Commit Graph

102 Commits

Author SHA1 Message Date
Roland Haeder f88345cbea
HTTPS support in validate_url()
Signed-off-by: Roland Haeder <roland@mxchange.org>
2016-05-01 14:58:41 +02:00
Roland Haeder 738e22d3eb
Some fixes:
- Letting $profile_link and $hcard always been overwritten with newly found
  entries has caused DFRN network detectio to fail and "connect" is no longer
  working. This fix makes sure that these both variables are only set if they
  are still empty.
- added more debug lines to see which URL was it before and after webfinger "detection"

Signed-off-by: Roland Haeder <roland@mxchange.org>
2016-05-01 14:52:48 +02:00
rabuzarus 95f129d9b6
xml: rename convert_element_to_array to element_to_array 2016-05-01 13:48:52 +02:00
rabuzarus 89f7a91907
move xml related function to the xml class + some doxygen docu 2016-05-01 13:48:52 +02:00
rabuzarus db581c08d0
rework autocomplete: some polishing 2016-05-01 13:47:24 +02:00
Michael Vogel 79fb05ff80
Removed moved function 2016-05-01 13:46:08 +02:00
rabuzarus 12479fb478
rework autocomplete: Access list of smilies over JSON 2016-05-01 13:46:08 +02:00
Matthew Exon 4ff8c9e3a1 Document new redirect_url return value 2016-01-25 11:38:36 +01:00
Matthew Exon 930da0aa44 Return redirect_url as part of the return array from z_fetch_url instead of via global state 2016-01-25 11:05:17 +01:00
Matthew Exon 3ada0093f9 Note the final URL used to retrieve content after all redirects 2016-01-25 07:41:56 +01:00
Matthew Exon 677903b316 Report errors during cURL fetch 2016-01-25 02:56:47 +01:00
Michael Vogel 5ec2c7e965 When a server isn't reachable there were timeout problems with fetching image data 2016-01-18 15:38:38 +01:00
rabuzarus 3df5c17124 reformat todo according to doxygen style 2015-12-25 23:17:34 +01:00
Michael Vogel 6a15dded0e Bugfix: The timeout value of z_fetch_url couldn't be set manually 2015-11-03 00:48:49 +01:00
Michael Vogel 2840bf2293 New option to permit crawlers 2015-10-03 23:16:40 +02:00
Michael Vogel 2038e65a57 Bugfix: The function "z_fetch_url" created warnings under certain circumstances. 2015-05-16 22:38:41 +02:00
fabrixxm 6258a21434 map bbcode tag from Red 2015-05-15 20:41:10 +02:00
Michael Vogel dbe8275ae5 Added some more timestamp measuring 2015-03-07 23:14:26 +01:00
Michael Vogel 7c4a1a059d New fields for the term table, improved query for the tag search. Changed the cache handling for rendered bbcode. 2015-03-07 21:24:39 +01:00
Michael Vogel 9e55dbf1db There is already a setting for that ... 2015-03-02 00:32:46 +01:00
Michael Vogel 5cb6d6e537 We have to disable the SSL checks, otherwise the communication will fail for servers with self-signed certificates 2015-03-01 10:53:49 +01:00
Fabrixxm caa21e6021 Merge branch 'master' into develop 2015-02-24 09:09:02 +01:00
Johannes Schwab 1cd1f7587c fix validate_email 2015-02-23 12:20:05 +01:00
Matthew Exon 2bb8b77123 Close cURL handle before recursing, so that cookie file is flushed 2015-02-11 03:42:07 +00:00
Matthew Exon d12b929c96 Accept lowercase location header 2015-02-10 03:14:08 +00:00
Michael Vogel aa6275356a The function to fetch the original url (without shortener) failed if a linked file was too large. 2015-01-04 15:31:47 +01:00
Michael Vogel 0c60eb18f7 Some more typical tracking parameter is removed 2014-12-01 22:52:51 +01:00
fabrixxm 6db73c0b6d fix update routine to support update from 3.2
disable PDO support
2014-09-07 17:28:38 +02:00
Michael Vogel 7af5c3c560 The useragent string doesn't contains of the words "Mozilla" anymore - but the name "Friendica" and the base url 2014-09-07 10:47:37 +02:00
Michael Vogel 07bc8ca166 fetch_url: Logging the data only when at the "data" loglevel 2014-08-07 08:22:32 +02:00
Michael Vogel e974e586cc Unified user agent string when fetching data. 2014-08-07 07:58:20 +02:00
Michael Vogel 6f5bc8d9a7 Setting the user agent to "Mozilla/5.0 (compatible; Friendica)" 2014-07-10 23:13:07 +02:00
Michael Vogel 6a28ae8b52 Just some better handling of tracker variables (url encoding) 2014-05-30 01:02:32 +02:00
Michael Vogel 135a403cec Added some tracking variables from facebook in the function to clean up URLs. 2014-05-29 11:43:37 +02:00
Michael Vogel 5ae9c5e538 Added some logging 2014-05-24 11:01:33 +02:00
Michael Vogel 6b461a74cb Cleaning up tracking parameters in urls. 2014-05-18 20:32:06 +02:00
Michael Vogel 426e0045ee The function "short_url" is used in several addons, so it deserves a central place. 2014-05-03 12:07:34 +02:00
Michael Vogel a64948b7f7 New function "original_url" (moved from an addon) that follows all redirects to the original url. 2014-04-04 10:55:30 +02:00
Michael Vogel 8c1d31decc Showing "view full size" is now configurable 2014-02-11 23:44:45 +01:00
fabrixxm ad5a47aecf add fetaure switch for acl-automention 2014-01-03 14:40:43 +01:00
Mats Sjöberg 19de78e511 Fixed subscription problem with new GNU Social. 2013-11-21 19:41:06 +01:00
Michael Vogel 759015e5fd Removed many "limit 1" at update and delete queries since they were superflous und generated massiv logfile entries 2013-11-02 16:13:11 +01:00
Matthew Exon 11163c6fb3 Allow calls to fetch_url to specify a cookiejar file, for sites that refuse to deliver content without setting a cookie 2013-08-05 10:59:22 +08:00
Matthew Exon 6a38705567 Replicate redirect URL building logic for post requests 2013-07-29 22:34:09 +08:00
Matthew Exon eb9bd03ef8 If the header returns an absolute path as a redirect, don't simply append the path to the old URL, remove the old absolute path first.
For example, original URL:

http://www.theregister.co.uk/2013/07/28/birmingham_uni_car_cracker_muzzled_by_lords/print.html

Redirect from header:

/Print/2013/07/28/birmingham_uni_car_cracker_muzzled_by_lords/

Incorrect result:

http://www.theregister.co.uk/2013/07/28/birmingham_uni_car_cracker_muzzled_by_lords/print.html/Print/2013/07/28/birmingham_uni_car_cracker_muzzled_by_lords/

Correct result after this patch:

http://www.theregister.co.uk/Print/2013/07/28/birmingham_uni_car_cracker_muzzled_by_lords/
2013-07-29 10:15:39 +08:00
friendica 886c3d1472 bad checkin 2013-05-02 22:19:16 -07:00
Matthew Exon 5b158dbcfb Make the content type available after calling fetch_url 2013-04-01 12:36:43 +08:00
tobiasd e7c705c9a1 Merge pull request #654 from mexon/filter_var
Use filter_var instead of parse_url to validate URLs, since the latter r...
2013-03-26 11:31:11 -07:00
Matthew Exon c7cfc6b170 Use filter_var instead of parse_url to validate URLs, since the latter returns true even for invalid URLs, in particular the empty string 2013-03-21 16:27:12 +08:00
Matthew Exon c53bdfb163 Restore redirect functionality removed by change 452ff19d 2013-03-14 14:09:42 +08:00