eqc documentation matching template (version 3) #415

julien2512 · 2016-09-15T22:40:21Z

Here is the third version, matching the template proposal.

As I got a better understanding of proj4 implementation, I think there might be a mistake in eqc implementation.

In the description of wikipedia, nor wolfram, there is only one parameter \phi_ts. Evenden do not use any.
But in proj4, we got two : \phi_ts and \phi_0.

That's why I tryed to document it on this pull request.
Although, I can't find how this value is valuated in proj4, and I find no documentation (wikipedia, evenden, wolfram, epsg) talking about a parameter phi_0 different than phi_ts.

May be phi_0 value is simply always 0 in proj4, and just leave in the code as a reminder ?

QuLogic

Mostly minor tweaks to stuff.

QuLogic · 2016-09-15T23:33:47Z

docs/source/projections/eqc.rst

@@ -4,7 +4,95 @@
 Equidistant Cylindrical (Plate Caree)
 ********************************************************************************

+The simplist of all projections. Standard parallels (0° when omitted) may be specified that determine latitude of true scale (k=h=1).


QuLogic · 2016-09-15T23:34:43Z

docs/source/projections/eqc.rst

+---------------------+----------------------------------------------------------+
+| `+lat_0`            | Center of the map                                        |
+---------------------+----------------------------------------------------------+
+
 .. image:: ./images/eqc.png
   :scale: 50%
   :alt:   Equidistant Cylindrical (Plate Caree)  


Carree, or even more correct, Carrée, especially because that's used later.

Even better !
done.

QuLogic · 2016-09-15T23:36:56Z

docs/source/projections/eqc.rst

+---------------------------------------------------------+--------------------------+
+| Projection Name                                         | (lat ts=) :math:`\phi_0` |
+---------------------------------------------------------+--------------------------+
+| Plain/Plane Chart                                       | 0◦                       |


This is the "White Bullet" symbol, not the degree sign, cf. ◦ vs. °.

It was from the old doc !
done.

QuLogic · 2016-09-15T23:39:29Z

docs/source/projections/eqc.rst

+---------------------------------------------------------+--------------------------+
+| Plate Carrée                                            | 0◦                       |
+---------------------------------------------------------+--------------------------+
+| Ronald Miller—minimum overall scale distortion          | 37°30'                   |


The "apostrophe" should be a "prime", cf. ' vs. ′ (though I can't really see a difference with my default font here, so it's rather minor, and probably not a big deal.)

Really ? I hope parsers will recognise both.
done.

QuLogic · 2016-09-15T23:42:34Z

docs/source/projections/eqc.rst

@@ -4,7 +4,95 @@
 Equidistant Cylindrical (Plate Caree)


Carree or Carrée

There is a point here. The plate carrée is just one case of equidistant cylindrical, isn't it ? So why using it on the title ...

done.

kbevers

Good work @julien2512. Thanks for doing this.

I've pointed out a few things I'd like to see adressed before I merge this. On top of this I think that @QuLogic has some valid points as well.

Regarding the phi_0 variable. It is the latitude of origin and can be set with +lat_0.

kbevers · 2016-09-16T08:43:02Z

docs/source/projections/eqc.rst

+---------------------+----------------------------------------------------------+
+| **Defined area**    | Global, but best used near the equator                   |
+---------------------+----------------------------------------------------------+
+| **Implemented by**  | Marinus of Tyre                                          |


This was meant to be a credit to the author of the code implementation. I believe this one was written by Gerald I. Evenden.

kbevers · 2016-09-16T08:52:05Z

docs/source/projections/eqc.rst

+Example using latitude of true scale::
+
+    $ echo 2 1 | proj +proj=eqc +a=6400000 +lat_1=0.5 +lat_2=2
+    223402.14       111701.07


Could you come up with a better example? This is the auto-generated test-case from the code. +lat_1 and +lat_2 has no effect on this particular projection. How about an example or two that uses the parameters mentioned in the table above?

kbevers · 2016-09-16T09:02:47Z

docs/source/projections/eqc.rst

+#######################
+
+The formulas describing the Equidistant Cylindrical projection are all taken from proj4 sources.
+


This section would be more clear if some of the variables are defined here. Like phi_ts is the latitude of true scale and can be set with +lat_ts.

I tryed something with the parameters.
You may have a look on it ?
I feel it is important to evaluate the quality before stepping to an other projection doc.

julien2512 · 2016-09-16T20:34:57Z

@kbevers

Regarding the phi_0 variable. It is the latitude of origin and can be set with +lat_0.

Can you help me to know where the lat_0 is taking a value in proj4 ? It can make me ask many less questions !

kbevers · 2016-09-16T20:45:18Z

Can you help me to know where the lat_0 is taking a value in proj4 ? It can make me ask many less questions !

Sure. It's done when calling pj_init. So when you initialize a projection.

julien2512 · 2016-09-16T20:51:02Z

Thank you, I found :
PIN->phi0 = pj_param(ctx, start, "rlat_0").f;

julien2512 · 2016-09-16T21:07:09Z

@QuLogic thank you for your help.

kbevers · 2016-09-16T21:27:08Z

docs/source/projections/eqc.rst

-:math:`lat_ts` is the latitude of true scale, that mean the standard parallels where the scale of the projection is true.
-:math:`lat_0` is the latitude of origin that match the center of the map.
+
+:math:`phi_{ts}` is the latitude of true scale, that mean the standard parallels where the scale of the projection is true. It can be set with "lat_ts".


I think there's a backslash missing here. Same for the line below.

julien2512 · 2016-09-17T18:47:22Z

This is working for me without backslashes. Is there a better way to format
these lines ?

Le 16 sept. 2016 23:27, "Kristian Evers" notifications@github.com a
écrit :

@kbevers commented on this pull request.

In docs/source/projections/eqc.rst
#415 (review):

@@ -71,8 +71,10 @@ Mathematical definition
#######################

The formulas describing the Equidistant Cylindrical projection are all taken from proj4 sources.
-:math:lat_ts is the latitude of true scale, that mean the standard parallels where the scale of the projection is true.
-:math:lat_0 is the latitude of origin that match the center of the map.
+
+:math:phi_{ts} is the latitude of true scale, that mean the standard parallels where the scale of the projection is true. It can be set with "lat_ts".

I think there's backslash missing here. Same for the line below.

—
You are receiving this because you authored the thread.
Reply to this email directly, view it on GitHub
#415 (review), or mute
the thread
https://github.com/notifications/unsubscribe-auth/AAzLSQ-ruW5kSydrgrG3SYCDFZUQ7dTJks5qqwmugaJpZM4J-bev
.

kbevers · 2016-09-18T13:17:16Z

Really? Do you build the docs with sphinx?

julien2512 · 2016-09-21T20:45:02Z

Of course !
Eventhough, I am open to your proposal.

kbevers · 2016-09-22T06:36:14Z

Great. I'll merge your PR once you change it.

kbevers · 2016-09-28T08:08:35Z

Thanks for writing this, Julien. It is very much appreciated!

julien2512 · 2016-09-28T20:36:02Z

Let's start other ones !

julien2512 · 2016-09-28T20:59:40Z

It will be Cassini.

Andre-J · 2018-04-08T17:55:22Z

There is a bug in the example:
Example using Plate Carrée projection with true scale at latitude 30° and central meridian 90°W:
$ echo -88 30 | proj +proj=eqc +lat_ts=30 +lat_0=90w

Central meridian should be lon_0, which is not included in the formulas, but changing values for lon_0 different from zero resolves to different results. Seems that Proj can do more than Snyder wrote in his formulas. Maybe this should be pointed out on the manpage.

kbevers · 2018-04-09T00:07:25Z

Good catch. Indeed it should be +lon_0. Fixed in 89af275

Seems that Proj can do more than Snyder wrote in his formulas. Maybe this should be pointed out on the manpage.

The +lon_0 adjust the longitude before it reaches the projection fwd/inv functions, as seen here: https://github.com/OSGeo/proj.4/blob/84c2fed374da46ba0ca3cae1125350b1aa3f73fd/src/pj_fwd.c#L86

I agree that this should be documented better. It's on my list for the next large overhaul of the docs.

eqc documentation matching template (version 3)

01bbea9

QuLogic suggested changes Sep 15, 2016

View reviewed changes

kbevers requested changes Sep 16, 2016

View reviewed changes

update eqc projection docs with @kbevers @QuLogic comments

8f1b2a7

eqc doc mathematical section update for parameters

26e3889

kbevers reviewed Sep 16, 2016

View reviewed changes

kbevers merged commit 0e01585 into OSGeo:master Sep 28, 2016

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

eqc documentation matching template (version 3) #415

eqc documentation matching template (version 3) #415

julien2512 commented Sep 15, 2016 •

edited

QuLogic left a comment

QuLogic Sep 15, 2016

julien2512 Sep 16, 2016

QuLogic Sep 15, 2016 •

edited

julien2512 Sep 16, 2016

QuLogic Sep 15, 2016

julien2512 Sep 16, 2016

QuLogic Sep 15, 2016 •

edited

julien2512 Sep 16, 2016 •

edited

QuLogic Sep 15, 2016

julien2512 Sep 16, 2016

kbevers left a comment

kbevers Sep 16, 2016

julien2512 Sep 16, 2016

kbevers Sep 16, 2016

julien2512 Sep 16, 2016

kbevers Sep 16, 2016

julien2512 Sep 16, 2016 •

edited

julien2512 commented Sep 16, 2016

kbevers commented Sep 16, 2016

julien2512 commented Sep 16, 2016

julien2512 commented Sep 16, 2016

kbevers Sep 16, 2016 •

edited

julien2512 commented Sep 17, 2016 •

edited

@kbevers commented on this pull request.

kbevers commented Sep 18, 2016

julien2512 commented Sep 21, 2016

kbevers commented Sep 22, 2016

kbevers commented Sep 28, 2016

julien2512 commented Sep 28, 2016

julien2512 commented Sep 28, 2016

Andre-J commented Apr 8, 2018

kbevers commented Apr 9, 2018

		#######################

		The formulas describing the Equidistant Cylindrical projection are all taken from proj4 sources.

eqc documentation matching template (version 3) #415

eqc documentation matching template (version 3) #415

Conversation

julien2512 commented Sep 15, 2016 • edited

QuLogic left a comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

QuLogic Sep 15, 2016 • edited

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

QuLogic Sep 15, 2016 • edited

Choose a reason for hiding this comment

julien2512 Sep 16, 2016 • edited

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

kbevers left a comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

julien2512 Sep 16, 2016 • edited

Choose a reason for hiding this comment

julien2512 commented Sep 16, 2016

kbevers commented Sep 16, 2016

julien2512 commented Sep 16, 2016

julien2512 commented Sep 16, 2016

kbevers Sep 16, 2016 • edited

Choose a reason for hiding this comment

julien2512 commented Sep 17, 2016 • edited

@kbevers commented on this pull request.

kbevers commented Sep 18, 2016

julien2512 commented Sep 21, 2016

kbevers commented Sep 22, 2016

kbevers commented Sep 28, 2016

julien2512 commented Sep 28, 2016

julien2512 commented Sep 28, 2016

Andre-J commented Apr 8, 2018

kbevers commented Apr 9, 2018

julien2512 commented Sep 15, 2016 •

edited

QuLogic Sep 15, 2016 •

edited

QuLogic Sep 15, 2016 •

edited

julien2512 Sep 16, 2016 •

edited

julien2512 Sep 16, 2016 •

edited

kbevers Sep 16, 2016 •

edited

julien2512 commented Sep 17, 2016 •

edited