remove jobs

UniStuttgart-VISUS · Jul 4, 2024 · 072eb1d · 072eb1d
1 parent 5586bfe
commit 072eb1d
Show file tree

Hide file tree

Showing 51 changed files with 26 additions and 1,362 deletions.
diff --git a/core/include/mmcore/job/AbstractJob.h b/core/include/mmcore/job/AbstractJob.h
diff --git a/core/include/mmcore/job/AbstractThreadedJob.h b/core/include/mmcore/job/AbstractThreadedJob.h
diff --git a/core/src/job/AbstractJob.cpp b/core/src/job/AbstractJob.cpp
diff --git a/core/src/job/AbstractThreadedJob.cpp b/core/src/job/AbstractThreadedJob.cpp
diff --git a/docs/manual.md b/docs/manual.md
@@ -629,62 +629,6 @@ It will need to wait if the correct data is not available, yet, which can reduce
 
 -->
 
-<!-- ###################################################################### -->
-<!-- TODO: UPDATE or DEPRECATED
-## Jobs
-
-This chapter discusses the job concept available in MegaMol. 
-Especially, how jobs can be used for data conversion. 
-Examples are based on the project script files available in the installed `examples` directory.
-
-### Job Instance
-
-Jobs are the second type of instances available at the MegaMol runtime (compare view instances in section [Views](#views)). 
-The primary difference is the `<job>` tag as primary instance tag. 
-Similarly, to the viewmod attribute, the `<job>` tag specifies a jobmod module as entry module.
-
-```xml
-    <job name="convjob" jobmod="job">
-```
-
-One significant limitation of this release is that the MegaMol Configurator is only able to edit view instance descriptions, see section [Configurator](#configurator). 
-If you want graphical assistance in creating a job description, the recommended way is to create a view instance description with all required modules and calls. 
-Use a `DataWriterJob` module as an entry point. 
-Save the corresponding project file and edit it manually with a text editor. 
-Replace the `<view>` tags with the similarly behaving `<job>` tags and adjust the corresponding attributes.
-
-### Converting to MMPLD
-
-The MegaMol Particle List Data file format (MMPLD) is a very fast loading binary memory dump of MegaMol meant for small and mid-sized datasets (1-10 mio. particles). MegaMol can convert most of its supported file formats to MMPLD file format. More precisely, all file formats which are loaded by a MegaMol module supporting the `MultiParticleDataCall` can be converted to MMPLD. For this, specify a converter job using a `DataWriterJob` module and a `MMPLDWriter` module. The content of the project file `makemmpld.mmprj` which can be used to convert data into the MMPLD file format, is shown below.
-
-```xml
-    <?xml version="1.0" encoding="utf-8"?>
-    <MegaMol type="project" version="1.3">
-
-    <job name="convjob" jobmod="job">
-        <module class="SIFFDataSource" name="data" />
-
-        <module class="DataWriterJob" name="job" />
-        <module class="MMPLDWriter" name="writer" />
-        <call class="DataWriterCtrlCall" from="job::writer" to="writer::control" />
-        <call class="MultiParticleDataCall" from="writer::data" to="data::getdata" />
-
-            </job>
-    </MegaMol>
-```
-
-The entry module for data conversion is of class `DataWriterJob`, see line 11. This job module controls writing several files into a new data set. The output is implemented in corresponding write modules, like the `MMPLDWriter`, see line 12. This writer module is then connected to a module providing the data. In the simplest scenario, this is directly a data loader module. The above example selects one module from several options, in the same way the data viewing project does (see section [Project Files](#project-files)). The job is instantiated similarly using the command line. The paths for `makemmpld.mmprj` and `inputfile.siff` might need to be adjusted:
-
-    $ mmconsole -p makemmpld.mmprj -i convjob j -v j::data::filename inputfile.siff -v j::writer::filename outputfile.mmpld
-
-The input file name and output file name are explicitly specified using the -v arguments. 
-The job execution starts immediately. 
-After data is written, MegaMol terminates itself. 
-
-To convert from other file formats, for which a corresponding loader does exist, you should be able to adjust this project file.
-
---> 
-
 <!-- ###################################################################### -->
 -----
 ## Reproducibility

diff --git a/plugins/cluster/include/cluster/ClusterController.h b/plugins/cluster/include/cluster/ClusterController.h
@@ -9,7 +9,6 @@
 
 #include "mmcore/CalleeSlot.h"
 #include "mmcore/Module.h"
-#include "mmcore/job/AbstractThreadedJob.h"
 #include "mmcore/param/ParamSlot.h"
 #include "vislib/net/DiscoveryListener.h"
 #include "vislib/net/DiscoveryService.h"
@@ -25,9 +24,7 @@ class ClusterControllerClient;
 /**
  * Class implementing the cluster rendering master job
  */
-class ClusterController : public job::AbstractThreadedJob,
-                          public Module,
-                          public vislib::net::cluster::DiscoveryListener {
+class ClusterController : public Module, public vislib::net::cluster::DiscoveryListener {
 public:
     /** DiscoveryService::PeerHandle */
     typedef vislib::net::cluster::DiscoveryService::PeerHandle PeerHandle;
@@ -113,7 +110,7 @@ class ClusterController : public job::AbstractThreadedJob,
      * @return The application dependent return code of the thread. This
      *         must not be STILL_ACTIVE (259).
      */
-    DWORD Run(void* userData) override;
+    DWORD Run(void* userData) /*override*/; // TODO old interfaced used by Jobs, currently has no trigger.
 
     /**
      * This method will be called, if a new computer was found

diff --git a/plugins/cluster/src/ClusterController.cpp b/plugins/cluster/src/ClusterController.cpp
@@ -41,8 +41,7 @@ const char* cluster::ClusterController::DEFAULT_CLUSTERNAME = "MM04RC";
  * cluster::ClusterController::ClusterController
  */
 cluster::ClusterController::ClusterController()
-        : job::AbstractThreadedJob()
-        , Module()
+        : Module()
         , vislib::net::cluster::DiscoveryListener()
         , cdsNameSlot("cdsName", "Name of the rendering cluster")
         , cdsPortSlot("cdsPort", "The ip port to be used by the cluster discovery service.")
@@ -227,7 +226,7 @@ void cluster::ClusterController::release() {
 DWORD cluster::ClusterController::Run(void* userData) {
     const unsigned int sleepTime = 250;
 
-    while (!this->shouldTerminate()) {
+    while (true /*!this->shouldTerminate()*/) { // TODO old interfaced used by Jobs.
 
         // update cluster discovery settings
         if (this->cdsNameSlot.IsDirty() || this->cdsPortSlot.IsDirty() || this->cdsRunSlot.IsDirty()) {

diff --git a/plugins/datatools/src/CSVWriter.cpp b/plugins/datatools/src/CSVWriter.cpp
@@ -86,9 +86,3 @@ bool megamol::datatools::CSVWriter::run() {
 
     return true;
 }
-
-
-bool megamol::datatools::CSVWriter::getCapabilities(core::DataWriterCtrlCall& call) {
-    call.SetAbortable(false);
-    return true;
-}