Create additional intermediate table to map node-based artery codes to TCL segments (related to #2)

[aharpalaniTO]

Present thinking is that this will consist of five steps:

1. Automated assignment of N, E, S, and W legs of intersection to specific TCL segments based on spatial distance to node and azimuth of segment relative to node (done)

2. Identification and "correction" of segments where the assigned direction is likely to be misaligned with that reflected in the turning movement count data (i.e. Kingston Rd.) (hopefully automated - to be looked at)

3. Additional segment corrections that are not picked up by the automated process in steps 1 and 2 (lookup table manually created - not done yet)

4. Translation of the results from 1, 2, and 3 into a detailed intemediate table that maps each (artery code-turning movement) combination to a (TCL from segment) and (TCL to segment) - (automated process, to be developed)

5. Result set from 4 is updated using a seperate table consisting of manual changes that reflect "strange" intersections (i.e. Lakeshore and Lower Spadina). - (manual process, to developed)

The table I envision looks something like this:

artery code	movement	tcl_from_segment	tcl_to_segment
5604	north_l	1235461	12348687
5604	north_t	1235461	186532
5604	north_r	1235461	832453
5604	east_l	12348687	186532
...	...	...	...
5801	north_l	83612	654123

[aharpalaniTO]

Have addressed Step 4 through creation of prj_volume.tmc_turns. Does not account for directionality of segment at the moment. Need to still build "manual mappings" table, subsequent update script, and porting over to final volumes table.

[aharpalaniTO]

Addressed through prj_volume.tmc_turns_corr.