Re: [code] [textadept] Textadept 7.0 beta 4

From: Ryan Pusztai <rpusztai.att.gmail.com>
Date: Wed, 2 Oct 2013 16:37:26 -0400

Hi Mitchell,

On Wed, Oct 2, 2013 at 4:26 PM, Ryan Pusztai <rpusztai.att.gmail.com> wrote:

> Hi Mitchell,
>
> On Tue, Oct 1, 2013 at 5:34 PM, Ryan Pusztai <rpusztai.att.gmail.com> wrote:
>
>> Hi Mitchel,
>>
>> On Tue, Oct 1, 2013 at 4:46 PM, Mitchell <m.att.foicica.com> wrote:
>>
>>> On Tue, 1 Oct 2013, Mitchell wrote:
>>>
>>> Hi Ryan,
>>>>
>>>> On Tue, 1 Oct 2013, Ryan Pusztai wrote:
>>>>
>>>> So I missed changing 'gui.' to 'ui.', but the error is not helpful at
>>>>> all
>>>>> when figuring out what I missed next. I do currently use Common, but I
>>>>> forked it because I went a different way with the display_name module.
>>>>> You
>>>>> can check out "beta 2" working Common here:
>>>>> https://github.com/**rjpcomputing/ta-common<https://github.com/rjpcomputing/ta-common>
>>>>>
>>>>> This is the hardest update I have had so far. So I think there might
>>>>> be a
>>>>> root cause to the error I am getting.
>>>>>
>>>>
>>>> I'm really sorry you're having so much trouble with the update. I am
>>>> going to take a look at your repo and try and help figure it out.
>>>>
>>>> The primary cause of this problem is that you are probably doing some
>>>> sort of operation at the module 'require' level (before ta is finished
>>>> initializing) that causes `events.UPDATE_UI` to be fired.
>>>> `buffer.get_lexer()` and `buffer.set_lexer()` do not exist until ta
>>>> finishes initializing.
>>>>
>>>
>>> To clarify: while the above statement is valid, it also happens when
>>> errors occur while loading modules (for example, indexing a nil value like
>>> `_M.textadept.whatever`). Sorry if this sounds obvious, but I hadn't
>>> thought about it before.
>>
>>
>> Thanks for your help. I worked on the issue a bit more and have most
>> everything working. What caused the most problem is if I:
>> 1. In init.lua you "require()" something to the _M table but then comment
>> it out.
>> 2. Use that modules functions in the keys.lua file and it indexes into _M
>> and that table does not exist.
>>
>> Example:
>> * init.lua
>>
>>> -- Version control integration
>>> --_M.version_control = require( "version_control" )
>>
>> ...
>>> -- Add custom keyboard shortcuts
>>> dofile( _USERHOME .. "/keys.lua" )
>>
>> * keys.lua
>>
>>> local keys = _G.keys
>>> ...
>>> -- Version control
>>> local vc = _M.version_control
>>> keys[not NCURSES and 'caj' or 'cmj'] = vc.snapopen_project
>>> keys[not NCURSES and 'aj' or 'mj'] = vc.command
>>> --keys[not NCURSES and 'cah' or 'cmh'] = {vc.snapopen_project,
>>> _HOME..'/'}
>>> --if NCURSES then keys.cmg = keys.cmh end
>>
>>
>> That causes a segfault on my system. I hope this helps.
>>
>
> Update: I think I got my modifications working again.
>
> Now onto a feature tweak. I used to make the lua run command start like
> this:
>
>> local m_run = textadept.run
>> m_run.run_command.lua = [[lua -e "io.stdout:setvbuf 'no'" %(filename)]]
>
>
> The reason is so the order of the prints come out correctly. Without it
> for example if you have a lua module that prints to the std::out it can
> come out of order and makes the output confusing. Thoughts?
> --
> Regards,
> Ryan
>
>
I think I found another bug in the "Toggle Block Comment" feature. I have
tested this with Lua and C++ files.

Steps to Follow:

   1. open a file with comments.
   2. Go to the the beginning of the line and press Ctrl + /
   3. the comment is removed, but the cursor moves the the previous line
   (by what appears to be 2 characters)

It does work if you are in the line farther than 2 characters. Please let
me know if you have any questions.

--
Regards,
Ryan
-- 
You are subscribed to code.att.foicica.com.
To change subscription settings, send an e-mail to code+help.att.foicica.com.
To unsubscribe, send an e-mail to code+unsubscribe.att.foicica.com.
Received on Wed 02 Oct 2013 - 16:37:26 EDT

This archive was generated by hypermail 2.2.0 : Thu 03 Oct 2013 - 06:49:38 EDT