Commit message (Collapse) | Author | Age | |
---|---|---|---|
* | add Build/BuildVersion, for use by autobuilders | 2014-06-18 | |
| | |||
* | don't start minimized | 2014-06-17 | |
| | |||
* | crazy quote for spaces | 2014-06-17 | |
| | |||
* | windows no-DOS-box and autostart, at last | 2014-06-17 | |
| | | | | | | | | | Using the crazy but apparently best approach of a VB Script that runs git-annex, in a hidden DOS window. Note that currently the git-annex messages are not directed to daemon.log. Would probably need another layer of script. Also problimatic since the repository may not exist yet. | ||
* | fix build | 2014-06-13 | |
| | |||
* | make EvilSplicer re-box file-embed's ByteStrings | 2014-06-13 | |
| | | | | | | | | | The armel autobuilder started failing: Assistant/WebApp/Types.hs:128:14: primitive string literal must contain only characters <= '\xFF' This fixes that. | ||
* | better heuristic for what is a package-qualified symbol | 2014-05-27 | |
| | | | | | Hopefully this will avoid mangling css files embedded in strings inside spliced code. | ||
* | proper fix for "foo"# -- need MagicHash extension | 2014-05-25 | |
| | |||
* | typo | 2014-05-25 | |
| | |||
* | put in hack for weird new problem with ghc's emitted splices | 2014-05-25 | |
| | |||
* | work around an EvilSplicer bug | 2014-05-10 | |
| | | | | | | This should fix the android build. The EvilSplicer cannot fix up ghc's misformatting of a case expression with a construstor with a lot of parameters, which spans multiple lines. | ||
* | Revert "add debugging for odd OOM or hang on 5 architectures, all where ↵ | 2014-04-22 | |
| | | | | | | | | configure runs sha224sum" This reverts commit dbc01939f3da3653df51f1fe8b2d5da891ea8828. The OOM was cabal's way of telling us it couldn't resolve dependencies. | ||
* | --detach-sign | 2014-04-21 | |
| | |||
* | sign files5.20140421 | 2014-04-21 | |
| | |||
* | fix distributionupdate build | 2014-04-21 | |
| | | | | Including avoiding needing cabal's defines for Utility.URI | ||
* | add debugging for odd OOM or hang on 5 architectures, all where configure ↵ | 2014-04-20 | |
| | | | | runs sha224sum | ||
* | use ustar format and omit filnames with colons. new hackage requirements | 2014-04-11 | |
| | | | | The tarball on hackage will no longer correspond to the git tag. Oh well. | ||
* | Remove wget from OSX dmg, due to issues with cert paths that broke git-annex ↵ | 2014-04-10 | |
| | | | | automatic upgrading. Instead, curl is used, unless the OSX system has wget installed, which will then be used. | ||
* | factor out library code (also used by propellor) | 2014-04-03 | |
| | |||
* | fix up old comments that used format=txt, which is no longer enabled | 2014-03-20 | |
| | |||
* | make configure -Wall clean | 2014-03-10 | |
| | |||
* | Probe for quvi version at run time. | 2014-02-28 | |
| | | | | | Overhead: git annex addurl runs quvi --version once. And more bloat to Annex state.. | ||
* | more distributionupdate fixes | 2014-02-27 | |
| | |||
* | commit before info file build, so that any modified annexed files get updated | 2014-02-21 | |
| | |||
* | windows: Fix process termination code. | 2014-02-13 | |
| | | | | | | | | | | | The ctrl-c hack used before didn't actually seem to work. No haskell libraries expose TerminateProcess. I tried just calling it via FFI, but got segfaults, probably to do with the wacky process handle not being managed correctly. Moving it all into one C function worked. This was hell. The EvilLinker hack was just final icing on the cake. We all know what the cake was made of. | ||
* | fix info file finding | 2014-02-10 | |
| | |||
* | check for out of date info files at end | 2014-02-10 | |
| | |||
* | deal with multiarch | 2014-02-07 | |
| | |||
* | remove debug info | 2014-01-30 | |
| | |||
* | fix install_name_tool library name replacement | 2014-01-15 | |
| | |||
* | fix warning | 2014-01-14 | |
| | |||
* | search DYLD_LIBRARY_PATH for libraries | 2014-01-14 | |
| | |||
* | Refuse to build with git older than 1.7.1.1, which is needed for git checkout -B | 2014-01-13 | |
| | |||
* | add back sha commands on OSX; SIGILL problem fixed | 2014-01-06 | |
| | |||
* | OSX: Avoid using external hash programs, since SIGILL is not fun | 2014-01-01 | |
| | |||
* | windows: add more libraries, these are used by gpg | 2014-01-01 | |
| | |||
* | include libgcc_s in case not pulled in by ldd | 2013-12-31 | |
| | | | | libc can demand-load libgcc_s in some situations, it seems | ||
* | Revert "include libgcc in linux standalone" | 2013-12-31 | |
| | | | | | | This reverts commit 90a28ca0ff7123cd939c90930dec6479da6bd136. ldd pulls it in | ||
* | include libgcc in linux standalone | 2013-12-31 | |
| | |||
* | add main type sig | 2013-12-30 | |
| | |||
* | add git-shell binary to bundles | 2013-12-27 | |
| | |||
* | avoid another encoding crash | 2013-12-27 | |
| | |||
* | -Wall clean | 2013-12-27 | |
| | |||
* | clean up | 2013-12-24 | |
| | |||
* | Include git-receive-pack, git-upload-pack, and git wrappers in the Linux ↵ | 2013-12-24 | |
| | | | | standalone build, and OSX app, so they will be available when it's added to PATH. | ||
* | fix gconvdir | 2013-12-24 | |
| | |||
* | Fix bug in Linux standalone build's shimming that broke git-annex-shell. | 2013-12-24 | |
| | |||
* | remove unused imports | 2013-12-24 | |
| | |||
* | resursively follow lib symlinks, just in case | 2013-12-24 | |
| | |||
* | convert hacky shell linux mklibs code to haskell ; fixing symlink bug | 2013-12-24 | |
| | | | | | | The shell code was nasty, and buggy. New haskell code is much nicer, and it's easy to do complicated calculations to properly convert possibly absolute symlinks between libraries into relative links using it. |