9 years agojason_leonard committed change 19977 into Adding latest patches for Swarm 2016.1 Adding: Client reset patch for ClientPool to prevent exception with dirty client. Adding: Lock review update... patch for preventing a race condition whilst Swarm updates or creates a review. Adding: Workers no tasks log patch for removing a "no tasks in queue" message whilst capturing debug logs. Updating: moderators email patch for allowing moderators to receive review change emails. Updating: projects caseinsensivie matching patch to workaround P4PHP case sensitivity issues when selecting a change's project. « | ||
Add a comment | ||
9 years agojason_leonard committed change 19194 into Adding project branch case insensitive matching. We were seeing issues with reviews not being assigned to projects if there was a case issue in eith...er the project branch depot path or the files in the review. This patch forces project branch paths to be matched case insensitive. This is useful for case insensitive Perforce servers when Swarm is run from Linux. The underlying P4PHP P4_Map class is case sensitive when run on Linux platforms. « | ||
9 years agojason_leonard committed change 15882 into Moderators Email patch by Jason Leonard (jason.leonard@citrix.com) This patch adds a new checkbox to the project to allow moderators to get all e...mails about changes to a review. This is equivilent (only for notification purposes) to automatically adding them as reviewers to a review. « | ||
10 years agojason_leonard requested review 13889 for perforce-software-p4transfer:main Fix for MoveAdd bug where the target file is sync'ed to the incorrect revision of the file before editing. This results in a pending resolve and a fa...iled submit. The sync -f is performed on the source server's file path and revision since "file.localIntegSyncSource()" is used. This is incorrect as far as I can see, and should be replaced with "source" since this holds the correct target file and head revision. #review @robert_cowham In a simple test scenario, where both source and target are initally empty, transfer of a move operation will result in identical revision numbers for both target and source, thus masking it's effect. The bug will only occur if the target is of a different head revision number and or path to the source. « | ||
10 years agojason_leonard committed change 13779 into Adding Initial version of swarm trigger in Perl. This script was ported from swarm-trigger.sh so that strict and enforce types function on Windows. N...o testing has been performed on Linux, and only limited on Windows. The perl script has dependencies on only core Perl components as far as I can tell. « | ||
Change | User | Description | Created | ||
---|---|---|---|---|---|
13888 | jason_leonard | Fix for MoveAdd bug where the target file is sync'ed to the incorrect revision of the file... before editing. This results in a pending resolve and a failed submit. The sync -f is performed on the source server's file path and revision since "file.localIntegSyncSource()" is used. This is incorrect as far as I can see, and should be replaced with "source" since this holds the correct target file and head revision. #review-13889 @robert_cowham In a simple test scenario, where both source and target are initally empty, transfer of a move operation will result in identical revision numbers for both target and source, thus masking it's effect. The bug will only occur if the target is of a different head revision number and or path to the source. « |
10 years ago | View Review |
Adjust when notifications are sent to you about reviews that you're associated with (as an author, reviewer, project member or moderator).