56077[BUGFIX] Harden deprecation log handling
Commit message
Review
-
Code review CR +2 Done
- Benjamin Franzke +2
- Markus Klein +1
-
Verified V +2 Done
- Benjamin Franzke +2
- Markus Klein +1
- TYPO3com +1 CI
Issues 3
Files 7
- typo3/sysext/core/Classes/Core/Bootstrap.php+1 −1
- typo3/sysext/core/Classes/Log/LogManager.php5+4 −0
- typo3/sysext/core/Configuration/DefaultConfiguration.php+2 −1
- typo3/sysext/core/Configuration/DefaultConfigurationDescription.yaml+1 −1
- typo3/sysext/core/Documentation/Changelog/9.5.x/Important-84105-StreamlineDeprecationLogHandling.rst10new+47 −0
- typo3/sysext/install/Classes/Configuration/Context/DebugPreset.php+4 −1
- typo3/sysext/install/Classes/Configuration/Context/LivePreset.php+4 −1
Review log 3948
Show 1823 earlier messages
-
Helmut Hummel uploaded patch set 1 03/09/18, 14:28
-
Helmut Hummel abandoned the change 03/09/18, 14:28
accidental push -
Helmut Hummel restored the change 04/29/20, 20:35
-
Helmut Hummel uploaded patch set 2 04/29/20, 21:00
-
Helmut Hummel rebased onto patch set 3 04/29/20, 21:00
-
Helmut Hummel updated the commit message in patch set 4 04/29/20, 21:02
-
TYPO3com voted on patch set 2 V +1 04/29/20, 21:19 Completed build in 19m 15s on Wed, 29 Apr, 09:19 PM Test Summary: 73663 passed Find logs and detail information at https://bamboo.typo3.com/browse/CORE-GTC-3110
-
TYPO3com voted on patch set 3 V -1 04/29/20, 21:23 Completed build in 14m 42s on Wed, 29 Apr, 09:23 PM Test Summary: 1 of 73815 failed Find logs and detail information at https://bamboo.typo3.com/browse/CORE-GTC-3111
-
TYPO3com voted on patch set 4 V +1 04/29/20, 21:27 Completed build in 19m 13s on Wed, 29 Apr, 09:27 PM Test Summary: 73663 passed Find logs and detail information at https://bamboo.typo3.com/browse/CORE-GTC-3112
-
Markus Klein commented on patch set 4 04/29/20, 21:57
LogManager.php:144!empty($logWriterOptions['disabled']) -
Markus Klein voted on patch set 4 CR +1 04/29/20, 22:04
-
Benni Mack commented on patch set 4 04/30/20, 06:00
that is cool! I guess we'd need some kind of Important.RST for that, or Feature RST (which is fine for me in this case, also for 9.5.x then). -
Helmut Hummel commented on patch set 4 04/30/20, 10:17
> Patch Set 4: > > I guess we'd need some kind of Important.RST for that, or Feature RST (which is fine for me in this case, also for 9.5.x then). Yepp. Makes sense! -
Helmut Hummel uploaded patch set 5 04/30/20, 10:47
-
TYPO3com voted on patch set 5 V +1 04/30/20, 11:03 Completed build in 15m 46s on Thu, 30 Apr, 11:03 AM Test Summary: 73663 passed Find logs and detail information at https://bamboo.typo3.com/browse/CORE-GTC-3122
-
Benni Mack commented on patch set 6 04/30/20, 11:26
Published edit on patch set 5. -
Benni Mack voted on patch set 6 V +1 CR +1 04/30/20, 11:26
-
TYPO3com voted on patch set 6 V +1 04/30/20, 11:37 Completed build in 11m 49s on Thu, 30 Apr, 11:37 AM Test Summary: 73663 passed Find logs and detail information at https://bamboo.typo3.com/browse/CORE-GTC-3123
-
Helmut Hummel commented on patch set 6 04/30/20, 11:54
LogManager.php:144ReplyDone -
Josef Glatz commented on patch set 6 04/30/20, 13:06
I would really love to see the new file in 10.4.x rST folder. (and for the backport moving rST to 9.5.x). -
Benni Mack commented on patch set 6 04/30/20, 13:48
> Patch Set 6: > > I would really love to see the new file in 10.4.x rST folder. (and for the backport moving rST to 9.5.x). What would be the reasoning for that? -
Markus Klein voted on patch set 6 V +1 CR +1 04/30/20, 14:44
-
Markus Klein commented on patch set 6 V -1 04/30/20, 14:52
syntax problem in rst
-
Markus Klein commented on patch set 7 04/30/20, 14:54
Published edit on patch set 6. -
Markus Klein voted on patch set 7 V +1 CR +1 04/30/20, 14:55
-
Daniel Siepmann commented on patch set 7 04/30/20, 15:09
I'm wondering why we introduce a new feature "disabled" / "enabled". Wouldn't it be possible to set the whole PHP array configuration for logging via the preset?Important-84105-StreamlineDeprecationLogHandling.rstAs this is going into master as well, the rst should also exist inside 10.4.x. We can use the `.. include::` to not duplicate the content, and ease contributions to the file. Not sure about the Backend features which render the rst in plain text (upgrade documentation + extension scanner). -
Josef Glatz commented on patch set 7 04/30/20, 15:17
> Patch Set 6: > > > Patch Set 6: > > > > I would really love to see the new file in 10.4.x rST folder. (and for the backport moving rST to 9.5.x). > > What would be the reasoning for that? Talked to Daniel. He already commented the possibility in rSt. Since this is an important change we have to add the important rst also in 10.4.x folder. Otherwise I'm not informed as a maintainer when updating from 10.4.1 to 10.4.2. -
Benjamin Franzke commented on patch set 7 CR -1 04/30/20, 15:19
sounds like something is merged inside LocalConfiguration.php. Better write "default configuration is merged with settings from LocalConfiguration.php". (Really had to wrap my head around this)Important-84105-StreamlineDeprecationLogHandling.rst:12Its unclear what this means. Something like "some specific log writers are configured by default, others are not"?Important-84105-StreamlineDeprecationLogHandling.rst:12This sounds like a bug report, not like an information. I would rather start with the second paragraph (with the Important change) and describe the issue afterwards.Important-84105-StreamlineDeprecationLogHandling.rst:34install tool -
TYPO3com voted on patch set 7 V +1 04/30/20, 16:32 Completed build in 21m 18s on Thu, 30 Apr, 04:32 PM Test Summary: 73663 passed Find logs and detail information at https://bamboo.typo3.com/browse/CORE-GTC-3130
-
Helmut Hummel commented on patch set 7 04/30/20, 17:00
> Patch Set 7: > > (1 comment) > > I'm wondering why we introduce a new feature "disabled" / "enabled". > Wouldn't it be possible to set the whole PHP array configuration for logging via the preset? Do you have something specific in mind, what could be set instead? -
Helmut Hummel commented on patch set 7 04/30/20, 17:05
@Benjamin: Feel free to add any improvement to the writings. -
Helmut Hummel uploaded patch set 8 05/01/20, 09:33
-
Helmut Hummel commented on patch set 8 05/01/20, 09:33
Reply
Done -
TYPO3com voted on patch set 8 V +1 05/01/20, 09:45 Completed build in 11m 46s on Fri, 1 May, 09:45 AM Test Summary: 73661 passed Find logs and detail information at https://bamboo.typo3.com/browse/CORE-GTC-3140
-
Helmut Hummel uploaded patch set 9 05/01/20, 21:00
-
Helmut Hummel commented on patch set 9 05/01/20, 21:06
Important-84105-StreamlineDeprecationLogHandling.rstReplyI don't really know where this file actually has to be. I put it into 10.4.x, Benni moved it ot 9.5.x, Josef also argued that it needs to be in both places. Can you help out? Maybe push a change with it being in both places with the include? Thanks! -
Helmut Hummel commented on patch set 9 05/01/20, 21:07
Important-84105-StreamlineDeprecationLogHandling.rst:12ReplyDoneImportant-84105-StreamlineDeprecationLogHandling.rst:12ReplyDoneImportant-84105-StreamlineDeprecationLogHandling.rst:34ReplyDone -
TYPO3com voted on patch set 9 V +1 05/01/20, 21:12 Completed build in 11m 10s on Fri, 1 May, 09:12 PM Test Summary: 73663 passed Find logs and detail information at https://bamboo.typo3.com/browse/CORE-GTC-3154
-
Benjamin Franzke commented on patch set 9 05/02/20, 05:29
Thanks Helmut for taking on the rewrite, I took a break yesterday and therefore couldn't add my suggestion on my own. Only one last (tiny) nitpick remains for me…LogManager.php:144I missed this during my first review, sorry. What's the reason to use empty() which uses loose comparison (instead of strict)? Do have any special configuration handling in mind that writes string values in here? I've seen that you mentioned "truthy" values in the RST. I'm asking because we're about to use phpstan more and more, and while it is still a loong way to enable phpstan-strict-rules, empty() would be considered disallowed by those strict rules. It's then likely to break the loose typing by refactoring this code. Anyway, if we really need loose typing here, I would still prefer (It looks cleaner and is easier to understand): if ($logWriterOptions['disabled'] ?? false)Important-84105-StreamlineDeprecationLogHandling.rstReplyThis should be discussed outside this patch. If you look in the 8.7.x or 9.5.x folders you see that RSTs were always added to the lowest backport version of the patch (only). Therefore, if you want to change this, lets discuss it outside, and before patches start to add duplicate files, all existing ("missing") RSTs would need to be duplicated in a patch that introduces this new policy. -
Benjamin Franzke uploaded patch set 10 05/02/20, 06:04
-
TYPO3com voted on patch set 10 V +1 05/02/20, 06:15 Completed build in 10m 43s on Sat, 2 May, 06:15 AM Test Summary: 73663 passed Find logs and detail information at https://bamboo.typo3.com/browse/CORE-GTC-3158
-
Helmut Hummel commented on patch set 10 05/02/20, 09:43
> Patch Set 7: > > (1 comment) > > I'm wondering why we introduce a new feature "disabled" / "enabled". > Wouldn't it be possible to set the whole PHP array configuration for logging via the preset?LogManager.php:144ReplyI'm fine with $logWriterOptions['disabled'] ?? false as well. If you look at the earlier patch sets, my initial comparison was isset($logWriterOptions['disabled']) && $logWriterOptions['disabled'] === true, which I changed to !empty() as desired/suggested by Markus. I'm fine either way. My gut feeling says, that loose comparison is more in line with the multitude of other checks for user provided configuratio within TYPO3. If we want to go strict, we should probably do it for the complete configuration set and maybe also add a validatation step for config -
Helmut Hummel commented on patch set 10 05/02/20, 09:44
Important-84105-StreamlineDeprecationLogHandling.rstReplyAck -
Benjamin Franzke commented on patch set 10 05/02/20, 10:39
LogManager.php:144ReplyAck -
Markus Klein voted on patch set 10 CR +1 05/04/20, 10:22
-
Markus Klein voted on patch set 10 V +1 05/04/20, 10:31
-
Benjamin Franzke voted on patch set 10 V +2 CR +2 05/04/20, 10:51
-
Benjamin Franzke merged patch set 11 05/04/20, 10:51