improve diff request

View: New views
4 Messages — Rating Filter:   Alert me  

improve diff request

by johannespschmidt :: Rate this Message:

Reply to Author | View Threaded | Show Only this Message

file a:

m_debug_string(NULL)

file b:

m_debug_string(NULL),
m_adjust_speed(false),
m_buffer_total(0)

the smartsvn diff compares file a's line with the 2nd line of file b (because the comma).  it would be much better to compare file a with the first line of file b.

probably smartsynchronize does the same thing (didn't try it).



------------------------------------

Yahoo! Groups Links

<*> To visit your group on the web, go to:
    http://groups.yahoo.com/group/smartsvn/

<*> Your email settings:
    Individual Email | Traditional

<*> To change settings online go to:
    http://groups.yahoo.com/group/smartsvn/join
    (Yahoo! ID required)

<*> To change settings via email:
    mailto:smartsvn-digest@...
    mailto:smartsvn-fullfeatured@...

<*> To unsubscribe from this group, send an email to:
    smartsvn-unsubscribe@...

<*> Your use of Yahoo! Groups is subject to:
    http://docs.yahoo.com/info/terms/


Re: improve diff request

by syntevo Support-3 :: Rate this Message:

Reply to Author | View Threaded | Show Only this Message

> the smartsvn diff compares file a's line with the 2nd line of file b
> (because the comma).  it would be much better to compare file a with the
> first line of file b.

SmartSVN does not try to find the best matching lines within two (probably
large) different blocks, but just "similar" lines based on an absolute
similarity measurement. We have adjusted these similarity conditions recently:
especially short lines are more likely to be considered as similar now.
Anyway, I agree that for your special case it doesn't work well. If you are
frequently getting such results, you may try to use "Character-based"
comparison in View|Settings.

--
Best regards,
Marc Strapetz
_____________
syntevo GmbH
www.syntevo.com



"johannespschmidt" <johannespschmidt@...> wrote:

> file a:
>
> m_debug_string(NULL)
>
> file b:
>
> m_debug_string(NULL),
> m_adjust_speed(false),
> m_buffer_total(0)
>
> the smartsvn diff compares file a's line with the 2nd line of file b (because
> the comma).  it would be much better to compare file a with the first line of
> file b.
>
> probably smartsynchronize does the same thing (didn't try it).
>
>
>
> ------------------------------------
>
> Yahoo! Groups Links
>
>
>
>








------------------------------------

Yahoo! Groups Links

<*> To visit your group on the web, go to:
    http://groups.yahoo.com/group/smartsvn/

<*> Your email settings:
    Individual Email | Traditional

<*> To change settings online go to:
    http://groups.yahoo.com/group/smartsvn/join
    (Yahoo! ID required)

<*> To change settings via email:
    mailto:smartsvn-digest@...
    mailto:smartsvn-fullfeatured@...

<*> To unsubscribe from this group, send an email to:
    smartsvn-unsubscribe@...

<*> Your use of Yahoo! Groups is subject to:
    http://docs.yahoo.com/info/terms/


Parent Message unknown Re: improve diff request

by Sebastien Daloz :: Rate this Message:

Reply to Author | View Threaded | Show Only this Message

Hello,

I don't understand what you call "absolute similarity measurement". Could
you give us more informations (without giving your algorithm) about the
method used to find similar lines ?
With a such example, where lines are quite identical and at the same place
into the file, I think it should be possible to compare better these lines.

Best regards,

Sebastien


2009/8/11 syntevo Support <smartsvn@...>

> > the smartsvn diff compares file a's line with the 2nd line of file b
> > (because the comma).  it would be much better to compare file a with the
> > first line of file b.
>
> SmartSVN does not try to find the best matching lines within two (probably
> large) different blocks, but just "similar" lines based on an absolute
> similarity measurement. We have adjusted these similarity conditions
> recently:
> especially short lines are more likely to be considered as similar now.
> Anyway, I agree that for your special case it doesn't work well. If you are
> frequently getting such results, you may try to use "Character-based"
> comparison in View|Settings.
>
> --
> Best regards,
> Marc Strapetz
> _____________
> syntevo GmbH
> www.syntevo.com
>
>
>
> "johannespschmidt" <johannespschmidt@...> wrote:
>
> > file a:
> >
> > m_debug_string(NULL)
> >
> > file b:
> >
> > m_debug_string(NULL),
> > m_adjust_speed(false),
> > m_buffer_total(0)
> >
> > the smartsvn diff compares file a's line with the 2nd line of file b
> (because
> > the comma).  it would be much better to compare file a with the first
> line of
> > file b.
> >
> > probably smartsynchronize does the same thing (didn't try it).
> >
> >
> >
> > ------------------------------------
> >
> > Yahoo! Groups Links
> >
> >
> >
> >
>
>
>
>
>
>
>
>
> ------------------------------------
>
> Yahoo! Groups Links
>
>
>
>


[Non-text portions of this message have been removed]



------------------------------------

Yahoo! Groups Links

<*> To visit your group on the web, go to:
    http://groups.yahoo.com/group/smartsvn/

<*> Your email settings:
    Individual Email | Traditional

<*> To change settings online go to:
    http://groups.yahoo.com/group/smartsvn/join
    (Yahoo! ID required)

<*> To change settings via email:
    mailto:smartsvn-digest@...
    mailto:smartsvn-fullfeatured@...

<*> To unsubscribe from this group, send an email to:
    smartsvn-unsubscribe@...

<*> Your use of Yahoo! Groups is subject to:
    http://docs.yahoo.com/info/terms/


Re: improve diff request

by syntevo Support-3 :: Rate this Message:

Reply to Author | View Threaded | Show Only this Message

> I don't understand what you call "absolute similarity measurement". Could
> you give us more informations (without giving your algorithm) about the
> method used to find similar lines ?

I was trying to explain that SmartSVN does not check whether it's better to
compare first left line against first right line *or* second right line. It
only check whether it's OK to compare a left line against a right line. This
usually takes linear time. Finding the best matching lines and comparing them
would in general require quadratic-time. So for the given example of 1 and 3
lines, it would be no problem to find the "best" line; however if files are
changed significantly (e.g. when comparing two different files) this could
take too long.

--
Best regards,
Marc Strapetz
_____________
syntevo GmbH
www.syntevo.com



Sebastien Daloz <sebastien.daloz@...> wrote:

> Hello,
>
> I don't understand what you call "absolute similarity measurement". Could
> you give us more informations (without giving your algorithm) about the
> method used to find similar lines ?
> With a such example, where lines are quite identical and at the same place
> into the file, I think it should be possible to compare better these lines.
>
> Best regards,
>
> Sebastien
>
>
> 2009/8/11 syntevo Support <smartsvn@...>
>
> > > the smartsvn diff compares file a's line with the 2nd line of file b
> > > (because the comma).  it would be much better to compare file a with the
> > > first line of file b.
> >
> > SmartSVN does not try to find the best matching lines within two (probably
> > large) different blocks, but just "similar" lines based on an absolute
> > similarity measurement. We have adjusted these similarity conditions
> > recently:
> > especially short lines are more likely to be considered as similar now.
> > Anyway, I agree that for your special case it doesn't work well. If you are
> > frequently getting such results, you may try to use "Character-based"
> > comparison in View|Settings.
> >
> > --
> > Best regards,
> > Marc Strapetz
> > _____________
> > syntevo GmbH
> > www.syntevo.com
> >
> >
> >
> > "johannespschmidt" <johannespschmidt@...> wrote:
> >
> > > file a:
> > >
> > > m_debug_string(NULL)
> > >
> > > file b:
> > >
> > > m_debug_string(NULL),
> > > m_adjust_speed(false),
> > > m_buffer_total(0)
> > >
> > > the smartsvn diff compares file a's line with the 2nd line of file b
> > (because
> > > the comma).  it would be much better to compare file a with the first
> > line of
> > > file b.
> > >
> > > probably smartsynchronize does the same thing (didn't try it).
> > >
> > >
> > >
> > > ------------------------------------
> > >
> > > Yahoo! Groups Links
> > >
> > >
> > >
> > >
> >
> >
> >
> >
> >
> >
> >
> >
> > ------------------------------------
> >
> > Yahoo! Groups Links
> >
> >
> >
> >
>
>
> [Non-text portions of this message have been removed]
>
>
>
> ------------------------------------
>
> Yahoo! Groups Links
>
>
>
>








------------------------------------

Yahoo! Groups Links

<*> To visit your group on the web, go to:
    http://groups.yahoo.com/group/smartsvn/

<*> Your email settings:
    Individual Email | Traditional

<*> To change settings online go to:
    http://groups.yahoo.com/group/smartsvn/join
    (Yahoo! ID required)

<*> To change settings via email:
    mailto:smartsvn-digest@...
    mailto:smartsvn-fullfeatured@...

<*> To unsubscribe from this group, send an email to:
    smartsvn-unsubscribe@...

<*> Your use of Yahoo! Groups is subject to:
    http://docs.yahoo.com/info/terms/