Project

General

Profile

Bug #2131

blogs - read more... not working (conditional browser dependent)

Added by Marko Kastelic about 9 years ago. Updated over 8 years ago.

Status:
Bug resolved
Priority:
Normal
Assignee:
Category:
-
Target version:
Start date:
11/10/2010
Due date:
% Done:

70%

Estimated time:
Complexity:
Normal
SCRUM pts - complexity:
?

Description

bad structured html span tags (cross sectioned span/h3) blogpost_introduction and blogpost_text prevents showing text section.
Generated html source is attached that show </span></h3> misplaced


Files

blog_intro_bad.JPG (132 KB) blog_intro_bad.JPG Marko Kastelic, 11/10/2010 16:21
blog_structure_bad.JPG (72 KB) blog_structure_bad.JPG Marko Kastelic, 18/10/2010 22:16
blog_structure_ok.JPG (83.7 KB) blog_structure_ok.JPG Marko Kastelic, 18/10/2010 22:16
Screenshot.png (34.1 KB) Screenshot.png Julio Montoya, 19/10/2010 11:15
not_good2.JPG (157 KB) not_good2.JPG Marko Kastelic, 19/10/2010 13:27
lorem.PNG (114 KB) lorem.PNG Marko Kastelic, 21/04/2011 22:12
lorem2.PNG (61.1 KB) lorem2.PNG Marko Kastelic, 21/04/2011 22:12
lorem3.PNG (129 KB) lorem3.PNG Marko Kastelic, 21/04/2011 22:12

Associated revisions

Revision b3f43281 (diff)
Added by Julio Montoya over 8 years ago

Fixing bug when clickin in the "read more" link in the blog tool see #2131

History

#1

Updated by Marko Kastelic about 9 years ago

two more pictures are attached that shows corrupted structures of the blog posts.
the bug is introduced in 1.8.7.1 and it exists in development 1.8.8. I do not know why, but the reason is in construction of blog introduction (line 742, blog.lib.php)

if we change

  $introduction_text .= " $tok";

to
  $introduction_text .= '&nbsp'.$tok;

the structure is fixed and read more works as expected.
This is a bit strange. I used tokenizer many times before as is used in original source and there were no problems at all. But my strings were not utf-8 encoded.

#2

Updated by Marko Kastelic about 9 years ago

nope, not good for 1.8.7.1. now blogpost_text section is messed ..., but at least read more works

#3

Updated by Julio Montoya about 9 years ago

it works well for me at least in FF 3.6

#4

Updated by Marko Kastelic about 9 years ago

this snaphot was taken with FF 3.5.1 from beeznests server right now. Same result, different structure i'm getting from mine (dedicated) server with ff 3.6, (php 5.2) and while using

$introduction_text .= '&nbsp'.$tok;

i noticed, that nbsp; is inserted inside some span tags like
<span&nbsp; style=> ... 

the reason for the mess inside strings could be a memory shortage..

#5

Updated by Marko Kastelic about 9 years ago

btw: first 5 posts from this blog just 'works', sixth, sevent and on are corrupted

#6

Updated by Yannick Warnier over 8 years ago

  • Target version set to 1.8.8 stable
#7

Updated by Julio Montoya over 8 years ago

  • Status changed from New to Needs more info
  • % Done changed from 0 to 70
#8

Updated by Marko Kastelic over 8 years ago

very nice: read more works now (the fix is good for now)

other:
a) button for saving new blog almost overlaps existing blog list if there is a previous blog; a little more space is needed
b) there is a case, and abnormal blog behaviour, when copy/paste from web page content is used (html formated text). Sample pictures are attached, content is taken from www.lipsum.com. First picture shows selected content, second 2 blog posts (upper with pasted unchanged html, lower with pasted pure text with no html formating). You can see read more link inactive. The 3rd shows expanded the same 2 posts with overlaping divs ...(read more was activated by manual removal of the first /empty/ line of the post)

#9

Updated by Marko Kastelic over 8 years ago

... from the social network, blog content(preview) formating is ok

#10

Updated by Yannick Warnier over 8 years ago

  • Status changed from Needs more info to Bug resolved
  • Assignee set to Julio Montoya

Well, copy-pasting doesn't always work great (mostly because some JavaScript code is trying to make sense from the copied code and when it cannot, it lets go without modification - and maybe the copied code didn't have the closing tags). I don't think that really a major bug. If we manage to fix it, we will probably break something else in the process.

Marko, if you find a common pattern that we can work on, please open a specific task for it. In the meantime and considering ur tight schedule, I'm closing this bug.

Thanks for reporting by the way!

Also available in: Atom PDF