It seems I was wrong about the cause of
the problem, the problem is more in Therion than in data entry.
Looks like Therion 5.3.15 cannot handle
survey dates in 2015 (as implied in the changes logs), and so 5.3.16 is
required to use coordinate systems with surveys dated from 2015 onwards.
That is OK, but what could be improved, is
the reporting of the error to something like…
magnetic
declination not available for survey date
and also to improve the variable
\magdecl in this circumstance.
Bruce
From: Bruce
[mailto:bruce@tomo.co.nz]
Sent: Monday, 19 January 2015 4:34
p.m.
To: 'List for Therion users'
Subject: Magnetic declination
meltdown
A random discovery with the new
If one uses a defined coordinate system
and forgets to provide a survey date for all surveys, then there is some
‘not very well trapped’ undesirable behaviour.
The log file produces something like
this…
“C:\Program
Files\Therion\therion.exe: warning -- unable
to determine magnetic declination for undated surveys
output coordinate system: EPSG:2193
meridian convergence (deg): -0.0696
“
which is appropriate behaviour, however it
is followed by…
“
#################### end of metapost log
file ####################
converting scraps ... done
making map ... done
######################## pdftex log file
#########################
This is pdfTeX, Version 3.1415926-1.40.10
(Web2C 2009) (format=pdftex 2010.1.2) 18 JAN 2015 20:21
entering extended mode
**data.tex
(./data.tex (./th_enc.tex)
(./th_texts.tex) (./th_resources.tex
(c:/Program
Files/Therion/texmf/tex/glyphtounicode.tex)) (./th_fontdef.tex{c:/P
rogram
Files/Therion/texmf/fonts/pdftex.map})
(./th_formdef.tex
Normal \count register pool exhausted,
switching to extended pool.)
(./th_pagedef.tex
Overfull \hbox (299.59392pt too wide) in
paragraph at lines 12--12
[]\thfb gridnorth -1261120821493519700000000000000000000000000000000000000000
00000000000000000000000000000000.00deg |
\hbox(8.33333+2.33333)x330.05206, glue set
- 1.0
.\hbox(0.0+0.0)x0.0
.\thfb g
.\thfb r
.\thfb i
.\thfb d
.etc.
) (./th_pages.tex)
253 numeric registers used out of
32768
[1] )
Here is how much of
1706 strings out of 95617
….
“
which is not so good, but
at least it does produce a generally OK looking pdf (with my experimental debug
like header layout).
In this case it might be
better to set \magdecl to a more
meaningful error value.
If it can take a string value then “unable to determine magnetic declination for undated
surveys” is probably good.
Bruce