keyboard/mouse context mgrs, printable width of sequences, sequence parsers #27
Conversation
- add unicode-derived 'Keystroke' class, which is yielded by the generator method _resolve_multibyte, adding additional properties, .is_sequence (bool), .code (int), and .name (str). - detect encoding and initialize incrementaldecoder in __init__, used by _resolve_multibyte. - inherit curses key capability names, values, and create multibyte sequence pattern lookup in __init__ - add _resolve_keycode method, returns key capability name for an integer. - add _resolve_multibyte(text) method, given a string of multibyte input, 'text', returns instances of Keystroke class, which may be checked for .is_sequence, .code, and .name.
Add wrap method to Terminal, with accompanying AnsiWrapper class, a
derived version of textwrap.TextWrapper, with a modified version of
_wrap_chunks that is sequence-safe.
A module-global function, 'ansiwrap' is exposed. Sequences are detected
using a range of byte matching and regular expressions. This sequence
padding code been tested thoroughly on ANSI Art from the "Dark Domains"
DVD by "ACiD Productions". Example usage:
>> import blessings
>> t = blessings.Terminal()
>>> for num, byte in enumerate('Pony Express, Choo! CHooooooOOoooOOo!'):
... seq += t.color(num % 7) + t.bold + byte + t.normal
>>> import textwrap
>>> print u'\n'.join(textwrap.wrap(seq, 15))
(as you can see, textwrap fails horribly !!)
>>> print u'\n'.join(t.wrap(seq, 15))
Pony Express,
Choo!
CHooooooOOoooOOo!
winning !!
Continuing the example from the previous commit, it is now possible:
>>> print t.center(seq)
Pony Express, Choo! CHooooooOOoooOOo!
Instantate AnsiString from Terminal, so AnsiString contains these padding methods, requires less instantiation.
also readability from pexpect
* add _conanical and _echo terminal state helpers and related functions. * remove another magic bit, '27', by explicitly setting to curses.ascii.ESC * fix for incremental decoder * add warning for failed decodes in _resolve_multibyte * placeholder for inkey() with docstrings of expected behavior
use tty.LFLAGS instead of magic attrs[3] bit. Also rename 'attrs' references to 'mode' to match python standard module naming conv.
in the usual style of mine, fully written but not yet run, this intermittent commit has complete docstrings for the interface in mind. I haven't a win32 platform, but diligent effort was made to support it. This commit also includes a possible fix for windows systems to determine the terminal height and width without calls to fcntl. - remove unused defaultdict import - refactor input stream key capability from __init__ to _init_keystrokes() - remove redundant input key capability definitions - store _state_echo and _state_canonical; necessary for win32 systems - refactor term.stream to term.o_stream to accompany term.i_stream - refactor documented @Property, do_style - refactor _resolve_multibyte to _resolve_msb, consistent reference 'MSB' can be either 'escape sequence' for application keys, or utf-8 bytes as part of a hamsterface emoticon.
getch() so far works in non-conanical mode, as unimpressive as it is. various issues with multibyte sequence decoding, currently looks like the loop timing and buffering at issue.
This is what the curses wrapper is missing; it is very obnoxious when developing curses when your program crashes, you have to type 'reset' after each crash That's fine, we simply check the state and always restore it on __del__. If python throws an exception and kicks you out while the terminal is in cbreak mode, a warning is issued to let you know it was restored after doing so. This will appear after your traceback. So it won't be any suprised as to what happened.
also sprinly various print debugs for MSB, there is a timing issue related to the 2nd byte of a '\x1b[A' sequence failing the kbhit() check. I think i've seen this before, has to do with flushing something correctly at a particular time? will look tomorrow.
| self._does_styling = ((self.is_a_tty or force_styling) and | ||
| # os.isatty returns True if output stream is an open file descriptor | ||
| # connected to the slave end of a terminal. | ||
| self.is_a_tty = (o_stream is not None and os.isatty(o_fd)) |
the default mode is a comprimise of the most compatible and most capable and documented well.
go ahead, draw a box ...
also brought a few changes to WILL_MOVE, WONT_MOVE also assert TypeError for non-ints for things like term.move(1.0)
my nose for python 2.5 doesn't seem to work
plus a test case, and disable the float exception testing, as it fails on older python where only a DeprecationWarning is emitted
|
Okay, got a chance to read over the rest of it. This is a BIG code dump, and I'd like to restructure some parts of it; the API feels geared toward experts in tty terminology and workings, but the draw of blessings is that it's dead-easy to pick up for anyone. However, the mechanics look sound—it's clear you have serious Terminal fu. :-) Rather than nit pick our way to release by commenting line by line on the pull (and enduring the communication lag for each exchange), I'd like to do a series of commits atop master myself, drawing from what you have here. That way, we can start releasing pieces of functionality without waiting for everything (mouse, Windows, etc.) to be perfect. How does that sound to you? Looking forward to 1.6, which will be prominently ballyhooed as the "Jeff Quast release"! |
|
Sounds good to me. Enjoy ! |
|
I am pumped for 1.6! |
|
To clarify on iso8859-1,
If the 8th bit is set high on a non-utf8 session, it is trying to indicate that "meta sends 8-bit output", that is, sets MSB to 1 to indicate Meta was depressed. If this value is decoded as utf-8, it raises UnicodeDecodeError, as it is not a valid UTF-8 start byte. If it is decoded as 'ascii', is also raises exception, as ascii is 7-bit. I made a gist that demonstrates what could happen if you attempted to decode this input through a utf-8 decoder: https://gist.github.com/jquast/5649654 It also displays the values of say, 'M-x' etc. Although there is not any support for binding Meta or Ctrl characters in the proposed patch, the input of any such characters (keyboard inputs 254, but not legal in utf-8), should not raise exception during multibyte sequence decoding. It would simply be a key event that is not currently mapped. If Ctrl and Meta were wanted, I would recommend something like adding additional functions or attributes to the Keypress class that matches a constant returned by Terminal.Meta('x') or Terminal.Ctrl('z'). as meta could be sent as \033[x ("meta sends escape") or b'248', as chr(ord('x') ^ 128) is b'248'. The user would enable/disable this option in their client, and through some configuration interface, such as XResources in X11, or through software UI, chose "meta sends escape" .. or not. when enabled, any meta would be escaped as \033 x. This is the preferred new input method for meta, as it supports both locale + meta. Some poorly programmed terminal software (i think 'tetradraw' does this), will simply check if(chr == 'ø'). It may be very hard to reproduce on OSX, Easier on xterm/linux. I seem to recall using both modes successfully in rxvt many years passed. In "irssi" you would switch windows by using alt+1 -- but for us UTF-8 folk, this now inputs ¡ instead. This can be worked around by simply pressing Escape (\e), then 1, which simulates "meta sends escape", though very slowly ! This has the same escape delay issue as the timer of inkey() presents when encoutering a single \e -- wait for more, or signal a single escape key?? Irssi seems to wait indefinitely for subsequent bytes; it does not provide any single "escape key" binding. For applications that support both Escape and Meta, some timing is required, as I think is already impl... Its up to the implementer what to do about bytes > 128. I could very well instruct the user to press alt+1, and seek '\e1' OR '¡'. But I'm making assumptions about UTF-8 and the user's keyboard. This is why emacs is so Control-Key heavy, and few terminal applications make use of meta -- the meta key is clearly a difficult thing to support ! I'm using python 3 now and noticing my patch doesn't work. Sorry about that.. I just thought I'd let you make any comments before I continued making more changes. Just let me know if you have trouble, no pressure. |
|
Ahh, from iTerm Profiles->Keys -> left/right Alt key acts as: so there you go. The idea is somebody using blessings with these older systems won't throw decoding exceptions when trying to match it with a MBS. Or if somebody really wanted to, they could program blessings to map them. |
|
I can try to bring this all in again, one stage/feature/pull request at a time sometime in the near future. I'm currently unemployed and homeless (I prefer 'nomadic', heh) and could use the distraction |
|
That would certainly make me less afraid of it and more likely to give it a timely review. :-) Sorry to hear about your troubles! Do you know any web stuff? Ever thought about applying at Mozilla? You're clearly a very smart guy. |
|
Unfortunately not any good at web stuff, really. avoided javascript/css/html 2,3,4, and now 5 for a long while, it's sadly a rapidly moving target that I've not kept up with much longer than since html 2.0 came out. the languages themselves I can do fine, its the rapidly evolving "everything else" thats around it. Appreciate the smart guy comment, though :-) (Crazy how employers expect you to do all of front and back-end, having such a diverse range of expertise to get your foot in the door -- seems when I'm actually there, working the back-end, I end up knowing more than the front-end teams, maybe I should just start lying on my resume like the rest of them, lol - http://jeffquast.com/resume.pdf btw) |
|
Two things!
|
|
An overview of the changes made: