WoMan generates different result from man?

WoMan generates different result from man?

[Hongxu Chen]

Hi all,

  I am using `WoMan' other than man to read manpages. But I find there
  are some slight differences when reading the content.  Take the
  manpage of `man' as an example:

  WoMan parses the result of the section number introduction like this:

     1@Executable programs or shell commands 
     2@System calls (functions provided by the kernel) 
     3@Library calls (functions within program libraries) 
     4@Special files (usually found in /dev) 
     5@File formats and conventions eg /etc/passwd 
     6@Games 
     7@Miscellaneous (including macro packages and conventions), e.g. man(7),
      groff(7) 
     8@System administration commands (usually only for root) 
     9@Kernel routines [Non standard] 

  While the result generates by `man' does have `@' but spaces `   ',
  just as follows:
  
       1   Executable programs or shell commands
       2   System calls (functions provided by the kernel)
       3   Library calls (functions within program libraries)
       4   Special files (usually found in /dev)
       5   File formats and conventions eg /etc/passwd
       6   Games
       7   Miscellaneous (including macro packages  and  conventions),  e.g.  man(7),
           groff(7)
       8   System administration commands (usually only for root)
       9   Kernel routines [Non standard]

  The margin in WoMan also seems a little odd.

  So is this a known issue?

  Thanks!
  Hongxu Chen

Re: WoMan generates different result from man?

[Jonathan Groll]

On Mon, 03 Jun 2013 19:39:07 +0800, Hongxu Chen <leftcopy.chx@gmail.com> wrote:

> 
> Hi all,
> 
>  I am using `WoMan' other than man to read manpages. But I find there
>  are some slight differences when reading the content.  Take the
>  manpage of `man' as an example:
> 
>  WoMan parses the result of the section number introduction like this:
> 
>     1@Executable programs or shell commands   2@System calls
> (functions provided by the kernel)   3@Library calls (functions within
> program libraries)   4@Special files (usually found in /dev)   5@File
> formats and conventions eg /etc/passwd   6@Games   7@Miscellaneous
> (including macro packages and conventions), e.g. man(7),
>      groff(7)   8@System administration commands (usually only for
> root)   9@Kernel routines [Non standard] 
> 
>  While the result generates by `man' does have `@' but spaces `   ',
>  just as follows:
>      1   Executable programs or shell commands
>       2   System calls (functions provided by the kernel)
>       3   Library calls (functions within program libraries)
>       4   Special files (usually found in /dev)
>       5   File formats and conventions eg /etc/passwd
>       6   Games
>       7   Miscellaneous (including macro packages  and  conventions),  e.g.  man(7),
>           groff(7)
>       8   System administration commands (usually only for root)
>       9   Kernel routines [Non standard]
> 
>  The margin in WoMan also seems a little odd.
> 
>  So is this a known issue?

Yes. Women are from venus.

Cheers,
Jonathan
--
jjg: Jonathan J. Groll : groll co za
has_one { :blog => "http://bloggroll.com" }
Any other Disclaimer in this mail is Wrong.

Re: WoMan generates different result from man?

[Hongxu Chen]

Jonathan Groll <lists@groll.co.za> writes:

> On Mon, 03 Jun 2013 19:39:07 +0800, Hongxu Chen <leftcopy.chx@gmail.com> wrote:
>
>> 
>> Hi all,
>> 
>>  I am using `WoMan' other than man to read manpages. But I find there
>>  are some slight differences when reading the content.  Take the
>>  manpage of `man' as an example:
>> 
>>  WoMan parses the result of the section number introduction like this:
>> 
>>     1@Executable programs or shell commands   2@System calls
>> (functions provided by the kernel)   3@Library calls (functions within
>> program libraries)   4@Special files (usually found in /dev)   5@File
>> formats and conventions eg /etc/passwd   6@Games   7@Miscellaneous
>> (including macro packages and conventions), e.g. man(7),
>>      groff(7)   8@System administration commands (usually only for
>> root)   9@Kernel routines [Non standard] 
>> 
>>  While the result generates by `man' does have `@' but spaces `   ',
>>  just as follows:
>>      1   Executable programs or shell commands
>>       2   System calls (functions provided by the kernel)
>>       3   Library calls (functions within program libraries)
>>       4   Special files (usually found in /dev)
>>       5   File formats and conventions eg /etc/passwd
>>       6   Games
>>       7   Miscellaneous (including macro packages  and  conventions),  e.g.  man(7),
>>           groff(7)
>>       8   System administration commands (usually only for root)
>>       9   Kernel routines [Non standard]
>> 
>>  The margin in WoMan also seems a little odd.
>> 
>>  So is this a known issue?
>
> Yes. Women are from venus.
Well, WoMan is really different from man...
Are there some workarounds to remedy?

BTW, is WoMan popular in Emacs, or Emacsers typically look up info
through info?
>
> Cheers,
> Jonathan
> --
> jjg: Jonathan J. Groll : groll co za
> has_one { :blog => "http://bloggroll.com" }
> Any other Disclaimer in this mail is Wrong.
>
>

-- 
Regards,
Hongxu Chen

Re: WoMan generates different result from man?

[Stefan Monnier]

> Well, WoMan is really different from man...
> Are there some workarounds to remedy?

woman is meant for those cases where you don't have "man" installed.
I.e. it's (much) better than nothing.

> BTW, is WoMan popular in Emacs, or Emacsers typically look up info
> through info?

Of course, if you have an info manual at hand, it's usually even better
than a man page.


        Stefan

Re: WoMan generates different result from man?

[Dan Espen]

Hongxu Chen <leftcopy.chx@gmail.com> writes:

> Jonathan Groll <lists@groll.co.za> writes:
>
>> On Mon, 03 Jun 2013 19:39:07 +0800, Hongxu Chen <leftcopy.chx@gmail.com> wrote:
>>
>>> 
>>> Hi all,
>>> 
>>>  I am using `WoMan' other than man to read manpages. But I find there
>>>  are some slight differences when reading the content.  Take the
>>>  manpage of `man' as an example:
>>> 
>>>  WoMan parses the result of the section number introduction like this:
>>> 
>>>     1@Executable programs or shell commands   2@System calls
>>> (functions provided by the kernel)   3@Library calls (functions within
>>> program libraries)   4@Special files (usually found in /dev)   5@File
>>> formats and conventions eg /etc/passwd   6@Games   7@Miscellaneous
>>> (including macro packages and conventions), e.g. man(7),
>>>      groff(7)   8@System administration commands (usually only for
>>> root)   9@Kernel routines [Non standard] 
>>> 
>>>  While the result generates by `man' does have `@' but spaces `   ',
>>>  just as follows:
>>>      1   Executable programs or shell commands
>>>       2   System calls (functions provided by the kernel)
>>>       3   Library calls (functions within program libraries)
>>>       4   Special files (usually found in /dev)
>>>       5   File formats and conventions eg /etc/passwd
>>>       6   Games
>>>       7   Miscellaneous (including macro packages  and  conventions),  e.g.  man(7),
>>>           groff(7)
>>>       8   System administration commands (usually only for root)
>>>       9   Kernel routines [Non standard]
>>> 
>>>  The margin in WoMan also seems a little odd.
>>> 
>>>  So is this a known issue?
>>
>> Yes. Women are from venus.
> Well, WoMan is really different from man...
> Are there some workarounds to remedy?
>
> BTW, is WoMan popular in Emacs, or Emacsers typically look up info
> through info?

I have no way of telling, but stopped using woman and use M-x man instead.
If the man page doesn't help and it mentions info, I'll try info.

-- 
Dan Espen

Re: WoMan generates different result from man?

[Emanuel Berg]

Hongxu Chen <leftcopy.chx@gmail.com> writes:

> BTW, is WoMan popular in Emacs, or Emacsers typically look up
> info through info?

WoMan also fails to draw tables.

Check out this dump:

http://user.it.uu.se/~embe8573/man_sv/dumps/woman_man_ascii.png

There you see the man and woman representation of the groff source.

It is absolutely not uncommon to browse the man pages (`M-x man')
in Emacs. You can navigate them with the familiar shortcuts, you
can yank quotes into mails or messages, and if you write source
code, you won't have to leave Emacs (you can have the
documentation in one window, and the source in another).

The exact same reasoning goes for info.

But there is no reason to put the man pages against info. But are
useful and you should learn to get comfortable with both.

-- 
Emanuel Berg - programmer (hire me! CV below)
computer projects: http://user.it.uu.se/~embe8573
internet activity: http://home.student.uu.se/embe8573

Re: WoMan generates different result from man?

[Hongxu Chen]

Emanuel Berg <embe8573@student.uu.se> writes:

> Hongxu Chen <leftcopy.chx@gmail.com> writes:
>
>> BTW, is WoMan popular in Emacs, or Emacsers typically look up
>> info through info?
>
> WoMan also fails to draw tables.
>
> Check out this dump:
>
> http://user.it.uu.se/~embe8573/man_sv/dumps/woman_man_ascii.png
>
> There you see the man and woman representation of the groff source.
Wow, WoMan really represents poorly for table.
I am using WoMan since the default `evil-lookup-func' in Evil-mode(well,
I am not a true Emacser) is woman, and when I

 (setq evil-lookup-func #'woman)

It doesn't work as expected and I would not spend too much to configure
it right.

Another concern is that while using `man' in emacs(which actually call
external program `man'), there seems a delay when I would get a
relatively big man page(such as `gcc'; I know I should switch to `info'
in this sense, but I am so customed to man in Vim...).
>
> It is absolutely not uncommon to browse the man pages (`M-x man')
> in Emacs. You can navigate them with the familiar shortcuts, you
> can yank quotes into mails or messages, and if you write source
> code, you won't have to leave Emacs (you can have the
> documentation in one window, and the source in another).
>
> The exact same reasoning goes for info.
>
> But there is no reason to put the man pages against info. But are
> useful and you should learn to get comfortable with both.

-- 
Regards,
Hongxu Chen

Re: WoMan generates different result from man?

[Emanuel Berg]

Hongxu Chen <leftcopy.chx@gmail.com> writes:

> I am using WoMan since the default `evil-lookup-func' in
> Evil-mode (well, I am not a true Emacser) is woman, and when I
>
>  (setq evil-lookup-func #'woman)
>
> It doesn't work as expected and I would not spend too much to
> configure it right.

Evil-mode? Haha, what is that?

I just hit `M-x man', what does that get you?

> Another concern is that while using `man' in emacs(which
> actually call external program `man'), there seems a delay when
> I would get a relatively big man page (such as `gcc'; I know I
> should switch to info' in this sense, but I am so customed to
> man in Vim...).

Again, there is nothing wrong with browsing the man pages in
Emacs, on the contrary. From where did you get that idea?

But, now that you say it, `M-x man RET gcc RET' takes a couple of
seconds in Emacs, but not so with man (the shell pager). But 1)
gcc is a very big manpage, and 2) you don't bring it up every day,
and you don't do it repeatedly. If you get stressed out over that,
perhaps you should program yourself, and not computers :)
Seriously, it depends what computer you have, of course. I have an
old laptop and for me that short-seconds pause is very
endurable. But yes - it is interesting that the Emacs mode is that
much slower than the shell tool.

-- 
Emanuel Berg - programmer (hire me! CV below)
computer projects: http://user.it.uu.se/~embe8573
internet activity: http://home.student.uu.se/embe8573

Re: WoMan generates different result from man?

[Hongxu Chen]

Emanuel Berg <embe8573@student.uu.se> writes:

> Hongxu Chen <leftcopy.chx@gmail.com> writes:
>
>> I am using WoMan since the default `evil-lookup-func' in
>> Evil-mode (well, I am not a true Emacser) is woman, and when I
>>
>>  (setq evil-lookup-func #'woman)
>>
>> It doesn't work as expected and I would not spend too much to
>> configure it right.
>
> Evil-mode? Haha, what is that?
>
> I just hit `M-x man', what does that get you?
OK, I've changed the keymaps for normal state in evil-mode to `man'(aha,
I didn't even notice that I could do this!)
>
>> Another concern is that while using `man' in emacs(which
>> actually call external program `man'), there seems a delay when
>> I would get a relatively big man page (such as `gcc'; I know I
>> should switch to info' in this sense, but I am so customed to
>> man in Vim...).
>
> Again, there is nothing wrong with browsing the man pages in
> Emacs, on the contrary. From where did you get that idea?
>
> But, now that you say it, `M-x man RET gcc RET' takes a couple of
> seconds in Emacs, but not so with man (the shell pager). But 1)
> gcc is a very big manpage, and 2) you don't bring it up every day,
> and you don't do it repeatedly. If you get stressed out over that,
> perhaps you should program yourself, and not computers :)
> Seriously, it depends what computer you have, of course. I have an
> old laptop and for me that short-seconds pause is very
> endurable. But yes - it is interesting that the Emacs mode is that
> much slower than the shell tool.

There is another problem: when using `man', it would firstly catch the
symbol/word at point by default. When there is NO manpage for this
symbol, it still displays as a candidate(I am using ido-ubiquitous and
don't know the behavior in other cases). And when I hit RET, it would
report like this:

error in process sentinel: user-error: Can't find the time_t manpage

This is annoying. But `woman' would ignore the symbol where the cursor
locates(it would not display `time_t' in this case) and I am informed
ahead that I cannot get the manpage of this symbol.

Also, the matching rule is confusing(possibly only for ido), for
instance, when I run `M-x man RET git', `git' does not appear at the
front of all candidates until I additionally add `(1)' (hence I have to
type the whole word `git(1)'), which is a bit inconvenient.

-- 
Regards,
Hongxu Chen

Re: WoMan generates different result from man?

[Emanuel Berg]

Hongxu Chen <leftcopy.chx@gmail.com> writes:

> There is another problem: when using `man', it would firstly
> catch the symbol/word at point by default. When there is NO
> manpage for this symbol, it still displays as a candidate (I am
> using ido-ubiquitous and don't know the behavior in other
> cases). And when I hit RET, it would report like this:
>
> error in process sentinel: user-error: Can't find the time_t
> manpage
>
> This is annoying. But `woman' would ignore the symbol where the
> cursor locates (it would not display `time_t' in this case) and
> I am informed ahead that I cannot get the manpage of this
> symbol.
>
> Also, the matching rule is confusing(possibly only for ido), for
> instance, when I run `M-x man RET git', `git' does not appear at
> the front of all candidates until I additionally add `(1)'
> (hence I have to type the whole word `git(1)'), which is a bit
> inconvenient.

OK, I'm sorry, I can't help you because nothing of this is even
the slightest familiar to me. I just `M-x man' and I've never had
any problems. Probably, some other guy on the list can help you.

-- 
Emanuel Berg - programmer (hire me! CV below)
computer projects: http://user.it.uu.se/~embe8573
internet activity: http://home.student.uu.se/embe8573

Re: WoMan generates different result from man?

[Hongxu Chen]

Emanuel Berg <embe8573@student.uu.se> writes:

> Hongxu Chen <leftcopy.chx@gmail.com> writes:
>
>> BTW, is WoMan popular in Emacs, or Emacsers typically look up
>> info through info?
>
> WoMan also fails to draw tables.
>
> Check out this dump:
>
> http://user.it.uu.se/~embe8573/man_sv/dumps/woman_man_ascii.png
>
> There you see the man and woman representation of the groff source.
>
> It is absolutely not uncommon to browse the man pages (`M-x man')
> in Emacs. You can navigate them with the familiar shortcuts, you
> can yank quotes into mails or messages, and if you write source
> code, you won't have to leave Emacs (you can have the
> documentation in one window, and the source in another).
>
> The exact same reasoning goes for info.
>
> But there is no reason to put the man pages against info. But are
> useful and you should learn to get comfortable with both.

Just find another difference: WoMan might miss some man entry.
For instance, man would find `malloc_stats' but WoMan fails.
Is it because the search path not set right or the flaw of WoMan?
I swear there is no sexism:-)

-- 
Regards,
Hongxu Chen

Re: WoMan generates different result from man?

[Emanuel Berg]

Hongxu Chen <leftcopy.chx@gmail.com> writes:

> Just find another difference: WoMan might miss some man entry.
> For instance, man would find `malloc_stats' but WoMan fails. Is
> it because the search path not set right or the flaw of WoMan?
> I swear there is no sexism :-)

If you notice, that's not really a man page. There is an
association to the mallopt man page, which you can confirm with
`apropos malloc_stats' that shows

mallopt (3)          - set memory allocation parameters

Now, run `whereis mallopt`

mallopt: /usr/share/man/man3/mallopt.3.gz

If you open that file in Emacs (no decompress needed), make a
search for "malloc_stats". There are no hits! So, how that
association is setup beats me, and, as you discovered, it beats
WoMan to :)

But I'm sure someone on the list knows this.

By the way, this is very interesting! *Keep digging!*

-- 
Emanuel Berg - programmer (hire me! CV below)
computer projects: http://user.it.uu.se/~embe8573
internet activity: http://home.student.uu.se/embe8573

Re: WoMan generates different result from man?

[Hongxu Chen]

Emanuel Berg <embe8573@student.uu.se> writes:

> Hongxu Chen <leftcopy.chx@gmail.com> writes:
>
>> Just find another difference: WoMan might miss some man entry.
>> For instance, man would find `malloc_stats' but WoMan fails. Is
>> it because the search path not set right or the flaw of WoMan?
>> I swear there is no sexism :-)
>
> If you notice, that's not really a man page. There is an
> association to the mallopt man page, which you can confirm with
> `apropos malloc_stats' that shows
>
> mallopt (3)          - set memory allocation parameters
>
> Now, run `whereis mallopt`
>
> mallopt: /usr/share/man/man3/mallopt.3.gz

Exactly! I also noticed that malloc_stats is NOT a man page; and man
even located it.

>
> If you open that file in Emacs (no decompress needed), make a
> search for "malloc_stats". There are no hits! So, how that
> association is setup beats me, and, as you discovered, it beats
> WoMan to :)
>
> But I'm sure someone on the list knows this.
Yeah, I hope people on the list can do me a favor.
>
> By the way, this is very interesting! *Keep digging!*

Thanks so much, Emanuel!

-- 
Regards,
Hongxu Chen

Re: WoMan generates different result from man?

[Glenn Morris]

Hongxu Chen wrote:

>   So is this a known issue?

It is know that woman.el is an incomplete implementation of the man page
protocol, yes. This can manifest itself in various ways, normally by
formatting glitches such as you noticed.

It's hard to see why much effort should be put into improving it, when
you can just use M-x man to run the real /usr/bin/man program. The only
answer that usually comes is that people on MS Windows might not have
the man program installed.

Re: WoMan generates different result from man?

[Eli Zaretskii]

> From: Glenn Morris <rgm@gnu.org>
> Date: Mon, 03 Jun 2013 13:18:27 -0400
> Cc: help-gnu-emacs <help-gnu-emacs@gnu.org>
> 
> It's hard to see why much effort should be put into improving it, when
> you can just use M-x man to run the real /usr/bin/man program. The only
> answer that usually comes is that people on MS Windows might not have
> the man program installed.

A Windows clone of 'man' is available from here:

  http://sourceforge.net/projects/ezwinports/files/man-1.4-bin.zip/download

(Groff, which is necessary to format man pages, is also available from
that site.)

Re: WoMan generates different result from man?

[Glenn Morris]

Eli Zaretskii wrote:

> A Windows clone of 'man' is available from here:
>
>   http://sourceforge.net/projects/ezwinports/files/man-1.4-bin.zip/download
>
> (Groff, which is necessary to format man pages, is also available from
> that site.)

Cool, so we can move woman.el to lisp/obsolete. :)