You can subscribe to this list here.
2003 |
Jan
|
Feb
(21) |
Mar
(9) |
Apr
(1) |
May
|
Jun
|
Jul
(1) |
Aug
(3) |
Sep
(3) |
Oct
(8) |
Nov
|
Dec
|
---|---|---|---|---|---|---|---|---|---|---|---|---|
2004 |
Jan
|
Feb
(1) |
Mar
|
Apr
|
May
(1) |
Jun
|
Jul
|
Aug
|
Sep
|
Oct
|
Nov
|
Dec
|
2005 |
Jan
|
Feb
|
Mar
|
Apr
|
May
|
Jun
|
Jul
|
Aug
|
Sep
(6) |
Oct
|
Nov
|
Dec
|
From: Joseph O. <jo...@en...> - 2003-02-04 16:25:24
|
Okay. Changes due: create Channel.image, textinput, items ahead of time. Create an "insertItem" to match "addItem"... or do you think it makes sense for addItem to prepend as default behaviour? Lastly, shoudl getItems() duplicate the list (deep copy, shallow copy) or just return the actual list itself? I'll let the userbase decide. As soon as I get some feedback worth measuring, I'll commit. :) On Tue, 4 Feb 2003, Joseph Ottinger wrote: > Noted. I'll have a new version out very soon. > > On Tue, 4 Feb 2003, Joseph Shelby wrote: > > > Joe Shelby wrote: > > > 2) in addition to addItem() (which, using the List interface does an > > > append), there should be an insertItem that adds it to the front, > > > following normal RSS for blog standards. Right now that can be done by > > > doing "getItems().add(0, new Item(...))", but to me that's risky. > > > > And a risk involved just happened to me. > > > > currently, the internal List is created lazily, and the result is that if > > you call getList().add(...) before a single item is added through the Channel's > > own methods, you get a NullPointerException. > > > > Joe > > > > --------------------------------------------------------- > Joseph B. Ottinger jo...@en... > http://enigmastation.com IT Consultant > > > > ------------------------------------------------------- > This SF.NET email is sponsored by: > SourceForge Enterprise Edition + IBM + LinuxWorld = Something 2 See! > http://www.vasoftware.com > _______________________________________________ > Rsslibj-developer mailing list > Rss...@li... > https://lists.sourceforge.net/lists/listinfo/rsslibj-developer > --------------------------------------------------------- Joseph B. Ottinger jo...@en... http://enigmastation.com IT Consultant |
From: Joseph O. <jo...@en...> - 2003-02-04 16:01:52
|
Noted. I'll have a new version out very soon. On Tue, 4 Feb 2003, Joseph Shelby wrote: > Joe Shelby wrote: > > 2) in addition to addItem() (which, using the List interface does an > > append), there should be an insertItem that adds it to the front, > > following normal RSS for blog standards. Right now that can be done by > > doing "getItems().add(0, new Item(...))", but to me that's risky. > > And a risk involved just happened to me. > > currently, the internal List is created lazily, and the result is that if > you call getList().add(...) before a single item is added through the Channel's > own methods, you get a NullPointerException. > > Joe > --------------------------------------------------------- Joseph B. Ottinger jo...@en... http://enigmastation.com IT Consultant |
From: Joseph S. <acr...@io...> - 2003-02-04 16:00:05
|
Joe Shelby wrote: > 2) in addition to addItem() (which, using the List interface does an > append), there should be an insertItem that adds it to the front, > following normal RSS for blog standards. Right now that can be done by > doing "getItems().add(0, new Item(...))", but to me that's risky. And a risk involved just happened to me. currently, the internal List is created lazily, and the result is that if you call getList().add(...) before a single item is added through the Channel's own methods, you get a NullPointerException. Joe -- ---------------------------------------------------------------------- Joseph Shelby mailto:acr...@io... 45533 Stablemates Ct. http://www.io.com/~acroyear Sterling, VA 20164 (703) 404-0314 | 247-7868 Software Engineer, ISX Corporation, Arlington, VA "The people of England will permit anything...except cruelty to horses and a rise in the price of beer." -- Return of the Musketteers (1989) ---------------------------------------------------------------------- |
From: Joe S. <acr...@io...> - 2003-02-04 15:20:22
|
i'm never any good with CVS and patch, so i'll leave this to someone else to do, but here goes. most of what I'd be doing would be loading an rss file, adding stuff to it, and saving it again. To do this, I would want to make sure to save it in the same format, and add it following typical "blog" format (most-recent-first). 1) if RSSReader is used to load an RSS file, it or the Channel it produces should remember what format (092, rdf, 200) was used in the original file in order to save it to the same state. A default getFeed() would do it automatically, or "rss = channel.getFeed(channel.getOriginalFormatType());". I realize, of course, that this requires a default value to be selected for when Channel was constructed in user code instead of RSSReader, and the selection of one standard as the default will annoy those who prefer the other (aren't holy wars fun?). 2) in addition to addItem() (which, using the List interface does an append), there should be an insertItem that adds it to the front, following normal RSS for blog standards. Right now that can be done by doing "getItems().add(0, new Item(...))", but to me that's risky. From a public API perspective, its rare for exported Lists and/or Collections to actually be the contained object, rather a copy (perhaps even a copy with synchronized or unmodifiable tagged to it). There's the issue of "messing with the Channel's internals" that I don't like about it, even if it is safe in this instance because we have the source code to see that it is. Its not safe in most of the JDK's instances (java.awt.dnd, java.security.cert, javax.imageio), so users following Java's own practices may not be used to doing that. For example, if one were to modify the List returned in SpinnerListModel, no event would be fired and the data in the model would be out of synch with the spinner. The javadocs don't tell you this, one has to infer it from the phrase "Note that list is not copied, the model just stores a reference to it", actually found in setList(), not getList(). Interfaces like that tend to make the experienced Java developer more cautious about modifying collections returned by objects. Its easily seen as a violation of encapsulation. Joe |