January 2018 - Therion

Scaling problems with backgrounds in th2 files
by John Stevens 21 Jan '18

21 Jan '18

A method I used to copy from a previous survey was to export the centreline as a xvi file. Then take a jpg of the old survey and scale it so the pixel count matches that of the xvi file. The next step was to use both images as background images in the th2 file. A bit of adjustment of the position location to make them match then it was simple to copy the old survey. Now I cant align the images as the zoom function now changes the scales of the two images differently. If I use a jpg to trace over and zoom in to trace a detailed area the scaling goes off. The resultant lines are to the wrong scale and/or are misplaced. Is there a way of fixing the scale of a background image so it scales properly. It seems to go wrong with the + – and scroll wheel. John

4 3

Redefined point continuation does not obey -align properly
by Bruce Mutton 21 Jan '18

21 Jan '18

Good morning Something I noticed some years ago, and I have noticed again recently. The alignment of point continuations does not properly adhere to the alignment conventions that other textual point labels (such as label, remark, date) do. -align [left | right] typically draws the symbol to the left or right of the 'point' that is drawn. For continuations, this occurs for the question mark, but not for the 'debug-like' text that is written describing the continuation. This debug-like text is always written to the right of the question mark. The debug-like text does responds to <br> text string codes, but not to <centre> or <left> codes, as other text entities do. This only affects point continuations that have been redefined like this; code metapost def p_continuation(expr pos,theta,sc,al) = % draw default continuation symbol p_continuation_UIS(pos,theta,sc,al); % if text attribute is set if known(ATTR__text) and picture(ATTR__text): % set labeling color to light orange push_label_fill_color(1.0, 0.9, 0.8); % draw filled label with text next to ? p_label.urt(ATTR__text,(.5u,-.25u) transformed T,0.0,8); % restore original labeling color pop_label_fill_color; fi; enddef; endcode So perhaps there is something to change in this metapost? I suspect around the line that starts with p_label.urt.. For example, here is a drawing with a point continuation I have tried to align top-left. point continuation -text "vegetable debris<br>gas bubbles up from sand" -attr who "B Mutton" -attr what passage -attr priority caution -attr ref plan -align tl Does anyone have a solution to this? Regards Bruce

1 0

Disable loop closure?
by Evaristo Quiroga 21 Jan '18

21 Jan '18

Is it possible to disable loop closure in Therion to export a 3d model? I want to compare two 3d models in Aven: one with loop closure and other without loop closure, to find where has the error loop.

4 9

LIDAR porvrch
by Michal Danko 19 Jan '18

19 Jan '18

Ahojte, hram sa s lidar datami a mam trosku problem. neviem ci som sa dostal na obmedzenie therionu alebo robim nieco zle. vstupne data som mal stvorec 48.99146, 19.67295 48.98282, 19.68312 vo formate tif, ktory som konvertoval cez gdal na txt - ArcInfo Asci Grid gdal_translate -of AAIGrid zlomiska.tif zlomiska-povrch.txt cez gdalinfo som zobrazil taketo hodnoty: Driver: GTiff/GeoTIFF Files: DMR_Zlomiska.tif Size is 1118, 2310 Coordinate System is: PROJCS["WGS 84 / UTM zone 34N", GEOGCS["WGS 84", DATUM["WGS_1984", SPHEROID["WGS 84",6378137,298.257223563, AUTHORITY["EPSG","7030"]], AUTHORITY["EPSG","6326"]], PRIMEM["Greenwich",0, AUTHORITY["EPSG","8901"]], UNIT["degree",0.0174532925199433, AUTHORITY["EPSG","9122"]], AUTHORITY["EPSG","4326"]], PROJECTION["Transverse_Mercator"], PARAMETER["latitude_of_origin",0], PARAMETER["central_meridian",21], PARAMETER["scale_factor",0.9996], PARAMETER["false_easting",500000], PARAMETER["false_northing",0], UNIT["metre",1, AUTHORITY["EPSG","9001"]], AXIS["Easting",EAST], AXIS["Northing",NORTH], AUTHORITY["EPSG","32634"]] Origin = (403129.719999999972060,5427465.219999999739230) Pixel Size = (0.500000000000000,-0.500000000000000) Metadata: AREA_OR_POINT=Area DataType=Generic Image Structure Metadata: INTERLEAVE=BAND Corner Coordinates: Upper Left ( 403129.720, 5427465.220) ( 19d40'32.80"E, 48d59'32.95"N) Lower Left ( 403129.720, 5426310.220) ( 19d40'33.79"E, 48d58'55.55"N) Upper Right ( 403688.720, 5427465.220) ( 19d41' 0.30"E, 48d59'33.26"N) Lower Right ( 403688.720, 5426310.220) ( 19d41' 1.29"E, 48d58'55.87"N) Center ( 403409.220, 5426887.720) ( 19d40'47.05"E, 48d59'14.41"N) Band 1 Block=1118x1 Type=Float32, ColorInterp=Gray z nich som sa pokusil vytvorit* surface.th <http://surface.th>.* encoding utf-8 surface * cs utm34* * grid 403129.72 5426310.22 0.5 0.5 1118 2310* input zlomiska-povrch.txt endsurface netusim ci mam spravny grid, tak ak sa dakto do toho vyzna prosim skontrolujte pretoze pri kompilacii dostanem chybu.. *error -- lidar/zlomiska-povrch.txt [7] -- line too long* *writing xtherion file ... done* 7- je riadok kde zacinaju vyskove data..(riadkov je 2310 ) S pozdravom Ing. Michal Danko

3 3

colour graduation lookup feature - carto-date, copyright-date, scrap-author
by Bruce Mutton 09 Jan '18

09 Jan '18

One of the uses I have found for the recent addition of 'colour map-fg' for explo-date and topo-date, together with lookups, is that I can check up on the cave drawing team members. It is now easy to find surveys that are missing metadata such as exploration date and survey date. This could be expanded to include 'colour map-fg carto-date', colour map-fg copyright-date. A metadata quality control mechanism for larger projects. And to keep up the competition pressure on the drawing team, how about 'colour map-fg scrap-author'? This way one can see qualitatively who has drawn the most passage length. And if 'statistics carto-count' was created, to be similar in behaviour to <https://therion.speleo.sk/wiki/faq#why_do_peoples_names_sort_alphabetically _when_they_used_to_sort_by_length_of_cave_explored_and_surveyed> 'statistics explo-length' (on, off, hide) , but counting the number of scraps, then with on or hide, one could see who has drawn the most scraps. (More and more I am finding that smaller scraps produce more versatile maps). Just keeping the development pressure up :) Regards Bruce

1 0

colour graduation lookup feature - choice of terminology
by Bruce Mutton 09 Jan '18

09 Jan '18

I'm liking the lookup feature added in October, in case you hadn't noticed by my earlier posts. However I wonder if the term 'lookup' is a good choice? When I think about what it does, or achieves, it is about creating collection(s) of map foreground colour palettes, which each map parameter values to specific colours. If a colour legend is produced, it controls the legend text, and the number of legend entries. I think 'lookup' is too generic, and it would be better to use a term that better conveys what it achieves for the user. What about replacing the term 'lookup' in Therion with 'palette' or 'palette-map-fg' (to allow for other possible palettes in future)? Before we get to imprinted on the word lookup. Thoughts?" Regards Bruce

1 0

Bug - select map -colour
by Bruce Mutton 07 Jan '18

07 Jan '18

I have managed to get some effect from 'select map -colour', but not entirely the desired effects. Just documenting the behaviour here. There are no lookups defined in this project. select AblutionInceptionPlanMap -colour [0 0 100] #blue select QuackeryPlanMap -color [0 100 0] #green select SouthBoundPlanMap -color [100 0 0] #red The above selects have no effect, the maps are all uncoloured, unless. I add, 'colour map-fg map', inside a layout, then I get all maps coloured with the last colour specified (red). I wasn't expecting 'colour map-fg map' to have an effect, as I thought the 'select map -colour' would override layout specifications, or at least have the option to override (to make it both 'easy' and allow combining other colour palettes (such as map-fg explo-date) with particular maps highlighted and overriding the palette). If I change the order of selects, I find the last colour selected is the one that colours all the maps. Eg, making green last colours all maps green. If I make all of the spelling consistent, ie color, or colour, there is no effect. If I add other select map statements with no -colour option, then, again, all maps are coloured with the last colour selected. If I remove the -colour option from select statements, and use only 'colour map-fg map', then I get appropriate automatically coloured maps, ie Bruce

1 0

colour graduation lookup feature - date and select
by Bruce Mutton 06 Jan '18

06 Jan '18

Stacho, and Therion users Tried to break lookups for dates, maps and scraps today, with only a small amount of success! Actually, it seems a pretty stable and useful feature. I have documented how I think they work in the wiki https://therion.speleo.sk/wiki/examples#colour_scales_-_lookups 1. I have a question about using - [0 0 0] "undated surveys" in a lookup list. As I noted in the wiki, the dash '-' seems to select all undated surveys if placed at the bottom of the list, and all surveys if placed at the top of the list. This would seem to suggest the legend colours are constructed the same as scrap drawings; items at the top of the list overwrite those below them, if there is any clash. It seems to be a useful and predictable behaviour, if that is actually the case - have I got it right? 2. Specifying colour with map selection. I could not get this to work at all, although have only tried it with one project so far. I have tried both spellings; select map1 -color [100 0 0] select map1 -colour [100 0 0] Both are silently ignored. I have checked to make sure there are no colour map-fig statements hidden in the project, as I assume these might override the map selection (although having 'select map colour' override 'colour map-fg xxxx' would be a nice behaviour). What might I be doing wrong here? 3. And lastly a minor comment. The default date colour legend looks like this. . with the most recent dates at the bottom. As with altitudes (which has already being adjusted) my preference would be to have the recent dates sort to the top (the colours - red for recent is good). I could make custom lookups of course, but I think the default should match what most people prefer. What do people think? Regards Bruce

2 1

Error morphing scanned extended profiles survey sketches
by Evaristo Quiroga 06 Jan '18

06 Jan '18

Hi, I have this problem: the xviresult file of the morphing processwith extended profiles is erroneous. The centerline and sketch don't are superposed. The xvi is at 0 0 position, but the sketch are very displaced in Y direction (X it's OK), far 350000000 units. I can do a manual solution but I think it's not normal to have to do this. Calculating the displacement error and manually editing the xvi file with and external editor, to correct the sketch Y coordinate. Thanks, Evaristo Quiroga

3 5

colour graduation lookup feature - altitude
by Bruce Mutton 06 Jan '18

06 Jan '18

Stacho I've had a wee play with colour graduation lookup feature, at least as far as altitudes go. First thing I noticed is that it does not work if the lookup is defined inside a layout. . LayoutMapThisCave.thc [89] -- unknown option -- lookup altitude -title "Altitude legend" But if I put it outside of the layout (effectively, even if not actually), inside the thconfig file (like a 'text' definition, it works just fine for altitudes. It feels to me like it should be inside a layout, to be consistent, but perhaps you have a reason? Perhaps it does not matter. The framework seems to be very flexible, allowing many pre defined lookup definitions, and works well for altitudes. I like it. If you are happy for me to do so, I will tidy up the wiki entry that I made on this topic earlier today. * As below, it seems that if an altitude range is defined, then a single altitude in the list will break the default text that Therion puts in the colour legend. I was attempting to get Therion to produce a constant colour over altitude ranges, and interpolate colours between those ranges. As below, I found that it does not work that way. lookup altitude:banded -title "Altitude legend (banded)" [105 150][] "top level" 90 75 [30 60] [] "lower level" endlookup It produces the legend below, and a map that is only partially coloured. The default legend item texts are incorrect. * Adding colours to the top and bottom improves things a little. lookup altitude:banded -title "Altitude legend (banded)" [105 150][100 0 0] "top level" 90 75 [30 60] [0 0 100] "lower level" endlookup The colour graduation is correct, but the map produced only colours the scraps that are within the 'top level' and 'lower level' altitude range. The others (60m to 105m) are uncoloured. * Adding colours to all altitudes does not improve things much. lookup altitude:banded -title "Altitude legend (banded)" [105 150][100 0 0] "top level" 90 [50 0 0] 75 [0 0 50] [30 60] [0 0 100] "lower level" endlookup The map is still uncoloured in the middle altitudes (60m to 105m). * If I leave out the 90 and 75, the result is similar, again with uncoloured scraps in the central altitude range. * Just to be sure, I attempted three bands. lookup altitude:banded -title "Altitude legend (banded)" [105 150][100 0 0] "top level" [60 105] [] "middle level" [30 60] [0 0 100] "lower level" endlookup This worked perfectly, colouring all scraps according to their altitude band. * And just to finish up, I removed the text overrides. lookup altitude:banded -title "Altitude legend (banded)" [105 150][100 0 0] [60 105] [] [30 60] [0 0 100] Endlookup Again, perfect! The default text is OK this time. Maybe tomorrow I'll see if I can break dates, maps and scraps! Regards Bruce

1 0