fixed prune date range calculation

Author: shukriadams

docs/development.md

@@ -16,6 +16,10 @@ If you want to develop Linux on Windows, Vagrant is an excellent and convenient
 - Dotnet 6.0 SDK
 - Visual Studio 2019 (Windows) or Visual Studio Code with C# extension. If you're using Visual Studio make sure you've update to the latest version. Visual Studio 2017 isn't supported.
 
+Unit test coverage is done with https://github.com/SteveGilham/altcover, install ReportGenerator with
+
+    dotnet tool install --global dotnet-reportgenerator-globaltool
+
 ## Running from Visual Studio
 
 - Start Visual Studio as administrator
@@ -73,15 +77,10 @@ To run tests in Visual Studio, open Test > Windows > Test Explorer, run desired
 
 ## Test coverage reports
 
-Unit test coverage is done with https://github.com/SteveGilham/altcover. To visualize your tests
-
-- Install ReportGenerator 
-
-        dotnet tool install --global dotnet-reportgenerator-globaltool
-
-- after testing run 
+After testing run 
 
     cd /src
+
     reportgenerator -reports:./Tetrifact.Tests/coverage.xml -targetdir:./Tetrifact.Tests/coverage -assemblyfilters:+Tetrifact.*;-Tetrifact.Tests -classfilters:-Tetrifact.Core.ThreadDefault;-Tetrifact.Web.DaemonProcessRunner;-Tetrifact.Web.Pager;-Tetrifact.Web.Program;-Tetrifact.Web.Startup;-Tetrifact.Web.ReadLevel;-Tetrifact.Web.WriteLevel;-*f__*
 
 The HTML report is at

docs/use.md

@@ -1,123 +1,158 @@
-# Using Tetrifact
-
-Tetrifact stores builds as packages. A package is a group of files which are added and retrieved as a unit, and should correspond to a single build in your CI system, which in turn should of course correspond to a COMMIT or REVISION in your VCS (version control system). It is highly recommended you name your Tetrifact packages after the revision number (or hash) in your VCS the package's contents were compiled from, as this allows you to easily trace a package back to the source code from which it was built.
-
-Don't worry if your revision hashes are difficult to pass around, share or remember, packages can be tagged with additional human-friendly names too.
-
-## Changing log level
-
-The default log level as of version 1.8.4 is `Warning`. You can override this with the environment variable
-
-    Logging__LogLevel__Microsoft=<YOUR LEVEL>
-
-Allowed levels are the standard internal Dotnet values : Trace|Debug|Information|Warning|Error|Critical|None
-
-## REST
-
-All Tetrifact's functionality is exposed via a REST API, so it should be familiar.
-
-## Adding a package
-
-### As individual files
-
-HTTP METHOD :
-
-    POST
-
-ENDPOINT :
-
-    /v1/packages/myPackageName
-
-HEADER :
-
-    Content-Type: multipart/form-data; boundary=-------------------------acebdf13572468 
-
-BODY :
-
-    ---------------------------acebdf13572468
-    Content-Disposition: form-data; name="Files"; filename="1.txt" 
-
-    file 1 text or binary content here
-    ---------------------------acebdf13572468
-    Content-Disposition: form-data; name="Files" ; filename="path/to/2.txt"
-
-    file 2 text or binary content here
-    ---------------------------acebdf13572468
-
-Each file's "name" property must be "Files". The filename property should be the path and filename of the file you want to store, relative to the root of the package.
-
-Were you posting actual files with CURL it should look like
-
-    curl -X POST \
-        -H "Content-Type: multipart/form-data" \
-        -H "Transfer-Encoding: chunked" \
-        -F "Files=@~/mybuild/1.txt;filename=1.txt" \
-        -F "Files=@~/mybuild/path/to/2.txt;filename=path/to/2.txt" \
-        http://myTetrifact.server/v1/packages/myPackageName 
-
-### As an archive
-
-To post your build as an archive, use the following.
-
-HTTP METHOD :
-
-    POST
-
-ENDPOINT :
-
-    /v1/packages/myPackageName?IsArchive=true
-
-HEADER :
-
-    Content-Type: multipart/form-data; boundary=-------------------------acebdf13572468 
-
-BODY :
-
-    ---------------------------acebdf13572468
-    Content-Disposition: form-data; name="FILES"; filename="whatever"
-    Content-Type: application/octet-stream
-
-    <@INCLUDE *~/path/to/archive.zip*@>
-    ---------------------------acebdf13572468
-
-When posting a zip, note the following
-
-- only a single file can be attached to your post. If you add more than one, you'll get an error.
-- filename doesn't matter, as it won't be used
-- the archive's root will be treated as the root of the project, meaning all file paths will be mapped relative to this.
-
-With CURL it would look like
-
-    curl -X POST -H "Content-Type: multipart/form-data" -H "Transfer-Encoding: chunked" -F "Files=@path/to/archive" http://tetriserver.example.com/v1/packages/myPackage?isArchive=true 
-
-Chunking the upload is important if your archive is large, as this might exceeded the multipart body attachment size.
-
-### Posting an archive from NodeJS
-
-This uses the request package (https://www.npmjs.com/package/request).
-
-    let request = require('request'),
-        fs = require('fs'),
-        formdata = {
-            Files : fs.createReadStream('path/to/archive')
-        };
-
-    request.post({url: 'http://tetriserver.example.com', formData: formdata}, function(err, httpResponse, body) {
-        if (err) {
-            return console.error('upload failed : ', err);
-        }
-
-        console.log('Upload succeeded : ', body);
-    });
-
-## Tagging
-
-### Curl
-
-To add the tag "MyTag" to the package "MyPackage, use
-
-    curl -X POST http://tetriserver.example.com/v1/tags/MyTag/MyPackage
-
-To remove the tag
-
-    curl -X DELETE http://tetriserver.example.com/v1/tags/MyTag/MyPackage
+# Using Tetrifact
+
+Tetrifact stores builds as packages. A package is a group of files which are added and retrieved as a unit, and should correspond to a single build in your CI system, which in turn should of course correspond to a COMMIT or REVISION in your VCS (version control system). It is highly recommended you name your Tetrifact packages after the revision number (or hash) in your VCS the package's contents were compiled from, as this allows you to easily trace a package back to the source code from which it was built.
+
+Don't worry if your revision hashes are difficult to pass around, share or remember, packages can be tagged with additional human-friendly names too.
+
+## Changing log level
+
+The default log level as of version 1.8.4 is `Warning`. You can override this with the environment variable
+
+    Logging__LogLevel__Microsoft=<YOUR LEVEL>
+
+Allowed levels are the standard internal Dotnet values : Trace|Debug|Information|Warning|Error|Critical|None
+
+## REST
+
+All Tetrifact's functionality is exposed via a REST API, so it should be familiar.
+
+## Adding a package
+
+### As individual files
+
+HTTP METHOD :
+
+    POST
+
+ENDPOINT :
+
+    /v1/packages/myPackageName
+
+HEADER :
+
+    Content-Type: multipart/form-data; boundary=-------------------------acebdf13572468 
+
+BODY :
+
+    ---------------------------acebdf13572468
+    Content-Disposition: form-data; name="Files"; filename="1.txt" 
+
+    file 1 text or binary content here
+    ---------------------------acebdf13572468
+    Content-Disposition: form-data; name="Files" ; filename="path/to/2.txt"
+
+    file 2 text or binary content here
+    ---------------------------acebdf13572468
+
+Each file's "name" property must be "Files". The filename property should be the path and filename of the file you want to store, relative to the root of the package.
+
+Were you posting actual files with CURL it should look like
+
+    curl -X POST \
+        -H "Content-Type: multipart/form-data" \
+        -H "Transfer-Encoding: chunked" \
+        -F "Files=@~/mybuild/1.txt;filename=1.txt" \
+        -F "Files=@~/mybuild/path/to/2.txt;filename=path/to/2.txt" \
+        http://myTetrifact.server/v1/packages/myPackageName 
+
+### As an archive
+
+To post your build as an archive, use the following.
+
+HTTP METHOD :
+
+    POST
+
+ENDPOINT :
+
+    /v1/packages/myPackageName?IsArchive=true
+
+HEADER :
+
+    Content-Type: multipart/form-data; boundary=-------------------------acebdf13572468 
+
+BODY :
+
+    ---------------------------acebdf13572468
+    Content-Disposition: form-data; name="FILES"; filename="whatever"
+    Content-Type: application/octet-stream
+
+    <@INCLUDE *~/path/to/archive.zip*@>
+    ---------------------------acebdf13572468
+
+When posting a zip, note the following
+
+- only a single file can be attached to your post. If you add more than one, you'll get an error.
+- filename doesn't matter, as it won't be used
+- the archive's root will be treated as the root of the project, meaning all file paths will be mapped relative to this.
+
+With CURL it would look like
+
+    curl -X POST -H "Content-Type: multipart/form-data" -H "Transfer-Encoding: chunked" -F "Files=@path/to/archive" http://tetriserver.example.com/v1/packages/myPackage?isArchive=true 
+
+Chunking the upload is important if your archive is large, as this might exceeded the multipart body attachment size.
+
+### Posting an archive from NodeJS
+
+This uses the request package (https://www.npmjs.com/package/request).
+
+    let request = require('request'),
+        fs = require('fs'),
+        formdata = {
+            Files : fs.createReadStream('path/to/archive')
+        };
+
+    request.post({url: 'http://tetriserver.example.com', formData: formdata}, function(err, httpResponse, body) {
+        if (err) {
+            return console.error('upload failed : ', err);
+        }
+
+        console.log('Upload succeeded : ', body);
+    });
+
+## Tagging
+
+### Curl
+
+To add the tag "MyTag" to the package "MyPackage, use
+
+    curl -X POST http://tetriserver.example.com/v1/tags/MyTag/MyPackage
+
+To remove the tag
+
+    curl -X DELETE http://tetriserver.example.com/v1/tags/MyTag/MyPackage
+
+
+## Pruning
+
+Tetrifact has automated pruning to delete older packages to prevent disk overuse. All prune settings are kept in config.yml.
+
+Pruning is disabled by default, to enable it use
+
+    ...
+    PruneEnabled : True
+    ...
+
+Pruning behaviour works in brackets. A bracket is a period for the age of packages, and x number of packages of a given age can be kept. Using the example below
+
+    PruneBrackets :
+    -   Days: 1
+        Amount: -1
+    -   Days : 3
+        Amount : 5
+    -   Days : 5
+        Amount : 4
+
+The first bracket is for all packages up to 1 day old, an amount of -1 means prune none, ie, keep all packages that are up to a day old. The second bracket indicates that for all packages created more than 1 ago and less than 3 days ago, keep 5 packages.
+
+Pruning can ignore packages marked with one or more tags, use the list 
+
+    PruneIgnoreTags:
+    -   MyKeepTag
+    -   MyImportTag
+
+Pruning runs at 2am server time by default, you can change this with
+
+    PruneCronMask: "0 3 * * *"
+
+

src/Tetrifact.Core/PruneService.cs

@@ -82,13 +82,15 @@ public void Prune()
 
             _log.LogInformation("*************************************** Finished prune exectution. **************************************************************************");
 
+            return;
         }
 
         public PrunePlan GeneratePrunePlan()
         {
-            // sort brackets oldest to most recent, clone to change settings without affecting original instances
+            // sort brackets newest to oldest, clone to change settings without affecting original instances
             IList<PruneBracketProcess> processBrackets = _settings.PruneBrackets
                 .Select(p => PruneBracketProcess.FromPruneBracket(p))
+                .OrderBy(p => p.Days)
                 .ToList();
 
             IList<string> taggedKeep = new List<string>();
@@ -104,7 +106,7 @@ public PrunePlan GeneratePrunePlan()
             foreach(PruneBracketProcess pruneBracketProcess in processBrackets) 
             {
                 pruneBracketProcess.Ceiling = ceiling;
-                pruneBracketProcess.Floor = pruneBracketProcess.Ceiling.AddDays(-1 * pruneBracketProcess.Days);
+                pruneBracketProcess.Floor = utcNow.AddDays(-1 * pruneBracketProcess.Days);
                 ceiling = pruneBracketProcess.Floor;
             }
 
@@ -137,7 +139,7 @@ public PrunePlan GeneratePrunePlan()
                     continue;
                 }
 
-                report.Add($"Package \"{packageId}\" fits into prune bracket {matchingBracket}. This bracket now contains {matchingBracket.Keep.Count} keep packages.");
+                report.Add($"Package \"{packageId}\" fits into prune bracket {matchingBracket}. This bracket currently contains {matchingBracket.Keep.Count} packages for keeping.");
                 matchingBracket.Found++;
 
                 // try to find reasons to keep package