28539f6aaf
search_to_sql: trim() input
2015-08-14 10:31:29 +03:00
dcbe36b2a2
fix some remaining old-style viewfeed() calls
2015-08-13 09:51:39 +03:00
45913edda0
fetch_file_contents: use http/1.1 when not using curl
2015-08-12 21:51:02 +03:00
0acc1ed76e
dijit style updates
2015-08-12 16:58:06 +03:00
6b0a17ad8b
viewfeed: switch to hash-based arguments
...
viewfeed: remove (obsolete) method parameter
add shortcut to debug viewfeed() query
2015-08-12 14:09:46 +03:00
884ae7a941
limit recently read feed to 1 day of read headlines
2015-08-12 10:02:46 +03:00
7d73cc4184
minor css fixes
2015-08-12 09:28:38 +03:00
b87744534a
add plugin-based filter actions (see example plugin in attic)
...
bump schema
2015-08-11 23:28:42 +03:00
6ee255fba3
use join between user_entries and entries table
2015-08-11 20:04:38 +03:00
c1ebb6cdf7
update notice/error css classes
2015-08-11 18:37:07 +03:00
6810a1de42
use dijit form widget for attachments dropdown
2015-08-11 14:53:07 +03:00
209ebdc644
queryfeedheadlines: only use disable_offsets hack in expanded mode
2015-08-11 09:19:47 +03:00
07877caf40
Revert "update: reenable first ref_id/id transaction"
...
This reverts commit 14ce0e2504
2015-08-10 13:15:37 +03:00
14ce0e2504
update: reenable first ref_id/id transaction
2015-08-10 12:29:27 +03:00
e07be79a8b
first id query: add last_read for recently read feed
2015-08-10 12:20:01 +03:00
dab229cd1f
update_rss_feed: disable transactions for the time being
2015-08-10 12:17:49 +03:00
6b3160cf1e
purge_feed: remove obsolete postgresql queries
2015-08-06 19:28:58 +03:00
c7edba791c
don't bother with topmost id if working with disabled offsets
2015-08-05 16:23:27 +03:00
0bd172cd6a
experimental: if using adaptive/unread mode w/ auto catchup don't calculate offsets because we're only interested in top $limit headlines anyway
2015-08-05 10:13:30 +03:00
332ff5cfb8
add a very obvious speed improvement to topmost_id calculation
2015-08-04 20:38:43 +03:00
8f135ce3b6
bump schema
2015-08-04 13:56:31 +03:00
60e680599d
allow selecting stemming language in search dialog
2015-08-04 13:53:48 +03:00
df659891ac
implement per-feed stemming language setting
2015-08-04 13:32:52 +03:00
e854442e1f
more fts stuff for simple index
2015-08-04 12:52:49 +03:00
b4f544d389
add some fts stuff
2015-08-04 09:53:43 +03:00
3ceb893f66
add one catchall function to make uniqids/keyhashes/etc used by tt-rss
2015-08-03 19:21:06 +03:00
97aa917c6d
smart_date_time: better eta_min handling
2015-07-18 00:01:37 +03:00
7680d44ef2
purge_orphans: optimize the query, maybe
...
https://tt-rss.org/forum/viewtopic.php?f=1&t=3417&p=20408#p20407
2015-07-15 13:29:03 +03:00
a950fbecb5
Revert "Performance breakthrough"
...
This reverts commit c291685654
2015-07-15 08:32:57 +03:00
c291685654
Performance breakthrough
...
Improvement in the Purge method for huge performance increase.
"WHERE ref_id IS NOT NULL" is needed by pgsql, not necessarily by mysql.
This is a more standard way to make deletions based on propagation (from ttrss_user_entries) with 1-to-N relations.
2015-07-13 12:03:50 +02:00
f45b67c53e
queryFeedHeadlines: remove filter stuff
2015-07-13 12:54:22 +03:00
0e4da73f06
do not allow commas in tags
2015-07-13 01:22:44 +03:00
f56e5a3504
add workaround for fresh feed first id calculation issue
2015-07-13 00:40:15 +03:00
f7fd1edb76
first_id default value: type is important
2015-07-12 22:54:55 +03:00
8831632905
add some more stuff to first_id query to fix virtual feeds
2015-07-12 22:50:21 +03:00
d5e0486e0c
minor first_id query fix
2015-07-12 18:47:06 +03:00
48fefe2f6b
fixes for first_id stuff
2015-07-12 18:08:02 +03:00
83ce77a2e8
functions: fix some phpstorm-reported warnings
2015-07-12 14:07:38 +03:00
3444020100
report top id changed in headlines buffer
2015-07-12 12:01:34 +03:00
f5a0fb8b64
queryFeedHeadlines: move to array-based arguments, optionally check if first element changed when paginating
2015-07-12 01:29:36 +03:00
9959acc87a
change version_static based on y.m
2015-07-09 21:47:42 +03:00
5a4074a9f8
Check for HTML in the first 100 bytes.
...
Some HTML pages have an XML header (XHTML), which alone is nearly 50 bytes.
Thus we need to check for the HTML or doctype tags in the first 100 bytes.
2015-07-09 13:39:07 +02:00
9090b874ab
simplify search, remove search_mode as useless
...
remove dialog to select by tags, simplify browsing by tag query
2015-07-07 15:59:32 +03:00
e2a3689a18
sanitize: disable <hr> for now
2015-07-06 23:09:51 +03:00
b90c4468fc
sanitize: allow <hr> elements
...
af_redditimgur: optionally add original content using Readability php implementation
2015-07-06 22:44:43 +03:00
35c3735494
fix article tags/labels not assigning properly on mysql because of the 4byte unicode workaround
2015-07-06 15:32:26 +03:00
6bb96beb3a
update_rss_feed: set basic feed info if title is [Unknown] (fixes batch subscribe)
2015-07-04 19:33:46 +03:00
b6714c77d2
limit smart_date_time stuff to headlines only
2015-07-03 23:36:47 +03:00
46973af5df
smart_date_time: add special case when article date is within current hour
2015-07-03 23:25:47 +03:00
3318d32410
move language detection to a plugin, remove config.php constant
2015-06-19 10:12:47 +03:00
Andrew Dolgov
Andrew Dolgov
GregThib
Christian Weiske