-
Bug
-
Resolution: Timed out
-
Medium
-
None
-
2.1.1
-
2
-
Severity 2 - Major
-
Obviously the colon is a special character in the Lucene search syntax, but can we handle it better? For instance, search for 'javax.mail.NoSuchProviderException: smtp' and nothing is returned, search for 'javax.mail.NoSuchProviderException smtp' and you get a lot of valuable information.
I don't think we can expect users to know the difference. Can we strip colons from the end of words which are not recognised field names?
Please note that this issue can only be reproduced on full search. If searching for a page title with a colon in it using quick search the page title is found.
- is related to
-
-
- Closed
-
![[Extranet] Page](/images/icons/generic_link_16.png "[Extranet] Page"){kind=icon}
[CONFSERVER-5152] Colons and other special characters (exlamation-marks, question-marks, etc) break searches
Tony Albans
added a comment - Search is deemed as BROKEN !!!
I tell it, how I see it...
There is no amount of platitude and/or work-around, that will mitigate a "FUNDAMENTAL FAIL" in Confluence Search.
URL is ubiquitous part of contemporary business operations.
Confluence borks at Search of content following special character, eg as contained within URL.
We do expect Search to return ALL relevant entries, irrespective of what character precedes it (special or otherwise)!
Even when we Search part of a URL, we expect to see results !!!
Atlassian decision makers, are missing the point as to why we use Search, and why we expect that it works "bug free".
A FIX for this BUG, is high priority, not something that gets pushed back.
We are in the reals of "product NOT FIT for purpose".
Default search default standard is Google.
Search comparison is enlightening.
Excerpt from support response CSP-186846...
At the meantime, I've replicated locally and I found following workarounds:
1. Dividing Internal Access Link to two different columns, http/https and IP address:
http/https IP Address
https 1.1.1.1
2. Adding a space after https:// : https:// 1.1.1.1
3. Removing the https:// from internal access link.
In all above workarounds, you can add Links to your text and make your IP Address as a hyperlink.
Business owners "do not accept rework" on content, such that it can circumvent product development failure!
FIX the product!
or
Customers globally share, such that REAL IMPACT of product shortcomings, are readily available to potential purchasers.
Such is "commercial reality" for Vendors that continue to peddle BROKEN product.
Want to get the status of this bug upgraded, then start to share of documented personal experience (ie impact of inability to globally search URL or part URL content), and will eventually impact share price, the only metric that decision makers will to take note of !!!
Tony Albans
added a comment - Search is deemed as BROKEN !!!
I tell it, how I see it...
There is no amount of platitude and/or work-around, that will mitigate a "FUNDAMENTAL FAIL" in Confluence Search.
URL is ubiquitous part of contemporary business operations.
Confluence borks at Search of content following special character, eg as contained within URL.
We do expect Search to return ALL relevant entries, irrespective of what character precedes it (special or otherwise)!
Even when we Search part of a URL, we expect to see results !!!
Atlassian decision makers, are missing the point as to why we use Search, and why we expect that it works "bug free".
A FIX for this BUG, is high priority, not something that gets pushed back.
We are in the reals of "product NOT FIT for purpose".
Default search default standard is Google.
Search comparison is enlightening.
Excerpt from support response CSP-186846...
At the meantime, I've replicated locally and I found following workarounds:
1. Dividing Internal Access Link to two different columns, http/https and IP address:
http/https IP Address
https 1.1.1.1
2. Adding a space after https:// : https:// 1.1.1.1
3. Removing the https:// from internal access link.
In all above workarounds, you can add Links to your text and make your IP Address as a hyperlink.
Business owners "do not accept rework" on content, such that it can circumvent product development failure!
FIX the product!
or
Customers globally share, such that REAL IMPACT of product shortcomings, are readily available to potential purchasers.
Such is "commercial reality" for Vendors that continue to peddle BROKEN product.
Want to get the status of this bug upgraded, then start to share of documented personal experience (ie impact of inability to globally search URL or part URL content), and will eventually impact share price, the only metric that decision makers will to take note of !!!
Jeff Barrett
added a comment - I was told by support that this is causing problems when we try to use @TAGS between the wiki and source code, for example searching for @AUTHOR when used within the wiki and Jira finds all occurrences of author within the wiki and not just the tags.
Jeff Barrett
added a comment - I was told by support that this is causing problems when we try to use @TAGS between the wiki and source code, for example searching for @AUTHOR when used within the wiki and Jira finds all occurrences of author within the wiki and not just the tags. 
Per Fragemann [Atlassian]
added a comment - This is actually quite annoying, raising severity because it reduces usability of search a lot.
Per Fragemann [Atlassian]
added a comment - This is actually quite annoying, raising severity because it reduces usability of search a lot.
Thank you for taking the time to raise, comment or vote on this Bug. We regret to inform you that due to a limited number of reports and based on our current backlog of higher impact issues that we are closing this issue as Timed Out.
If this issue is still impacting you on a recent version please feel free to comment with the affected version. Any further details you may be able to provide regarding reproduction or impact of this issue may help us better address this issue.
Thanks again.
Regards,
Confluence Development