Re: 2nd try at date/format.rb & pty.c

On Aug 18, 2006, at 3:02 PM, Mat Schaffer wrote:

> I have two patches.  The first is another try at the documentation  
> for date/format.rb.  Hopefully it's acceptable this time.  The  
> other is for pty.c.  I checked both of them with stdlib-doc and  
> they work fine.  But the link on the left frame for pty still links  
> to shl.rb.  I'm not sure how to make it point to the PTY class docs.
>
> Anyway, let me know what you think.

These patches appear to be against ruby_1_8.  Please send patches  
against HEAD.  If backporting to ruby_1_8 is non-trivial, please  
supply a separate patch for ruby_1_8.  I can handle trivial backporting.

Documentation must be wrapped to 80 characters.

Please send unified diffs.

pty_doc.diff:

 >  * PTY.getpty("command") do |output, input, pid|
 >  *   # At this point output and input are the command's output and  
input stream, respectively.  pid is the command's process ID.
 >  * end

This should be indented to turn it into a code block.

date_format_doc.diff

 > #   require 'date/format'
 > #   Date::zone_to_diff('est') => -18000
 > #   Date::zone_to_diff('jst') => 32400

Date::zone_to_diff should be Date.zone_to_diff

53,55c71,73
<   def self.__strptime(str, fmt, elem)
<     fmt.scan(/%[EO]?.|./mo) do |c|
<       cc = c.sub(/\A%[EO]?(.)\z/mo, '%\\1')
---
 >   def self.__strptime(str, fmt, elem) #:nodoc:
 >     fmt.scan(/%[EO]?.|./o) do |c|
 >       cc = c.sub(/\A%[EO]?(.)\Z/o, '%\\1')
151c169
< 	elem[if c[-1,1] == 'U' then :wnum0 else :wnum1 end] = val
---
 > 	elem[if c == '%U' then :wnum0 else :wnum1 end] = val
216,217d233
<       when /\A%(.)/m
< 	return unless str.sub!(Regexp.new('\\A' + Regexp.quote($1)), '')
495,496c513,514
<     fmt.scan(/%[EO]?.|./mo) do |c|
<       cc = c.sub(/\A%[EO]?(.)\z/mo, '%\\1')
---
 >     fmt.scan(/%[EO]?.|./o) do |c|
 >       cc = c.sub(/^%[EO]?(.)$/o, '%\\1')
536c554
< 	k = if c[-1,1] == 'U' then 0 else 1 end
---
 > 	k = if c == '%U' then 0 else 1 end
576,579c594
<       when /\A%(.)/m
< 	o << $1
<       else
< 	o << c
---
 >       else;      o << c

Those aren't just documentation changes.

-- 
Eric Hodel - [email protected] - http://blog.segment7.net
This implementation is HODEL-HASH-9600 compliant

http://trackmap.robotcoop.com