• Bernard Helyer (7/7) Jan 30 2012 I've found myself using dman more than I thought I would, but it
• Walter Bright (7/11) Jan 30 2012 I decided against a text file that dman reads because, inevitably, peopl...
• Jesse Phillips (4/19) Jan 31 2012 Couldn't it be a text file that is compiled in with import()? Or
• Walter Bright (2/5) Jan 31 2012 It already does that to some extent.

"Bernard Helyer" <b.helyer gmail.com> writes:

I've found myself using dman more than I thought I would, but it 
is a little annoying when it fails to find a page and I navigate 
to it manually -- I wouldn't mind if I could improve dman for the 
next release.

Perhaps a text file on GitHub we could submit pull requests to?

Thanks,
-Bernard.

Walter Bright <newshound2 digitalmars.com> writes:

On 1/30/2012 12:04 AM, Bernard Helyer wrote:
 I've found myself using dman more than I thought I would, but it is a little
 annoying when it fails to find a page and I navigate to it manually -- I
 wouldn't mind if I could improve dman for the next release.

 Perhaps a text file on GitHub we could submit pull requests to?

I decided against a text file that dman reads because, inevitably, people will 
have problems with dman not finding that file, or finding the wrong one, etc. I 
thought it would be more bulletproof if it was a standalone executable.

Definitely its internal database, though, is too small.

Eventually, it should have everything in it that has an html anchor name on the 
web site.

"Jesse Phillips" <jessekphillips+D gmail.com> writes:

On Monday, 30 January 2012 at 11:12:36 UTC, Walter Bright wrote:
 On 1/30/2012 12:04 AM, Bernard Helyer wrote:
 I've found myself using dman more than I thought I would, but 
 it is a little
 annoying when it fails to find a page and I navigate to it 
 manually -- I
 wouldn't mind if I could improve dman for the next release.

 Perhaps a text file on GitHub we could submit pull requests to?

 I decided against a text file that dman reads because, 
 inevitably, people will have problems with dman not finding 
 that file, or finding the wrong one, etc. I thought it would be 
 more bulletproof if it was a standalone executable.

 Definitely its internal database, though, is too small.

 Eventually, it should have everything in it that has an html 
 anchor name on the web site.

Couldn't it be a text file that is compiled in with import()? Or 
a d file with a single function containing add(name, url); for 
each entry. Some stand alone and easy to read/edit.

Walter Bright <newshound2 digitalmars.com> writes:

On 1/31/2012 8:19 AM, Jesse Phillips wrote:
 Couldn't it be a text file that is compiled in with import()? Or a d file with
a
 single function containing add(name, url); for each entry. Some stand alone and
 easy to read/edit.

It already does that to some extent.