Transcripts in Spark do not show recent messages

That’s me, who occasionally scrolls years history I’m fine with month pagination. Also i’m fine with the current search.

There was an error in the patch if you try to open the transcript in the chat window, this is fixed in the new patch.

So we stay by paging with last week/month? (4580 Bytes)

If nobody replies, than i say, yes, we can go with last week/month.

no complaints from me!

any comments on the patch provided by Tim? Should he commit to the SVN?

The last patch is not clear to me. How it divides the history? I now see 81 pages in stead of ~20. But i see the same < and > buttons. Is it first showing one week and if i click > it shows the last month and so on? Maybe we should have more buttons to make it clear? Or not?

Ok, i’d worked a bit on the patch.

Now you can select time periods of one week, three weeks and one month.

I guess, the look is not yet final and i’m still testing a with old transcripts and different timezones, maybe that could cause bugs.

But here is the patch and i’d like to get some suggestions (8660 Bytes)

Can you explain better how every setting works? because if i select week.three it shows every message on the new page (1400 pages).

“week.three” should be named “Three Weeks” sounds like the default spark_i18n.propies is missing the keys.

Maybe i should put an info box to the elements to explain their function.

It should list the messages, grouped in a time period of three weeks per page, for example “Week 1 to 3 of 2011 on one page, week 4 to 6 of 2011 on the next” and so on. Works fine with my transcript.

The “” option will group all messages that where send at the same month and the “” all that where send at the same week.

Strange that your messages are shown one per page… do you use test data or a real transcript?

I use real transcript with 4 years history. One week looksok. Three weeks as i said separates every message on a different page. One month setting has only 4 pages… Well probably i have talked with that contact only once per year But one of that 4 pages shows more months.

Tim, Walter, any update on this patch? Maybe we can have a simple patch reversing the 5000 messages limit to show latest messages, not the first ones, until the whole patch is complete? We are going to test a few clients to replace our old client. The main complaint about the old client was about not saving the history, so with current bug it would be hard to sell Spark to my bosses

We’re testing the patch in our client for some time now and, except there seems to be a bug in the xml parser, the transcripts working fine.

I’m going to upload the patch the next days.

Great We are planning to test with a pilot group next week, so maybe i will have a better version to provide for testers.

Hate to whine, but i suppose i won’t be able to get updated build for tomorrow pilot project start? Any new when can i have my hands on the updated build in Bamboo?

Hey, here is the patch for the transcript plugin we’re using in our client.

Feedback is, as always, welcome. (9557 Bytes)

Once again bumping this We are deploying Spark company wide this and next week. We probably won’t hist the 5000 messages limit anytime soon, but would want to have new history gui soon, so users won;t have to re-learn using it. Btw, check the comment on history orientation on the ticket. It is opposite to the way history is displaying in the chat window itself (and from the old behavior). Would be more logical to have old messages on top, or at least to have a sorting switch.

Newest patch committed to trunk.

Options for pagination is reduced to month/year/all per page and, on default, date is sorted ascending.

The option to switch sorting is located in the spark chat properties.


Any news on when 2.7.0. will be out? Or at least a beta? I want to try this out asap (and no, I can’t compile to save my life).

Thanks Tim.

Here’s your “beta” l4j/ first link

Thanks Wr00t, got it installed to test out before deployment.

I thought this was coming out with 2.7.0 like it’s stated in the ticket (SPARK-1407) not implemented in the current version. Not that I’m complaining, this is much better.

Anyway, thanks guys