Re: [Therion] Conflict between "extend ignore" command and inverse legs
I have commented all the splays "extend ignore", and Therion is doing now a good job, ignoring the appropriate leg (extend ignore 83 92). All the splays "extend ignore" was confusing the Therion compiler. This data is an export from Topodroid, that incorporates the "extend ignore" command before the splays shots. I will comment to Marco Corvi on the problem. @ Evaristo & Bruce: what about a command "extend auto", which means let therion decide how to extend the splay shot ? for the moment therion parser could behave as if the command weren't there. marco
Hi Marco I am not sure I understand, let me test my thinking, for everyone to pick apart :) I think the extend options provide control of the non-splay survey legs. In general that control amounts to left, right, vertical, not at all, or not visible. The splay legs should just ‘wobble around’ as a second order effect, in response to the extend parameters applied to the non-splay legs. So the only sensible behaviour for splays is ‘extend auto’, and my observation is that this is already coded into PocketTopo and Therion, and so there is no reason to require the user to invoke ‘extend auto’ manually. As noted in previous threads, Therion and PocketTopo do manage to orient splay legs differently. Quite often I have to mentally reorient splays when drawing in Therion Drawing Editor. Neither application seems to do it perfectly, but then because the shape of the extended centreline network is as much an arbitrary artistic choice on the part of the user, as it is based on hard numeric data, there is perhaps no ‘right approach’ for the software to take. I suspect that I have observed that TopParser can affect the orientation of splays as well, but I am not sure of that. I think the right approach is to aim for Therion (and PocketTopo, TopoDroid) to orient splays automatically, based on the extend directions of the incoming and outgoing non-splay legs. If the applications were to allow direct user control of splay extend direction, then the statements should not be able to be confused with existing statements that control the non-splay legs (ie adopt extend splay left, extend splay right, extend splay vertical, extend splay auto). Perhaps this topic is something the authors should collaborate on to achieve a consistent behaviour? What do you think? Bruce From: Therion <therion-bounces@speleo.sk> On Behalf Of Marco Corvi via Therion Sent: Thursday, 7 June 2018 11:29 PM To: therion@speleo.sk Cc: Marco Corvi <marco.corvi@gmail.com> Subject: Re: [Therion] Conflict between "extend ignore" command and inverse legs I have commented all the splays "extend ignore", and Therion is doing now a good job, ignoring the appropriate leg (extend ignore 83 92). All the splays "extend ignore" was confusing the Therion compiler. This data is an export from Topodroid, that incorporates the "extend ignore" command before the splays shots. I will comment to Marco Corvi on the problem. @ Evaristo & Bruce: what about a command "extend auto", which means let therion decide how to extend the splay shot ? for the moment therion parser could behave as if the command weren't there. marco
Hi, In order to move forward with the issue of extend splay control, I will give some examples. After a download of DistoX2 to Topodroid, and check the extend leg direction, I have the next centerline, where I can sketch . You can see in the next screenshot that only the leg shot have the extend direction set [<]. Topodroid have automatically oriented the splays depending on its angle with respect to the leg shot angle. It's very easy to sketch, with this help. Now this data in Therion, after removing the "extend ignore" statement for splays write by Topodroid.
extend left 56 55 7.04 123.5 -1.7 56 . 2.15 196.9 -5.0 56 . 0.69 32.1 -7.0 56 . 1.06 176.0 45.1 56 . 0.70 303.3 -84.0 56 . 2.31 229.4 21.3 56 . 2.39 244.0 20.5 56 . 3.37 240.2 22.0 56 . 3.60 230.6 22.4 56 . 3.80 244.2 41.0 56 . 2.43 240.7 -6.1 56 . 2.97 160.0 -7.5 56 . 3.62 111.1 -6.8 56 . 2.56 269.3 -7.0 extend left 57 56 6.37 130.1 2.2
I export to xvi, and open the exported .th2 sketch by Topodroid. You can see that the splays shots are affected by the "extend left" statement. All the splays are to left, independent of its angle with respect to the leg shot angle. Now we can see the manually reorganized data to properly orient the splays.
55 . 4.01 275.9 -2.6 extend right 55 . 3.34 235.6 67.3 extend left 56 55 7.04 123.5 -1.7 56 . 2.15 196.9 -5.0 56 . 0.69 32.1 -7.0 56 . 1.06 176.0 45.1 56 . 2.31 229.4 21.3 56 . 3.37 240.2 22.0 56 . 3.60 230.6 22.4 56 . 2.97 160.0 -7.5 56 . 3.62 111.1 -6.8 extend right 56 . 3.80 244.2 41.0 56 . 0.70 303.3 -84.0 56 . 2.39 244.0 20.5 56 . 2.43 240.7 -6.1 56 . 2.56 269.3 -7.0 extend left 57 56 6.37 130.1 2.2
The result is better, but splay are not a the same position in the sketch. Now we see something else that Topodroid does with the splays. Not only is it orienting the splays, it is also calculating its projection on the plane of the leg, which is better for a better extend sketch. Seen these examples I think the same as Marco, that Therion should have a "extend splay-auto" statement, that projects the splays on the plane of the leg shots, depending on its angle with respect to the leg shot angle. This can only be done from Therion, since if you have to do by Topodroid export, it could only do it by modifying the splay data. And in the meantime Therion does not have that function its compiler should ignore an "extend splay-auto" command, which the data capture and drawing programs could already incorporate into their Therion exports. To be recognized in the future when Therion incorporates this new function Best regards, Evaristo. El 07/06/2018 a las 22:02, Bruce Mutton via Therion escribió:
Hi Marco
I am not sure I understand, let me test my thinking, for everyone to pick apart :)
I think the extend options provide control of the non-splay survey legs. In general that control amounts to left, right, vertical, not at all, or not visible.
The splay legs should just ‘wobble around’ as a second order effect, in response to the extend parameters applied to the non-splay legs.
So the only sensible behaviour for splays is ‘extend auto’, and my observation is that this is already coded into PocketTopo and Therion, and so there is no reason to require the user to invoke ‘extend auto’ manually.
As noted in previous threads, Therion and PocketTopo do manage to orient splay legs differently. Quite often I have to mentally reorient splays when drawing in Therion Drawing Editor. Neither application seems to do it perfectly, but then because the shape of the extended centreline network is as much an arbitrary artistic choice on the part of the user, as it is based on hard numeric data, there is perhaps no ‘right approach’ for the software to take. I suspect that I have observed that TopParser can affect the orientation of splays as well, but I am not sure of that.
I think the right approach is to aim for Therion (and PocketTopo, TopoDroid) to orient splays automatically, based on the extend directions of the incoming and outgoing non-splay legs. If the applications were to allow direct user control of splay extend direction, then the statements should not be able to be confused with existing statements that control the non-splay legs (ie adopt extend splay left, extend splay right, extend splay vertical, extend splay auto). Perhaps this topic is something the authors should collaborate on to achieve a consistent behaviour?
What do you think?
Bruce
*From:*Therion <therion-bounces@speleo.sk> *On Behalf Of *Marco Corvi via Therion *Sent:* Thursday, 7 June 2018 11:29 PM *To:* therion@speleo.sk *Cc:* Marco Corvi <marco.corvi@gmail.com> *Subject:* Re: [Therion] Conflict between "extend ignore" command and inverse legs
I have commented all the splays "extend ignore", and Therion is doing now a good job, ignoring the appropriate leg (extend ignore 83 92). All the splays "extend ignore" was confusing the Therion compiler.
This data is an export from Topodroid, that incorporates the "extend ignore" command before the splays shots. I will comment to Marco Corvi on the problem.
@ Evaristo & Bruce:
what about a command "extend auto", which means let therion decide how to extend the splay shot ?
for the moment therion parser could behave as if the command weren't there.
marco
_______________________________________________ Therion mailing list Therion@speleo.sk https://mailman.speleo.sk/listinfo/therion
participants (3)
-
Bruce Mutton -
Evaristo Quiroga -
Marco Corvi