diff options
Diffstat (limited to 'README.DOS')
| -rw-r--r-- | README.DOS | 20 |
1 files changed, 20 insertions, 0 deletions
diff --git a/README.DOS b/README.DOS new file mode 100644 index 0000000..d522b81 --- /dev/null +++ b/README.DOS | |||
| @@ -0,0 +1,20 @@ | |||
| 1 | |||
| 2 | Windows 95 & Windows NT users: | ||
| 3 | |||
| 4 | 1. There's a pre-built executable, bzip2.exe, which | ||
| 5 | should work. You don't need to compile anything. | ||
| 6 | You can run the `test.bat' batch file to check | ||
| 7 | the executable is working ok, if you want. | ||
| 8 | |||
| 9 | 2. The control-C signal catcher seems pretty dodgy | ||
| 10 | under Windows, at least for the executable supplied. | ||
| 11 | When it catches a control-C, bzip2 tries to delete | ||
| 12 | its output file, so you don't get left with a half- | ||
| 13 | baked file. But this sometimes seems to fail | ||
| 14 | under Windows. Caveat Emptor! I think I am doing | ||
| 15 | something not-quite-right in the signal catching. | ||
| 16 | Windows-&-C gurus got any suggestions? | ||
| 17 | |||
| 18 | Control-C handling all seems to work fine under Unix. | ||
| 19 | |||
| 20 | 7 Aug 97 | ||