Decided to pick it up again, maybe finish it. Seven levels.
Decided to pick it up again, maybe finish it. Seven levels.
Given my too short stay in Australia, it was deemed necessary I check out Melbourne’s festival scene as Canada’s state of affairs has led to attending headlining tours-only. With the geographical isolation of Oz, a rarity of international acts coming through leads to excessive admission charges to make up for visa and travel costs. Next month Welsh band Future of the Left is $47AUS at The Corner. I paid $15CDN in Toronto at ElMo. $10US at Austin’s Beauty Bar in 2009. So dropping $150 for Harvest Festival of more than a handful of desirable bands hooks into my opportunity cost appreciation. And the lineup for this festival was a doozy for overseas acts Down Under.
My prospective schedule for the day:
Actual happenings:
Read the rest of this entry »
Working with the Seam Framework and RichFaces for over a year, I found their respective components work only for the simplest of HTML forms. Nested AJAX views cause all sorts of difficulties once you get into EL expressions for iterated components (e.g., from <rich:datatable>) within <f:subview>-named scopes that require usage of rich:element.
Read the rest of this entry »
Even after I move to Australia, Rogers manages to get my ire. Emailed below to their support, likelihood of a proper response…?
My most recent wireless invoice contains billing detail:
Oct 02 Early Cancellation Fee 500MB Data Option 100.00
I was under contract with Rogers wireless May 2008 to May 2011. I purchased an unlocked iPhone4 direct from apple.com and activated Mt. Sinai’s iPhone corporate plan in September 2010. At no point was another time-sensitive service agreement signed or even communicated to me by a CSR.
On top of this assumed contract, the most current billing cycle ends Oct 7 yet this ETF appears Oct 2 while I’m also being invoiced for another period into November due to Rogers’ draconian 30 day notice for cancellation policy. If the iPhone plan is for one year, the cancellation date would fall outside my service time.
When I called in to cancel, at no point did the retentions rep indicate an iPhone plan ETF of $100 would be applied. And I’m traveling in Australia now and won’t be calling again using international long distance to clear this up.
I sent the below letter to TSN’s Audience Relations, where relations means you get an auto-reply indicating a cog will read the message but not personally respond. If there’s enough volume on a topic, someone in power at Bell Media may hear about it at the end of the day so you may be better off calling them at 416-384-7660. Also, come to my club’s AFL Grand Final party in Toronto. We’ll figure out a way to show it on TV whether TSN is involved or not.
I’ve come to learn that Bell Media doesn’t plan to air the AFL Grand Final on TSN2, instead showing Rugby World Cup games of Australia v Russia and France v Tonga. The March 18th press release stated,
“The top two teams face off in the AFL GRAND FINAL – the Aussie Super Bowl – airing live on TSN2.”
Football clubs across the country have planned Grand Final parties with the expectation of a live broadcast, just as many have done for more than five years now when it was once annually broadcast by Setanta Sport. The AFL’s party tracker has 11 registered events in Canada that will have hundreds of attendees each.
Last minute scheduling news of a tape-delayed broadcast on TSN this Saturday has lead to a non-announcement through social media of it now being streamed live on tsn.ca. Fans of AFL in Canada already had an ad-free web site to stream the Grand Final and such a haphazard handling of the AFL license will tarnish good will toward the TSN brand while also diminishing the opportunity to attract future viewership to the sport.
At this point, the best case scenario would be for Bell Media to offer a pay-per-view option to all the pubs holding event nights this Friday. AFL Grand Final parties have a social atmosphere akin to Canada’s run at Olympic hockey gold and that setting certainly isn’t attained hunched over a laptop.
I thought about posting this on Craigslist but it’s seriously not even worth the effort meeting people to off-loading such junk. To storage the lot goes!
Also four CaseLogic flip books for storing CD/DVDs. 30x, 100x, and two 200x.
Are landlines relevant anymore? No. But I have a GE 26920GE2 cordless phone.
Yes, I made this ridiculous cookie-within-a-cookie for our Monday dessert club at work. The recipe from Picky Palate indicates it’s portioned for two dozen cookies but I only got a baker’s dozen out of it. I guess I had a biggie ice cream scoop? Either way, it’s time we try a cookie-within-a-cookie-within-a-cookie and just wait for the kick.
When using RichFaces’ <a4j:support> to reRender components in the view upon an event, I’ve come across an XML parsing error in the AJAX response which prevents the refresh from completing. The AJAX progress indicator icon continues into infinity despite the request completing its cycle, albeit in an error state. Firebug’s panel indicates the issue is “duplicate attribute”, where I track down the id=”” attribute as being added twice to at least one of the components being redrawn.
Read the rest of this entry »
Swear I’ve been meaning to post more on this blog but every item is in a scattered draft form. It doesn’t help that last week my 1TB Seagate hard drive decided to stop powering up so I had to rebuild my computer while losing over a year’s worth of music along with 10 hours+ of Doom editing which was the tedious details-oriented sort. Otherwise, I tend to keep all my files online somewhere so the incident wasn’t as disastrous as it’s been in the past. But I’m going to try finding a circuit board to hopefully repair the hard drive because my meticulously organized music collection is a labour of love (and I cry myself to sleep at night.)
Ingredients:
1/2 cup butter or margarine
1 1/2 cup graham cracker crumbs
1 cup semi-sweet chocolate chips
1 1/2 cup sweetened coconut flakes
1 can Eagle Brand milk
Melt butter in a 13″x9″x2″ pan in the oven while it pre-heats at 350°F. Sprinkle Graham cracker crumbs on top of butter, add chocolate chips, coconut, then pour Eagle Brand milk evenly on top for a layered effect. Bake for 25 minutes or until lightly brown. Let cool before cutting.
Makes 3 dozen.
Like last year, I tracked my plate appearances for my softball team.
Spring: Multitards, 7-1 for 1st
May 11 won 23-17, 7-7, 2B, 3B
May 18 won 21-8, 5-5, 2B, HR
May 25 won 33-14, didn’t play (sick)
June 1 won 28-13, didn’t play (TTC subway delay due to St. George Stn. police investigation)
June 8 lost 21-8, 2-5 (warning track FO with the bases juiced was disappointing)
June 15 won 15-14, didn’t play (Thrice concert!)
June 12 won 8-7, 1-2, SF
June 29 won 16-15, 5-5, 3B (pitched my teammate the hit that was the winning RBI in top 8th)
AVG 0.833 (20-24), SLG 1.208
I didn’t end up playing the summer session because games were Monday nights which interfered with my Aussie Rules Football team’s training at Cherry Beach. Then for autumn, our team (named Catch the Taste!) forfeited half our games due to lack of numbers and for the other half I was either out of town or too damned hungover to make it. Mid-afternoon Sunday leagues played in seasonally-deteriorating weather are just not a cool idea. But I did fill in a for a couple friends with spot starts:
June 13 for Kieno’s team, won 10-9, 2-3, 2B, GW 1B
June 17 for Zeff’s team, lost 15-9, 3-4, 2B
Aug 8 for Kieno’s team, lost 16-6, 1-3, BB
My commitment to hard hit grounders between SS/3B or liners to shallow outfield paid dividends in keeping my on-base percentage on the up-and-up. Here’s to hoping a free evening schedule in 2011 will allow me to keep this momentum going against my age’s grain.
Creating TestNG casesI was having issues creating proper data sets for our test suite since the data model is so complex. I was manually instantiation each class’ entity, populating the object, persisting, then committing the transaction manually. It became obscene for certain objects that referenced up to ten different classes with multiple constraints each. We already have SQL scripts to pre-populate our local JBoss environment so it was a possible solution to use that same data for our tests. I posed the question of a better solution on the interweb and it was suggested I try another DBMS that supports DB2 used by our site and also operates in-memory to increase speed. We already were using Hypersonic but it doesn’t support SQL import from DB2. So I tried H2, using this JDBC connect string to enable the DB2 compatibility mode and trigger our create + populate scripts on the test suite’s startup.
jdbc:h2:mem:localDB;MODE=DB2;INIT=RUNSCRIPT FROM ‘~/workspace/database/create.sql’\;RUNSCRIPT FROM ‘~/workspace/database/populate.sql’
As the H2 features page indicates, “for certain features, this database can emulate the behavior of specific databases. Not all features or differences of those databases are implemented.” Enter what I found out was in fact DB2 incompatibility mode.
|
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 |
CREATE TABLE MAIL_RECEIPT ( MAIL_RECEIPT_KEY INTEGER NOT NULL, DATE TIMESTAMP NOT NULL, USER VARCHAR(255) NOT NULL, CONTENT VARCHAR(32) NOT NULL, FROM VARCHAR(255) NOT NULL, TO VARCHAR(1000), CC VARCHAR(1000), BCC VARCHAR(1000), SUBJECT VARCHAR(255) NOT NULL, BODY VARCHAR(2000) NOT NULL, PRIMARY KEY (MAIL_RECEIPT_KEY) ); |
org.h2.jdbc.JdbcSQLException: Syntax error in SQL statement
DB2 allows FROM as a column name. H2 does not. And no, I didn’t originally write this schema so don’t blame me for its poor naming.
|
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 |
CREATE VIEW REPORT_CAGE_MOVEMENT AS SELECT CURRENT_ROOM.ROOM_NAME AS CURRENT_ROOM_NAME, CURRENT_ROOM.ROOM_ID AS CURRENT_ROOM_ID, ANIMAL_PROTOCOL.PROTOCOL_NUMBER, PERSON.LAST_NAME, PERSON.FIRST_NAME, CAGE.BARCODE_VALUE, MOVE_FROM.CAGE_ID AS CAGE_ID, MOVE_FROM.ROOM_ID AS FROM_ROOM_ID, MOVE_TO.ROOM_ID AS TO_ROOM_ID, MOVE_FROM.TO_DATE AS EXIT_DATE, MOVE_TO.FROM_DATE AS ENTRY_DATE, FROM_ROOM.ROOM_NAME AS FROM_ROOM_NAME, TO_ROOM.ROOM_NAME AS TO_ROOM_NAME FROM CAGE_MOVEMENT AS MOVE_FROM INNER JOIN CAGE_MOVEMENT AS MOVE_TO ON MOVE_FROM.CAGE_ID = MOVE_TO.CAGE_ID AND MOVE_FROM.TO_DATE + 1 DAY = MOVE_TO.FROM_DATE INNER JOIN ROOM AS FROM_ROOM ON MOVE_FROM.ROOM_ID = FROM_ROOM.ROOM_ID INNER JOIN ROOM AS TO_ROOM ON MOVE_TO.ROOM_ID = TO_ROOM.ROOM_ID INNER JOIN CAGE ON MOVE_FROM.CAGE_ID = CAGE.CAGE_ID INNER JOIN PERSON ON CAGE.CAGE_OWNER_KEY = PERSON.PERSON_KEY INNER JOIN ANIMAL_PROTOCOL ON CAGE.ANIMAL_PROTOCOL_ID = ANIMAL_PROTOCOL.ANIMAL_PROTOCOL_ID INNER JOIN ROOM AS CURRENT_ROOM ON CAGE.ROOM_ID = CURRENT_ROOM.ROOM_ID |
org.h2.jdbc.JdbcSQLException: Column “CURRENT_ROOM.ROOM_NAME” not found
Aliases defined in an INNER JOIN clause cannot be referenced in the SELECT expression (isn’t that how Hibernate builds all its join queries?)
|
1 2 3 |
GRANT CONNECT ON DATABASE TO USER SOMEUSER; |
org.h2.jdbc.JdbcSQLException: Table “DATABASE” not found
Uh…
|
1 2 3 4 5 6 7 8 9 |
CREATE TRIGGER BI_STRG_ALLCTIN NO CASCADE BEFORE INSERT ON STORAGE_ALLOCATION REFERENCING NEW AS NEW_STORAGE FOR EACH ROW MODE DB2SQL WHEN (NEW_STORAGE.STORAGE_ALLOCATION_KEY IS NULL) SET (NEW_STORAGE.STORAGE_ALLOCATION_KEY) = (NEXTVAL FOR TCP_POID_SEQUENCE); |
org.h2.jdbc.JdbcSQLException: Syntax error in SQL statement “…”; expected “AFTER”
Well it looks like a DB2 trigger doesn’t need that AFTER keyword.
So we’ve switched back to Hypersonic and I’m now creating an XML-based data set using DBUnit.
The Polkaroo and I are connected. A few years ago I subletted a flat in Guelph for the summer; a four bedroom apartment to myself. (Un)fortunately one month in the renters found more subletters to fill the remaining rooms. One guy had already settled in (who we coined Creeperface for various reasons) when two girls started moving their own furniture and belongings into the place. It happened to be a crunch time week at work so I was around the apartment intermittently. During three different moving trips, they got to the place just after I had left. The mystery of my actual existence would be confronted by the question, “we missed him again?”
Hence I earned the moniker of the Polkaroo. It’s a reference to the 80s Canadian kid’s show Polka Dot Door, where a whack-looking costumed characters comes on screen for minutes at a time only repeating the line “polkaroo” over and over. The show’s host is off-screen at the same time and the Polkaroo always leaves before the host returns. However, I had no clue what my new roomies were talking about since I grew up in Nova Scotia and this show aired on TVO, an Ontario-only channel. I’ve come to accept my fate. The above photo is my birthday cake they made in summer 2008, decorated by the polka dot door. (Although the enjoyment of its sugar goodness was thwarted when we got locked out of our apartment after my birthday dinner due to the cheaply-made doorknob suddenly becoming the kryptonite to our keys’ Superman. Richmond Property Ltd’s handling of the situation was also so messed up that we ended up staying in a hotel room that night with no comp. Stay away from pseudo-student housing.)
This nickname managed to stick for the whole summer, spreading to their friends I met along the way. So there’s about a dozen people out there that call me Polkaroo. I’m not sure if that’s better than more than 50 people in Toronto using my middle name (that they believe is my first) since I was introduced to my footy team as such by my current housemate that declared he didn’t like “Derek”.
Somehow the no longer airing show is all over social networking web sites, including Polkaroo on Tumblr, where they’re running a contest for free t-shirts to promote the brand. I don’t know if I could bring myself to wear one in public, but I rep Threadless the odd time which isn’t far off on the cred scale.
Since I prefer to dilute my online brand, you can now also follow me on Tumblr. Longer items that are less microbloggy will still be appearing on this domain.
Cut potatoes int bite size pieces and place in 2 quart (8 litre) casserole dish.
Mix remaining ingredients in saucepan. Heat and stir until it boils and thickens. Pour over potatoes. Cover and bake in 350°F oven for 20 minutes. Remove cover for last few minutes if desired. Serves 8.
For some reason Seam Framework doesn’t include form validation to ensure at least one checkbox is selected out of a group of checkboxes. Like a radio button, but more than one may be selected. One custom code sample was dodgy because validation takes place in JSF lifecycle before the model is updated, so you must wait for the component tree to be built before such a validation can take place.
You may use jQuery validation but what if the user has JavaScript disabled in their web browser? Or the JS payload hasn’t yet arrived? You still require backend assertion for the correct entity state.
The solution I found wasn’t optimal but hey, I’m not the only one wanting this simple form functionality.
<mytaglib:atLeastOneValidator> tag/resources/WEB-INF/web.xml
|
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 |
<?xml version="1.0"?> <web-app xmlns="http://java.sun.com/xml/ns/javaee" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xsi:schemaLocation="http://java.sun.com/xml/ns/javaee http://java.sun.com/xml/ns/javaee/web-app_2_5.xsd" version="2.5"> . . . <context-param> <param-name>facelets.LIBRARIES</param-name> <param-value>/WEB-INF/compositions.taglib.xml</param-value> </context-param> </web-app> |
/resources/WEB-INF/compositions.taglib.xml
|
1 2 3 4 5 6 7 8 9 10 11 12 13 |
<?xml version="1.0"?> <!DOCTYPE facelet-taglib PUBLIC "-//Sun Microsystems, Inc.//DTD Facelet Taglib 1.0//EN" "facelet-taglib_1_0.dtd"> <facelet-taglib> <namespace>http://mytaglib.com/jsf</namespace> <tag> <tag-name>atLeastOneValidator</tag-name> <validator> <validator-id>atLeastOneValidator</validator-id> </validator> </tag> </facelet-taglib> |
AtLeastOneValidator validator specific for that tag/src/main/validator/AtLeastOneValidator.java
|
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 |
package yourpackage.validator; import java.io.Serializable; import java.util.ArrayList; import java.util.HashMap; import java.util.List; import java.util.Map; import javax.faces.application.FacesMessage; import javax.faces.component.UIComponent; import javax.faces.component.html.HtmlSelectBooleanCheckbox; import javax.faces.context.FacesContext; import javax.faces.validator.ValidatorException; import org.jboss.seam.ScopeType; import org.jboss.seam.annotations.Name; import org.jboss.seam.annotations.Scope; import org.jboss.seam.annotations.faces.Validator; import org.jboss.seam.annotations.intercept.BypassInterceptors; import org.jboss.seam.ui.component.html.HtmlLabel; @Name("atLeastOneValidator") @Validator @BypassInterceptors @Scope(ScopeType.CONVERSATION) public class AtLeastOneValidator implements javax.faces.validator.Validator, Serializable { private static final long serialVersionUID = -4249428843435574402L; private static Map<String, Map<String, Boolean>> formCheckboxes; public void validate(FacesContext context, UIComponent component, Object value) throws ValidatorException { List<HtmlSelectBooleanCheckbox> checkboxes = new ArrayList<HtmlSelectBooleanCheckbox>(); String groupWith = ""; UIComponent rootComponent = FacesContext.getCurrentInstance().getViewRoot(); if (!(component instanceof HtmlSelectBooleanCheckbox)) { throw new ValidatorException(createErrorMessage("atLeastOneValidator can only be used on HtmlSelectBooleanCheckbox components.")); } if (component.getAttributes().get("groupWith") != null) { groupWith = (String)component.getAttributes().get("groupWith"); } if (formCheckboxes == null) { formCheckboxes = new HashMap<String, Map<String, Boolean>>(); } if (formCheckboxes.get(groupWith) == null) { formCheckboxes.put(groupWith, new HashMap<String, Boolean>()); } // Store this component's value, queuing to be checked at the last // checkbox in the group to see of any component has the value // set to "true" formCheckboxes.get(groupWith).put((String)component.getAttributes().get("id"), (Boolean)value); // retrieve all checkboxes under this group in the component tree getCheckboxes(rootComponent, checkboxes, groupWith); // Last checkbox in the component tree of this group, so // now check whether at least one is checked. if (component.equals(checkboxes.get(checkboxes.size()-1))) { boolean atLeastOneChecked = false; String styleClass = ""; for (String componentId : formCheckboxes.get(groupWith).keySet()) { if (formCheckboxes.get(groupWith).get(componentId)) { atLeastOneChecked = true; break; } } // highlight or unmark checkboxes/labels if (!atLeastOneChecked) { styleClass = "required"; } for (HtmlSelectBooleanCheckbox checkbox : checkboxes) { if (checkbox.getParent() instanceof HtmlLabel) { ((HtmlLabel)checkbox.getParent()).setStyleClass(styleClass); } else { checkbox.setStyleClass(styleClass); } } if (!atLeastOneChecked) { throw new ValidatorException(createErrorMessage("At least one " + ((groupWith == null || groupWith.length() == 0) ? "" : "\"" + groupWith + "\" ") + "checkbox must be selected.")); } } } /** * Recursively traverse a component tree to retrieve all the * HtmlSelectBooleanCheckbox objects containing a specific groupWith="" * attribute. * * @param component Initial invocation should be the root component. * @param checkboxes Initial invocation should be instantiated blank list. * @param groupWith Name/label given to the group of checkboxes. null * or a blank string returns ALL checkboxes in the tree. * @return List of child checkboxes in a component tree. */ protected List<HtmlSelectBooleanCheckbox> getCheckboxes(UIComponent component, List<HtmlSelectBooleanCheckbox> checkboxes, String groupWith) { if (component != null) { for (UIComponent childComponent : component.getChildren()) { if (childComponent instanceof HtmlSelectBooleanCheckbox) { if ((groupWith == null || groupWith.length() == 0) || groupWith.equals(childComponent.getAttributes().get("groupWith"))) { checkboxes.add((HtmlSelectBooleanCheckbox)childComponent); } } else if (childComponent.getChildCount() > 0) { getCheckboxes(childComponent, checkboxes, groupWith); } } } return checkboxes; } private FacesMessage createErrorMessage(String s) { FacesMessage message = new FacesMessage(); message.setDetail(s); message.setSummary(s); message.setSeverity(FacesMessage.SEVERITY_ERROR); return message; } } |
<mytaglib:atLeastOneValidator>/view/atLeastOneValidatorTest.xhtml
|
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 |
<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd"> <ui:composition xmlns="http://www.w3.org/1999/xhtml" xmlns:s="http://jboss.com/products/seam/taglib" xmlns:ui="http://java.sun.com/jsf/facelets" xmlns:f="http://java.sun.com/jsf/core" xmlns:h="http://java.sun.com/jsf/html" xmlns:rich="http://richfaces.org/rich" xmlns:mytaglib="http://mytaglib.com/jsf"> <h:form> <h:messages /> . . . <s:label> <h:selectBooleanCheckbox value="male" groupWith="Sex"> <mytaglib:atLeastOneValidator /> </h:selectBooleanCheckbox> <h:outputText value="Duder" /> </s:label> <s:label> <h:selectBooleanCheckbox value="female" groupWith="Sex"> <mytaglib:atLeastOneValidator /> </h:selectBooleanCheckbox> <h:outputText value="Dudette" /> </s:label> <s:label> <h:selectBooleanCheckbox value="unspecified" groupWith="Sex"> <mytaglib:atLeastOneValidator /> </h:selectBooleanCheckbox> <h:outputText value="Unspecified" /> </s:label> <s:label> <h:selectBooleanCheckbox value="other" groupWith="Sex"> <mytaglib:atLeastOneValidator /> </h:selectBooleanCheckbox> <h:outputText value="Other" /> </s:label> <s:label> <h:selectBooleanCheckbox value="unknown" groupWith="Sex"> <mytaglib:atLeastOneValidator /> </h:selectBooleanCheckbox> <h:outputText value="???" /> </s:label> . . . </h:form> </ui:composition> |
DEBUG [SequenceGenerator]Sequence identifier generated: 1025
DEBUG [AbstractBatcher]about to close PreparedStatement (open PreparedStatements: 1, globally: 1)
DEBUG [ConnectionManager]aggressively releasing JDBC connection
DEBUG [ConnectionManager]releasing JDBC connection [ (open PreparedStatements: 0, globally: 0) (open ResultSets: 0, globally: 0)]
DEBUG [AbstractSaveEventListener]generated identifier: 1025, using strategy: org.hibernate.id.SequenceGenerator
DEBUG [AbstractSaveEventListener]generated identifier: 1025, using strategy: org.hibernate.id.ForeignGenerator
DEBUG [AbstractFlushingEventListener]processing flush-time cascades
DEBUG [AbstractEntityManagerImpl]mark transaction for rollback
It’s a one-to-one foreign key generated by a DB2 sequence to persist multiple entities in one transaction. It seems to be created and assigned to the two entities mapped to each other but I have no clue why the JPA/Hibernate transaction decides it’s a no go as all the required fields are assigned in the objects being persisted. I’ve been working on this Seam Framework wizard for a week and this is an example of roadblocks in the dozens that have taken me to varying edges of sanity when it comes to troubleshooting this project.
I can only take solace in knowing once these niggles are straightened out, the rest of the system will be easy to implement. But the technological learning curve and unseen issues derived from requirement complexities make me believe Seam is a terrible framework solution. Absolutely terrible.
Update July 2: This error was occurring because a RunTimeException was thrown in the Hibernate library which wasn’t displaying in the server’s log or on the JBoss stack trace because the entrance method was invoked using EL. Wrapping my persist() call with a try {} containing a generic Exception catch revealed it was a constraint violation occurring on a value set as a Hibernate TrueFalseType (char(1) with ‘T’ or ‘F’ values) when it should’ve been a YesNoType (char(1) with ‘Y’ or ‘N’ values).
My expectations of this reunion were pretty low but wow, Cornell still has the vocal chops and this performance highlights how Kim Thayil has been sorely missed for more than a decade.
You can’t. But you can use <h:outputText> for anonymous visitors (read-only) and <rich:inplaceInput> for known logged-in users (write-enabled.) Customize the inplaceInputTest.xhtml s:hasRole() parameter to give the proper boolean value on whether the logged-in user is an administrator or not of the field bound to the input.
<rich:inplaceInput>/resources/WEB-INF/web.xml
|
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 |
<?xml version="1.0"?> <web-app xmlns="http://java.sun.com/xml/ns/javaee" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xsi:schemaLocation="http://java.sun.com/xml/ns/javaee http://java.sun.com/xml/ns/javaee/web-app_2_5.xsd" version="2.5"> . . . <context-param> <param-name>facelets.LIBRARIES</param-name> <param-value>/WEB-INF/compositions.taglib.xml</param-value> </context-param> </web-app> |
/resources/WEB-INF/compositions.taglib.xml
|
1 2 3 4 5 6 7 8 9 10 11 |
<?xml version="1.0"?> <!DOCTYPE facelet-taglib PUBLIC "-//Sun Microsystems, Inc.//DTD Facelet Taglib 1.0//EN" "facelet-taglib_1_0.dtd"> <facelet-taglib> <namespace>http://mytaglib.com/jsf</namespace> <tag> <tag-name>inplaceInput</tag-name> <source>../compositions/inplaceInput.xhtml</source> </tag> </facelet-taglib> |
Below, evaluation of the rendered="" attribute decides which of the two elements appear at one time.
/view/compositions/inplaceInput.xhtml
|
1 2 3 4 5 6 7 8 9 10 11 |
<ui:composition xmlns="http://www.w3.org/1999/xhtml" xmlns:ui="http://java.sun.com/jsf/facelets" xmlns:h="http://java.sun.com/jsf/html" xmlns:rich="http://richfaces.org/rich"> <rich:inplaceInput value="#{value}" showControls="true" editEvent="ondblclick" layout="block" rendered="#{isManager}" /> <h:outputText value="#{value}" rendered="#{not isManager}" /> </ui:composition> |
<mytaglib:inplaceInput> Usage/view/inplaceInputTest.xhtml
|
1 2 3 4 5 6 7 8 9 10 11 12 13 14 |
<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd"> <ui:composition xmlns="http://www.w3.org/1999/xhtml" xmlns:s="http://jboss.com/products/seam/taglib" xmlns:ui="http://java.sun.com/jsf/facelets" xmlns:f="http://java.sun.com/jsf/core" xmlns:h="http://java.sun.com/jsf/html" xmlns:rich="http://richfaces.org/rich" xmlns:mytaglib="http://mytaglib.com/jsf"> <mytaglib:inplaceInput value="Example text" isManager="#{s:hasRole('module_manager')}" /> </ui:composition> |
Yeah, it’s all happening. Migrating a CMS to Java’s Seam framework using an existing DB2 schema/dataset, the Lucene search indexes on a few tables caused DB2 error SQL1584N. The bizarre part was it was only happening locally on my machine, but not for two others at work, who each have the same amount of RAM. It may have been because I was on Ubuntu and the other devs use Red Hat Enterprise Linux.
System temporary tablespace with page size of at least pagesize could not be found.
Explanation:
A system temporary table space was required to process the statement. There was no system temporary table space available that had a page size of pagesize or larger.The statement cannot be processed.
User Response:
Create a system temporary table space with a page size of at least pagesize.sqlcode: -1584
sqlstate: 57055
Following the instructions for ensuring system temporary table spaces page sizes meet requirements, I ran DB2 query:
|
1 2 3 |
LIST TABLESPACES SHOW DETAIL |
…checking for the entry with Type = System managed space, Contents = System Temporary data, I confirmed the page size used was 32k, which is the highest possible with DB2.
After googling a bit, it seemed the solution was to increase the bufferpool size used by the tablespace giving me this error. The instructions for adjusting the bufferpool are to simply run DB2 query:
|
1 2 3 |
SELECT * FROM syscat.bufferpools |
Confirm the bufferpool size, then increase the number of pages available until the Lucene indexing doesn’t shit the bed. e.g.,
|
1 2 3 |
ALTER BUFFERPOOL ibmdefaultbp SIZE 2000 |
During this process, you may get the error:
SQL20169W The buffer pool is not started.
sqlcode: +20169
sqlstate: 01654
This just means the change won’t take effect until DB2 is restarted, so execute:
|
1 2 3 4 5 |
db2 force applications all db2stop db2start |
Now, if you have the Seam observer Java annotation setup for Lucene indexing, this DB2 error will begone on the next deploy. And yes, this whole post makes me yearn for a non-Java project started from scratch.
