Geodatabase Fundamentals

Video Statistics and Information

Video
Captions Word Cloud
Reddit Comments
Captions
okay hello everyone and welcome to today's webinar titled geo database fundamentals my name is Ian green Smith that I'm a higher education Alif working in the education research group of our Toronto office those of you who have intended one of our webinars before I'm recording this session and we'll make recording and PDF copy of the slides available to you shortly after the webinar finishes in this webinar will discuss the primary data storage format used in the ArcGIS platform specifically the geo database and discuss the benefits of using it in your teaching and research first we'll discuss what it is why you would use them then take a look at the different types of geodatabases then we'll look at how you go about creating one and loading a schema and we'll discuss some data loading techniques and approaches for working with vector and raster data and lastly we'll look at how you can invoke behavior rules that will help users seeking to make edits to the data across the ArcGIS platform in this first section I'll start by introducing the geo database discuss its role in the platform and then discuss methods for creating them and schema loading data techniques the urk GIS platform has a number of different components as you start to explore the geo database it will help you have a general understanding of how things fit together storing your data inside the geo database allows you to access your data across the entire platform meaning that you are not limited to the traditional desktop applications like arcmap in our catalog to manage your data create maps in the form analysis this data can be published the ArcGIS server or ArcGIS online to build web mapping applications that can help extend the reach of your data to a wider audience these same maps can be consumed on mobile devices like tablets or smartphones for easy access in a mobile environment so while we'll just be focusing on the geo database and how we can use it it's important to keep in mind the end-use or actually the end user that helps you in mind simply put the geo database can be defined as a collection of geographic data that are easily managed and scalable depending on the intended use the geo database is for ArcGIS data model and the primary means for organizing editing and analyzing spatial information within ArcGIS based on the relational database model a geo database stores the database schema or its design which is essentially rules that govern its properties and behavior that will let it we'll cover a little bit in this in the seminar so while it's true that arcgis for desktop can read and write to numerous spatial data formats none of them are as functionally capable on their own as data stored inside the geo database to follow up on this notion of functionality this includes store and restoring of fundamental data sets such as tables for non spatial data feature classes for vector data and raster datasets for storing or even referencing imagery and continuous data the geo database provides behavior rules and relationships to maintain attributed integrity it also contains tools to help maintain the integrity of spatial data by implementing advanced data set select apology from modeling coincident features geometric networks for utility modeling and networked datasets for modeling traffic flows the geodatabase can also store custom tool boxes containing geoprocessing tools Python scripts and models built in model builder it is you can create and store an address locator directly within the geo database to use for geocoding tasks the last point I'll make relates to scalability and integration so kind of on the bottom of the slide towards the bottom left a geo database can store and manage spatial data from the smallest project right up to multiple terabytes of vector rest or in tabular data in doing so it supports a range of simultaneous users and editors and this can range from a single user managing data in a file geodatabase to complex workflows involving multiple editors in an enterprise geo database stored on single server oracle or PostgreSQL continuing on with this idea of scalability there are three basic types of geodatabases the first is personal which was introduced to back at that really you the inception of our GIS for desktop at version 8.0 and this is a Microsoft Access database the next the next geo database that came along as actually as an evolution of the personal geo database and this is the file geodatabase so this is for just a little bit later at our GIS 9.2 and rather than mean Microsoft Access based it's actually a collection of files in a system folder that's stored on disk so not not dependent upon access at all and the third one that that all I'll mention here and we'll actually discuss it a completely different webinar relates to the enterprise or arc ste geo database which is a collection of tables views and functions stored inside a database management system but the types of geo database is listed on screen the file and personal geodatabase are most similar and that they are both designed for single user editing or the plot however the file geodatabase is recommended over the possibility to handle larger data and how it handles multiple users personal jus databases have a two gigabyte storage limit compared to ethology database that has a one terabyte default stores on it per data set so it's a massive jump already and the reality is it's actually even a limit that's hard to reach you're actually going to be limited more by your infrastructure your hardware than by the geo database itself so there's a massive massive improvement between the two already in fact especially for for large data sets but raster data the maximum size can be increased to 256 terabytes per form based on configuration keywords within the file geodatabase built on top of this are that it's actually it's possible to have more than one editor at the same time editing data within a file junit it show database provided that it is that the users are editing different feature data sets standalone feature classes or tables within the geo database it basically means you can't be editing the same feature well how does this compare to other data types that you might be familiar with pathology databases actually also preferred over the shapefile as it outperforms the shapefile for operations involving attributes and its data size can scale way beyond the limits of the shapefile which is actually debased based including those additional dentists that I mentioned just earlier for the purposes of this webinar though I'm just going to be focusing the discussion today demonstrations that I show in and around the file geodatabase creating a geo database is actually relatively straight forward in fact out of the box ArcGIS desktop creates a geo database for you the services default for every math document or MXD and default working directory for any geoprocessing work that you want to do in practice however it's always a good idea to set your own working directory which could or would house the geo database that you're working in now for those of you who are working with ArcGIS pro you'll notice that pro actually works in a project style environment and therefore actually creates a new default geodatabase for each project that you create creating a geo data this can be done either in arcmap art catalog or arcgis pro depending on your preference and can be manually or automated or menu clear is created rather or automated using the tools available in arc toolbox once the geo database is created you'll need to decide how to model your data by giving it some structure this is known as creating the geotubes geodatabases schema schema refers to the physical structure of the database along with its relationships and optional behavior rules for each constituent data set to define and implement a schema you can either load existing data create it from scratch or use a data model template like the Canadian municipal data model which is a proposed standard model for Canadian Missa palette ease regardless of the method that you choose all of these options will work but you still have to consider how you design your geo database as you develop it once the skin is created an easy and popular way to share it is through an extensible markup language workspace document or XML workspace document which is designed to transport and store your schema and data now in our chess space an XML workspace document is used specifically to hold the contents of the geo database so using this format you can import an export all items on the geo database such as any behavior and using the data advanced data types such as topology and geometric networks in fact the transferability of the Excel workspace document is a significant benefit of the geo database as a collection of over a collection of non-related shape files as once it has been created it can be easily shared and potentially modified or customized by others so I'll transition out of the slides here we'll just let me show you just some basics of the geo database just to get started so here we are in arcmap the first thing I want to show you is every time you start arcmap you'll you'll miss to come to the same home directory that is right here so tied to your user name here is I grain Smith that that's me within your documents and that are in my car GIS folder and you'll have something similar to this now depending is pros installed if you have any Python add-ins that have been added in this directory might look a little bit different but what I want to draw your attention to is the default geodatabase now I can go into here I can see that over time there are a wide range of datasets that I've written to this geo database and that's fine it's it suits it suits the purpose but remember as I said earlier a common practice is to change the working directory so that is creating a new geo database or pointing to a geo database that you're already working with you notice here that as the default is set here as my eyes even my default geodatabase I can go into my workspace environments here under geoprocessing environments click on workspace and you'll see that this is also a set so it doesn't matter really how I can really go about adding data in even you still need to account for the workspaces or the workspace environment that you're working in so having if you're doing any geo processing tasks specifically having this set ahead of time will drastically reduce the amount of time you spend either retyping the workspace or moving moving data around after the fact so I just wanted to show you that really quick now beyond that you can kind of you can see that a geodatabase can be created just from here in arcmap so I'm working in the catalog window I can right click within the catalog window scroll down to new and just choose new file geodatabase I can give it a name in this case I'll give it I'll give it the name that relates to something else I want to be a little bit later on I'll give it that name there and that's fine but it's also a system folder as well so I could actually go out to my system folder here find out where to find where it's saved Oh backup of it and here I have the same folder that I can access I can go into it but you're not going to see anything that's a bits of use - these are all proprietary things that are tied to tie to the geo database but the point I want to make here is that it's a system folder which is different than the personal geodatabase which is the access-based so i could actually open up microsoft access and review the contents of the geo database moving on from that we can actually switch our focus on to the schema itself and giving the geo database a bit of structure so I've got a browser open here I'm just going to drag this over so you guys can see it now this is I'll share this link with you if you want but this is actually a home page for the ArcGIS for Canadian Municipalities where every there are a number of apps that have been built based on this this Canadian municipal data model here so that I read it to take a look at that just as a just as an example I just want to show you one quick way that I can actually download the Canadian municipal data model so I'm actually downloading this ahead of time and you can actually take elements of it and continue on and actually work with it on your own modify it and do whatever you want the point is that you're taking a pre-existing schema applying it to your own geo database and doing your work with it so here I have the Canadian municipal data mall see MDM in an XML workspace document now ahead of time just to save a little bit of time for the webinar I imported this into this file geodatabase so if I double-click on here you'll see how to expand this a bit for you you'll see that there are a wide range of feature data sets tables relationship classes you sponsor rosters as well if I go into any of these or this is where I go from the feature dataset level down to the future class level so the future data set would store feature classes as well as relationship classes and advanced data types the reason I bring it up to you is that I can actually export elements of this schema and use it for my own purposes and so the second example what I'll do is I'll take the address feature data set the parcel publishing features data set here and one more the reference data set I'm just going to right click and choose export I'm going to export that to this XML workspace document reminisce to see this is the file that we can really use to share content on or extract content or in this case import and export content so I'll let that go just a couple of things I want to point out here this the CM VM geo database is actually empty but you can store all of your data in the XML workspace document as well you'll just notice that the file size jump substantially for the purposes of this webinar all I really want to do is just take the schema so I'll take this all I'll export it out to where we're working I'll just call it C mdm2 or Mesa a bit more descriptive small every cm VM small I'll save that now it's going to go through and say exactly what it's going to be extracting from here so you'll see there's the addresses parts of publishing and reference data so there's a lot of information in terms of schema and design this getting extracted but no data this poster should go relatively quick there we go so come back up here refresh my page and I've got my cm DM small here what I want to do is as I mentioned before is I want to take this and import it in here so you see here that it's empty I can click into it and choose import and I'm going to import an XML workspace document I just want the schema to come in point to this it's going to tell me exactly what's going to bring in now click finish I love that import and there we go there's no need to refresh the to refresh the geo database you can see here that I just have a small set of a much larger data model in here that I can actually start to work with and that's really what I wanted to show you in this first demonstration so hopefully you get the good idea of the default geodatabase the default working directory how to create a geo database and how to move schema around so now that we have reviewed some geo database types and modeling considerations we can now move on to topics related to loading and storing vector and raster data the following table describes the various methods for loading data into geo database now the point of the table is not to overwhelm you but rather present options for loading data based on your needs the simple data loader is exactly what it sounds like it's simple it's simple to use it's a tool access from the catalog window or actually in our catalog as well we're loading data into existing data sets and works well with simple data sets like points lines and polygons but can't be used for complex features such as topology geometric networks or network data sets if you want to use something like that then you use the next one you use the object loader and the object loader is actually used or an access during an added session and this is where you'd use what you'd use to load in complex data but there are there are other ways to load data in as well so you have conversion tools so this can be loading between different data types so whether it's not the ones that I'm going to show you today relates to working with shape files CAD files and so on but if you had other ones or even if you had data that was already in was already in another geo database you can move that data around with dragging and dropping it between geodatabases or you can use conversion tools like any of them that said that are left on the conversion tool set an easy one to give you that love is the projective tool so that's just switching between switching Pidgey switching let me just catch my might wording here switching between projections immersion is done on a one-to-one basis so you won't combine datasets in this operation as opposed to compare to the simple and object loaders which they append to data in this is so it's going to create a brand new data set each time the last one I'll say is actually probably the most useful this is a geoprocessing tool that you may already be familiar with but it's actually ideal for quickly loading data into an existing feature class within the geo database if you know that the scheme is between the data sets match and I say that but the reality is it doesn't matter if they match either because you can you can actually just force force the data in as well on the other side of things we we need to consider raster data the best way to manage raster data in the geo database is through the mosaic data set now the storage was introduced back at ArcGIS 10.1 it allows you to manage display serve and distribute raster data they're created edited and managed with the tools of the mosaic data set toolset but there's a lot you can do right from the catalog window as well one of the key points of the mosaic data set model is that it does not load the rosters into the geo database the mosaic data sets stores references to where the data resides on disk but allows the geodatabase itself to manage them there are actually a few benefits to this first it reduces duplication of your data I'm sure those of you are on the session today and those who you are going to be listening to this later fully appreciate what it's like to have multiple copies of the exact same data set or just clips or smaller versions of the exact same data that or something that follows the same schema mosaic data set takes care of that another one is that it doesn't actually take long to add imagery to them as a data set the last it doesn't it actually allows you to combine data of different types resolutions and formats all into one was a data set so whether you're working in desktop or ArcGIS Pro you'll have access to a mosaic data set as an image with all the required processing applied on the fly as a catalog of footprints and metadata making it one seamless data set so what I'll do here is I'll actually switch back to arcmap just show you some of these data loading techniques there we go so now that we've actually have our geo database created now granted I created one for for a different example but what I want to show you is how we can lower it there's some data that's been actually downloaded from the city of Kamloops pcs open data site I want to show you how we can actually bring that in so I have an existing geo database here I can double click into that here I've already got one data set added into it that's just that's just an urban area but what I want to show you now is how I can create a couple of different things that are that are useful I'll show you how we can easily start bringing in data the first one is something that you heard me refer to but you may not have caught the significance of it so the first one is the feature data set I can do that just by so I'm working here if you understand what you're looking at within the catalog window so you have your directory tree up at the top and what's actually in the directory underneath so I'm working within the canvas geo database I right click on it I click new and because I'm inside the geo database I can choose let's create a new feature data set now for this example I just want to give it a name of assessment I can accept a number of the defaults because I know where the data is coming from I know the projection that they're working in I know that it's an at4 UTM zone 10 so I have that saved in my favorites so I can easily grab it and keep them keep working I'm not working with with any vertical coordinate system so I can actually ignore that and as a best practice it's a good idea not to change any of the tolerances that the geo database provides you these are actually quite quite accurate so I'll click finish and you'll see here in the catalog window that the assessment feature data sets been created so now what I'll do is I'll actually right click on the featured data set and choose import and what this is going to do is open the simple data loader de minimis of a misnomer really it's not you you don't see anything called simple data loader it's really more more of a workflow I suppose you get to put it that way but for those of you who aren't over the road with geo processing tasks this is a common tool that that or layout that you would see so you'd have your input features your output features and in any additional details that get mapped what I like about how ESRI has really built these tools into ArcGIS is that for each text box for each field that you're required to do something you have to help that comes here so it's going to tell you exactly what it's looking for so for the input feature it wants to know okay well what what actually do you want to add in here I want to add the property information here this gets coming from a shapefile now it's going to populate what that what the fields actually look like these are the attributes I'll just give it a name property just keep it simple and click OK it's a bit of it's a big data set but it actually goes quite quick and it adds it into the map as well so I can close this right click on right click on my attribute table and I can see here I have all that information loaded in and in my table of contents here as I'm listing by source you can see here that this property feature class is stored within the assessment feature data set which is stored in the Kamloops geo database so now let's set let's let's get some other data in here because it's well it's need to talk about just just one specific aspect that as well show you how to have some other things work so I'm going to right click again choose import but this time I'm actually going to choose import multiple so in this case I'm going to browse to some different folders the first one here I need I want to bring in these address points so I'll click that I'll add that in that I have some planimetric data so I have buildings and I also have street data that I want to add in and I'll just click OK now you'll notice that I didn't have to add a output feature class name all it's going to do is just going to take the input name and just make that the make that the output name click OK now take just slightly a little bit longer but still it's actually quite fast and there we go so this time unfortunately didn't add it to the map but that's not that's not the end of the world so I can just come in here grab all this data for me and just add that in so now I have quite a bit of information that's in here that is that's actually layered appropriately as well that's simple data logger and that's what that's all I've done the next thing I want to do is actually want to simulate if you will what would happen if focusing on this amis address points for example if I had an update if the exam I was working a municipality for example or if I was working on a student project and I there were multiple students working together we're all working in the same area and at some point we need to bring all this data together to do that what I can do is actually use the append tool so easiest way to find tools in is then arc map or ArcGIS Pro sexy G's the search functionality so I just type in append and it's going to give me a couple of options the one that we're actually specifically interested in is the append data management tool so I'll click on that again it's going to give me the tool help that I'm interested in and I go from there I'm actually getting ahead of myself because I needed to show you the data set that I wanted to bring in the data set I actually want to bring end is right here so just some simple points we can simulate that this is a this is a new subdivision that's going in I'll open the tool back up the input data set in this case is actually going to be these oh let's close this and reopen it cuz it's not picking up the one data set real the input data set is the address update the target data set is my address points now if I minimize this I actually open the address attribute table in the address point attribute table you'll see here that these from the most part match now one is coming from a shapefile and the other one and another one stored in a geo database so the address point is now in the geo database but in the shapefile they had a field ID so it looks they look different but actually it's just the the object ID that's changed so it's pretty safe to assume that they have the exact same schema so going back to the append so I can actually say no test set the schema types and no test and no test means it's not required to this the schema match because I actually already know that the schema match I can click OK this will happen quick as well click close and actually what I'll do is I'll just turn off the address update and you'll see that the exact all the points that we're out of it were appended and directly to the data set so that quite quite useful tools to use to spend especially helpful within that within a geodatabase and loading data into the geo database we zoom back out again and I'll give you an example another example here of working with a different type of data that is that working with Tod data so here we have our cameras to Davis as well but I also have some CAD data that I want to bring it now I'm not going to take you guys through all of the CAD data being imported but for the same example what I want to show you is how we can easily bring in one one dxs file and import it right into the geo database what I'm going to search for here is on the search for CAD so there's a CAD tool set now these are two cats so that's implying that it is an export or conversion but what I'm actually looking for is the CAD two geo database tool so my input here I'm going to browse to it specifically just going to show you the the neighborhood's DXF file here you'll notice here that that's actually going to bring it into the canvas geo database and it's going to create a new data set called neighborhood in this instance but it could be it could be anything that you wanted what the tool is actually going to do is whole let this run is import the data as the data to the map but if we go back to the geo database here you'll see that that data set is actually a feature data set that it's a it's referring to and what does that an in is all of your CAD data here and it adds it in here as a group so that's the last thing I wanted to show you in regards to vector data the next thing I'll show you relates to raster data and now I want to show you a little bit more of the of the mosaic data set so to do that what I'm going to show you here is this directory here so still working in my folder my working directory here I have a folder of six rosters and these are all from Kamloops BC as well now you'll remember that the mosaic data set it doesn't actually import any of the imagery into the geo database what it does is just references it so what I'm going to do is I'm going to go back to my Kamloops to your database right click on it choose new and new mosaic data set I'll give it a name I'll just call it imagery much like the feature data set you want to assign a projection to it so that's our zone 10 and I'll just let that run and you'll see here that it's been created and my default added to added to the map as well but there's no data in it it's really just a reference that's that's that it's loaded so I'll go into my canvas new database and you'll see that I have my imagery to expand this a bit my imagery mosaic data set right there that's great but what about these six images that I just showed you it's as simple as really just right clicking on the image or mosaic data set and choosing add rosters and then this what you would do is you would set the type so in this case it's a raster data set but we all know that ArcGIS can work with a number of different digits formats I have all of it actually stored in a workspace so that's my vector workspace I'll browse to that here select it and then I'll just click OK I'll let that finish and the next thing I just want to show you is if I actually hit the full extent here I now have six images and six footprints add to the map at my entire boundary that delineates the the full extent of the of the mosaic data set and then I have footprints and beyond that now this can be clipped beyond you can do further analysis you can you can stretch and skew and modify this beyond it but for the time that we had today the goal is to really just show you the mosaic data set as an interesting tool and a huge benefit of working within the file geodatabase so with that I'll just get out of this map document here let's get back to the slides a GIS is actually only as good as the data its doors in this last section we'll discuss the concept of behavior and how different rules can be implemented to enforce both spatial and attribute integrity of features stored in the geo database data integrity is a crucial element in the evaluation of a do data visit trustworthiness functionality built into the do database such as attribute domains and featured subtypes help ensure attribution standards are enforced while leveraging advanced data types will handle complex relationships between features of participating feature classes such as subtypes are designed to represent a subset of features as a method to categorize I'm going to stress categorize or stress upset and stressed categorize they're really synonyms in this in this case so categorize features within a feature class or value in a table that had the same characteristics these subtypes are implemented by assigning coded values associated with either a short or long integer field type where the assigned subtype value represents a group or category of features from a specific teacher class in the example as you can see on the slide all secondary streets would exhibit one behavior and all the highways would exhibit a different behavior because they've been assigned different subtype values one more thing to mention here and it actually goes for any majority of actually the things that I've shown you here today that it's possible that you can automate the adding removing and controlling of subtypes directly within direct from our toolbox or if you're savvy with modern builder or Python scripting you can do that you can do that as well another behavior rule that can be used to manage attribute data or attribute domains is the domains rules that define the accessible values that a field can have data integrity is ensured by limiting the choice of attribute domains only to those fields whose data type match compared to those features subtypes that I talked about in the previous slide domains are applied to the entire geo database meaning that they can actually be reused as rules for multiple feature classes tables and even applied to specific subtypes there are two types of attribute domains that I want to talk to you about first is range domains these specify a valid range of values for a numeric data so just like the table on the screen shows you an acceptable range of new variant of numeric data values for service connections within a water distribution network you can set the minimum value at 40 psi for example and the maximum value at 100 psi and then in the editing workflow if someone added an to the value that's above or below that range when validating they would receive an error so they know that something's wrong the other side of this are coded value domains these specify values for each class a common application of domains is to enforce color value data classifications such as zoning codes or less for land use types so zero for industrial one for residential at two for commercial they what it shows on on the slides you can also work so doesn't necessarily have to be a numeric value for coded values actually you just put in an abbreviation as well the do database can also store relationships between data and attributes in the form of a relationship class now this is not a new concept as the majority of us are already familiar with joins and late however compared to a one to one join or a one-to-many relate that we typically do to enrich our data within arcmap a relationship class can be created within a geo database that serves as a permanently linked I'll stress permanent and you know you'll see where I'm going in a second to manage associations between contents of one feature class or table that is the origin and another so its destination feature class or table that's within the same geo database relationship classes help ensure referential integrity between related content this means that a value in the foreign key field must reference of valid existing and equivalent primary key in the related table impaired temporary joins and relates that only persist within a one map document relationship classes are stored in the geo database so this is the permanent determinant that I was getting at now depending on the intended use there are a few types of relationships classes and a simple relationship related objects can exist independently of each other for example in a parcel feature class like the one I was using for Kamloops you can have parcels that have one or more buildings contained on them by comparison a competent relationship relates to instances where the destination object cannot exist independently of the related objects so continuing with the parcel and building examples from above if the parcel or the origin rather is deleted than any related building that's associated with that parcel is also deleted one last slide here and then we'll jump into our last demonstration relates to advanced data types that can be stored within the geo database itself these are intended to handle complex relationships between data sets due to the nature of connectivity needed these data are stored in a common feature data set so examples of advanced data types that can be stored in a native geo database are topology so to more broadly defined what topology is it's a collection of rules that enable geodatabases to accurately model geometric relationships that are stored in a feature class and network data sets are well-suited to model transportation networks as I mentioned before they're created from simple source features such as lines and points but can also incorporate turns and store that connectivity of source features so if you're doing an analysis with the network analyst extension that analysis is always taking place on a network data set similarly is similar to the network data set is the geometric network so this is comprised of a network of connected edges and junctions used to model behaviors of common municipal network infrastructures such as gas power water wastewater networks so I'll jump I'll jump back to this this last demo and then and then we'll come back and we'll regroup so we're going to load it into the Kamloops geodatabase what I want to focus on here is this street data set I want to add it to the map if I open up the attribute table and scroll all the way to the right you'll notice that I have a classic I here but it has a classification now remember I mentioned there you can categorize subtypes within within a future class or I want to use this classification field to do that but to get now I've got 4,000 records here get a count and summarize all of those potential classifications I'm just going to right click on the classification field choose summarize I'm just going to write this back to my camelids geo database just call it class which is okay I'll add it to the map I'm going to open it so now you can see here let's make sure my screen refreshes the same times yours now you can see here that I actually have ten different classes now at the time that we have left I'm not going to go through and add all these sub sub types in but there's potential for you to do that and this would drastically improve the attribute integrity of the feature class if you incorporated subtypes like this because if you what you can do is by setting an individual subtypes you can say okay for that subtype it has specific values and only those values apply to that specific subtype and go on and on for the nine remaining classifications of roads what I want to show you is how we can go back to the j/s streets database here and create a new field I'm just going to call it code I'll leave it as short because remember that subtypes are picked up by as a short or a long integer field I've got that created now I'm actually going to go in my catalog window right click on streets go to the properties and click on my subtypes tab I'll actually come back to the fields properties here when I talk about domains but come to subtypes tab and now I want to let arcmap know or let ArcGIS know that when we're working in the geo database that the code field is my contains all my subtypes so here just just to show you I'm going to add in one and then like them local to just just a few just to give you a give the idea of how it works and the last one is rural I'll ply that now these are ready to these subtypes are ready to be used it is not necessarily represented in a data set yet so what we do is just add some good old fashioned select by attributes so using my select by attributes I'll scroll down to the classification field I want to select everything that is related to local which is ok so now I have over half of the features selected and without going into an edit session I'm going to open a field calculator let's just ignore this this message here I'm going to be they get a bit code value of 1 this remember my code the subtype value that I entered for local was 1 so I'll click OK and now notice here if I expand this that anything that has been given a subtype of 1 has this code now of local or produced that process two more times there's major arterial so it's just over five hundred okay I can just do this by showing the selected features right click give the value of two okay again same as before is your arterial and then the last one is that just rural so although do that really quick as well so here I've got ten ten features calculate that to be three and there we go so initially you won't see anything that says it's different so what actually going to do is I'm going to remove this streets layer and I'm going to add it back in now what you should see here on the table of contents is that you have three classes these are three categories of features if I actually go into our properties for the layer and go to the symbology tab you'll see here I have these these features here and if actually turned off those that aren't symbolized and click OK you'll see that those actually aren't they don't have a classification at all will be removed from the map now that's that's just a symbology example but the point is it's an easy way to classify your data because the further the further I go with this I can actually define ok for each subtype I can say ok well for every every suffix can have I can assign domains for that as well and that's the next thing I want to show you so let's turn this off and what I'm going to show you is actually relates to the property a feature class that we were working with before if we open up the attribute table of and scroll all the way over here you'll notice an owner field now you notice that some of these things so some of these things are blank now if you're part of it anything you're looking at it from an editing perspective you don't want to have missing data so domains can often help in an editing workflow just to ensure that if data is being is being typed in for example there are no spelling mistakes it's a consistent look and feel so what we can do is actually go it's a similar workflow that we did for the subtypes is summarize what this looks like it's going to summarize I didn't call this owner allow this to the map you can see I've got four classes so city federal private and province and what I can do here because I'm actually working at the feature or rather the geo database level and I can go right into the geo database properties and go to the domains tab and actually create a new a new domain so quickly I'll create a new domain it's actually going to be a text field type and because it's a text it's going to be coded values so in this case I want it to be a city city and that is that's how you create the domain so just like the subtype initially when you set it up you you're making it available in it within the geo database to be consumed but it hasn't been applied to anything where that comes is actually back at the feature class level so I'll open the feature class for properties for the property data set and go back to the fields tab and scroll down to my owners field here and notice that I have additional field properties to pop up I have a domain field here if I click that I can choose the owner domain and now apply that so if I go back and open up my attribute table and start an edit session and scroll over here and just click on one of these empty cells here you'll see here that I have a pop up these are the acceptable values that can be used for the four domains so for the specific field rather and it's all powered by a domain stop that at it the last thing I want to show you is actually still using the the same data set but in this case we're working within the within the assessment feature data set and because what we want to do is try to model on advanced data type and because we're working within the teach data sets I can right click into the siege of data set click new and then scroll down to apology and here you can see all of the advanced data types that can be created so from additional feature classes to relationship classes and so on just choose topology I'll accept the cluster tolerance in the default name at this point I only have one feature class that that is in the suture data set but if I had more than one for example like in a geometric network or network data set or even for a topology itself you can say well these I want these feature classes to participate and it's just as simple as checking them on if you had more than one you can rank them as well but just because I just just for the same example here I have one and that that makes it simple to demonstrate so click Next now you'll recall that topology is a collection of rules that accurately model the geometric relationships so the geometric relationships are here under this add rules section so under property if I had more than one I could apply this sauce there were there was if there was a line participating or if there was a point feature class participating different rules would show up but for this I would just want to show that we had I can say that none of the parcels can have gaps so I don't want to be any gaps between and then here it can say you know it will say exactly it will flag an error if it does not meet these conditions we'll click OK and click Next just gives me a recap of everything that I've created that but I said I'll click finish and I'll let that run just it when asked if you want to validate it that's that's entirely up to you just for the savings I'm just going to validate it on the site this might take a little bit longer but still we're not working with that many parcels that the city of campus is not not too big compared to some of the large anticip alleys across the country so it shouldn't take too long once that's done we're going to add it to the map and we're actually going to see the errors have been highlighted so again this is looking through looking at Rosslyn spatial integrity and interconnectedness of all features and vertices and determining if there are any gaps between any of the parcels that exist in the data set and every goal so the topology has been created allow this to my map yeah when asked if you want to add all the feature classes that participate in this case I can click no because you'll see that I already have my property feature class added and the pink or they have their base both they're both pink I'll change the color here just to help the pink represents errors that exist they said and to edit your errors or to review the errors you actually need to be in an edit session so right click on the property layer start editing and you'll notice here that I have two toolbars open one is my editor toolbar and the other one is that's apology toolbar so if you're not familiar with working with apology this is actually the little bar that is going to allow you to do all the work so here I can zoom into an area of interest for example this area looks like it has a bunch of a bunch of errors so that's that that's kind of interesting here I can open my an error inspector and based on the visible extent of that so now I've got a small rectangle to work through but based on that I can I can search I can look through all of these and it's actually identifying that there are errors all the way along here now if I want I can I can go through and inspect these I can right-click on it and marked them as an expected as an exception or using my using my topology tools I can actually even edit the topology right away and what that's going to do is edit the underlying data set as well so it's a great way to based on rules that you define to really have that spatial integrity it kept in check and that's the last thing I wanted to show you some of what we've covered over the past hour we've learned about some fundamental aspects of the geo database in regards to functionality and some advantages of using it we've also looked at creating a geo database in loading schemas we spent 11 a little bit of time we'll probably a little bit more time than I wanted on data loading techniques specifically regarding sector and raster data and then the last one what I think what I think you'll get a lot of value of is working with behavior rules that is both to manage the attribute integrity and keep these spatial integrity intact when working with and with invest with advanced data sets within the geo database a great starting point for geo database resources can be found at desktop at arcgis calm where you can find a quick tour of the geo database and a number of different tutorials like the raster data tutorial which is actually a good one for getting your feet wet with the mosaic data set some homegrown can the examples actually invite you to check out Azra canada's lesson planner and the URLs there on the screen where if you're not already aware you'll find a wide range of resources that relate what we've discussed today whether you're looking for desktop online or solution focused resources the tools built into the search functionality allow you to quickly refine and find learning resources that you can use a relevant example is actually based on the webinar that we've picture sitting in on are the geo database labs that cover a lot of material that we cover today these include a suite of tutorials and actual tasks assignments that walk you through the implementation of the K in municipal data model using both file geodatabase and an enterprise geo database a quick plug for the webinar happening next month titled working with multi-user geodatabases this SS is going to provide an overview of working with multi-user geo database in the ArcGIS platform and discuss the benefits of using them in your teaching and research will also discuss workflows will help you create an enterprise geo database add users and define privileges through roles and introductory aspects to versioning and replication workflows so there's a lot of a lot of content being covered there but remember one of the first things we talked about in the slide when it came to keeping the end-use in mind to the end user in mind those of you who are already comfortable with one workflow I'm not looking to I'm not seeking to compound things and make them harder to understand think about how you want to use it and then reverse engineer how you're going to go about getting that data out there if you're comfortable using the Jew database already that's great if you want to start using it more that's great too if you're wondering how few database can be incorporated into ArcGIS online and the apps like a collector for ArcGIS that's great we're going to cover that in the next in the next webinar but on behalf of Ezra Canada I want to thank you for attending the webinar my contact information is there on the screen so feel free to contact me if you have any questions and I'll make this slides now recording available is shortly after this so thanks a lot everybody
Info
Channel: Esri Canada Higher Education & Research
Views: 36,630
Rating: 4.681818 out of 5
Keywords: geodatabase, domains, subtypes, relationship classes, topology
Id: ocSd7NxPvBg
Channel Id: undefined
Length: 63min 26sec (3806 seconds)
Published: Tue Nov 22 2016
Related Videos
Note
Please note that this website is currently a work in progress! Lots of interesting data and statistics to come.