail-archive.com This email sent to [EMAIL PROTECTED]
Re: WO 5.4 Linux Deployment
This does not cause any problems ... It's an obnoxious error that isn't really an error. It's been a couple months since I tracked down what exactly was going on, but I _believe_ it's trying to lookup the jdbc datasource in your tomcat server config, and if it's not there, it displays this warning. You can safely ignore it. ms On Apr 2, 2008, at 5:14 PM, Jeff Smith wrote: FWIW, I'm getting the "java name not found in context" as well and I haven't figure out a way around getting the messages either including specifying my data source in the web.xml. Doesn't seem to hurt anything but it does clog up the log file. On Apr 2, 2008, at 10:34 AM, Aleksey Novicov wrote: Yes, I will post what I've learned on the WO wiki as soon as I think I have everything worked out. Aleksey On Apr 2, 2008, at 2:05 AM, Andrew Lindesay wrote: Hello Aleksey; It would be really great if you could write-up your experiences an tips in the WO wiki for others to read about. http://wiki.objectstyle.org/confluence/display/WO/Home 4. For the static resources to be served correctly via port 80 of the Apache web server (as opposed to Tomcat's port 8080), I had to install the Tomcat-Apache plugin, mod_jk. Actually Apache 2.2 has AJP support built-in as well as a neat little balancer web app to monitor the deploy. I have some unrelated, but possibly helpful notes on the Apache 2.2 setup for this in this document; http://homepage.mac.com/andrewlindesay/le/manuals/lewostuff/LEWOStuffOverview.pdf See 9.5 and 9.6. I find the Apache 2.2 balancer setup much more elegant than mod_jk. 5. In the Tomcat log file (catalina.out), I keep getting an exception (below) related to JDBC but everything runs fine regardless (I'm using MySQL). Tomcat seems to have it's own data source mechanism I think I read something about that in an old mailing list post. I think you need to give the data source the same name as the model or something -- perhaps somebody can clarify? cheers. ___ Andrew Lindesay www.lindesay.co.nz ___ Do not post admin requests to the list. They will be ignored. Webobjects-dev mailing list (Webobjects-dev@lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/webobjects-dev/jas35%40mac.com This email sent to [EMAIL PROTECTED] ___ Do not post admin requests to the list. They will be ignored. Webobjects-dev mailing list (Webobjects-dev@lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/webobjects-dev/mschrag%40mdimension.com This email sent to [EMAIL PROTECTED] ___ Do not post admin requests to the list. They will be ignored. Webobjects-dev mailing list (Webobjects-dev@lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/webobjects-dev/archive%40mail-archive.com This email sent to [EMAIL PROTECTED]
Re: Why no 1to1 relationships?
On Apr 2, 2008, at 1:09 PM, netBrackets wrote: Thanks for the great comments. And 1:1, owned/mandatory data etc. is what I want. I perhaps have an unusual case in that for a small minority of days I experience bursts very heavy activity, with changes occurring on only a small subset of my data during this period And exactly what problem does this result in? and furthermore, changes to any one record can be guaranteed (through disabling a few features of the site temporarily) to be made by only a single session. I have an idea that I'd like to segregate this changing data into its own table, and then on the "heavy days", turning off all locking on the table (along with disabling the required site features to assure single session changes) to speed things up. Which leads to my next question, might there be a way to easily disable lock checking on a single table, and then turn it back on without having to create a whole new build and or re-create my database schema? I think it would be fair to characterize that method of handling this situation as "insane", "crazy", "wacked out", and "totally nuts". Unless your goal is data inconsistency. Maybe we should look at the real problem? Chuck On Wednesday, April 02, 2008, at 02:12PM, "Art Isbell" <[EMAIL PROTECTED] > wrote: On Apr 2, 2008, at 8:45 AM, Robert Walker wrote: Yes, it is possible to create one-to-one relationships, joined on the primary keys. One side of the relationship will need "Propagates primary key," as has been mentioned. I just wanted to add one caveat to this configuration: The relationship will always be 1-to-1 and will never be 1-to-0. In other words if you created an instance of the entity on the side propagating the key, you will get an instance of the other side of the relationship. This will happen even if no fields are set on the other side. This means that if any attributes (other than the PK) on the other side of the relationship are set to required, neither object will save. Also if you attempt to create a new instance on other other side (not propagating the key) you will also have problems saving. As far as I recall this is the experience that I've had when attempting to do this. Because of these complications I now opt for the more flexible one-to-one style by setting it up in the model as a one-to-many relationship. Then force the many-side to allow only one or zero related rows though validation logic. Another way of stating this is that a one-to-one relationship must be mandatory (i.e., it cannot be optional). This makes sense because the foreign key in a one-to-one relationship is the primary key which cannot be null and is guaranteed to be unique. If a foreign key is not null, the destination object must exist. If a foreign key is not the primary key, it is not guaranteed to be unique, so an attribute that is not the primary key cannot be the foreign key of a one-to-one relationship. I think that this agrees with relational database theory; i.e., all one-to-one relationships must be 1:1, not 1:0. A 1:0 relationship is just a special case of a 1:many relationship. Aloha, Art ___ Do not post admin requests to the list. They will be ignored. Webobjects-dev mailing list (Webobjects-dev@lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/webobjects-dev/jeffandmonica%40mac.com This email sent to [EMAIL PROTECTED] ___ Do not post admin requests to the list. They will be ignored. Webobjects-dev mailing list (Webobjects-dev@lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/webobjects-dev/chill%40global-village.net This email sent to [EMAIL PROTECTED] -- Practical WebObjects - for developers who want to increase their overall knowledge of WebObjects or who are trying to solve specific problems. http://www.global-village.net/products/practical_webobjects ___ Do not post admin requests to the list. They will be ignored. Webobjects-dev mailing list (Webobjects-dev@lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/webobjects-dev/archive%40mail-archive.com This email sent to [EMAIL PROTECTED]
Re: an interesting lockRowComparingAttributes:
On Apr 2, 2008, at 7:38 AM, David Avendasora wrote: Couldn't this be the same problem though? If the SELECT ... FOR LOCK can't find any rows that have fields with values that match exactly with what the DB has, then it wouldn't get any rows either. No? Maybe, but I think Chuck's point is that SELECT ... FOR LOCK shouldn't be generated unless the fetch specification was configured to lock objects (i.e., fetchSpec.locksObjects() returns true). This is not recommended. Aloha, Art ___ Do not post admin requests to the list. They will be ignored. Webobjects-dev mailing list (Webobjects-dev@lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/webobjects-dev/archive%40mail-archive.com This email sent to [EMAIL PROTECTED]
Re: Adaptor problems in Eclipse. No Instance Available
Have their been any network changes? Have you made the WOHost changes here: http://wiki.objectstyle.org/confluence/display/WO/Development+Tools-Running+Through+Apache On Apr 2, 2008, at 12:56 PM, Helmut Schottmüller wrote: Hi Group, since a few hours my WO httpd adaptor seems to be broken. I'm getting getting constant "No Instance Available" messages when I am trying to run my projects with webserver connect from Eclipse. I already checked the adaptor log file (enabled it with Debug in the .conf file) but nothing strange in there. The httpd access log is the only log which shows something (bunches of 500 (Internal Server Error) responses). I tried restarting apache and the computer for several times. The strange thing is that I did not change anything besides adding one or two new components to my project. I thought it might be memory related but I increased the VM settings and even with some simple projects the webserver connect isn't running as supposed when I run the projects from within Eclipse. With DirectConnect everything runs fine and even if I deploy my applications on the same Mac webserver connect is working. I am working with Eclipse 3.3.1.1, Leopard 10.5.2 and WO 5.4 using the system httpd and adaptor. Does anyone know if there is some other place where I might find some more information why I am getting this "No Instance Available"? Is there any more logging that might be activated besides the usual? Maybe it's a timeout problem. Are there any settings which I could increase to prevent the "No Instance Available" problem? I'll be happy for any hints which might help to solve my problem. Thanks in advance, Helmut ___ Do not post admin requests to the list. They will be ignored. Webobjects-dev mailing list (Webobjects-dev@lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/webobjects-dev/chill%40global-village.net This email sent to [EMAIL PROTECTED] -- Practical WebObjects - for developers who want to increase their overall knowledge of WebObjects or who are trying to solve specific problems. http://www.global-village.net/products/practical_webobjects ___ Do not post admin requests to the list. They will be ignored. Webobjects-dev mailing list (Webobjects-dev@lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/webobjects-dev/archive%40mail-archive.com This email sent to [EMAIL PROTECTED]
Re: WO 5.4 Linux Deployment
FWIW, I'm getting the "java name not found in context" as well and I haven't figure out a way around getting the messages either including specifying my data source in the web.xml. Doesn't seem to hurt anything but it does clog up the log file. On Apr 2, 2008, at 10:34 AM, Aleksey Novicov wrote: Yes, I will post what I've learned on the WO wiki as soon as I think I have everything worked out. Aleksey On Apr 2, 2008, at 2:05 AM, Andrew Lindesay wrote: Hello Aleksey; It would be really great if you could write-up your experiences an tips in the WO wiki for others to read about. http://wiki.objectstyle.org/confluence/display/WO/Home 4. For the static resources to be served correctly via port 80 of the Apache web server (as opposed to Tomcat's port 8080), I had to install the Tomcat-Apache plugin, mod_jk. Actually Apache 2.2 has AJP support built-in as well as a neat little balancer web app to monitor the deploy. I have some unrelated, but possibly helpful notes on the Apache 2.2 setup for this in this document; http://homepage.mac.com/andrewlindesay/le/manuals/lewostuff/LEWOStuffOverview.pdf See 9.5 and 9.6. I find the Apache 2.2 balancer setup much more elegant than mod_jk. 5. In the Tomcat log file (catalina.out), I keep getting an exception (below) related to JDBC but everything runs fine regardless (I'm using MySQL). Tomcat seems to have it's own data source mechanism I think I read something about that in an old mailing list post. I think you need to give the data source the same name as the model or something -- perhaps somebody can clarify? cheers. ___ Andrew Lindesay www.lindesay.co.nz ___ Do not post admin requests to the list. They will be ignored. Webobjects-dev mailing list (Webobjects-dev@lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/webobjects-dev/jas35%40mac.com This email sent to [EMAIL PROTECTED] ___ Do not post admin requests to the list. They will be ignored. Webobjects-dev mailing list (Webobjects-dev@lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/webobjects-dev/archive%40mail-archive.com This email sent to [EMAIL PROTECTED]
Re: Adaptor problems in Eclipse. No Instance Available
On Apr 2, 2008, at 9:56 AM, Helmut Schottmüller wrote: since a few hours my WO httpd adaptor seems to be broken. I'm getting getting constant "No Instance Available" messages when I am trying to run my projects with webserver connect from Eclipse. If a "No instance available" response occurs after 30 seconds, the default timeout to receive a response from an instance, then the instance is taking more than 30 seconds to process the request. This is usually due to database activity. Frequently, lots of faults are firing resulting in many one-object fetches (batch fetch). Or maybe there are fetches on huge DB tables without good indexes (add indexes). Or maybe fetches involve huge many-table joins. If you can't fix the performance problem, then you can increase the HTTP adaptor's receive timeout. I don't know how to do this in Web server connect mode during development. I have worked around my ignorance by configuring an instance of this app using JavaMonitor, increasing the receive timeout for this app in JavaMonitor, and then launching from Eclipse with the "-WOPort " launch argument in which is the port of the instance that JavaMonitor assigned. This will allow the HTTP adaptor to get the receive timeout setting for the process listening on this port from the SiteConfig.xml file. If a "No instance available" response occurs almost immediately, then something else is wrong. I don't know what that "something else" might be. Aloha, Art ___ Do not post admin requests to the list. They will be ignored. Webobjects-dev mailing list (Webobjects-dev@lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/webobjects-dev/archive%40mail-archive.com This email sent to [EMAIL PROTECTED]
Re: an interesting lockRowComparingAttributes:
Couldn't this be the same problem though? If the SELECT ... FOR LOCK can't find any rows that have fields with values that match exactly with what the DB has, then it wouldn't get any rows either. No? Maybe log the SQL and then compare what WO is looking for with what is actually in the DB? Dave On Apr 2, 2008, at 1:19 PM, Chuck Hill wrote: Close, but no rabbit stamp. What you are talking about produces an optimistic locking expection and "update operation failed to update any rows" from the channel. What Johan is hitting is an SQL statement like SELECT ... FOR LOCK. That is pessimistic locking. ___ Do not post admin requests to the list. They will be ignored. Webobjects-dev mailing list (Webobjects-dev@lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/webobjects-dev/archive%40mail-archive.com This email sent to [EMAIL PROTECTED]
Re: Why no 1to1 relationships?
On Apr 2, 2008, at 9:52 AM, Sacha Michel Mallais wrote: It is definitely possible to have an optional to-one, just not one where the FK is also the PK. But if the FK isn't the PK, then the FK isn't guaranteed to be unique. But it must be unique because it joins with the PK of another object which, by definition, is unique. I suppose that you could add a unique constraint to the FK, but that's cheating :-) (at least, to relational DB purists) So an optional to-one relationship is really a to-many relationship constrained to contain no more than 1 object (i.e., 0 or 1 object). From the "EOModeler User Guide": "Cardinality Every relationship has a cardinality. The cardinality defines how many destination records can potentially resolve the relationship. In relational database systems, there are generally two cardinalities: • to-one relationship — for each source record, there is exactly one corresponding destination record • to-many relationship — for each source record, there may be zero, one, or more corresponding destination records" Aloha, Art ___ Do not post admin requests to the list. They will be ignored. Webobjects-dev mailing list (Webobjects-dev@lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/webobjects-dev/archive%40mail-archive.com This email sent to [EMAIL PROTECTED]
Re: Locking Session.defaultEditingContext() Question
On Apr 2, 2008, at 9:10 AM, Chuck Hill wrote: What the LockScreamingEditingContext won't tell you is when you are using the EC but it is NOT locked. Unless you are doing the locking using ERXEC and friends or the MultiECLockManager (whatever its real name is), there is a very good chance that you are using unlocked editing contexts at some point. I've have been using MultiECLockManager without problems for several years. It's at http://wocode.com/WOCode/Files/MultiECLockManager.java. Aloha, Art ___ Do not post admin requests to the list. They will be ignored. Webobjects-dev mailing list (Webobjects-dev@lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/webobjects-dev/archive%40mail-archive.com This email sent to [EMAIL PROTECTED]
Re: Why no 1to1 relationships?
Thanks for the great comments. And 1:1, owned/mandatory data etc. is what I want. I perhaps have an unusual case in that for a small minority of days I experience bursts very heavy activity, with changes occurring on only a small subset of my data during this period and furthermore, changes to any one record can be guaranteed (through disabling a few features of the site temporarily) to be made by only a single session. I have an idea that I'd like to segregate this changing data into its own table, and then on the "heavy days", turning off all locking on the table (along with disabling the required site features to assure single session changes) to speed things up. Which leads to my next question, might there be a way to easily disable lock checking on a single table, and then turn it back on without having to create a whole new build and or re-create my database schema? Thanks, Jeff On Wednesday, April 02, 2008, at 02:12PM, "Art Isbell" <[EMAIL PROTECTED]> wrote: >On Apr 2, 2008, at 8:45 AM, Robert Walker wrote: > >> Yes, it is possible to create one-to-one relationships, joined on >> the primary keys. One side of the relationship will need "Propagates >> primary key," as has been mentioned. >> >> I just wanted to add one caveat to this configuration: The >> relationship will always be 1-to-1 and will never be 1-to-0. In >> other words if you created an instance of the entity on the side >> propagating the key, you will get an instance of the other side of >> the relationship. This will happen even if no fields are set on the >> other side. This means that if any attributes (other than the PK) on >> the other side of the relationship are set to required, neither >> object will save. >> >> Also if you attempt to create a new instance on other other side >> (not propagating the key) you will also have problems saving. >> >> As far as I recall this is the experience that I've had when >> attempting to do this. Because of these complications I now opt for >> the more flexible one-to-one style by setting it up in the model as >> a one-to-many relationship. Then force the many-side to allow only >> one or zero related rows though validation logic. > > Another way of stating this is that a one-to-one relationship must be >mandatory (i.e., it cannot be optional). This makes sense because the >foreign key in a one-to-one relationship is the primary key which >cannot be null and is guaranteed to be unique. If a foreign key is >not null, the destination object must exist. If a foreign key is not >the primary key, it is not guaranteed to be unique, so an attribute >that is not the primary key cannot be the foreign key of a one-to-one >relationship. > > I think that this agrees with relational database theory; i.e., all >one-to-one relationships must be 1:1, not 1:0. A 1:0 relationship is >just a special case of a 1:many relationship. > >Aloha, >Art > > ___ >Do not post admin requests to the list. They will be ignored. >Webobjects-dev mailing list (Webobjects-dev@lists.apple.com) >Help/Unsubscribe/Update your Subscription: >http://lists.apple.com/mailman/options/webobjects-dev/jeffandmonica%40mac.com > >This email sent to [EMAIL PROTECTED] > > ___ Do not post admin requests to the list. They will be ignored. Webobjects-dev mailing list (Webobjects-dev@lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/webobjects-dev/archive%40mail-archive.com This email sent to [EMAIL PROTECTED]
Adaptor problems in Eclipse. No Instance Available
Hi Group, since a few hours my WO httpd adaptor seems to be broken. I'm getting getting constant "No Instance Available" messages when I am trying to run my projects with webserver connect from Eclipse. I already checked the adaptor log file (enabled it with Debug in the .conf file) but nothing strange in there. The httpd access log is the only log which shows something (bunches of 500 (Internal Server Error) responses). I tried restarting apache and the computer for several times. The strange thing is that I did not change anything besides adding one or two new components to my project. I thought it might be memory related but I increased the VM settings and even with some simple projects the webserver connect isn't running as supposed when I run the projects from within Eclipse. With DirectConnect everything runs fine and even if I deploy my applications on the same Mac webserver connect is working. I am working with Eclipse 3.3.1.1, Leopard 10.5.2 and WO 5.4 using the system httpd and adaptor. Does anyone know if there is some other place where I might find some more information why I am getting this "No Instance Available"? Is there any more logging that might be activated besides the usual? Maybe it's a timeout problem. Are there any settings which I could increase to prevent the "No Instance Available" problem? I'll be happy for any hints which might help to solve my problem. Thanks in advance, Helmut smime.p7s Description: S/MIME cryptographic signature ___ Do not post admin requests to the list. They will be ignored. Webobjects-dev mailing list (Webobjects-dev@lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/webobjects-dev/archive%40mail-archive.com This email sent to [EMAIL PROTECTED]
Re: Why no 1to1 relationships?
On Apr 2, 2008, at 12:02 PM, Art Isbell wrote: I think that this agrees with relational database theory; i.e., all one-to-one relationships must be 1:1, not 1:0. A 1:0 relationship is just a special case of a 1:many relationship. It is definitely possible to have an optional to-one, just not one where the FK is also the PK. sacha -- Sacha Michel Mallais Senior Developer / President Global Village Consulting Inc. http://www.global-village.net/ PGP Key ID: 7D757B65 AIM: smallais ObAd: read "Practical WebObjects" http://www.global-village.net/products/practical_webobjects ___ Do not post admin requests to the list. They will be ignored. Webobjects-dev mailing list (Webobjects-dev@lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/webobjects-dev/archive%40mail-archive.com This email sent to [EMAIL PROTECTED]
Re: Locking Session.defaultEditingContext() Question
On Apr 2, 2008, at 12:04 PM, Dov Rosenberg wrote: Anywhere we created a new EOEditingContext - we replaced it with the LockScreamingEditingContext so we can track lock and unlocks properly. All of our locks and unlocks are in TRY/FINALLY methods. Not a perfect solution but it certainly helped clean up our stability issues. What the LockScreamingEditingContext won't tell you is when you are using the EC but it is NOT locked. Unless you are doing the locking using ERXEC and friends or the MultiECLockManager (whatever its real name is), there is a very good chance that you are using unlocked editing contexts at some point. I removed the lock we were doing on a session based default editing context and it fixed the issue we were having. Definitely not a good idea to lock the default editing context. I agree. :-) Chuck Dov Rosenberg On 4/2/08 1:29 PM, "Chuck Hill" <[EMAIL PROTECTED]> wrote: On Apr 1, 2008, at 8:17 AM, Dov Rosenberg wrote: I have been tracking down a strange error in our app. java.lang.IllegalStateException: There is no database snapshot available for the object In a lot of places we use a new EditingContext instead of the default Editing Context for the session. When we use a new editing context we have been very careful to lock and unlock before and after use. How? What do you use? Where / how do you do this? In some places we use the session default EditingContext (we are probably going to stop doing that in the future). A couple of questions come to mind: • I have read that it is not necessary to lock the session default EditingContext because EOF will automanage it. What happens if it does get manually locked? The error above was thrown when trying to do an ec.lock() on a Session.defaultEditingContext(). I'd really go a long way to avoid locking and unlocking the default EC. This can lead to serious problems, especially around session termination. • What physically happens when an EC is locked? It seems it is some sort of reference counter. In an earlier post Chuck mentioned that locking the EODatabaseContext will prevent fetches/saves from occurring, what happens when an EC is locked? Mostly it prevents other threads from updating its state (e.g. when an EC in those threads saves changes to objects in this EC). The changes are queued for later processing. • Is there any compelling reason to use the session default editing context? We aren’t using Project Wonder (yet) I use them for things whose lifetime matches the session: logged in user, their permissions, etc. Chuck -- Practical WebObjects - for developers who want to increase their overall knowledge of WebObjects or who are trying to solve specific problems. http://www.global-village.net/products/practical_webobjects ___ Do not post admin requests to the list. They will be ignored. Webobjects-dev mailing list (Webobjects-dev@lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/webobjects-dev/archive%40mail-archive.com This email sent to [EMAIL PROTECTED]
Re: Locking Session.defaultEditingContext() Question
Anywhere we created a new EOEditingContext - we replaced it with the LockScreamingEditingContext so we can track lock and unlocks properly. All of our locks and unlocks are in TRY/FINALLY methods. Not a perfect solution but it certainly helped clean up our stability issues. I removed the lock we were doing on a session based default editing context and it fixed the issue we were having. Definitely not a good idea to lock the default editing context. Dov Rosenberg On 4/2/08 1:29 PM, "Chuck Hill" <[EMAIL PROTECTED]> wrote: > > On Apr 1, 2008, at 8:17 AM, Dov Rosenberg wrote: >> I have been tracking down a strange error in our app. >> >> java.lang.IllegalStateException: There is no database snapshot >> available for the object >> >> In a lot of places we use a new EditingContext instead of the >> default Editing Context for the session. When we use a new editing >> context we have been very careful to lock and unlock before and >> after use. > > How? What do you use? Where / how do you do this? > > >> In some places we use the session default EditingContext (we are >> probably going to stop doing that in the future). A couple of >> questions come to mind: >> >> I have read that it is not necessary to lock the session default >> EditingContext because EOF will automanage it. What happens if it >> does get manually locked? The error above was thrown when trying to >> do an ec.lock() on a Session.defaultEditingContext(). > > I'd really go a long way to avoid locking and unlocking the default > EC. This can lead to serious problems, especially around session > termination. > > > >> What physically happens when an EC is locked? It seems it is some >> sort of reference counter. In an earlier post Chuck mentioned that >> locking the EODatabaseContext will prevent fetches/saves from >> occurring, what happens when an EC is locked? > > Mostly it prevents other threads from updating its state (e.g. when an > EC in those threads saves changes to objects in this EC). The changes > are queued for later processing. > > >> Is there any compelling reason to use the session default editing >> context? We aren¹t using Project Wonder (yet) > > > I use them for things whose lifetime matches the session: logged in > user, their permissions, etc. > > Chuck ___ Do not post admin requests to the list. They will be ignored. Webobjects-dev mailing list (Webobjects-dev@lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/webobjects-dev/archive%40mail-archive.com This email sent to [EMAIL PROTECTED]
Re: Why no 1to1 relationships?
On Apr 2, 2008, at 8:45 AM, Robert Walker wrote: Yes, it is possible to create one-to-one relationships, joined on the primary keys. One side of the relationship will need "Propagates primary key," as has been mentioned. I just wanted to add one caveat to this configuration: The relationship will always be 1-to-1 and will never be 1-to-0. In other words if you created an instance of the entity on the side propagating the key, you will get an instance of the other side of the relationship. This will happen even if no fields are set on the other side. This means that if any attributes (other than the PK) on the other side of the relationship are set to required, neither object will save. Also if you attempt to create a new instance on other other side (not propagating the key) you will also have problems saving. As far as I recall this is the experience that I've had when attempting to do this. Because of these complications I now opt for the more flexible one-to-one style by setting it up in the model as a one-to-many relationship. Then force the many-side to allow only one or zero related rows though validation logic. Another way of stating this is that a one-to-one relationship must be mandatory (i.e., it cannot be optional). This makes sense because the foreign key in a one-to-one relationship is the primary key which cannot be null and is guaranteed to be unique. If a foreign key is not null, the destination object must exist. If a foreign key is not the primary key, it is not guaranteed to be unique, so an attribute that is not the primary key cannot be the foreign key of a one-to-one relationship. I think that this agrees with relational database theory; i.e., all one-to-one relationships must be 1:1, not 1:0. A 1:0 relationship is just a special case of a 1:many relationship. Aloha, Art ___ Do not post admin requests to the list. They will be ignored. Webobjects-dev mailing list (Webobjects-dev@lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/webobjects-dev/archive%40mail-archive.com This email sent to [EMAIL PROTECTED]
Re: Why no 1to1 relationships?
Yes, it is possible to create one-to-one relationships, joined on the primary keys. One side of the relationship will need "Propagates primary key," as has been mentioned. I just wanted to add one caveat to this configuration: The relationship will always be 1-to-1 and will never be 1-to-0. In other words if you created an instance of the entity on the side propagating the key, you will get an instance of the other side of the relationship. This will happen even if no fields are set on the other side. This means that if any attributes (other than the PK) on the other side of the relationship are set to required, neither object will save. Also if you attempt to create a new instance on other other side (not propagating the key) you will also have problems saving. As far as I recall this is the experience that I've had when attempting to do this. Because of these complications I now opt for the more flexible one-to-one style by setting it up in the model as a one-to-many relationship. Then force the many-side to allow only one or zero related rows though validation logic. This also makes me think of the Ruby on Rails associations mechanism for creating one-to-one relationships: One-to-one Master: has_one :detail Detail: belongs_to :master as opposed to the one to many setup... One-to-many Master: has_many :details Detail: belongs_to :master It seems to me that adding something along this line to EOF would be very useful, and not particularly difficult. I don't mean in the syntax above, but rather having a "built-in" way to constrain the many-side to allow only one. Come to think of it: Core Data supports this in its relationship configuration. It even goes a step further by allowing you to specified the number of related objects allowed. On Apr 2, 2008, at 1:48 PM, Jacky Gagnon wrote: Hi Jeff, I've decided to fully EOize my app to replace my current use of flat files, mainly to better deal with concurrency issues. In creating my analogous tables, I found that there's no way to specify a "1 to 1" relationship between tables using the Entity Modeler tool. Is there a specific reason for this? Or is it just figured that you might as well make it one to many anyway even if it is just one to one? You can specify «1 to 1» relationship with Entity Modeler. If the destination cannot be null, use the same id in the two tables. -In Properties > Basic select : the cardinality «To One», your destination, and add your primary keys. -In Properties > Advanced select : for the source object, check Propagates primary key and Owns Destination (if the destination object cannot exist without an owner). Good luck! Jacky If you're wondering why I want to do this, it's to hopefully reduce conflicts of multiple people updating the same record in a table. Is that generally a good reason to split 1 to 1 data into two tables? Thanks, Jeff ___ Do not post admin requests to the list. They will be ignored. Webobjects-dev mailing list (Webobjects-dev@lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/webobjects-dev/jgagnon% 40druide.com This email sent to [EMAIL PROTECTED] ___ Do not post admin requests to the list. They will be ignored. Webobjects-dev mailing list (Webobjects-dev@lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/webobjects-dev/robert.walker% 40bennettig.com This email sent to [EMAIL PROTECTED] Robert Walker [EMAIL PROTECTED] ___ Do not post admin requests to the list. They will be ignored. Webobjects-dev mailing list (Webobjects-dev@lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/webobjects-dev/archive%40mail-archive.com This email sent to [EMAIL PROTECTED]
Re: an interesting lockRowComparingAttributes:
Yes, but... SELECT ... FOR LOCK is _pessimistic_ locking and EOF should be using optimistic locking. If EOF is issuing pessimistic locking SQL, then something has gone wrong. Ah. I see. Who cares why it isn't returning rows. It shouldn't be doing this in the first place. Maybe log the SQL and then compare what WO is looking for with what is actually in the DB? That is not the issue. The important symptom is that EOF is doing a SELECT ... FOR LOCK. Didn't you see this when you were using Vertical Inheritance? *whistling, while looking around the room* I have no idea what you're talking about. Vertical who? ___ Do not post admin requests to the list. They will be ignored. Webobjects-dev mailing list (Webobjects-dev@lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/webobjects-dev/archive%40mail-archive.com This email sent to [EMAIL PROTECTED]
Re: an interesting lockRowComparingAttributes:
On Apr 2, 2008, at 10:38 AM, David Avendasora wrote: Couldn't this be the same problem though? No. :-) If the SELECT ... FOR LOCK can't find any rows that have fields with values that match exactly with what the DB has, then it wouldn't get any rows either. No? Yes, but... SELECT ... FOR LOCK is _pessimistic_ locking and EOF should be using optimistic locking. If EOF is issuing pessimistic locking SQL, then something has gone wrong. Maybe log the SQL and then compare what WO is looking for with what is actually in the DB? That is not the issue. The important symptom is that EOF is doing a SELECT ... FOR LOCK. Didn't you see this when you were using Vertical Inheritance? Chuck On Apr 2, 2008, at 1:19 PM, Chuck Hill wrote: Close, but no rabbit stamp. What you are talking about produces an optimistic locking expection and "update operation failed to update any rows" from the channel. What Johan is hitting is an SQL statement like SELECT ... FOR LOCK. That is pessimistic locking. -- Practical WebObjects - for developers who want to increase their overall knowledge of WebObjects or who are trying to solve specific problems. http://www.global-village.net/products/practical_webobjects ___ Do not post admin requests to the list. They will be ignored. Webobjects-dev mailing list (Webobjects-dev@lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/webobjects-dev/archive%40mail-archive.com This email sent to [EMAIL PROTECTED]
Re: Why no 1to1 relationships?
Hi Jeff, I've decided to fully EOize my app to replace my current use of flat files, mainly to better deal with concurrency issues. In creating my analogous tables, I found that there's no way to specify a "1 to 1" relationship between tables using the Entity Modeler tool. Is there a specific reason for this? Or is it just figured that you might as well make it one to many anyway even if it is just one to one? You can specify «1 to 1» relationship with Entity Modeler. If the destination cannot be null, use the same id in the two tables. -In Properties > Basic select : the cardinality «To One», your destination, and add your primary keys. -In Properties > Advanced select : for the source object, check Propagates primary key and Owns Destination (if the destination object cannot exist without an owner). Good luck! Jacky If you're wondering why I want to do this, it's to hopefully reduce conflicts of multiple people updating the same record in a table. Is that generally a good reason to split 1 to 1 data into two tables? Thanks, Jeff ___ Do not post admin requests to the list. They will be ignored. Webobjects-dev mailing list (Webobjects-dev@lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/webobjects-dev/jgagnon% 40druide.com This email sent to [EMAIL PROTECTED] ___ Do not post admin requests to the list. They will be ignored. Webobjects-dev mailing list (Webobjects-dev@lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/webobjects-dev/archive%40mail-archive.com This email sent to [EMAIL PROTECTED]
Re: Locking Session.defaultEditingContext() Question
On Apr 1, 2008, at 8:17 AM, Dov Rosenberg wrote: I have been tracking down a strange error in our app. java.lang.IllegalStateException: There is no database snapshot available for the object In a lot of places we use a new EditingContext instead of the default Editing Context for the session. When we use a new editing context we have been very careful to lock and unlock before and after use. How? What do you use? Where / how do you do this? In some places we use the session default EditingContext (we are probably going to stop doing that in the future). A couple of questions come to mind: • I have read that it is not necessary to lock the session default EditingContext because EOF will automanage it. What happens if it does get manually locked? The error above was thrown when trying to do an ec.lock() on a Session.defaultEditingContext(). I'd really go a long way to avoid locking and unlocking the default EC. This can lead to serious problems, especially around session termination. • What physically happens when an EC is locked? It seems it is some sort of reference counter. In an earlier post Chuck mentioned that locking the EODatabaseContext will prevent fetches/saves from occurring, what happens when an EC is locked? Mostly it prevents other threads from updating its state (e.g. when an EC in those threads saves changes to objects in this EC). The changes are queued for later processing. • Is there any compelling reason to use the session default editing context? We aren’t using Project Wonder (yet) I use them for things whose lifetime matches the session: logged in user, their permissions, etc. Chuck -- Practical WebObjects - for developers who want to increase their overall knowledge of WebObjects or who are trying to solve specific problems. http://www.global-village.net/products/practical_webobjects ___ Do not post admin requests to the list. They will be ignored. Webobjects-dev mailing list (Webobjects-dev@lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/webobjects-dev/archive%40mail-archive.com This email sent to [EMAIL PROTECTED]
Re: an interesting lockRowComparingAttributes:
On Apr 2, 2008, at 4:30 AM, David Avendasora wrote: The fact that Chuck didn't come up with this makes me doubt my memory, but... It seems that when I've run into this in my applications, it usually means that between the time the EO was read out of the database, and the time it is written back in, something in one of the locking fields changed, or the field isn't "locking-friendly" in that WO has a difficult time reading the datatype of the DB field and converting it into the requested Java datatype and then converting it back and coming up with the original value. I know with MS SQL Server that long text fields are a problem. I _believe_ (please correct me if I'm wrong) when EOF does an update, it does it along the lines of "UPDATE TABLE WHERE ATTRIBUTE_1 = x, ATTRIBUTE2 = y, etc" where the attribute values are from the original reading of the record. The problem is that the UPDATE can't find any records matching the original attributes, and therefore didn't "select" any rows for updating. Dave On Apr 1, 2008, at 6:06 PM, Johan Henselmans wrote: Error: com.webobjects.eoaccess.EOGeneralAdaptorException: lockRowComparingAttributes -- com.webobjects.jdbcadaptor.JDBCChannel: lock operation failed to select any rows Close, but no rabbit stamp. What you are talking about produces an optimistic locking expection and "update operation failed to update any rows" from the channel. What Johan is hitting is an SQL statement like SELECT ... FOR LOCK. That is pessimistic locking. That said, I think that we did look at a case where you had this problem. I just can't recall what the cause was. Chuck -- Practical WebObjects - for developers who want to increase their overall knowledge of WebObjects or who are trying to solve specific problems. http://www.global-village.net/products/practical_webobjects ___ Do not post admin requests to the list. They will be ignored. Webobjects-dev mailing list (Webobjects-dev@lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/webobjects-dev/archive%40mail-archive.com This email sent to [EMAIL PROTECTED]
Re: WO 5.4 Linux Deployment
Yes, I will post what I've learned on the WO wiki as soon as I think I have everything worked out. Aleksey On Apr 2, 2008, at 2:05 AM, Andrew Lindesay wrote: Hello Aleksey; It would be really great if you could write-up your experiences an tips in the WO wiki for others to read about. http://wiki.objectstyle.org/confluence/display/WO/Home 4. For the static resources to be served correctly via port 80 of the Apache web server (as opposed to Tomcat's port 8080), I had to install the Tomcat-Apache plugin, mod_jk. Actually Apache 2.2 has AJP support built-in as well as a neat little balancer web app to monitor the deploy. I have some unrelated, but possibly helpful notes on the Apache 2.2 setup for this in this document; http://homepage.mac.com/andrewlindesay/le/manuals/lewostuff/LEWOStuffOverview.pdf See 9.5 and 9.6. I find the Apache 2.2 balancer setup much more elegant than mod_jk. 5. In the Tomcat log file (catalina.out), I keep getting an exception (below) related to JDBC but everything runs fine regardless (I'm using MySQL). Tomcat seems to have it's own data source mechanism I think I read something about that in an old mailing list post. I think you need to give the data source the same name as the model or something -- perhaps somebody can clarify? cheers. ___ Andrew Lindesay www.lindesay.co.nz ___ Do not post admin requests to the list. They will be ignored. Webobjects-dev mailing list (Webobjects-dev@lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/webobjects-dev/archive%40mail-archive.com This email sent to [EMAIL PROTECTED]
Re: [OT] Postgresql capital accentuated characters problem
Hi! I see... I specially like this quote: "The drawback of using locales other than C or POSIX in PostgreSQL is its performance impact. It slows character handling and prevents ordinary indexes from being used by LIKE. For this reason use locales only if you actually need them." In this situation, I really prefer to drop the accentuation (it's an username). How can we be in 2008 and accentuated characters are still a problem!? Must become farmer, must become farmer... Yours Miguel Arroz On 2008/04/02, at 12:33, Ralf Schuchardt wrote: Hi Miguel, I think you must initialize your database cluster with the correct locale, to get this working (see http://www.postgresql.org/docs/8.3/interactive/locale.html) . Ralf Schuchardt Am 02.04.2008 um 03:34 schrieb Miguel Arroz: I created an UNIQUE INDEX using the lower() function to create a case-insensitive UNIQUE constraint. I tested this and I noticed it's failing on accentuated characters. The problem is this: select lower('JoÃo'); lower --- joÃo (1 row) lower() is not "lowering" the accentuated characters. ___ Do not post admin requests to the list. They will be ignored. Webobjects-dev mailing list (Webobjects-dev@lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/webobjects-dev/arroz%40guiamac.com This email sent to [EMAIL PROTECTED] Miguel Arroz http://www.terminalapp.net http://www.ipragma.com smime.p7s Description: S/MIME cryptographic signature ___ Do not post admin requests to the list. They will be ignored. Webobjects-dev mailing list (Webobjects-dev@lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/webobjects-dev/archive%40mail-archive.com This email sent to [EMAIL PROTECTED]
Re: WO 5.4 Linux Deployment
> See: the "Data Source for Model Database Configuration" section of > http://wiki.objectstyle.org/confluence/display/WO/Web+Applications-Deployment- > Tomcat+Deployment#WebApplications-Deployment-TomcatDeployment-DataSourceforMod > elDatabaseConfiguration btw: I have deployed a Wonder App as a Servlet and tried to specified the db connection via context.xml. It still uses the information from the eo model. I have played around with resource-refs in web.xml but no luck. Does anyone here have a working configuration? Oliver ___ Do not post admin requests to the list. They will be ignored. Webobjects-dev mailing list (Webobjects-dev@lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/webobjects-dev/archive%40mail-archive.com This email sent to [EMAIL PROTECTED]
Re: WO 5.4 Linux Deployment
Hey Aleksey, You are right Tomcat has it's own DB connection stuff, which can be a pain in the a** at times. I ran into this and it actually was a show- stopper when I had two (or more) EOModels. See: the "Data Source for Model Database Configuration" section of http://wiki.objectstyle.org/confluence/display/WO/Web+Applications-Deployment-Tomcat+Deployment#WebApplications-Deployment-TomcatDeployment-DataSourceforModelDatabaseConfiguration . This wiki page is somewhat out-of-date due to the improved project setup options in WOLips, but the core information in it still holds true, so read through it to see if you can find any other changes you may need to make to your build files. Specifically, it will help you get all the frameworks imbedded. On Apr 2, 2008, at 5:05 AM, Andrew Lindesay wrote: 5. In the Tomcat log file (catalina.out), I keep getting an exception (below) related to JDBC but everything runs fine regardless (I'm using MySQL). Tomcat seems to have it's own data source mechanism I think I read something about that in an old mailing list post. I think you need to give the data source the same name as the model or something -- perhaps somebody can clarify? ___ Do not post admin requests to the list. They will be ignored. Webobjects-dev mailing list (Webobjects-dev@lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/webobjects-dev/archive%40mail-archive.com This email sent to [EMAIL PROTECTED]
Re: [OT] Postgresql capital accentuated characters problem
Hi Miguel, I think you must initialize your database cluster with the correct locale, to get this working (see http://www.postgresql.org/docs/8.3/ interactive/locale.html). Ralf Schuchardt Am 02.04.2008 um 03:34 schrieb Miguel Arroz: I created an UNIQUE INDEX using the lower() function to create a case-insensitive UNIQUE constraint. I tested this and I noticed it's failing on accentuated characters. The problem is this: select lower('JoÃo'); lower --- joÃo (1 row) lower() is not "lowering" the accentuated characters. ___ Do not post admin requests to the list. They will be ignored. Webobjects-dev mailing list (Webobjects-dev@lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/webobjects-dev/archive%40mail-archive.com This email sent to [EMAIL PROTECTED]
Re: an interesting lockRowComparingAttributes:
The fact that Chuck didn't come up with this makes me doubt my memory, but... It seems that when I've run into this in my applications, it usually means that between the time the EO was read out of the database, and the time it is written back in, something in one of the locking fields changed, or the field isn't "locking-friendly" in that WO has a difficult time reading the datatype of the DB field and converting it into the requested Java datatype and then converting it back and coming up with the original value. I know with MS SQL Server that long text fields are a problem. I _believe_ (please correct me if I'm wrong) when EOF does an update, it does it along the lines of "UPDATE TABLE WHERE ATTRIBUTE_1 = x, ATTRIBUTE2 = y, etc" where the attribute values are from the original reading of the record. The problem is that the UPDATE can't find any records matching the original attributes, and therefore didn't "select" any rows for updating. Dave On Apr 1, 2008, at 6:06 PM, Johan Henselmans wrote: Error: com.webobjects.eoaccess.EOGeneralAdaptorException: lockRowComparingAttributes -- com.webobjects.jdbcadaptor.JDBCChannel: lock operation failed to select any rows ___ Do not post admin requests to the list. They will be ignored. Webobjects-dev mailing list (Webobjects-dev@lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/webobjects-dev/archive%40mail-archive.com This email sent to [EMAIL PROTECTED]
Re: Locking Session.defaultEditingContext() Question
Hi, I'm not entirely sure the problem is related to locking the editing context. This exception can be produced by objects, which are disconnected from their editing context but keep hovering about. Could it be your application holds onto an object, which has been deleted (in the EO sense), but some component maintains a reference? Or created one without inserting it into an editing context in the first place? J.L.Simon On 01.04.2008, at 20:13, Tonny Staunsbrink wrote: If I should take a guess, I would say it's a concurrency issue. Are you running a multithreaded app, do you have custom RequestHandler or are you running any background jobs? /T On Tue, Apr 1, 2008 at 5:17 PM, Dov Rosenberg <[EMAIL PROTECTED]> wrote: I have been tracking down a strange error in our app. java.lang.IllegalStateException: There is no database snapshot available for the object In a lot of places we use a new EditingContext instead of the default Editing Context for the session. When we use a new editing context we have been very careful to lock and unlock before and after use. In some places we use the session default EditingContext (we are probably going to stop doing that in the future). A couple of questions come to mind: I have read that it is not necessary to lock the session default EditingContext because EOF will automanage it. What happens if it does get manually locked? The error above was thrown when trying to do an ec.lock() on a Session.defaultEditingContext(). What physically happens when an EC is locked? It seems it is some sort of reference counter. In an earlier post Chuck mentioned that locking the EODatabaseContext will prevent fetches/saves from occurring, what happens when an EC is locked? Is there any compelling reason to use the session default editing context? We aren't using Project Wonder (yet) Thanks in advance for any help Dov Rosenberg ___ Do not post admin requests to the list. They will be ignored. Webobjects-dev mailing list (Webobjects-dev@lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/webobjects-dev/mezz.dk%40gmail.com This email sent to [EMAIL PROTECTED] ___ Do not post admin requests to the list. They will be ignored. Webobjects-dev mailing list (Webobjects-dev@lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/webobjects-dev/simon%40webtecc.com This email sent to [EMAIL PROTECTED] ___ Do not post admin requests to the list. They will be ignored. Webobjects-dev mailing list (Webobjects-dev@lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/webobjects-dev/archive%40mail-archive.com This email sent to [EMAIL PROTECTED]
Re: WO 5.4 Linux Deployment
Hello Aleksey; It would be really great if you could write-up your experiences an tips in the WO wiki for others to read about. http://wiki.objectstyle.org/confluence/display/WO/Home 4. For the static resources to be served correctly via port 80 of the Apache web server (as opposed to Tomcat's port 8080), I had to install the Tomcat-Apache plugin, mod_jk. Actually Apache 2.2 has AJP support built-in as well as a neat little balancer web app to monitor the deploy. I have some unrelated, but possibly helpful notes on the Apache 2.2 setup for this in this document; http://homepage.mac.com/andrewlindesay/le/manuals/lewostuff/LEWOStuffOverview.pdf See 9.5 and 9.6. I find the Apache 2.2 balancer setup much more elegant than mod_jk. 5. In the Tomcat log file (catalina.out), I keep getting an exception (below) related to JDBC but everything runs fine regardless (I'm using MySQL). Tomcat seems to have it's own data source mechanism I think I read something about that in an old mailing list post. I think you need to give the data source the same name as the model or something -- perhaps somebody can clarify? cheers. ___ Andrew Lindesay www.lindesay.co.nz ___ Do not post admin requests to the list. They will be ignored. Webobjects-dev mailing list (Webobjects-dev@lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/webobjects-dev/archive%40mail-archive.com This email sent to [EMAIL PROTECTED]
D2WS: Can't obtain rules because of protected access!
Hi, I posted this message to the WOLips list at first. But since this seems to be more of a WebObjects isue, I'll cross post it here. I'm trying to develop a D2WS application on Eclipse with WOLips. Starting the web service assistant and connecting to the application fails. The error message looks like this: java.lang.IllegalStateException: Can't obtain rules because of protected access! [2008-3-31 14:39:16 CEST] java.lang.IllegalStateException: Can't obtain rules because of protected access! at com .webobjects .webservices .generation._WSDirectAction.loadD2WModelAction(_WSDirectAction.java:224) at com.kontrast.cms.index.app.Session.awake(Session.java:15) at com.webobjects.appserver.WOSession._awakeInContext(WOSession.java: 834) at com .webobjects .appserver .WOApplication._initializeSessionInContext(WOApplication.java:2111) at com .webobjects .appserver ._private .WOComponentRequestHandler ._dispatchWithPreparedApplication(WOComponentRequestHandler.java:319) at com .webobjects .appserver ._private .WOComponentRequestHandler ._handleRequest(WOComponentRequestHandler.java:369) at com .webobjects .appserver ._private .WOComponentRequestHandler .handleRequest(WOComponentRequestHandler.java:442) at com .webobjects.appserver.WOApplication.dispatchRequest(WOApplication.java: 1678) at com .webobjects .appserver._private.WOWorkerThread.runOnce(WOWorkerThread.java:144) at com .webobjects.appserver._private.WOWorkerThread.run(WOWorkerThread.java: 226) at java.lang.Thread.run(Thread.java:613) I checked the project out on an older machine, where it worked under XCode and Eclipse/WOLips (after I added the server.wsdd and client.wsdd to the resources). So i have to assume it's a problem with 5.4. Any ideas? J.L.Simon ___ Do not post admin requests to the list. They will be ignored. Webobjects-dev mailing list (Webobjects-dev@lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/webobjects-dev/archive%40mail-archive.com This email sent to [EMAIL PROTECTED]
Re: Multiple EOModels
Thanks for that, obviously frameworks are the way to go, I haven't really needed to make my own up to now but clearly I should have :-) Back to the bookshelf then :-) Regards, David. On 27 Mar 2008, at 18:52, Chuck Hill wrote: On Mar 27, 2008, at 1:57 AM, David Griffith wrote: Hi all, I have two apps that have two distinct EOModels. I now need to link these two apps so that they have access to each other's data, so I was going to add both models to each app and recompile. Frameworks. You want frameworks. Frameworks are WebObjects (and, well, OS X's) way of sharing things between applications. Put the models and the related code in a framework and use the framework in the application. I _never_ _ever_ put a model in an application. However, it transpires that both models have a couple of entity names in common. Are they the same entity (same database, same table)? If so, create a framework and model to hold the common entities. Use this framework in the frameworks holding the models with the other entities. Include all three frameworks in your application. If they are not the same entity but just happen to have the same name, then you need to rename one of them. I can understand why this would cause a problem, what I need to know is if there is a way around this without having to completely rename the entities in one of the models. The other strange thing is that it tells me there is a name conflict between the EOPrototypes entity in each model, but I don't think there is any other way to name this entity, correct? You probably guessed this already, but "framework". The prototypes go into a model by themselves and this framework is included in (referenced in) the frameworks that have models and in the application. Thus Application -> Prototypes framework -> Common entities framework -> Other EO framework 1 -> Other EO framework 2 Chuck -- Practical WebObjects - for developers who want to increase their overall knowledge of WebObjects or who are trying to solve specific problems. http://www.global-village.net/products/practical_webobjects ___ Do not post admin requests to the list. They will be ignored. Webobjects-dev mailing list (Webobjects-dev@lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/webobjects-dev/archive%40mail-archive.com This email sent to [EMAIL PROTECTED]