no commit message given

Author: veltzer

README.md

@@ -1,8 +1,23 @@
+## kcpp
+
+version: 0.0.1
+
+description: C++ framework for the linux kernel
+
+website: https://veltzer.github.io/kcpp
+
+## build
+
+![build](https://github.com/veltzer/kcpp/workflows/build/badge.svg)
+
+
+## contact
+
+chat with me at [![gitter](https://badges.gitter.im/Join%20Chat.svg)](https://gitter.im/veltzer/mark.veltzer)
+
 POC for writing C++ modules for the Linux kernel
 ================================================
 
-The problem:
-------------
 C++ has lots of issues which are not dealt with in the kernel.
 * Problems in linking.
 * Exception handling that needs to be turned off.
@@ -11,8 +26,6 @@ C++ has lots of issues which are not dealt with in the kernel.
 since they conflict with many C++ constructs ('class' for instance and more).
 * The kernel build system does not compile C++ correctly, or even at all.
 
-The idea:
----------
 * separate the "top part" - the one that is getting called by Linux, like driver
 entry and exit points, interrupt functions, tasklets, whatever and handle them in a C layer.
 expose APIs for all of these facilities and other low level facilities (printk, hardware access
@@ -22,8 +35,6 @@ not include the kernel headers and are C++ friendly. In the top part write a fra
 device driver writing based on OO concepts (inherit from Device and implement the relevant
 methods).
 
-Future ideas:
--------------
 * Add ability to include header files directly in C++ code.
 Progress: have 'copy_headers' target in makefile
 which also applies patch but headers are still hard to
@@ -32,12 +43,10 @@ This is to be developed as a patch to the linux kernel (big project - watch out!
 may never happen...)
 * add a framework for writing char drivers
 
-Related projects:
------------------
 * http://code.google.com/p/kernelcpp/
 They do not have a buffer layer but instead do "extern C" before kernel headers.
 Try to see if this works for a big example.
 They have all kinds of other ideas about constructors for global object and so on.
 * http://www.drdobbs.com/cpp/c-exceptions-the-linux-kernel/229100146
-	
-	Mark Veltzer <[email protected]>, 2011-2020
+
+Mark Veltzer, Copyright © 2011, 2012, 2013, 2014, 2015, 2016, 2017, 2018, 2019, 2020, 2021, 2022, 2023

config/version.py

@@ -0,0 +1,2 @@
+""" version of this software """
+tup = (0, 0, 1)

snipplets/main.md.mako

@@ -0,0 +1,37 @@
+POC for writing C++ modules for the Linux kernel
+================================================
+
+## The problem:
+C++ has lots of issues which are not dealt with in the kernel.
+* Problems in linking.
+* Exception handling that needs to be turned off.
+* Memory management functions (like new and delete) which need to be implemented.
+* The kernel headers are not suitable for inclusion for a C++ compiler
+since they conflict with many C++ constructs ('class' for instance and more).
+* The kernel build system does not compile C++ correctly, or even at all.
+
+## The idea:
+* separate the "top part" - the one that is getting called by Linux, like driver
+entry and exit points, interrupt functions, tasklets, whatever and handle them in a C layer.
+expose APIs for all of these facilities and other low level facilities (printk, hardware access
+and more) via C functions.
+* write a top part that only accesses these services by well controlled C functions which do
+not include the kernel headers and are C++ friendly. In the top part write a framework for
+device driver writing based on OO concepts (inherit from Device and implement the relevant
+methods).
+
+## Future ideas:
+* Add ability to include header files directly in C++ code.
+Progress: have 'copy_headers' target in makefile
+which also applies patch but headers are still hard to
+work with...
+This is to be developed as a patch to the linux kernel (big project - watch out!
+may never happen...)
+* add a framework for writing char drivers
+
+## Related projects:
+* http://code.google.com/p/kernelcpp/
+They do not have a buffer layer but instead do "extern C" before kernel headers.
+Try to see if this works for a big example.
+They have all kinds of other ideas about constructors for global object and so on.
+* http://www.drdobbs.com/cpp/c-exceptions-the-linux-kernel/229100146

templates/README.md.mako

@@ -0,0 +1,40 @@
+<%!
+    import pydmt.helpers.project
+    import pydmt.helpers.misc
+    import pydmt.helpers.signature
+    import pydmt.helpers.urls
+    import config.project
+    import config.personal
+    import os.path
+    import glob
+    import yaml
+%>## ${pydmt.helpers.project.get_name()}
+
+version: ${pydmt.helpers.misc.get_version_str()}
+
+description: ${config.project.description_short}
+
+website: ${pydmt.helpers.urls.get_website()}
+
+${"##"} build
+
+<%
+	action_files = glob.glob(".github/workflows/*.yml")
+	for action_file in action_files:
+		with open(action_file, "r") as stream:
+			action_name=yaml.safe_load(stream)["name"]
+			context.write(f"![{action_name}](https://github.com/{config.personal.github_username}/{pydmt.helpers.project.get_name()}/workflows/{action_name}/badge.svg)")
+%>
+
+% if hasattr(config.project, "description_long"):
+${config.project.description_long}
+% endif
+
+${"##"} contact
+
+chat with me at [![gitter](https://badges.gitter.im/Join%20Chat.svg)](https://gitter.im/veltzer/mark.veltzer)
+
+% if os.path.isfile("snipplets/main.md.mako"):
+<%include file="../snipplets/main.md.mako" />
+% endif
+${config.personal.fullname}, Copyright © ${pydmt.helpers.signature.get_copyright_years_long()}