So, none of us like it, but should we do it? :-) I decided to ping Mike Keith (co-lead of the spec) about this. Here's what he said...
You are correct. Lobs are to be treated like other Basic attributes and should default to EAGER. Your colleague is somewhat justified in that lobs can be costly to load, however, once we put lobs under the category of basic mappings we needed to be consistent and have them use the same defaults. If they are expected to be both large and not referenced often then @Basic(fetch=LAZY) should be used. Note that if one of these is not true then you probably do want them to be eagerly loaded. -Mike So, it sounds like we need to change the default behavior to be EAGER. I'll add the necessary remarks to the JIRA Issue and pursue the proper fix. Thanks for the discussion! Kevin On 7/9/07, Craig L Russell <[EMAIL PROTECTED]> wrote:
Ok, I don't like it either. Craig On Jul 9, 2007, at 9:34 AM, Patrick Linskey wrote: > Hi, > > I think that 2.1.6 makes the behavior relatively clear: > > "If the type of the field or property is one of the following, it is > mapped in the same way as it > would if it were annotated as Basic:" > > That definitely implies that a field with no annotation should be > mapped as if the Basic annotation were present. > > However, I still don't like it. > > -Patrick > > On 7/9/07, Craig L Russell <[EMAIL PROTECTED]> wrote: >> As I've already said, the spec is pretty opaque. Before assuming that >> a missing @Basic annotation still has a default fetch=EAGER element, >> I'd want to see a spec clarification. Clearly there is no TCK test >> possible for this... >> >> Craig >> >> On Jul 9, 2007, at 9:17 AM, Kevin Sutter wrote: >> >> > :-) The only reason why I would want to eagerly fetch LOBs is to >> > satisfy >> > the intent of the spec. Since the @Basic annotation is optional >> > and the >> > default fetchType is EAGER, then I assert that the spec indicates >> > that LOBs >> > need to be fetched EAGERly. Do you read the spec differently? >> > >> > I will agree that "in practice" LOBs should not be fetched >> > EAGERly. But, we >> > need to be consistent with the spec so as not to surprise customers >> > as they >> > move from one JPA implementation to another. >> > >> > Kevin >> > >> > On 7/9/07, Patrick Linskey <[EMAIL PROTECTED]> wrote: >> >> >> >> Indeed you do... why would we want to eagerly fetch LOBs? I >> >> understand >> >> the logic if @Basic is specified, but not if it is omitted. >> >> >> >> -Patrick >> >> >> >> On 7/9/07, Kevin Sutter <[EMAIL PROTECTED]> wrote: >> >> > I have opened JIRA Issue 281 ( >> >> > https://issues.apache.org/jira/browse/OPENJPA-281) for this >> >> problem. It >> >> > seems that we have agreed that enums should be EAGER by >> >> default. I just >> >> > have to convince everybody that lobs also have to be EAGER, by >> >> default... >> >> > :-) >> >> > >> >> > Thanks, >> >> > Kevin >> >> > >> >> > On 7/8/07, Kevin Sutter <[EMAIL PROTECTED]> wrote: >> >> > > >> >> > > Craig, >> >> > > Comments below... >> >> > > >> >> > > On 7/7/07, Craig L Russell <[EMAIL PROTECTED]> wrote: >> >> > > > >> >> > > > Hi Kevin, >> >> > > > >> >> > > > If you can figure out what the spec (9.1.18 and 9.1.19) >> sez, my >> >> hat's >> >> > > > off to you. What a mudbake this is. >> >> > > >> >> > > >> >> > > Isn't that just standard order of business with these >> specs? :-) >> >> > > >> >> > > Part of the issue is the annotation definition. If @Basic is >> >> > > > specified, and the user doesn't explicitly override the >> >> fetch type, >> >> > > > it appears to our annotation processor as if the user >> specified >> >> > > > EAGER. Even if @Basic is used with @Lob, if lazy is wanted, >> >> it has >> >> to >> >> > > > be explicitly stated. >> >> > > >> >> > > >> >> > > That's how I read the spec. The first paragraph of 9.1.18 >> >> indicates >> >> that >> >> > > @Basic can be applied to any those types (enums and lob types >> >> included). >> >> > > The default fetchType is EAGER. It also states that @Basic is >> >> optional. >> >> > > So, the way I read this is that we should be doing EAGER >> >> fetching for >> >> all of >> >> > > those listed types unless explicitly told to do otherwise via >> >> the LAZY >> >> > > fetchType via an @Basic annotation. >> >> > > >> >> > > Maybe we should discuss @Lob in more detail. It isn't obvious >> >> to me >> >> > > > that @Basic can always be used and we might have an option >> >> to choose >> >> > > > a better default for the fetch behavior if @Basic >> annotation is >> >> omitted. >> >> > > >> >> > > >> >> > > It seems to me that the spec is clear on the use and >> >> expectations of >> >> > > @Basic and the default fetchType of EAGER. Unless there are >> >> other >> >> spec >> >> > > references that contradict the statements in 9.1.18... >> >> > > >> >> > > Craig >> >> > > > >> >> > > > On Jul 6, 2007, at 11:37 AM, Kevin Sutter wrote: >> >> > > > >> >> > > > > I guess the spec is a bit clearer on this than I first >> >> thought. >> >> > > > > Section >> >> > > > > 9.1.8 of the JPA spec indicates that @Basic is optional >> and >> >> applies >> >> > > > > to the >> >> > > > > following types: >> >> > > > > >> >> > > > > "..Java primitive types, wrappers of the primitive types, >> >> > > > > java.lang.String, >> >> > > > > java.math.BigInteger, >> >> > > > > java.math.BigDecimal, java.util.Date, java.util.Calendar, >> >> > > > > java.sql.Date, >> >> > > > > java.sql.Time, java.sql.Timestamp, byte[], Byte[], char[], >> >> Character >> >> > > > > [], >> >> > > > > enums, and any other type that implements Serializable." >> >> > > > > >> >> > > > > And, since the default fetch type for @Basic is EAGER, it >> >> looks >> >> > > > > like we need >> >> > > > > to do eager fetching for both @Enumerated and @Lob fields >> >> unless >> >> > > > > otherwise >> >> > > > > overridden by a LAZY fetch type (ie. @Basic(fetch=LAZY)). >> >> Agree? >> >> > > > > >> >> > > > > Kevin >> >> > > > > >> >> > > > > On 7/6/07, Craig L Russell <[EMAIL PROTECTED] > wrote: >> >> > > > >> >> >> > > > >> >> >> > > > >> On Jul 6, 2007, at 10:52 AM, Patrick Linskey wrote: >> >> > > > >> >> >> > > > >> > The spec doesn't seem to discuss it, but I think that >> >> lazy is a >> >> > > > >> pretty >> >> > > > >> > good default behavior for @Lob field types -- typically >> >> @Lob >> >> things >> >> > > > >> > are big, so you often don't want them in the default >> fetch >> >> graph. >> >> > > > >> >> >> > > > >> Enum is different, though. Enum should be eager >> fetching by >> >> default. >> >> > > > >> >> >> > > > >> Lazy fetching is optional, so we can decide what we want >> >> to do. >> >> It >> >> > > > >> seems that the existence of @Basic should not change our >> >> strategy. >> >> > > > >> And we should default to lazy fetching for Lob and eager >> >> fetching >> >> for >> >> > > > >> >> > > > >> Enum. >> >> > > > >> >> >> > > > >> Craig >> >> > > > >> > >> >> > > > >> > -Patrick >> >> > > > >> > >> >> > > > >> > On 7/6/07, Kevin Sutter <[EMAIL PROTECTED]> wrote: >> >> > > > >> >> I am finding that the supposed default action of Eager >> >> fetching is >> >> > > > >> >> not >> >> > > > >> >> happening with @Enumerated and @Lob fields. If I >> >> explicitly >> >> > > > >> >> specify the >> >> > > > >> >> @Basic annotation, then the fields are eagerly >> >> fetched. But, >> >> > > > >> >> without this >> >> > > > >> >> extraneous @Basic, these fields are lazily loaded. >> This >> >> action >> >> > > > >> >> does not >> >> > > > >> >> seem to be consistent with the spec. Nor, can I >> find any >> >> mention >> >> > > > >> >> of this >> >> > > > >> >> alternate behavior in our OpenJPA manual. Sounds like >> >> a bug >> >> to >> >> > > > >> >> me. Any >> >> > > > >> >> other insights? >> >> > > > >> >> >> >> > > > >> >> This works (eager loading kicks in): >> >> > > > >> >> >> >> > > > >> >> @Basic @Enumerated(EnumType.STRING) >> >> > > > >> >> private Gender gender; >> >> > > > >> >> >> >> > > > >> >> This does not work (lazy loading kicks in): >> >> > > > >> >> >> >> > > > >> >> @Enumerated(EnumType.STRING) >> >> > > > >> >> private Gender gender; >> >> > > > >> >> >> >> > > > >> >> I have also tried to use defaults (without any >> >> annotations), >> >> > > > >> and lazy >> >> > > > >> >> loading still kicks in: >> >> > > > >> >> >> >> > > > >> >> private Gender gender; >> >> > > > >> >> >> >> > > > >> >> Thanks, >> >> > > > >> >> Kevin >> >> > > > >> >> >> >> > > > >> > >> >> > > > >> > >> >> > > > >> > -- >> >> > > > >> > Patrick Linskey >> >> > > > >> > 202 669 5907 >> >> > > > >> >> >> > > > >> Craig Russell >> >> > > > >> Architect, Sun Java Enterprise System >> >> http://java.sun.com/products/ >> >> > > > >> jdo >> >> > > > >> 408 276-5638 mailto: [EMAIL PROTECTED] >> >> > > > >> P.S. A good JDO? O, Gasp! >> >> > > > >> >> >> > > > >> >> >> > > > >> >> >> > > > >> >> > > > Craig Russell >> >> > > > Architect, Sun Java Enterprise System >> >> http://java.sun.com/products/jdo >> >> > > > 408 276-5638 mailto:[EMAIL PROTECTED] >> >> > > > P.S. A good JDO? O, Gasp! >> >> > > > >> >> > > > >> >> > > > >> >> > > >> >> > >> >> >> >> >> >> -- >> >> Patrick Linskey >> >> 202 669 5907 >> >> >> >> Craig Russell >> Architect, Sun Java Enterprise System http://java.sun.com/products/ >> jdo >> 408 276-5638 mailto:[EMAIL PROTECTED] >> P.S. A good JDO? O, Gasp! >> >> >> > > > -- > Patrick Linskey > 202 669 5907 Craig Russell Architect, Sun Java Enterprise System http://java.sun.com/products/jdo 408 276-5638 mailto:[EMAIL PROTECTED] P.S. A good JDO? O, Gasp!
