[Baypiggies] Hi all -can anyone convert this code to python---many thanks
Russell Whitaker
whitaker at google.com
Thu Dec 14 22:33:23 CET 2006
On 12/14/06, Mike Cheponis <mac at wireless.com> wrote:
> As I see it, the correct answer is to wrap this C function so it's callable from Python.
>
> That way, no recoding, maximum performance.
>
> -Mike
>
>
> p.s. I'll send a bill later. You gotta know exactly where to tap the hammer...
>
Charles Proteus Steinmetz! A classic piece of engineering lore:
http://en.wikipedia.org/wiki/Steinmetz%2C_Charles_Proteus
:)
Russell
>
>
> On Thu, 14 Dec 2006, Sharon Kazemi wrote:
>
> > Date: Thu, 14 Dec 2006 10:16:31 -0800
> > From: Sharon Kazemi <sharonk at gisc.berkeley.edu>
> > To: baypiggies at python.org
> > Subject: [Baypiggies] Hi all -can anyone convert this code to python---many
> > thanks
> >
> > Hi all,
> >
> > have below code that needs to be recoded into Python for use in ESRI ArcGIS
> > software.
> > I would greatly appreciate any assistance.
> > Thanks
> > Sheri
> >
> > /* The Inner Loop of the Douglas-Peucker line generalization
> > algorithm is the process of finding, for a section of a polyline,
> > the vertex that is furthest from the line segment joining the
> > two endpoints. The method coded below in C (or C++) is the most
> > efficient, in terms of operation counts, that I have seen. */
> > /* _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ */
> >
> > long FurthestFromSegment ( /* Return index of furthest point. */
> > long startindex, /* Index of start vertex in arrays. */
> > long endindex, /* Index of end vertex in arrays. */
> > double *x , /* Array, abscissae of polyline vertices. */
> > double *y , /* Array, ordinates of polyline vertices. */
> > double *distMaxSquare /* Return square of maximum distance. */
> > )
> > /*
> > This function, given a section of a polyline in arrays,
> > will return the index of the intermediate node that is furthest
> > from the segment joining the two endpoints of the section,
> > and the square of the distance from the segment.
> >
> > If no intermediate point exists, then the returned index will
> > be the index of the start vertex and the returned distance
> > squared will be -1.0 . Caution: Do not calculate the square
> > root of this returned value without ruling out the possibility
> > that it may have defaulted to -1.0 . In a normal
> > Douglas-Peucker application, you should never have to calculate
> > the square root of this output value, and you should never
> > need to invoke this function without intermediate points.
> > */
> > {
> > /*
> > The variable names below assume we find
> > the distance of point "A" from segment "BC" .
> > */
> > long index, outindex ;
> > double distSquare, bcSquare ;
> > double cx, cy, bx, by, ax, ay ;
> > double bcx, bcy, bax, bay, cax, cay ;
> >
> > *distMaxSquare = -1.0 ;
> > if ( endindex < startindex + 2 ) return startindex ;
> > outindex = startindex ;
> > bx = x[startindex] ;
> > by = y[startindex] ;
> > cx = x[endindex] ;
> > cy = y[endindex] ;
> > /* Find vector BC and the Square of its length. */
> > bcx = cx - bx ;
> > bcy = cy - by ;
> > bcSquare = bcx * bcx + bcy * bcy ;
> > /* The inner loop: */
> > for ( index = startindex + 1 ; index < endindex ; index++ )
> > {
> > /* Find vector BA . */
> > ax = x[index] ;
> > ay = y[index] ;
> > bax = ax - bx ;
> > bay = ay - by ;
> > /* Do scalar product and check sign. */
> > if ( bcx * bax + bcy * bay <= 0.0 )
> > {
> > /* Closest point on segment is B; */
> > /* find its distance (squared) from A . */
> > distSquare = bax * bax + bay * bay ;
> > }
> > else
> > {
> > /* Find vector CA . */
> > cax = ax - cx ;
> > cay = ay - cy ;
> > /* Do scalar product and check sign. */
> > if ( bcx * cax + bcy * cay >= 0.0 )
> > {
> > /* Closest point on segment is C; */
> > /* find its distance (squared) from A . */
> > distSquare = cax * cax + cay * cay ;
> > }
> > else
> > {
> > /* Closest point on segment is between B and C; */
> > /* Use perpendicular distance formula. */
> > distSquare = cax * bay - cay * bax ;
> > distSquare = distSquare * distSquare / bcSquare ;
> > /* Note that if bcSquare be zero, the first
> > of the three branches will be selected,
> > so division by zero will not occur here. */
> > }
> > }
> >
> > if ( distSquare > *distMaxSquare )
> > {
> > outindex = index ;
> > *distMaxSquare = distSquare ;
> > }
> > }
> > /*
> > Note that in the inner loop above, if we follow
> > the most common path where the perpendicular
> > distance is the one to calculate, then for each
> > intermediate vertex the float operation count is
> > 1 divide, 7 multiplies, 5 subtracts, 1 add, and 2 compares.
> > */
> >
> > return outindex ;
> > }
> >
> > Sharon Kazemi
> > Visiting Scholar/GIS Analyst
> > Geographic Information Science Center
> > 412 Wurster Hall
> > University of California Berkeley
> > Berkeley, CA 94720-1820
> > Phone: +1-510-642-2812
> > Fax: +1-510-643-3412
> > Email: sharonk at gisc.berkeley.edu
> > http://www.gisc.berkeley.edu/
> >
> >
> >
> > -----Original Message-----
> > From: baypiggies-bounces at python.org
> > [mailto:baypiggies-bounces at python.org]On Behalf Of
> > baypiggies-request at python.org
> > Sent: Thursday, December 14, 2006 3:00 AM
> > To: baypiggies at python.org
> > Subject: Baypiggies Digest, Vol 14, Issue 16
> >
> >
> > Send Baypiggies mailing list submissions to
> > baypiggies at python.org
> >
> > To subscribe or unsubscribe via the World Wide Web, visit
> > http://mail.python.org/mailman/listinfo/baypiggies
> > or, via email, send a message with subject or body 'help' to
> > baypiggies-request at python.org
> >
> > You can reach the person managing the list at
> > baypiggies-owner at python.org
> >
> > When replying, please edit your Subject line so it is more specific
> > than "Re: Contents of Baypiggies digest..."
> >
> >
> > Today's Topics:
> >
> > 1. Re: Dec. 14 Meeting "Programming Productivity: What Really
> > Matters?" (Anna Ravenscroft)
> > 2. Re: Dec. 14 Meeting "Programming Productivity: What Really
> > Matters?" (Shannon -jj Behrens)
> > 3. Re: Dec. 14 Meeting "Programming Productivity: What Really
> > Matters?" (Dennis Reinhardt)
> > 4. Re: Dec. 14 Meeting "Programming Productivity: What Really
> > Matters?" (Anna Ravenscroft)
> > 5. Re: Dec. 14 Meeting "Programming Productivity: What Really
> > Matters?" (Marilyn Davis)
> > 6. Re: Dec. 14 Meeting "Programming Productivity: What Really
> > Matters?" (Anna Ravenscroft)
> > 7. Re: Dec. 14 Meeting "Programming Productivity: What Really
> > Matters?" (Alex Martelli)
> >
> >
> > ----------------------------------------------------------------------
> >
> > Message: 1
> > Date: Wed, 13 Dec 2006 17:22:03 -0800
> > From: "Anna Ravenscroft" <annaraven at gmail.com>
> > Subject: Re: [Baypiggies] Dec. 14 Meeting "Programming Productivity:
> > What Really Matters?"
> > To: "Mike Cheponis" <mac at wireless.com>
> > Cc: Python <baypiggies at python.org>
> > Message-ID:
> > <cb361a610612131722v7c6fef90ha4504a604b9b5246 at mail.gmail.com>
> > Content-Type: text/plain; charset=ISO-8859-1; format=flowed
> >
> > On 12/13/06, Mike Cheponis <mac at wireless.com> wrote:
> >> "Social Time" is really just "Random Mapping".
> >>
> >> So, if nobody stands up during "Mapping" you get all "Social Time". Also,
> > those that don't "map" are, by default, in the "Social" mapping...
> >>
> >> Don't worry, the meetings aren't nearly as formal as our discussions
> > _about_ the meetings are(!).
> >>
> >> -Mike
> >
> > I've been to the meetings. Social TIme != mapping.
> >
> > I don't have time or energy to wait for the formal "mapping" to get
> > over with before I can socialize with people. I have a final tomorrow
> > and expect to be pretty exhausted and wanting to go to bed early. I
> > was looking forward to some generic, informal socializing.
> >
> > OR can we have some freeform socializing BEFORE the official part of
> > the meeting - that way those who want (and have the energy) to do the
> > mapping can stay for that?
> >
> > Anna
> >
> >
> > ------------------------------
> >
> > Message: 2
> > Date: Wed, 13 Dec 2006 18:09:35 -0800
> > From: "Shannon -jj Behrens" <jjinux at gmail.com>
> > Subject: Re: [Baypiggies] Dec. 14 Meeting "Programming Productivity:
> > What Really Matters?"
> > To: "Anna Ravenscroft" <annaraven at gmail.com>
> > Cc: Python <baypiggies at python.org>, Mike Cheponis <mac at wireless.com>
> > Message-ID:
> > <c41f67b90612131809q37612254i1007ffe3e5fb7667 at mail.gmail.com>
> > Content-Type: text/plain; charset=ISO-8859-1; format=flowed
> >
> > On 12/13/06, Anna Ravenscroft <annaraven at gmail.com> wrote:
> >> On 12/13/06, Mike Cheponis <mac at wireless.com> wrote:
> >>> "Social Time" is really just "Random Mapping".
> >>>
> >>> So, if nobody stands up during "Mapping" you get all "Social Time".
> > Also, those that don't "map" are, by default, in the "Social" mapping...
> >>>
> >>> Don't worry, the meetings aren't nearly as formal as our discussions
> > _about_ the meetings are(!).
> >>>
> >>> -Mike
> >>
> >> I've been to the meetings. Social TIme != mapping.
> >>
> >> I don't have time or energy to wait for the formal "mapping" to get
> >> over with before I can socialize with people. I have a final tomorrow
> >> and expect to be pretty exhausted and wanting to go to bed early. I
> >> was looking forward to some generic, informal socializing.
> >>
> >> OR can we have some freeform socializing BEFORE the official part of
> >> the meeting - that way those who want (and have the energy) to do the
> >> mapping can stay for that?
> >
> > It sounds like we have a pretty vocal contingent of people who want 30
> > minutes of informal socializing followed by an hour for the actual
> > talk. That's fine by me.
> >
> > -jj
> >
> > --
> > http://jjinux.blogspot.com/
> >
> >
> > ------------------------------
> >
> > Message: 3
> > Date: Wed, 13 Dec 2006 18:17:51 -0800
> > From: Dennis Reinhardt <DennisR at dair.com>
> > Subject: Re: [Baypiggies] Dec. 14 Meeting "Programming Productivity:
> > What Really Matters?"
> > To: Python <baypiggies at python.org>
> > Message-ID: <5.1.0.14.0.20061213173953.023a51d0 at localhost>
> > Content-Type: text/plain; charset="us-ascii"; format=flowed
> >
> > At 05:22 PM 12/13/2006, Anna Ravenscroft wrote:
> >
> >> OR can we have some freeform socializing BEFORE the official part of
> >> the meeting -
> >
> > Anna,
> >
> > I have regularly attended the pre-dinner meetings Brian has organized (at
> > least the ones at IronPort). They may well be the social time you are
> > looking for and in the time frame you ask for.
> >
> > If one looks at the agenda template we have had this year, it looks like:
> >
> > 7:00 - 7:30 unstructured pre-meeting
> > 7:30 - 9:00 meeting program
> >
> > Nothing has prevented people from socializing in the 7:00 - 7:30
> > pre-meeting time frame. By my observation, not that much socialization
> > gets done then. Whatever limiters there are in socializing in the 7:00 to
> > 7:30 time frame are likely to affect tomorrow's meeting as well. Suppose
> > we revise the schedule to read
> >
> > 7:00 - x:yz social
> > x:yz - 9:00 program
> >
> > I don't think moving the 7:00 earlier is that easy without checking the
> > room schedule or someone from Google taking responsibility for being
> > there. Could happen but the fact is that we already have an time preceding
> > every meeting where completely unstructured socialization is possible.
> >
> > Suppose we set x:yz to 7:40, allowing JJ's presentation to take the entire
> > 80 minutes. IMO, we have run that experiment as well and it is called
> > starting the meeting 10 minutes late. I have not seen the demand for that.
> >
> > The major alternative here is set x:yz to 9:00 and shut down JJ's talk
> > entirely. My sense is that JJ has a great talk planned and while meet and
> > greet was preferred to no meeting at all, JJ's unabridged talk and
> > free-for-all discussion (i.e. what passes for socialization for some of us)
> > is a preferred use of the time available.
> >
> > If you have not attended the pre-meeting dinners, I can recommend them as a
> > good way to socialize.
> >
> > Dennis
> >
> >
> > ---------------------------------
> > | Dennis | DennisR at dair.com |
> > | Reinhardt | http://www.dair.com |
> > ---------------------------------
> >
> >
> >
> > ------------------------------
> >
> > Message: 4
> > Date: Wed, 13 Dec 2006 18:57:16 -0800
> > From: "Anna Ravenscroft" <annaraven at gmail.com>
> > Subject: Re: [Baypiggies] Dec. 14 Meeting "Programming Productivity:
> > What Really Matters?"
> > To: "Dennis Reinhardt" <DennisR at dair.com>
> > Cc: Python <baypiggies at python.org>
> > Message-ID:
> > <cb361a610612131857r7a3a18c5u20344f1f13398cb7 at mail.gmail.com>
> > Content-Type: text/plain; charset=ISO-8859-1; format=flowed
> >
> > Okay - let me describe what I mean by "socializing".
> >
> > To me, socializing is not:
> >
> > "mapping" (mapping is a formal structure grouping up people and the
> > grouping process seems to take forever.)
> >
> > dinner (getting stuck next to someone I've never met and have nothing
> > in common with and having to make smalltalk with them for the next
> > hour is my idea of hell.)
> >
> > extending the Q&A session after a presentation - leaving those of us
> > who have heard enough stuck with the choice of being *rude* or waiting
> > patiently in hopes that eventually we can stand up and go talk to
> > someone we have been wanting to talk with - which usually happens
> > about 5 minutes before we're kicked out of the room...
> >
> > Milling around in the lobby of google waiting to fill out a badge and
> > be shepherded up to a room just in time for the presentation to start.
> > Not everyone is there, and we're usually rather distracted.
> >
> > Socializing - what I'm asking for - is a chance to talk to people, _by
> > choice_ -- not by enforced proximity or structured format -- including
> > the chance to walk around and talk to more than one person. Having
> > snacks and beverages is nice.
> >
> > I would like a chance to socialize.
> >
> > Cordially,
> > Anna
> >
> >
> > ------------------------------
> >
> > Message: 5
> > Date: Wed, 13 Dec 2006 21:56:50 -0800
> > From: Marilyn Davis <marilyn at deliberate.com>
> > Subject: Re: [Baypiggies] Dec. 14 Meeting "Programming Productivity:
> > What Really Matters?"
> > To: baypiggies at python.org
> > Message-ID: <20061214055653.9296D1E4006 at bag.python.org>
> > Content-Type: text/plain; charset="ISO-8859-1"
> >
> >
> > ----- On Wednesday, December 13, 2006 annaraven at gmail.com wrote:
> >
> >> Okay - let me describe what I mean by "socializing".
> >>
> >> To me, socializing is not:
> >>
> >> "mapping" (mapping is a formal structure grouping up people and the
> >> grouping process seems to take forever.)
> >>
> >> dinner (getting stuck next to someone I've never met and have nothing
> >> in common with and having to make smalltalk with them for the next
> >> hour is my idea of hell.)
> >>
> >> extending the Q&A session after a presentation - leaving those of us
> >> who have heard enough stuck with the choice of being *rude* or waiting
> >> patiently in hopes that eventually we can stand up and go talk to
> >> someone we have been wanting to talk with - which usually happens
> >> about 5 minutes before we're kicked out of the room...
> >>
> >> Milling around in the lobby of google waiting to fill out a badge and
> >> be shepherded up to a room just in time for the presentation to start.
> >> Not everyone is there, and we're usually rather distracted.
> >
> > Yeh. We are not allowed in early, as far as I know. The lobby is not the
> > right thing. We're not there yet.
> >
> > Google is providing snacks? How do the snacks get snacked upon if the
> > program starts when people first arrive?
> >
> > So far, it seems to me, this little disagreement we are having is divided
> > against the gender axis. Just commenting, I don't have any judgement or
> > conclusion about it, I just find it interesting.
> >
> > I'm looking forward to whatever part I'm able to get to.
> >
> > Marilyn
> >
> >>
> >> Socializing - what I'm asking for - is a chance to talk to people, _by
> >> choice_ -- not by enforced proximity or structured format -- including
> >> the chance to walk around and talk to more than one person. Having
> >> snacks and beverages is nice.
> >>
> >> I would like a chance to socialize.
> >>
> >> Cordially,
> >> Anna
> >> _______________________________________________
> >> Baypiggies mailing list
> >> Baypiggies at python.org
> >> http://mail.python.org/mailman/listinfo/baypiggies
> >
> >
> >
> >
> > ------------------------------
> >
> > Message: 6
> > Date: Wed, 13 Dec 2006 22:06:25 -0800
> > From: "Anna Ravenscroft" <annaraven at gmail.com>
> > Subject: Re: [Baypiggies] Dec. 14 Meeting "Programming Productivity:
> > What Really Matters?"
> > To: "Marilyn Davis" <marilyn at deliberate.com>
> > Cc: baypiggies at python.org
> > Message-ID:
> > <cb361a610612132206i163aad6frfebd13274094cdac at mail.gmail.com>
> > Content-Type: text/plain; charset=ISO-8859-1; format=flowed
> >
> > On 12/13/06, Marilyn Davis <marilyn at deliberate.com> wrote:
> >>
> >> ----- On Wednesday, December 13, 2006 annaraven at gmail.com wrote:
> >>
> >> Yeh. We are not allowed in early, as far as I know. The lobby is not the
> > right thing. We're not there yet.
> >>
> >> Google is providing snacks? How do the snacks get snacked upon if the
> > program starts when people first arrive?
> >>
> >> So far, it seems to me, this little disagreement we are having is divided
> > against the gender axis. Just commenting, I don't have any judgement or
> > conclusion about it, I just find it interesting.
> >>
> >
> > The main disagreement I really see is a different definition of
> > socializing. Which is why I posted my definition so we could have the
> > discussion be more clear. Some people apparently consider the lobby
> > milling or the mapping to be socializing. That's okay for them to have
> > that definition - but it's not what I was asking for when I asked for
> > a chance to socialize, and so I figured it would be easier to discuss
> > if we had a definition on the table. (whether for people to agree or
> > disagree with - at least we're not assuming what it means.) Once we
> > agree on what it means to "socialize" we can agree on whether we want
> > a socializing time.
> >
> >> I'm looking forward to whatever part I'm able to get to.
> >
> > Me too. I'm sure the presentation will be great. I hope I get a chance
> > to chat with you. I haven't had much opportunity for that.
> >
> > --
> > cordially,
> > Anna
> > --
> > It is fate, but call it Italy if it pleases you, Vicar!
> >
> >
> > ------------------------------
> >
> > Message: 7
> > Date: Wed, 13 Dec 2006 22:54:46 -0800
> > From: "Alex Martelli" <aleax at google.com>
> > Subject: Re: [Baypiggies] Dec. 14 Meeting "Programming Productivity:
> > What Really Matters?"
> > To: "Marilyn Davis" <marilyn at deliberate.com>
> > Cc: baypiggies at python.org
> > Message-ID:
> > <55dc209b0612132254y6ea79c01h257c2d515fbe6fd6 at mail.gmail.com>
> > Content-Type: text/plain; charset=ISO-8859-1; format=flowed
> >
> > On 12/13/06, Marilyn Davis <marilyn at deliberate.com> wrote:
> > ...
> >> So far, it seems to me, this little disagreement we are having is divided
> > against the gender axis. Just commenting, I don't have any judgement or
> > conclusion about it, I just find it interesting.
> >
> > OK, so let me weigh in...: I _know_ I'm going to be seriously tired
> > tomorrow night, except (perhaps, with some luck:-) very early on, as
> > I'm going to have a long and intense workday before it; so, from my
> > subjective POV the tradeoff is -- am I going to be dozing at the end
> > of the presentation (if we have socialtime _before_), or am I going to
> > be dozing through socialtime (if we have it _after_). Overall, I'd
> > (marginally) prefer the first option.
> >
> > Sorry to spoil the nice gender-axis alignment, but then, I tend to do
> > that (I guess that, according to our great State's great governor,
> > this makes me a "girlyman":-).
> >
> >
> >> I'm looking forward to whatever part I'm able to get to.
> >
> > Likewise, except s/get to/be awake for/ :-).
> >
> >
> > Alex
> >
> >
> > ------------------------------
> >
> > _______________________________________________
> > Baypiggies mailing list
> > Baypiggies at python.org
> > http://mail.python.org/mailman/listinfo/baypiggies
> >
> >
> > End of Baypiggies Digest, Vol 14, Issue 16
> > ******************************************
> >
> >
> >
> > _______________________________________________
> > Baypiggies mailing list
> > Baypiggies at python.org
> > http://mail.python.org/mailman/listinfo/baypiggies
> >
>
> _______________________________________________
> Baypiggies mailing list
> Baypiggies at python.org
> http://mail.python.org/mailman/listinfo/baypiggies
>
--
Russell Whitaker
Sysops Tools Team Lead
http://www.corp.google.com/~whitaker/
"From the point of view of any orthodoxy, myth might be defined as
'other people's
religion'... to which an equivalent definition of religion might be
'misunderstood mythology'."
- Joseph Campbell, "The Hero with a Thousand Faces"
More information about the Baypiggies
mailing list