Bug#633297: funguloids: FTBFS everywhere: error: reference to 'map' is ambiguous
Hi Paul, thanks for your elaboration. Am 02.10.2012 17:11, schrieb Paul Wise: was lost. At this point I got a bit de-motivated and slack, moved on to other things. So we have to treat the pre-rendered files as source, replace them where possible over time and live with them until then. I don't consider the mpak file format such a critical issue, as long as the file contents are declared free and we have a free tool to handle it. Currently, the mpak.py script does not include a license, but we could ask upstream for one. IIRC you said you were in contact with him? Then we could repair the data files in the Debian package on the fly and finally upload a new package. Please remember that the game is licensed under the zlib license, which does not require a prefered form for modification. If you would like to join upstream, then I will try to fix the remaining issues in the tar2git script (which are because upstream developed Windows and Linux versions separately) and push to git. After that the Arch, Debian, Maemo and Mandriva patches can be merged, the code adapted to not using the mpak format and a new release made. I am not that much interested in joining upstream, but more in fixing what we already have in Debian. However, I'd by happy to help if there's anything I can do. - Fabian -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#633297: funguloids: FTBFS everywhere: error: reference to 'map' is ambiguous
On Thu, 2012-10-04 at 09:26 +0200, Fabian Greffrath wrote: Currently, the mpak.py script does not include a license, but we could ask upstream for one. IIRC you said you were in contact with him? I've sent him a mail just now, I don't expect a response. The alternative to using mpak.py btw is to use the C++ code in mpak.cpp to do the packing/unpacking. Please remember that the game is licensed under the zlib license, which does not require a prefered form for modification. The license is not what I was concerned about. I was concerned about the practicalities of developing the game, without data source it is harder to improve the artwork, just like it is harder to improve the executable/engine if you don't have non-obfuscated C++ code. In addition, Debian has promised to ship source to our users. I am not that much interested in joining upstream, but more in fixing what we already have in Debian. However, I'd by happy to help if there's anything I can do. Ok, thanks for your interest anyway. -- bye, pabs http://wiki.debian.org/PaulWise signature.asc Description: This is a digitally signed message part
Bug#633297: funguloids: FTBFS everywhere: error: reference to 'map' is ambiguous
On Thu, 2012-10-04 at 17:04 +0800, Paul Wise wrote: I've sent him a mail just now, I don't expect a response. Wow, he replied immediately, see the attached files. -- bye, pabs http://bonedaddy.net/pabs3/ ---BeginMessage--- Hi Paul, What is the license for mpak.py? Can you release it under a free license such as the GNU GPL or the BSD/MIT/Expat licenses? Here you go, brand new mpak.py version licensed under the MIT license.. :) Cheers, -- Mika mpak.py Description: Binary data ---End Message--- #!/usr/bin/env python MPAK package handling utility Version 1.5 (Python-implementation) Copyright (c) 2008, 2012, Mika Halttunen. http://www.mhgames.org Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the Software), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions: The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software. THE SOFTWARE IS PROVIDED AS IS, WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - Description follows: This command line tool allows creation and extraction of MPAK (.mpk) packages used in several of my games. MPAK is a simple WAD-like file format of mine, that allows storing the game data in one single .mpk file. I originally had a very crude command line program bit like this one (written in C++), and decided to write this Python-implementation as an Python-programming excercise. So, it's my first Python program. :) Version history: v1.5: Licensed under MIT license (no functional changes) v1.4: The first Python version v1.0 -- v1.31: The original C++ implementation import getopt, sys import os import traceback import struct import binascii import fnmatch import shutil from ctypes import c_uint32 def usage(): Prints the program usage. print MPAK package handling utility print Version 1.5 (Python-implementation) print Copyright (c) 2008, 2012, Mika Halttunen. print print Usage:, sys.argv[0],[switch],-f pakfile.mpk,[file1],[file2], [...], [fileN] print where [switch] is one of the following: print -f, --file=FILE Use package FILE print -c, --create Create a new package with files 'file1' to 'fileN' print -l, --listList the files from given package print -e, --extract Extract all files (by default) from given package. If you print want to only extract some specific files, you can name print them individually, and/or use wildcards (i.e. *.png). print You can supply path where to extract with -p. print -p, --path=PATH Extract to PATH (created if doesn't exist) print -h, --helpPrint this usage text print def errorMsg(msg): Prints an error message and exits. try: pos = traceback.extract_stack(limit=2) if pos: print ERROR: In %s:%s, line %d: % (pos[0][0], pos[0][2], pos[0][1]) else: print ERROR: print \t,msg except: if __debug__: traceback.print_exc() pass sys.exit(2) def separator(): Prints the separator line. print -*75 def computeCRC(file, offset): Computes the CRC32 for the file, starting at given offset. f = open(file, rb) f.seek(offset) crc = 0 # Compute a running CRC32 for the file in 16kb chunks while True: buffer = f.read(16384) if buffer == : break # End of file crc = binascii.crc32(buffer, crc) f.close() return crc def createPackage(pakFile, files): Creates a new MPAK package. This copies the given files into the new package file, writes the file table and closes the package. MPAK doesn't support adding new files to an existing package. print Creating '%s'.. % pakFile if len(files) 1: errorMsg(No input files specified!) separator() # Open the package file for writing out = open(pakFile, wb) # Write the header and reserve 4+4 bytes for CRC32 and file table offset out.write(MPK1) out.write(.*8) # Write each file package = { fileNames: [], fileOffsets: [] } count = 0 for file in files: # Get the basename filename = os.path.basename(file) print ,filename,..., package[fileNames].append(filename) # Get the file size in bytes stats = os.stat(file) # Store the current offset
Bug#633297: funguloids: FTBFS everywhere: error: reference to 'map' is ambiguous
Am 04.10.2012 11:23, schrieb Paul Wise: Wow, he replied immediately, see the attached files. o_O Awesome! I guess we can go and take this to fix the Debian package now, right? -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#633297: funguloids: FTBFS everywhere: error: reference to 'map' is ambiguous
On Thu, 2012-10-04 at 11:38 +0200, Fabian Greffrath wrote: I guess we can go and take this to fix the Debian package now, right? I guess so, I'm going to focus on upstream first though. -- bye, pabs http://bonedaddy.net/pabs3/ signature.asc Description: This is a digitally signed message part
Bug#633297: funguloids: FTBFS everywhere: error: reference to 'map' is ambiguous
Am 04.10.2012 11:48, schrieb Paul Wise: I guess so, I'm going to focus on upstream first though. Alright, would you mind me hacking on it in SVN, though? -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#633297: funguloids: FTBFS everywhere: error: reference to 'map' is ambiguous
On Thu, 2012-10-04 at 12:13 +0200, Fabian Greffrath wrote: Alright, would you mind me hacking on it in SVN, though? I never mind people doing useful work :) -- bye, pabs http://bonedaddy.net/pabs3/ signature.asc Description: This is a digitally signed message part
Bug#633297: funguloids: FTBFS everywhere: error: reference to 'map' is ambiguous
Am 04.10.2012 12:18, schrieb Paul Wise: I never mind people doing useful work :) I never mind a review. ;) -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#633297: funguloids: FTBFS everywhere: error: reference to 'map' is ambiguous
Hi Paul, Upstream has finally gotten back to me, forking it will not be needed since he is going to give me admin access to the project. this was in February. Has this turned out any good? I'd like to see funguloids revived in Debian and got the current package to work with a handfull of patches and the mpak.py script from opensuse. So how should we proceed? - Fabian -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#633297: funguloids: FTBFS everywhere: error: reference to 'map' is ambiguous
On Tue, 2012-10-02 at 15:14 +0200, Fabian Greffrath wrote: this was in February. Has this turned out any good? I'd like to see funguloids revived in Debian and got the current package to work with a handfull of patches and the mpak.py script from opensuse. So how should we proceed? So, the mpak stuff is basically like a proprietary archive file format, like zip/tar. Inside it I discovered a bunch of files, including some XML that was precompiled into a specific format with Ogre tools. Basically it wasn't in any sane format for modification. The plan was to ditch it completely. So I've been working on a script to take the source and binary tarballs, unpack them and the mpak files, strip out non-free bits, prebuilt or generated stuff and embedded code/data copies, then import the result into git. Then I noticed the DirectDraw surface files and that there was no way to generate them in Debian, so that was a blocker for a while, but is resolved now. Then there are all the usual source issues with the graphics and sound; an unknown font was used for the menu, the Photoshop files were lost, the Cubase audio mixing data was lost. At this point I got a bit de-motivated and slack, moved on to other things. So we have to treat the pre-rendered files as source, replace them where possible over time and live with them until then. If you would like to join upstream, then I will try to fix the remaining issues in the tar2git script (which are because upstream developed Windows and Linux versions separately) and push to git. After that the Arch, Debian, Maemo and Mandriva patches can be merged, the code adapted to not using the mpak format and a new release made. -- bye, pabs http://wiki.debian.org/PaulWise signature.asc Description: This is a digitally signed message part
Bug#633297: funguloids: FTBFS everywhere: error: reference to 'map' is ambiguous
On Sat, 2012-02-25 at 12:23 +, peter green wrote: IMO if you have made reasonable attempts to contact upstream and have seen no signs of life then forking (and announcing your fork to other distro packagers) is the corect thing to do. Upstream has finally gotten back to me, forking it will not be needed since he is going to give me admin access to the project. -- bye, pabs http://wiki.debian.org/PaulWise signature.asc Description: This is a digitally signed message part
Bug#633297: funguloids: FTBFS everywhere: error: reference to 'map' is ambiguous
mpak.py needs to be re-implemented or re-licensed because the current implementation as distributed by Arch does not have any license attached to it, so it is not distributable by Debian and Arch folks are violating copyright law. Thanks for explaining why the bug has not been fixed. Ideally though, upstream needs to be revived and the mpak stuff ripped out and replaced with either filesystem accesses or some standard pack format like zip/tar. Seems like a good idea to me. I've been trying to do that for a while but upstream hasn't responded. Since upstream is not responding I am not sure how to deal with the situation. Either mpak.py has to be run at orig.tar.gz creation time or a second tarball needs to be added with the pre-fixed data. Alternatively the project could be completely forked, but that does not seem to be a good way to do things to me. IMO if you have made reasonable attempts to contact upstream and have seen no signs of life then forking (and announcing your fork to other distro packagers) is the corect thing to do. -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#633297: funguloids: FTBFS everywhere: error: reference to 'map' is ambiguous
Back in august pabs tagged this bug as pending, Looking at the SVN (which took a while to find since the PTS still points at the git repo) it seems he did commit it to the repo but soon afterwards put a notice in the changelog saying. * DO NOT UPLOAD, needs to reimplement the mpak.py from archlinux http://aur.archlinux.org/packages/fu/funguloids/PKGBUILD Can anyone explain why mpak.py needs to be reimplemented, why there doesn't seem to be a bugreport about the issue (I searched all the packages open bug reports for mpak.py and found nothing) and why it justifies leaving a FTBFS bug unfixed for so long? Following the link to arch didn't seem to reveal anything to my untrained eye. -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#633297: funguloids: FTBFS everywhere: error: reference to 'map' is ambiguous
The FTBFS bug and the other RC bugs have been open for a long time. There is no point in fixing the FTBFS bug without fixing the crashes that prevent the game from working at all. The package is not in testing or stable anyway so it can basically be considered a new package. To fix the crashes you need to modify the data and for that you need to run mpak.py or maybe some of the C code from the game can be used. mpak.py needs to be re-implemented or re-licensed because the current implementation as distributed by Arch does not have any license attached to it, so it is not distributable by Debian and Arch folks are violating copyright law. Ideally though, upstream needs to be revived and the mpak stuff ripped out and replaced with either filesystem accesses or some standard pack format like zip/tar. I've been trying to do that for a while but upstream hasn't responded. Since upstream is not responding I am not sure how to deal with the situation. Either mpak.py has to be run at orig.tar.gz creation time or a second tarball needs to be added with the pre-fixed data. Alternatively the project could be completely forked, but that does not seem to be a good way to do things to me. -- bye, pabs http://wiki.debian.org/PaulWise signature.asc Description: This is a digitally signed message part
Bug#633297: funguloids: FTBFS everywhere: error: reference to 'map' is ambiguous
Package: funguloids Version: 1.06-8 Severity: serious Justification: fails to build from source (but built successfully in the past) From my build log: | g++ -DHAVE_CONFIG_H -I. -I../include -I../include -I../include/SimpleIni -I/usr/include/lua5.1 -I/usr/include/OIS -pthread -I/usr/include/OGRE -O2 -g -Wall -MT funguloids-ballworm.o -MD -MP -MF .deps/funguloids-ballworm.Tpo -c -o funguloids-ballworm.o `test -f 'ballworm.cpp' || echo './'`ballworm.cpp | In file included from ../include/soundsystem.h:39:0, | from ../include/player.h:33, | from ballworm.cpp:27: | ../include/openalsoundsystem.h:50:9: error: reference to 'map' is ambiguous | /usr/include/c++/4.6/bits/stl_map.h:88:11: error: candidates are: templateclass _Key, class _Tp, class _Compare, class _Alloc class std::map | /usr/include/OGRE/OgrePrerequisites.h:497:9: error: templateclass K, class V, class P, class A struct Ogre::map | ../include/openalsoundsystem.h:50:9: error: 'map' does not name a type | ../include/openalsoundsystem.h:81:2: error: 'SoundMapType' does not name a type | make[2]: *** [funguloids-ballworm.o] Error 1 | make[2]: Leaving directory `/tmp/buildd/funguloids-1.06/src' | make[1]: *** [all-recursive] Error 1 | make[1]: Leaving directory `/tmp/buildd/funguloids-1.06' | make: *** [build-stamp] Error 2 | dpkg-buildpackage: error: debian/rules build gave error exit status 2 Full builds logs are available from: https://buildd.debian.org/status/package.php?p=funguloidssuite=sid -- System Information: Debian Release: wheezy/sid APT prefers unstable APT policy: (500, 'unstable'), (1, 'experimental') Architecture: amd64 (x86_64) Kernel: Linux 2.6.39-2-amd64 (SMP w/4 CPU cores) Locale: LANG=fr_FR.UTF-8, LC_CTYPE=fr_FR.UTF-8 (charmap=UTF-8) Shell: /bin/sh linked to /bin/bash -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org