[TYPO3-core] RFC: Resubmission of feature #1585: Logfile by Request URI + configuration of Empty Referer handling + logline hook

[Steffen Kamper]

"Martin Kutschker" <martin.kutschker-n0spam at no5pam-blackbox.net> schrieb im 
Newsbeitrag 
news:mailman.1.1204472123.827.typo3-team-core at lists.netfielders.de...
> Steffen Kamper schrieb:
>> "Michael Stucki" <michael at typo3.org> schrieb im Newsbeitrag 
>> news:mailman.1.1204468185.28864.typo3-team-core at lists.netfielders.de...
>>> Steffen Kamper wrote:
>>>
>>>>> The reason why it was hardcoded: When for example a page is not found,
>>>>> there
>>>>> is also no template loaded. So the setting had to be hardcoded, but of
>>>>> course this was also the reason why I didn't consider the patch to be
>>>>> completed :-)
>>>> sure, that makes it complicated. The logfile name could vary so match 
>>>> is
>>>> not easy. Sounds like a complex thing:
>>>> * scan lodir for logfiles
>>>> * compare with domain if name match
>>>> * hope to get the right one
>>> As I said: You can just skip this part. It is no longer needed!
>>>
>>> - michael
>>> -- 
>>> Use a newsreader! Check out
>>> http://typo3.org/community/mailing-lists/use-a-news-reader/
>>
>> ok, so lets have a try, here is the patch
>>
>> vg  Steffen
>
> Michael, Steffen, I fear you lost me somewhere in your long thread, so I 
> will only comment on small detail.
>
> It's a nice idea to go away from plain mode integer arguments to something 
> else. But if you define constants you should IMHO define them where they 
> belong to. So TYPO3_SEVERITY_INFO belong to t3lib_div no 
> config_default.php. Especially it's prefixed with TYPO3 and has no 
> reference to sysLog() at all.
>
> I played with a similar idea, but to use class constants:
>
> class t3lib_div {
>   const SYSLOG_SEVERITY_INFO = 0;
> }
>
> Whatever is done in the end. Please makes such changes only globally in 
> the complete Core. Ie change all occurences if sysLog() accordingly.
>
> Masi

Hi Masi,

good idea to move them to t3lib_div. I will make the patch now after 
chatting with michael and i post new patch as answer to first post so it's 
easier to follow.
I will do a second patch with substitution of other syslog calls to have 
this patch readable.

vg  Steffen