Meeting 2012 07 11 - QGEP/QGEP GitHub Wiki

Time: 9:15 - 12 - Place: Zurich, Josefstrasse

Agenda

  1. review geo-summit

  2. [archive of talks http://www.geosummit.ch/de/Vortragsarchiv.html]

  3. [Nutzen und Neuerungen von PostGIS http://geoconf.geosummit.ch/share/public/GEOSummit/web/2012/Vortraege/Neumann.pdf]

  4. sia 405 et al.: any influence ?

  5. [Neue Merkblätter zu SIA Norm 405: Geodaten zu Ver- und Entsorgungsleitungen http://geoconf.geosummit.ch/share/public/GEOSummit/web/2012/Vortraege/Burckhardt_Hunziker.pdf]

  6. [Die revidierte Norm SIA 405: Geodaten zu Ver- und Entsorgungsleitungen http://geoconf.geosummit.ch/share/public/GEOSummit/web/2012/Vortraege/Noack_Duetschler.pdf]

-> additional attribute "upper_elevation" in addition to VSA model (already added)

  1. OID: what's missing ?

Decision to add another table is_oid_prefixes where one record is set to active. QGIS will check with a project init function to see which is one is active will ask which one should be active if there are several entries.

This is the OID generator function. It takes two arguments: the OID prefix and the table name. It outputs the 16 character OID. I am not totally sure if 999999 (1 million) IDs per table will be enough for all cases. We can discuss this tomorrow.

To generate an ID for the table 'od_reach' for the Uster organization one calls:

SELECT qgep.generate_oid('ch11h8mw','od_reach')

and it returns something like:

ch11h8mwRE000027

Three additional changes need to be done in the model:

  1. the table qgep.is_dictionary needs to exist and contain the correct metadata.

  2. Each table that has OIDs should have a separate sequence:

CREATE SEQUENCE qgep.seq_od_reach_oid INCREMENT 1 MINVALUE 1 MAXVALUE 999999 START 1 CACHE 1;

  1. The default value should for the OID should be: qgep.generate_oid('ch11h8mw','od_reach')

e.g. ALTER TABLE qgep.od_reach ALTER COLUMN obj_id SET DEFAULT qgep.generate_oid('ch11h8mw'::text, 'od_reach'::text); One issue I see is that we kind of need to hardcode the OID prefix in the SQL code, which is not very nice.

Maybe we should store the prefix somewhere in an is_table?

  1. importing existing projects: the fischer-way

TO DO SBU (19.7.2012): Change numeric definitions of datatypes like (5,2) 12345.67 is (7,2) Timestamp: http://www.postgresql.org/docs/8.1/static/functions-formatting.html to_date('05 Dec 2000', 'DD Mon YYYY')

TO DO AN (19.7.2012): Write Function to generate valid STANDARDOID for existing DABASE Data CAB_NS1234 -> ch12dhkdxx001234 CAB_ = DABASE Prefix -> replace by INTERLIS Prefix, NS = class shortcut German -> xx = class shortcut english digits - fill up to 6 with zeros ahead

  1. editing-tool

KF describes process and from where data comes in each mask -> publish docu on github (KF) DR is responsible for pushing forward this tool, support by KF and UK

  • needed: data set
  1. next step for extend database (catchment area, organisation, weir/pump, ...)

tell SB if needed (create statement)

  1. next events

  2. November 2012: meeting AfU TG: KF has the chance to talk to all waste water engineers. Check-Service in TG: What could we present? Webinterface? (QGIS Web-Client),

  3. milestones ?

  4. Basic paper with project descriptions, steps done, future steps and milestones, funding tasks, priorities

  5. DR: QGIS 0 .. n / n .. n relations - for maintenance_events table connected to waster_water_structure

  6. next date

  7. ...

@4: importing an access-dabase.mdb by odbc to postgis, copy attributs etc to qgep-structure @5: modify geometry (move wastewater-node), modify attribute (change diameter) @7: 09.11.12: meeting AfU TG @8: OID, mod geo, mod att, dxf-export, webgis, LKmap