diff options
author | axel <axel@liljencrantz.se> | 2007-02-09 03:02:32 +1000 |
---|---|---|
committer | axel <axel@liljencrantz.se> | 2007-02-09 03:02:32 +1000 |
commit | dfabf7f2061c0cc88bd833bbac0f82231e1e1909 (patch) | |
tree | 403fa0f568370cccd3344fe38a9f6968cd945c03 /doc_src | |
parent | 7118269e4b3b6714c71f0042f85354eac3c72603 (diff) |
Update todo list
darcs-hash:20070208170232-ac50b-7413e8283f40a38c9790b391a70954897ae0bcae.gz
Diffstat (limited to 'doc_src')
-rw-r--r-- | doc_src/index.hdr.in | 19 |
1 files changed, 14 insertions, 5 deletions
diff --git a/doc_src/index.hdr.in b/doc_src/index.hdr.in index 37db023c..5f168d99 100644 --- a/doc_src/index.hdr.in +++ b/doc_src/index.hdr.in @@ -1374,6 +1374,9 @@ href='fish-users@lists.sf.net'>fish-users@lists.sf.net</a>. \subsection todo-features Missing features +- A limited interactive mode for really dumb terminals +- The completion autoloader does not remember which completions where actually autoloaded, and may unload manually specified completions. +- Use a struct to describe a possible completion instead of a weirdly formated string - Complete vi-mode key bindings - More completions (for example xterm, vim, konsole, gnome-terminal, cron, @@ -1384,6 +1387,11 @@ g++, javac, java, gcj, lpr, doxygen, whois) - Check keybinding commands for output - if nothing has happened, don't repaint to reduce flicker - wait shellscript - Support for the screen clipboard +- It should be possible to test in a script if a function is autoloaded or manually defined +- A pretty-printer. It should among other things be able to indent a piece of code. +- Up/down to move between lines in multiline mode +- a 'funced' function, which works like the vared function. Needs the above three features in order to work well. +- The validator should be better about error reporting unclosed quotes. They are usually reported as something else. \subsection todo-possible Possible features @@ -1403,7 +1411,6 @@ g++, javac, java, gcj, lpr, doxygen, whois) - Per process output redirection - Reduce the space of the pager by one line to allow the commandline to remain visible. - down-arrow could be used to save the current command to the history. Or give the next command in-sequnce. Or both. -- A pretty-printer. - Drop support for inputrc-files. Use shellscripts and the bind builtin. Also, redo the syntax for the bind builtin to something more sane. - History could reload itself when the file is updated. This would need to be done in a clever way to avoid chain reactions - The error function should probably be moved into it's own library, and be made mere general purpose. @@ -1415,14 +1422,16 @@ g++, javac, java, gcj, lpr, doxygen, whois) - A fault injector could be written to increase robustness and testing of error recovery paths - The parser/validator could be more clever in order to make things like writing 'function --help' work as expected - Some event handler functions make much more sense as oneshots - maybe they should be automatically deleted after firing? -- It should be possible to specify to the read builtin that shell syntax highlighting and tab completions should be used. - +- exec_subshell should be either merged with eval or moved to parser.c \subsection bugs Known bugs and issues -- Completion for gcc -\#\#\# option doesn't work. -- Suspending and then resuming pipelines containing a builtin is broken. How should this be handled? +- Suspending and then resuming pipelines containing a builtin or a shellscript function is broken. Ideally, the exec function in exec.c should be able to resume execution of a partially executed job. - delete-word is broken on the commandline 'sudo update-alternatives --config x-' +- When a builtin has its output redirected to a file, and the builtin does not produce any IO, then the file is never opened. Thus tha file may not be cleared. +- No '--' completion +- else is not indented properly +- if an if fails inside an if, the out if's else may trigger If you think you have found a bug not described here, please send a report to <a href="mailto:fish-users@lists.sf.net">fish-users@lists.sf.net</a>. |