Merge documentation / RfC regarding application installation

This commit is contained in:
Fischlurch 2011-02-13 20:28:53 +01:00
commit 802831a8b8
16 changed files with 575 additions and 31 deletions

View file

@ -316,7 +316,7 @@ function change_state()
local state="$2"
local nl=$'\n'
local comment="$state$nl//add reason$nl $(date +%c) $(git config --get user.name) <$(git config --get user.email)>$nl"
local comment=".State -> $state$nl//add reason$nl $(date +%c) $(git config --get user.name) <$(git config --get user.email)>$nl"
edit_state "$name" "$state" "$comment"
edit "$name" -4 "endof_comments"
process "$name"

View file

@ -1,2 +1,2 @@
documentation
source for documentation and on some cases generated documentation in its own subdirs
source for documentation and in some cases generated documentation in its own subdirs

View file

@ -5,7 +5,7 @@ GUI Brainstorming Reviewed
State of the GUI
----------------
image::images/screenshot090124-resources.jpg
image:{l}/media/img/design.gui/screenshot090124-resources.jpg[Screenshot 090124]
Updated 18/01/09

View file

@ -10,7 +10,7 @@ GUI concepts and -features and proposals.
* link:Proposal.RichardSpindler.html[Richard Spindler]
* link:Workspaces.html[about Workspace organsiation]
* link:scrolling.html[about scrolling..]
* link:MenuAndShortcuts.html[Ideas for menues and shortcuts]
* link:MenuAndShortcuts.html[Ideas for menus and shortcuts]
- link:TimelineDiscussion.html[Conclusions by Joel Holdsworth]
[icon="warning.png"]

View file

@ -3,7 +3,9 @@ Design Documents: GUI
In the early stages of the project, there was a lot of debate regarding
GUI concepts and -features and proposals.
link:GuiDiscussion/index.html[here] is a collection of documents from these discussions.
link:GuiDiscussion/index.html[Here] is a collection of documents from these early
discussions.
In our discussions, *Workflow* denotes all considerations how certain tasks
can be achieved within the application in the most suitable and stringent fashion.

View file

@ -32,11 +32,11 @@ The Processing layer (or proc) is where the elements from the Session are assemb
==== link:backend/index.html[The Backend]
The backend uses the Low-level model. It allows the rendering and makes a heavy use of the Input/Output System to the screen or to external monitors.
//link:workflow/index.html[The Workflow]
Extra components
----------------
link:application/index.html[The Application]
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

View file

@ -1,7 +1,22 @@
Design Process
==============
*TODO* describe how the design process works and what we denote by *RfC*
How it Works
------------
* Every proposal starts out as _*Idea*_, allowing other people to review and comment on it, while still working out details.
* When the _Idea_ is in a proper shape and worked out in most details it becomes a _*Draft*_. This _Draft_ need to be carefully reviewed, commented, perhaps corrected and rated by the other Developers.
* We decide on some proposals to talk about that another time, these get a _*Parked*_ state.
* At some point we may decide that a _Draft_ becomes a _*Final*_. Usually, this happens in our link:MonthlyMeetings[]. _Final_ Documents will be imported into the git repository (currently to the tiddlywiki).
* Sometimes proposals will become dropped for some reason, this is indicated by changing their state to _*Dropped*_, they still stay in the system for further reference.
Generally you should refrain from editing someone others proposals (except for typo and gramatic fixes) when this is not acknowleded with the original author. Rather only add comments to the pages and end them with `@SIG@` to make it easier to find out who is the contact person for some note.
.Add a new Proposal:
. Check if there is no similar proprosal below! If yes, contact it's author and contribute to that one.
. Think of a good/descriptive *WikiName* for the Proposal. It will be used to create a RfC subpage.
. Read the instructions and add your proposed text.
* link:rfc[RfC accepted]
* link:rfc_pending[RfC pending]

View file

@ -0,0 +1,225 @@
ApplicationInstall
==================
// please don't remove the //word: comments
[grid="all"]
`------------`-----------------------
*State* _Draft_
*Date* _Di 11 Jan 2011 17:00:55 CET_
*Proposed by* Ichthyostega <prg@ichthyostega.de>
-------------------------------------
[abstract]
*********************************************************************************
Lumiera should be a _freely relocatable_ application bundle.
Relying only on the relative folder structure within this bundle, the application
will be fully functional at any location, provided that the external library
dependencies are resolvable using the standard mechanisms of the platform.
The setup mechanism must be obvious, self-explanatory and must not rely
on compiled in magic or buildsystem trickery. Yet packaging into a FSH conforming
installation location should be supported by the same mechanisms.
*********************************************************************************
Description
-----------
//description: add a detailed description:
Lumiera is planned to become a large professional application bundle, relying
on several external resources for proper operation. An installed Lumiera
application will be more like Gimp, Blender, OpenOffice or Eclipse,
not like bash, autotools or emcas.
Besides that, it can be expected that Lumiera frequently will be used in a
project or studio like setup, where the application isn't installed, but just
unZIPped / unTARed and used as-is. Thus, it should be sufficient to unpack
the application bundle and point it to the session file and maybe the
media storage.
The Eclipse platform can serve as a model for the setup of an modern
application of that style: It can be just unpacked, and when looking
into the folder structure, the meaning of the parts is obvious, and the
basic bootstrap is controlled by two short text based INI files.
While Lumiera presumably won't get _that_ heavyweight and is clearly
not intended to become a general business application platform like OSGi --
the underlying principles can serve as a point of reference for modern
development standards.
This leads to the following conclusions:
- we need to define a standard folder layout for the bundle
- the application must not rely on any compiled-in absolute paths
- the executable should fetch the directly accompanying shared modules
- all other lib dependencies should be handled by the system mechanisms
- the bootstrap configuration likewise must not be compiled-in
- this configuration must be human readable and clearly exhibit its purpose
- the same system must be able to adapt to a FSH conforming installation layout
Judging from our current planning and the existing codebase, Lumiera
is on a good way in that direction, yet some cleanup needs to be done,
especially removing convenience shortcuts from the early days of development
and catching up with the repair of some traits of sloppyness here and there.
Library resolution
~~~~~~~~~~~~~~~~~~
In former days, it was common habit to compile-in a hard wired absolute
+RPATH+. This can be considered obsolete practice; for example, the Debian
policy forbids doing so. This is the result from numerous maintainability
problems in the past. On the other hand, the GNU linker and other modern
linkers support a relative resolution of shared modules directly accompanying
an specific executable. The Debian policy allows this, if and only if these
shared modules are installed with the same binary package and only used by
this specific executable(s). Together, this is exactly what we need to
solve our requirement.
Thus, the build process enables the new-style DT-tags in the Elf binary
and sets the +DT_RUNPATH+ with an value relative to +$ORIGIN+, which resolves
to the path of the currently executing binary. Moreover, it is _sufficient_
to set this on the initial executable _only,_ because this creates a common
searchpath for all lib resolution events in the scope of that loaded executable.
Besides that, we need to care that our private libraries have a unique +SONAME+,
in this case all starting with the prefix +liblumiera*+. Note moreover that this
new-style +DT_RUNPATH+ indeed _can_ be overridden by an +LD_LIBRARY_PATH+ in the
environment, should there be the need for very special experiments.
Bootstrap location
~~~~~~~~~~~~~~~~~~
Thus, a single relative library folder becomes the only hard wired start
configuration. In our case, the folder +$ORIGIN/modules+ was chosen. The
root of the package then holds all the binaries depending on these common
internal libraries, that is the +lumiera+ executable and any accompanying
special tools. As usual with such large application bundles, these get
only _symlinked_ into the +/usr/bin+ folder on installation.
For sake of clarity, after starting the executable, the _same location_
is used to load the bootstrap configuration. This configuration in turn
defines all further locations like the extended configuration, project
templates, plugin search path, the GUI module to load, the search path
for icons and GUI resources, project templates and similar basics.
Relative paths and the location of the executable
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
According to the general requirements outlined above, most paths should
be given in a relative fashion. Unfortunately there is no portable solution
for self-discovering the currently running executable. But at least there
is a solution for all current major platforms. Under Linux, this information
can be retrieved from the kernel through the +/proc+ filesystem.
Again for sake of clarity, the same token +$ORIGIN+ was chosen to denote
this executable location (note: this is _not_ the current working directory).
Moreover, due to the folder layout detailed above, this coincides with the
root of the application bundle, thus making for a self-explanatory convention.
Besides +$ORIGIN+, these search paths later on likely will contain locations
below the user's home directory, e.g. +~/.lumiera/themes+
Tasks
~~~~~
// List what needs to be done to implement this Proposal:
* identify what impedes such a modern setup procedure ([green]#✔ done#)
* rectify the folder structure created in the build target
directory ([green]#✔ done#)
* build the executables in a way to allow relative resolution of the
internal shared modules ([green]#✔ done#)
* replace the compiled-in path definitions for plugin loading by a
configurable bootstrap ([green]#✔#)
* add an working library implementation for a config loader ([green]#✔ done#)
* add a mechanism for establishing the path of the current execubable. +
This is _non-portable_ ([green]#✔ done#)
* wire the prepared API in the GUI to use this working config loader
for resolving GUI resources ([green]#✔ done#)
* try to extract the path search code from the existing config loader,
or build a new solution based on standard libraries ([green]#✔ done#)
* introduce an output root directory into the buildsystem, allowing
for package builds ([green]#✔#)
* define a _Debian packaging_ as proof-of-concept ([green]#✔ done#)
Discussion
~~~~~~~~~~
Pros
^^^^
* self-contained
* self-explanatory
* based on _best practices_
* conforming with FSH and Debian policy
Cons
^^^^
* requires work
* raises the bar at the implementation side
* requires an bootstrap sequence to be explicitly performed
on application startup
* breaks with some beloved habits of the Unix community
Alternatives
^^^^^^^^^^^^
//alternatives: explain alternatives and tell why they are not viable:
I can think of two alternatives
. dealing with all those problems _later_
. not making an concept, rather sticking to UNIX habits
The first alternative is indeed worth considering, because we're settling
some things to be really implemented way later, which bears some dangers.
But, on the other hand, it is a common practice known from extreme programming
to deliver early and regularly, which effectively means to set up the deploy
path of an application really early in the development cycle. The rationale
is that -- according to general experience -- the deployment always turns
up some very specific problems and constraints, which can be a serious
threat when discovered late in the development process.
The second alternative isn't really applicable IMHO. The original UNIX philosophy
breeds on an academic setup and really excels with small nifty commandline utils
combined by pipes, each specialised to do a single thing very well. These utils
are more like the objects within our implementation. The concept of large
application software bundles and desktop software was always a bit alien
within the classic UNIX environment.
Rationale
---------
//rationale: Give a concise summary why it should be done *this* way:
This RfC can be seen as an commitment to an professional approach and as
clarification: Traditionally, the Unix community hailed a lot of _black magic_
practices like compiled-in installation paths, macro magic, +sed+ and +awk+
trickery, inline code compiled on-the-fly, relying on very specific and
un-obvious behaviour of some build script, configuration via environment
variables and a lot of similar idioms. These practices might be adequate
in a quickly moving Research & Development setup, but turned out to be
not so helpful when it comes to industrial strength development,
as they are known to lead to maintenance problems.
//Conclusion
//----------
//conclusion: When approbate (this proposal becomes a Final)
// write some conclusions about its process:
Comments
--------
//comments: append below
.State -> Draft
There is now a complete implementation of this concept on my ``proc'' branch. +
Moreover, I was able to define an initial Debian packaging for Lumiera on top
of that implementation.
During that work, I had opportunity to visit various areas of the existing codebase,
which reminded me of several small issues, which seem to become unhealthy when lying
around unfixed for such a long time. Probably I'll start a clean-up initiative and
try to bring these points to discussion separately.
So 13 Feb 2011 20:04:00 CET Ichthyostega <prg@ichthyostega.de>
//endof_comments:

View file

@ -9,7 +9,7 @@ This documentation section contains documentation for both users and developers.
The user manual containing the full reference for using Lumiera :
* link:manual.html[User Manual] _(planned)_
* link:user/manual.html[User Manual] _(planned)_
=== Developer
Informations about the structure of Lumiera, they give an overview on the main components of the application :
@ -23,6 +23,4 @@ Programming-oriented details about the functional layers of Lumiera as well as a
=== Presentation documents
Documents that you can use as you wish to present or promote Lumiera :
* link:white-papers-presentations.html[Whitepapers and Presentations]
* link:/media/index.html[Whitepapers and Presentations]

View file

@ -0,0 +1,64 @@
GDB Python pretty printers
==========================
Since some time, GDB supports Python written extension modules, especially
for pretty printing of data structures. A selection of pre-defined pretty printers
for STL containers is part of the standard distribution. The graphical debugger frontend
provided by the Eclipse CDT automatically uses this debugger provided presentation
to show the value of variables in the detail pane of the variables view, while the
individual variable entries always show the expandable structure view.
Installation notes
------------------
This feature requires an python enabled GDB (>6.8.50). Debian/Lenny isn't enough,
but compiling the GDB package from Debian/Squeeze (GDB-7.1) is straightforward.
Moreover, you need to check out and install the python pretty-printers and
you need to activate them through your +~/.gdbinit+
[source,python]
----
#
# .gdbinit : GDB Config file
#
# 3/2010 - Ichthyo: add python pretty printers for STL
python
import sys
sys.path.insert(0, '/opt/gdb/stlPrettyPrinter')
from libstdcxx.v6.printers import register_libstdcxx_printers
register_libstdcxx_printers (None)
end
----
For more informations link:http://sourceware.org/gdb/wiki/STLSupport[see here..]
.Hello World
[source,C]
----
#include <iostream>
#include <string>
#include <vector>
using std::string;
using std::cout;
using std::endl;
int
main (int, char**)
{
std::string str = "hullo wöld";
std::vector<int> vec (1000);
for (uint i = 0; i < vec.size(); ++i)
vec[i] = i;
cout << str << "\n.gulp.\n";
return 0;
}
----
When selecting the string or the vector in the Eclipse variables view
(or when issuing "print str" at the GDB prompt), the GDB python pretty-printer
should be activated.

View file

@ -0,0 +1,18 @@
Developer HOWTOs
================
This section contains a loose collection of instructions, recipes, tutorials and
similar usefull pieces of information targeted at Lumiera developers. See also
- the general link:{l}/project/faq.html[Lumiera FAQ]
- the link:{ldoc}/user/index.html[User documentation]
== Notepad
- link:DebugGdbPretty.html[Python pretty printers for GDB]
[icon="warning.png"]
WARNING: Website under construction

View file

@ -1,40 +1,34 @@
////
Technote Documentation
======================
Technical Documentation
=======================
* link:http://www.lumiera.org/doxy/[Doxygen generated documentation]
* link:gui/index.html[GUI-Layer]
* link:proc/index.html[Proc-Layer]
* link:backend/index.html[Backend]
* link:build/index.html[Buildsystem]
[icon="warning.png"]
WARNING: Website under construction
////
== Technical Documentation
This documentation section contains technical documentation about Lumiera.
To get an overview of all the internals and components, you might want to read
link:innerCore/the_inner_core.html[Lumiera the Inner Core]
=== Three Layers
== Three Layers
The technical documentation is split in three parts, one for each of the three main layers of Lumiera.
You may want to read the link:../design/index.html[Design Documents] first to get an overview of all the components.
* link:gui/index.html[*Graphical User Interface*] : Documents about the default GTK GUI that comes with Lumiera.
* link:gui/index.html[*Graphical User Interface*] : Documents related to the GTK based Lumiera GUI
* link:proc/index.html[*Processing layer*]
* link:proc/index.html[*Processing layer*] : Documents regarding Session, models and the engine structure
* link:backend/index.html[*Backend system*]
* link:backend/index.html[*Backend system*] : Documents about the data and processing backend
== Tools
=== Tools
.Development
* link:http://www.lumiera.org/doxy/[*Doxygen generated documentation*] : Complete reference for the code of Lumiera.
.Building
* link:build/index.html[*Buildsystem*] : Installation & compilation tools, dependencies and packaging.
.HOWTO
* link:howto/index.html[*Developer HOWTOs*] : Collection of instructions, recipes, hints and similar
materials intended for developers
[icon="warning.png"]
WARNING: Website under construction

View file

@ -5,7 +5,7 @@ User Documentation
* link:manual.html[We need to write a user manual....]
* The following document might become an introductory overview: +
link:outerSpace/lumiera_from_outer_space.html[Lumiera from Outer Space]
link:outerSpace/lumiera_from_outer_space.html[Lumiera (as seen) from Outer Space]
* link:outerSpace/Glossary.html[Glossary of common terms]
[icon="warning.png"]

View file

@ -0,0 +1,195 @@
Building Lumiera from source
============================
Currently, after building the application, you can try out the Lumiera GUI
and you can run the test suite.
For building Lumiera, besides the GNU C/C++ compiler (Version 4.X), you will
need:
* http://git-scm.com/[git]
* http://www.gnu.org/software/libtool/[libtool]
* http://www.boost.org/[Boost libraries]
* http://gmerlin.sourceforge.net/[GAVL library]
* *NoBug* (see below)
TIP: Generally speaking, when you want to build software, you need the
_development_ version of the packages, containing the headers and pre-built
libraries to link against. Usually, these packages are named `-devel` or `-dev`
.Notes for Ubuntu:
On Hardy you need also:
* libglade2-dev
* libgtkmm-2.4-dev
* libxv-dev
For most Debian based systems, e.g. Ubuntu, you can install these dependencies
with:
------------------------------------------------------------
sudo apt-get install build-essential autoconf git-core libboost-dev \
libboost-program-options-dev libboost-regex-dev libgavl-dev libgdl-1-dev \
libglade2-dev libgtkmm-2.4-dev librsvg2-dev libtool libxv-dev scons valgrind
------------------------------------------------------------
Build directory
---------------
You need a directory for checking out the sources and running the build. This
could be a temp directory, or some "workspace" directory below your home
directory. We'll call it _workspace directory_ from now on.
Lumiera specific libraries
--------------------------
Now that you have your basic build system setup, Lumiera needs a few more
special support libraries. Currently you need to build those from source and
install them, so the Lumiera build process can pick them up.
WARNING: Note that the following procedures will try to install files into your
base system below `/usr/local`.
To do so, you'll need administrative permissions for the machine you're working
on. These additions might interfere with other libraries installed by your
package manager (if you get into trouble updating your system later on,
sometimes you need to remove these manually built libraries).
Installing NoBug
~~~~~~~~~~~~~~~~
*NoBug* is an instrumentation and diagnostics library. +
Go into the mentioned _workspace direcory_ (maybe create a new
directory). First, let's get the NoBug source code:
------------------------------------------------------------
git clone git://git.pipapo.org/nobug
------------------------------------------------------------
This will create a (sub)directory called nobug with source code in your current
directory. +
Compile NoBug with the following commands
------------------------------------------------------------
cd nobug
autoreconf -i
mkdir build
cd build
../configure
make
make install
------------------------------------------------------------
Installing GDL
~~~~~~~~~~~~~~
The *GNOME Docking library* is generally available through your package manager,
but we contributed some improvements, which are only available in the very
recent development versions of GDL. Thus, for now we created a special package,
which doesn't interfere with an existing (older) installation of GDL.
Ubuntu 9.04 note: intltool-update is not patched, you must add
/usr/share/intltool-debian/ to get the gdl-package to configure correctly (JSC).
------------------------------------------------------------
git clone git://git.lumiera.org/gdl-package
cd gdl-package
./configure
make
sudo make install
------------------------------------------------------------
For more detailed instructions on how to build GDL (also how to build it into a
Debian package) see
http://lists.lumiera.org/pipermail/lumiera/2009-April/000891.html[this message +
on the Lumiera mailing list].
check library linkage
^^^^^^^^^^^^^^^^^^^^^
The compile will warn you to add various directories to /etc/ld.so.conf and then
run ldconfig. This will allow your dynamic liker to pick up the newly built
libraries later when you try to start Lumiera. If you don't want to reconfigure
your system and add `/usr/local/lib` to the mentioned linker configuration, you
may alternatively just add the directories to your LD_LIBRARY_PATH environment
variable.
Either way, check that all libraries are accessible and OK:
------------------------------------------------------------
sudo ldconfig -v | grep 'gdl-lum\|nobug'
------------------------------------------------------------
and you should get a list of the libraries, part of which should look like this:
------------------------------------------------------------
libnobug.so.0 -> /usr/local/lib/libnobug.so.0.0.0
libnobugmt.so.0 -> /usr/local/lib/libnobugmt.so.0.0.0
libgdl-lum.so.0 -> /usr/local/lib/libgdl-lum.so.0.0.0
------------------------------------------------------------
or similar. If any if this libs are not listed, investigate why before continuing.
Building Lumiera
----------------
Lumiera has two maintained (and equivalent) build systems: *scons* and
*autotools*. You can pick the one you feel more comfortable with.
When trying to build a development version of Lumiera, it might well be that at
times one of the builds doesn't work temporarily. It is always a good idea to
check the current build stats from our *builddrone*, which automatically builds
the latest version from master repository. + Please have a look at this
http://lumiera.org/builddrone/table.html[current build stats]-page.
Next, after having built and installed the prerequisite libraries, go into the
_workspace directory_ to retrieve the Lumiera source code and build it.
* to retrieve the source code with git and then build with *scons*:
+
(see available build options by issuing `scons -h` )
+
-----------------
git clone git://git.lumiera.org/LUMIERA
cd LUMIERA
scons
-----------------
+
* alternatively, if you prefer building with *autotools*:
+
-----------------
git clone git://git.lumiera.org/LUMIERA
cd LUMIERA
autoreconf -fi
mkdir build
cd build
../configure
make
-----------------
maybe build and run the test suite by issuing `scons check` or `make check`
This will take some time.
NOTE: you can not _install_ Lumiera currently. Do not try it, it will not work.
Just run it from the build directory.
After the build has finished successfully, you should be able to start Lumiera.
Currently, this will bring up the GUI, without any further functionality
* for autotools build: issue `./lumiera` (from within the `build` subdirectory)
* for scons build:
+
-------------------
cd bin
./lumiera
-------------------
You should see something like (screenshot from 1/2009):
image:/media/img/design.gui/screenshot090124-resources.jpg[]

View file

@ -0,0 +1,26 @@
Contributing to Lumiera
=======================
Nothing is easier, follow the
http://www.kernel.org/pub/software/scm/git/docs/gittutorial.html[the basic
instructions], notably the following parts:
------------------------------------------------------------
$ git config --global user.name "Your Name Comes Here"
$ git config --global user.email you@yourdomain.example.com
------------------------------------------------------------
Then you are ready to go, you can edit and commit the lumiera code locally in
your cloned repository. Please do small commits which fix/improve only one
single thing and use meaningful commit messages.
Check that you didn't break anything, by running the testsuite (see above)
Finally you can push your changes to the lumiera server to the 'mob' repository:
------------------------------------------------------------
$ git push git://git.lumiera.org/lumiera/mob master:refs/heads/YOURNAME
------------------------------------------------------------
This creates a new branch 'YOURNAME' on the mob repository. Then you notify the
other devs on the mailinglist and they may merge your code into the mainline.

View file

@ -0,0 +1,7 @@
Tutorials
=========
This page contains tutorials for beginners interested in Lumiera development.
* link:building.html[Building Lumiera from source]
* link:contributing.html[Contributing to Lumiera]