I am writing a game engine and I'd like it to have Python scripting as well as support for mods using PhysFS.
My game data is stored something like this:
/
native
scripts
sprites
...
mods
mymodname
scripts
What I want is for the mod scripts to be able to 'import' the native scripts as if they were in the same directory. Is something like that possible using PhysFS?
You could create a symbolic link so that you can link those files/folders are in a higher directory, with PhysFS you can do:
PHYSFS_permitSymbolicLinks()
Then have PhysFS follow your symbolic links, hope this help :-)
EDIT: What I would do is symbolically link /mods/scripts to /native/mods-scripts so that /native/scripts can call mods-scripts (which actually points to /mods/scripts)
[I am the same person who asked the question.]
The solution I used eventually was to modify Python's sys.path when my program starts. This does not pollute the game's data directories with symbolic links and is overall much cleaner.
Related
this question might have been asked before, but I could not find it.
I am on a Linux box. I have py app that runs from a folder called /avt. (example)
I did not write this code, and it has about 12 modules that go with it. I was the lucky engineer to inherit this mess.
this app imports other modules that live under this dir /avt/bin
I want to be able find my modules in the /bin dir no matter where the current working dir is. sometimes the app changes dir to some other sub folders to perform some file I/O. Then should return, but seems like sometimes it does not make it back, because the code will error out with "no such file or directory" error. so I want to test for working dir each time before I do any file I/O to the /bin dir.
As an example, I want to create files in /bin, and then later open those files and read data from them. How can I test to make sure my current working dir is always /avt? and if it is not, then ch.dir to it? Note: it also has to be portable code meaning if must run on any directory structure on any Linux machine.
I tried this code, but it is not very clean I think. Python is not my main language. Is this coding proper and will it work for this? forgive me I don't know how to format it for this forum.
Avtfolder = os.path.realpath(os.path.abspath(os.path.split(inspect.getfile( inspect.currentframe() ))[0]))
if Avtfolder not in sys.path:
sys.path.insert(0, Avtfolder)
if Avtfolder.__contains__('/avt'):
modfilespath = Avtfolder + '/bin'
print 'bin dir is ' + modfilespath
else:
print 'directory lost...'
#write some code here that changes to the root /avt dir
I have a few notes.
First, I'm afraid you are mixing up two problems (or I couldn't tell from the question which one you're facing). These problems are:
I/O to files that can reside in different directories on different machines
Importing Python modules used by your app that can also be in slightly different locations.
The title of the question and some of the text suggests you're dealing with problem 2, whereas references to I/O and "no such file or directory" error point to problem 1.
Those are, however, separate problems and are treated separately. I won't be able to give the exact recipes on both, but here are some suggestions:
For problem 1: I don't think it's a good idea to do some I/O, create files, etc. in the folder where the user installs the Python libraries. It's a folder for Python modules, not data. Also, if the library is installed via setup.py, using pip or easy_install (if it isn't the case now, that can change in the future) then the program will probably habe insufficient permissions to write there, unless invoked as root. And that's right. Create files somewhere else.
As to "how to track the directory changes" part: I must confess I don't quite understand what you mean. Why do you even using the concept of "current directory"? In my mind you should just have some variable such as write_path, data_path, etc. and the code would be
data = open(os.path.join(data_path, 'data.foo'))
dump = open(os.path.join(write_path, 'dump.bar'), 'w')
etc.
Why do you even care where are your libraries located? I don't think it's right, I'd change that. This inspect.currentframe() stuff smells like you really need to rethink the design of the library.
Now, what the location of the libraries matters for is Problem 2. But again, the absolute path shouldn't matter (if it does, change that!). You only need all the modules to be inside one folder (or its subfolders). If they are in the same folder, you're good. import foo will just work. If some are in subfolders, those subfolders should have a file named __init__.py in them, and then they will be seen as modules by Python interpreter, so you'll be able to do from foo import bar, where foo is a subfolder with __init__.py and bar.py in it.
So, try to rewrite it so that you don't depend on where the .py files are. You really shouldn't need to use inspect there at all. On another note, don't use special methods like __contains__ directly unless you really need to. if '/avt' in Avtfolder will do the same.
I want to make a program that moves certain named folders (and all files contained) from directory A to directory B. It was suggested to code in C++. So I was wondering if anyone knew of a simple way to do this, if they could give me a link, and if anyone know's where it's possible to set the directories as variables that can be loaded from a text file. I'm asking this question, because I want to basically have all my program settings and whatnot from the appdata folder since I move between computers alot, be easily transferable.
settings.txt (This is an example of what I mean.)
fldrget = (Folder Name)
fldrdir = (Path to Folder)
fldrplc = (Folder Destination)
Would creating an xml document be a better idea as far as the txt document goes?
Additional Information: OS: Windows(XP, Vista, 7) and I'd like to make this a GUI, but as I'm not familiar with any C language, I'll settle for basics first if anyone can give me a push in the right direction.
Can you provide a bit more information? Are you wanting to write a console application or a GUI? What platform are you targeting? You could use the Win32 API, Boost, Qt, Wx, or a number of others.
You may want to look into using PowerShell. It is much more friendly for a new comer and built-in to Windows 7 and freely available for Windows XP. Just do a google search for moving a directory using PowerShell to get started
Is there a library that will allow easier generation of a simple website using C++ code. This 'website' will then be compiled into a CHM help file (which is the final goal here). Ideally, it will allow generation of pages easily and allow links to be generated between pages easily. I can do this all by hand, but that is going be very tedious and error prone.
I know about bigger libraries such as Wt, but am more interested in smaller ones with little or no dependencies and a need for installation.
You can try CTPP template engine. It is written in C++ is small and quite fast.
Do you need this project to be written in c++? Because if you just need to prepare documentation in CHM I would go with Sphinx. Sphinx is a set of tools written in Python that generate manuals in few formats (chm, html, LaTeX, PDF) from text files (formated using reStructuredText markup language). Those text files could be created by hand or using some application and then combined into one manual using Sphinx. In my work right now we are using this solution to write documentation, because it is very easy to maintain text files (merging, tracking changes etc.) than for example html or doc. Sphinx is used to generate Python language documentation (chm), so it is capable to handle really large project.
I've used the FLATE library every day for ten years and it works flawlessly. It's a piece of cake to use; I can't recommend it enough.
It will definitely do the trick, though probably at a much lower level than you have in mind. It is a C-language source library that you can link with a C++ caller. It's also available as a Perl module, but I haven't used that.
FLATE library
Flate is a template library used to deal with html code in CGI applications. The library includes C and Perl support. All html code is put in an external file (the template) and printed using the library functions: variables, zones (parts to be displayed or not) and tables (parts to be displayed 0 to n times). Using this method you don't need to modify/recompile your application when modifying html code, printing order doesn't matter in your CGI code, and your CGI code is much cleaner.
HTH and good luck!
Are this CHM lib and the related links what you're looking for?
My problem is that I'm having trouble specifying paths for Lua to look in.
For example, in my script I have a require("someScript") line that works perfectly (it is able to use functions from someScript when the script is run standalone.
However, when I run my app, the script fails. I believe this is because Lua is looking in a location relative to the application rather than relative to the script.
Hardcoding the entire path down to the drive isn't an option since people can download the game wherever they like so the highest I can go is the root folder for the game.
We have XML files to load in information on objects. In them, when we specify the script the object uses, we only have to do something like Content/Core/Scripts/someScript.lua where Content is in the same directory as Debug and the app is located inside Debug. If I try putting that (the Content/Core...) in Lua's package.path I get errors when I try to run the script standalone.
I'm really stuck, and am not sure how to solve this. Any help is appreciated. Thanks.
P.S. When I print out the default package.path in the app I see syntax like ;.\?.lua
in a sequence like...
;.\?.lua;c:...(long file path)\Debug\?.lua; I assume the ; means the end of the path, but I have no idea what the .\?.lua means. Any Lua file in the directory?
You can customize the way require loads modules by putting your own loader into the package.loaders table. See here:
http://www.lua.org/manual/5.1/manual.html#pdf-package.loaders
If you want to be sure that things are nicely sandboxed, you'll probably want to remove all the default loaders and replace them with one that does exactly what you want and nothing more. (It will probably be somewhat similar to one of the existing ones, so you can use those as a guide.)
I am currently developing/hacking an image analyzing/transforming tool.
The filters therein will be loaded at runtime using dlopen&co.
My question is where do *nix tools usually put plugins (*.so files) when installed?
bin/program
lib/program/plugins/thisandthat.so
maybe?
Secondly how do I use it and where do I put it during development without installing it. (this is probably the tricky part)
I want to avoid shell-scripts if possible.
thanks in regard
Ronny
Usually /usr/lib/programmname should be a good spot
During development I'd create a command line paramter to specify the plugin search path and just leave the plugins in the build-dir for example.
Consider:
/usr/lib/program/*.so
A good guide for choosing is Filesystem Hierarchy Standard.
Most Linux distribuitions use this standard.
Here is a very short summary.
Place application binary in:
/usr/bin/progname, /usr/local/bin/progname or /opt/progname
Place plugins or library files in:
/usr/lib/progname, /usr/local/lib/progname or /opt/progname/lib
Place host configuration for the application in:
/etc/progname or /etc/opt/progname
Place user configuration in:
$HOME/.progname
Place application manual page in:
/usr/shar/man/man1/
There is separate hierachy for /var. As an example use /var/log/progname for logging.
In responce to caf's comment. I find it very usefull to choose target directory at compile time. Using a $PREFIX also makes it easy to separate devellopment build's from shippment.
Most use /usr/progname, /usr/lib/progname and /etc/progname
Do not forget:
$HOME/.program/
The layout seems sensible. You can, for instance, look in current directory, look up environment variable or command line switch during development. It depends on the details of your development environment and workflow.