489898 - Some calendar entries not showing or findable by searching

Status: RESOLVED INCOMPLETE

(Calendar :: Calendar Frontend, defect)

Description

[David Stein]

User-Agent:       Mozilla/5.0 (Windows; U; Windows NT 6.0; en-US; rv:1.9.0.3) Gecko/2008092417 Firefox/3.0.3
Build Identifier: Mozilla/5.0 (Windows; U; Windows NT 6.0; en-US; rv:1.8.1.18pre) Gecko/20080917 Sunbird/0.9

I've got an .ICS that's exhibiting a strange problem: several of my recent entries are not visible. I've created these new entries a few times, and they remain visible as long as I have the instance open; but if I close Sunbird and start a new instance pointed at the calendar, the latest entries don't show up in the calendar view. Also, I can't find them via search. This remains true even though the entries are clearly visible in the ICS.

Here's one recent entry that is visible in the ICS:

BEGIN:VEVENT
CREATED:20090422T200121Z
LAST-MODIFIED:20090422T200308Z
DTSTAMP:20090422T200208Z
UID:2522cc7e-0536-49a6-b87e-ca36c7fe6b33
SUMMARY:CSU Intro to Swing (Andy Lammers)
RECURRENCE-ID;TZID=America/New_York:20090312T184500
RRULE:FREQ=WEEKLY;COUNT=5;INTERVAL=1;BYDAY=TH
DTSTART;TZID=America/New_York:20090402T184500
DTEND;TZID=America/New_York:20090402T194500
LOCATION:Cleveland State University Rec Center
X-MOZ-GENERATION:5
END:VEVENT

I don't see any obvious problem with this entry. However, none of the entries appear in the calendar view, and a title search doesn't find it (even if I search for the exact and full title.)

I can imagine that this problem might involve one or two aspects of my particular ICS:

(1) I'm using an unusually large ICS - 784kb of data. (Is that a lot? I don't know... but I do know that it takes Sunbird about eight seconds to start up and parse the ICS!)

(2) The entries that I'm noticing are problematic are recurring entries. There's nothing else peculiar about the entry, though: no strange recurrence pattern (just weekly for (x) instances), no file attachments or unusual use of fields, etc.


Reproducible: Always

Steps to Reproduce:
1. Load my ICS.
2. View the contents.

Actual Results:  
Expected calendar entries don't appear.

Expected Results:  
Expected calendar entries *should* appear.

n/a

Comment 1

[Pete Riley]

I don't know if it matters, but in my ics files I never have both an RRULE and a RECURRENCE-ID inside a single BEGIN:VEVENT...END:VEVENT section.

Comment 2

[Philipp Preißer-Kewisch [:Fallen][☁️📆]]

Strange enough, but the file validates using the iCalendar Validator, but: having a RRULE together with e a RECURRENCE-ID would mean that the specific occurrence of a series itself recurrs. We don't support this and I don't think the spec really covers this.

I'm not sure if this event doesn't go through our libical parser or if it fails later on, but in the long run, I think we should decide if we'd rather listen to the RRULE or the RECURRENCE-ID. Maybe we can do this depending on if there is a master item that belongs to the RECURRENCE-ID.

Comment 3

[David Stein]

Hmm. This entry was created solely within Sunbird (not imported from anywhere, hand-edited, etc.), so I presume that Sunbird's UI has a latent bug here.

Quick update on how I created this entry - this may lead to a diagnosis. First, I created a single entry with a weekly recurrence of five weeks. Then, I edited the very first entry (only) to change the title (specifically, to remove the name of the instructor, since I was stepping in to teach the first class of the course.)

I'm guessing that there might be a bug relating to editing the first entry of a recurring item.

However - I don't think that the other recent entries that aren't showing up involved this second editing step... I think I just entered them in a completely normal manner. I'll check the ICS this week to see if I can identify and post another recent entry that's not appearing.

- David Stein

Comment 4

[Philipp Preißer-Kewisch [:Fallen][☁️📆]]

Does this still happen with the latest 1.0b2pre nightlies?

Comment 5

[Wayne Mery (:wsmwk)]

no response to questions in previous comment, so => incomplete.
if you still see the problem in the newest version, please comment in the bug.