Closed
Bug 122386
Opened 24 years ago
Closed 24 years ago
when syncing newsgroup for offline, should remove expired/cancelled articles from db
Categories
(MailNews Core :: Networking: NNTP, defect)
MailNews Core
Networking: NNTP
Tracking
(Not tracked)
VERIFIED
FIXED
People
(Reporter: Bienvenu, Assigned: Bienvenu)
Details
Attachments
(1 file)
|
854 bytes,
patch
|
naving
:
review+
|
Details | Diff | Splinter Review |
When syncing newsgroups for offline, we should remove expired/cancelled articles
from db, so that the next time we sync that newsgroup, we don't try to download
the body of the expired articles again. I developed the upcoming patch with seth
over AIM so I just need an r=.
| Assignee | ||
Comment 1•24 years ago
|
||
Comment 2•24 years ago
|
||
Comment on attachment 66884 [details] [diff] [review]
proposed fix
r=naving
Attachment #66884 -
Flags: review+
| Assignee | ||
Comment 3•24 years ago
|
||
fix checked in.
Status: NEW → RESOLVED
Closed: 24 years ago
Resolution: --- → FIXED
QA Contact: stephend → gchan
David,
I am trying to verify this bug. I noticed I had about 85 total
mesgs in a newsgrp. 57 of them were expired articles. When I downloaded
that newsgroup, it downloaded only my new/unread mesgs and as result
the expired articles are no longer viewable in the thread pane and
now the mesg count is like 25.
Is this the correct result?
Also I was trying to read a news protocol log and trying to figure
out whether it worked as well. I can't 'interpret' the results.
What should I be looking for? thnks in advance.
small snipet of my log:
0[931be0]: (49f1af0) Next state: NNTP_READ_ARTICLE
0[931be0]: (49f1af0) Next state: NNTP_READ_ARTICLE
0[931be0]: (49f1af0) Next state: NNTP_READ_ARTICLE
49f1af0) Next state: NNTP_READ_ARTICLE
0[931be0]: (49f1af0) Next state: NEWS_DONE
0[931be0]: (49f1af0) Next state: NEWS_FREE
0[931be0]: (49f1af0) CleanupAfterRunningUrl()
0[931be0]: (47291c0) setting busy to 1
0[931be0]: (47291c0) setting busy to 1
0[931be0]: (47291c0) ParseURL
0[931be0]: (47291c0) original message spec =
news-message://g@news/alt.sports.basketball.nba.gs-warriors#19412
0[931be0]: (47291c0) m_messageID = (null)
0[931be0]: (47291c0) group = (null)
0[931be0]: (47291c0) commandSpecificData = (null)
0[931be0]: (47291c0) m_key = 19412
0[931be0]: (47291c0) Next state: SEND_FIRST_NNTP_COMMAND
0[931be0]: (47291c0) current group = alt.sports.basketball.nba.gs-warriors,
desired group = alt.sports.basketball.nba.gs-warriors
0[931be0]: (47291c0) Next state: NNTP_SEND_ARTICLE_NUMBER
0[931be0]: (47291c0) Sending: ARTICLE 19412
0[931be0]: (49f1af0) setting busy to 0
0[931be0]: (47291c0) Next state: NNTP_RESPONSE
0[931be0]: (47291c0) Receiving: 220 19412 <3c7c58b6.809138@netnews.attbi.com>
article
0[931be0]: (47291c0) Next state: SEND_FIRST_NNTP_COMMAND_RESPONSE
0[931be0]: (47291c0) Next state: SETUP_NEWS_STREAM
0[931be0]: (47291c0) Next state: NNTP_BEGIN_ARTICLE
| Assignee | ||
Comment 5•24 years ago
|
||
yes, if I understand you correctly, that's the correct result.
As far as reading the log is concerned, you don't have to worry about it if you
don't want to. What you would look for is that the second time you download
articles for offline use, we won't get any "No - article doesn't exist
responses" from article fetches for articles that are expired. The error code is
423, I believe. The first time, you would get such responses, and we would
notice it and remove the hdrs from the local db, so that the second time, we
wouldn't try to fetch the expired articles.
Thanks David.
one note: i see possible side effect. Say I have newsgrp
with 10 mesgs. Maybe 4 are read/expired and 6 brand new.
If download that newsgroup, it expunges the 4 expired articles.
Well I notice that the 'Total' count in the folder pane and in
the lower right hand corner of mesgr (unread:#, Total:#)
isn't right anymore. For my example above the total count is zero.
I can get correct 'total' by 'downloading' the newsgrp. But
as soon as I click another non-downloaded newsgrp (either online or
offline) and revisit the downloaded newsgrp the 'total' count is
zero again.
Would you like me to file a new bug on this?
| Assignee | ||
Comment 7•24 years ago
|
||
sounds like maybe our handling of the counts when articles expire isn't right in
general - that might explain a few other problems people have seen with expired
articles in general...
Using
2002-02-27-06-trunk/ nt 4.0
2002-02-22-09-trunk linux 2.2
2002-02-22-07-trunk mac 9.2.2, mac 10.1.3
Verified when you now try to download a newsgroup with expired
mesgs, it results in the expired mesg headers in the thread
pane being removed and the only the new/unexpired mesgs are downloaded.
I'll file a new bug about total counts maybe not syncing correctly
when we download and as a result remove expired headers.
Marking as verified.
Status: RESOLVED → VERIFIED
Updated•21 years ago
|
Product: MailNews → Core
Updated•17 years ago
|
Product: Core → MailNews Core
You need to log in
before you can comment on or make changes to this bug.
Description
•