heads up: headline list behavior changes

[pcause]

I was seeing it as slower as well. Just pulled the rollback and it seems faster now.

[pcause]

oops, spoke to soon. grouping is gone.

[edit] meant the grouping of headlines by feed.

[Maru]

yeah, i've just rolled it back, so we might have new and exciting bugs now. it also seems faster, a bit.

I just installed the rollback and now I see the following error every time I click on a category..

Exception: TypeError: reply.headlines.content.each is not a function
Function: headlines_callback2()

Furthermore search no longer has the dropdown menu to select All Feeds

[fox]

yeah, i've just rolled it back, so we might have new and exciting bugs now. it also seems faster, a bit.

I just installed the rollback and now I see the following error every time I click on a category..

Exception: TypeError: reply.headlines.content.each is not a function
Function: headlines_callback2()

Furthermore search no longer has the dropdown menu to select All Feeds

ctrl-r

[fox]

and yes search dropdowns are gone

those were remnants of antediluvian times when dinosaurs roamed the earth and virtual feeds were not implemented yet. these days if you want to search all articles you can just click on the all articles feed and then search.

[fox]

oops, spoke to soon. grouping is gone.

[edit] meant the grouping of headlines by feed.

that thing again

should work now

[pcause]

just pulled and the grouping is fixed. thanks.

BTW, noticed the feed header look changed. a bit cleaner.

[JustAMacUser]

I just updated my production deployment and things seem fast using Safari 8.0.7. Using the same browser this morning with my dev virtual machine was noticeably slower (never bothered to update dev with the latest commits since fox started patching/reverting this later in the day).

[fox]

BTW, noticed the feed header look changed. a bit cleaner.

yeah things that i don't personally use tend to get broken/forgotten until someone reminds me they exist

[martywd]

Not a new issue for me but I'm still seeing similar errors as posted earlier by 'pcause' and 'nameless' in this thread.

As I type I'm running: tt-rss v15.7 (91837f0), PHP 5.5, postgresql-9.3 ... Enabled plugins: auth_internal, af_comics, af_unburn, close_button, feedstatistics, note, swap_jk

This started back on July 12, showing up every couple of days in tt-rss' system log in the browser client and postgresql's log. After a week or so, every day. Sometimes a couple of times a day. Here's the error from tt-rss' system log:

Code: Select all

Error:
E_USER_ERROR (256)

Filename:
classes/db/pgsql.php:46

Message:
Query SELECT DISTINCT ttrss_feeds.title, date_entered, guid, ttrss_entries.id, ttrss_entries.title, updated, score, marked, published, last_marked, last_published FROM ttrss_entries,ttrss_user_entries LEFT JOIN ttrss_feeds ON (feed_id = ttrss_feeds.id) WHERE ttrss_user_entries.ref_id = ttrss_entries.id AND ttrss_user_entries.owner_uid = '2' AND date_entered >= NOW() - INTERVAL '1 hour' AND unread = false AND last_read IS NOT NULL ORDER BY last_read DESC LIMIT 1 failed: ERROR: for SELECT DISTINCT, ORDER BY expressions must appear in select list LINE 24: ...unread = false AND last_read IS NOT NULL ORDER BY last_read ... ^


And the error(s) I typically see in postgresql's log. This the latest:

Code: Select all

2015-08-09 18:41:51 CDT ERROR:  for SELECT DISTINCT, ORDER BY expressions must appear in select list at character 592
2015-08-09 18:41:51 CDT STATEMENT:  SELECT DISTINCT
                                                                ttrss_feeds.title,
                                                                date_entered,
                                                                guid,
                                                                ttrss_entries.id,
                                                                ttrss_entries.title,
                                                                updated,
                                                                score,
                                                                marked,
                                                                published,
                                                                last_marked,
                                                                last_published
                                                        FROM
                                                                ttrss_entries,ttrss_user_entries
                                                        LEFT JOIN ttrss_feeds ON (feed_id = ttrss_feeds.id)
                                                        WHERE
                                                       
                                                        ttrss_user_entries.ref_id = ttrss_entries.id AND
                                                        ttrss_user_entries.owner_uid = '2' AND
                                                       
                                                       
                                                       
                                                        date_entered >= NOW() - INTERVAL '1 hour' AND
                                                        unread = false AND last_read IS NOT NULL ORDER BY last_read DESC LIMIT 1

As an aside, I ran the re-indexing as instructed on Aug 4. That completed without error.

No issues with using tt-rss in a browser. Login, subscribing, reading, updating, etc. all seem to be working without problems that I notice.

I don't know?
.

[fox]

this is already fixed in trunk

https://tt-rss.org/gitlab/fox/tt-rss/co ... b5d0b0d674

[martywd]

this is already fixed in trunk

https://tt-rss.org/gitlab/fox/tt-rss/co ... b5d0b0d674

Yes, I saw that commit...

Doing a git pull indicates I'm up-to-date:

Code: Select all

$ sudo -u www-data git pull origin master -v
[sudo] password for marty:
From https://tt-rss.org/git/tt-rss
 * branch            master     -> FETCH_HEAD
 = [up to date]      master     -> origin/master
Already up-to-date.

And. My current '../include/functions2.php' file includes that commit change. Here's lines 739-746 of 'functions2.php':

Code: Select all

                                                        marked,
                                                        published,
                                                        last_marked,
                                                        last_published,
                                                        last_read
                                                FROM
                                                        $from_qpart
                                                WHERE


If this is issue with my database(?), my postgresql skills suck. Any help by others in the forum would be much appreciated.

.

[fox]

if you read the error message you'll see that query select part ends with last_published instead of "last_published, last_read". so you shouldn't see this error any more from now on unless you've somehow managed to update wrong tt-rss directory (this has happened before).

this happened if you clicked on recently read feed so you can try doing that and see if error pops up in the log.

in any case other than generating a system log entry this doesn't really break anything.

[martywd]

if you read the error message you'll see that query select part ends with last_published instead of "last_published, last_read". so you shouldn't see this error any more from now on unless you've somehow managed to update wrong tt-rss directory (this has happened before).

I see. I suck at postgresql _and_ reading, too! As far as 'update wrong tt-rss'? There's _only_one_ working instance of a tt-rss directory on my server! I've even done searches for 'fucnctions2.php' (and other assorted files) to be sure. Other than a nightly rsync of the web server's root to another partition there's just the one instance of tt-rss.

this happened if you clicked on recently read feed so you can try doing that and see if error pops up in the log.

Yeah, I'll keep poking it to see if I can figure what's triggering the error.

in any case other than generating a system log entry this doesn't really break anything.

True. Thanks for your help.

.