Release Notes

TeamCity plugin: The repository revisions needed to build a server-side checkout patch will be retrieved using a single call from now on.

Command line configuration tools: The clconfigureclient command line utility can now be used to configure your Plastic SCM client with a Cloud server.

Command Line Interface: The cm cat command will request revision information in block in a single server-call if revision ID-based specs are provided.

Mac OS GUI: The focus and the scroll position is now kept after checking out, renaming and moving files in the 'Items' view.

Mac OS GUI: Dynamic views sizes (such as 'history', 'shelves', 'changesets',... ) are now preserved among application restarts.

plasticlogstats server utility gets a 'concurrency' subcommand to calculate the number of concurrent calls inside the server at a given moment.

The new command implements an interactive mode that:

* Allows you to list the call segments with concurrency (periods with 1 or more calls in the server).

* Sort the segments by max concurrent calls, date, or segment length.

* Quickly identify slow (>500ms) and super slow (>1sec) calls.

* Navigate the calls inside each segment.

* Group the segment by minute and navigate segments in the minute.

Additionally, parsing logs is about 40% faster now, which not only helps with this feature but with regular plasticlogstats sessions.

plasticlogstats is now able to export to XML and CSV for easy Excel analysis of channel call files.

Example: grep ChannelCall of several files and analyze them all together:

grep -a -h ChannelCall plastic.debug.log.txt.20161124 plastic.debug.log.txt.20161125 | plasticlogstats.exe
export --csv=201611-analysis.csv

plasticlogstats gains a new 'analyzer' mode. It lets you easily and quickly analyze ChannelCall logs to find hot spots, slow calls and more. It is a text based GUI, fully interactive, that basically replaces the analysis we used to do with Excel and pivot tables. plasticlogstats uses less RAM and is much, much faster.

Features:

* Group all calls by day.

* Group by hour.

* Group by method.

* See methods inside each group - and regroup from here in days and hours.

It calculates interesting params like average milliseconds per call, which give you a very good understanding of how fast your Plastic SCM server is responding.

It also calculates the % of fast (<500ms), slow (500-1000ms) and super-slow (>1000ms) calls.

Just invoke it as follows:

plasticlogstats.exe analyzer --file=allchannelcall.txt