I'm in the process where I need to store data under a specific users account with Firebase. I'm fairly new working with the backend and I'm primarily looking for a second option before I start typing away.
This is how I currently have it structured:
Providers
City
Cincinnati
Company 1
Jobs
History
Company 2
Jobs
History
Columbus
Company 1
Company 2
My thought is it would be better to have each company listed within a specific city node when a user is requesting from that city. However, if the city was at the first level then we'd have to go through the city first. But instead of having the city at the higher level we can store the cities their in under the companies node essentially in an array.
So here's the user flow:
The user makes a request based on their location (city). The provider accepts that specific job which moves into the 'jobs' node. Once the job is complete it then moves into their 'history' node.
My question:
Should I keep the structure as is or place the city within the companies node?
Your structure is right, as city is your primary key in this case, since a company has a city, you will need to specify first the city where you want to pull companies from. I assume that this structure is the right way to go to solve your problem.
If your parent node is not City and just the name of the cities you will end up with a lot of children in your main tree, and that is what you don't want, you want to structure your database in order to do it readable when data is big enough.
PS: Think in scalability, if you think this way you end up with a good organized database, now, as you are doing it it's OK, but if you have another doubt, think in this way to improve your structure.
Related
I am currently exploring MongoDB.
I built a notes web app and for now the DB has 2 collections: notes and users.
The user can create, read and update his notes.
I want to create a page called /my-notes that will display all the notes that belong to the connected user.
My question is:
Should the notes model has an ownerId field or the opposite - the user model will have a field of noteIds of type list.
Points I found relevant for the decision making:
noteIds approach:
There is no need to query the notes that hold the desired ownerId (say we have a lot of notes then we will need indexes and search accross the whole notes collection). We just need to find the user by user ID and then get all the notes by their IDs.
In this case there are 2 calls to DB.
The data is ordered by the order of insertion to the notesIds field in the document.
ownerId approach:
We do need to find the notes by their ownerId field across the notes collection which might be more computer "intensive".
We can paginate / sort the data as we want - more control over the data.
Are there any more points you can think of?
As I can conclude this is a question of whether you want less computer intensive DB calls vs more control over the data.
What are the "best practices"?
Thanks,
A similar use case is explained in the documentation. If there is no limit on number of notes a user can have, it might be better to store a userId reference field in notes document.
As you've figured out already, pagination would be easier in the second approach. Also when updating notes, you can simply updateOne({ _id: "note_id", userId: 1 }) instead of checking user's document if the note actually belong to the user.
I'm an intern student at a company that does both wiring and aircon services. The job that they gave me was to make a database for them. I don't have any experience in anything related to databases.
So, I started to look up videos and stuff to at least learn a bit about databases and made something that works and I made it after 1.5 months of learning.
in the database that I created,
I have 1 table (CustomerDetailsT):
CustomerID (pk)
CustomerName
PhoneNumber
Address
Aircond (type and model of ac,ex: WM daikin 1.0HP)
AcDetails (what has been done for the ac.)
Others (yes/no) (Wiring, installing a fan and so on)
WhatHasBeenDone (shows what has been done for others)
Then 3 queries (CustomerOthersDetailsQ, CustomerAcDetailsQ, CustomerDetailsQ).CustomerAcDetailsQ has CustomerName, PhoneNumber, Address, Aircond and AcDetails. CustomerOthersDetailsQ has CustomerName, PhoneNumber, Address, Others, and WhatHasBeenDone.CustomerDetailsQ has CustomerID, CustomerName, PhoneNumber and Address
And 1 form with 3 subforms.
it's a search form, which would search for customers as we're typing in their name/phone number and it will show what has been done for the customer.
With this, I have created what the company wants, but now they want to add dates. Dates which would show when we have done something for a customer. Dates for Aircond and the Others stuff.
I've tried with what I know and it didn't work. tried searching it on youtube and google, but still couldn't find it.
how can I go about doing this?. I have tried having separate tables for each service, but it became a hassle when I wanted to create a new customer. . I hope I could some help, I could send pictures if someone needs them.
[1]: https://i.stack.imgur.com/mtrmC.png [The Customer search form] [1]: https://i.stack.imgur.com/A3Y9d.png [example of a customer that has ac installation] [1]: https://i.stack.imgur.com/dsGL5.png [example of a customer that has both ac and wiring done]
Acknowledging the question is too broad, here is some guidance. One of the nice things about Access is that each database is a single file. First protect your work by finding that file and make two copies. Make a backup and a play around version. Only mess with the play around version.
Your question indicates you are still learning Table Normalization and 1 to many relationships. Both of these topics are general to all databases, so you don't have to restrict yourself to just Access when looking for guides and Youtube videos.
Part of normalization is putting separate entities into their own tables. Also, in Access there is a big payoff for using the Relationship Tool, so here is a rather lame example of normalization:
Make sure to select the checkboxes when setting up relationships.
WhatHasbeenDone should also have WhatHasbeenDoneDate. I've wrapped AC and Other as Unit because later it will be easier than having two WhatHasBeenDone tables(AC)(Other).
Now imagine someone taking the customer request call. They just want to see a form to enter the customer details, request, unit-type, etc. They don't want to see those tables. Even with training entering data in the tables is error prone. The person fulfilling the request just wants to enter what they did and when. That's how you start to figure out what your final Data entry forms will look like.
Since we normalized the tables and used the relationships tool, the payoff is Access can give us an assortment of working starter forms. Select Each Table and then hit Create and then hit Form. Choose your Favorites and start playing around from there. While playing, keep in mind that Access will not let you add an item on the many side of a relationship unless there is an item on the 1 side.
For example I selected the customers table and hit create form:
Access uses a concept of form and subform based on separate but related tables. So, to get a form that shows what has been done for each customer I created a form for the What has been done table, and dragged it onto the customers form:
Unless an ID is also being used as a part number or something there is probably no reason for the person entering data to see it. So I removed the texboxes bound to ID's. Except for UnitTypeID, where I replaced the textbox with a combobox that displays the userfriendly UnitDescription. The ID's are still part of the form recordsources, Access is still adding new IDs and using those IDs to put the appropriate data in the right tables.
Oh, didn't we need dates (went back and added a date to the table, and adjusted the subform accordingly). Also changed the subform format from single record to continuous records to show multiple dates:
In conclusion and in my opinion your final forms will use VBA behind the scenes to insert data from the forms into the tables. This is because either you will want to rapidly insert multiple records or How the end users think about the data will not match the default forms and subforms approach Access depends upon to figure out how to insert the data. However, the default approach is fast and I always use it for version 1 of my Access Databases.
P.S. For simplicity I avoided including any Many to Many relationships
Scenario: a simple address book where a user can create his own contacts and organize them by adding them in groups. A contact may have multiple addresses.
I have created the following diagram:
![schema-design][1]
I want to query all the contacts who are placed in group x and live in country y.
Is this schema design good enough for those purposes (I want to use the neo4j database)?
It looks like the notion of country should be a first class citizen in your graph since your query depends on it. Graph model design typically gets influenced a lot by your query patterns.
So I suggest to have a node labeled Country for each country and connect the Address node with :LOCATED_IN relationships to the country. (consequently drop the country property from the address nodes).
With that change your query is as easy as:
MATCH (:Group{name:'family'})<-[:placed_in_group]-(contact)-[:lives-at]->()-[:LOCATED_IN]->(:Country{name:'US'})
RETURN contact
One option is to have another node from address for country, as pointed out by stefan Armbruster. If you do not want to change the data structure, just add an index to the field "country" of Address. Then you can have a query
MATCH (:Group{name:'family'})<-[:placed_in_group]-(contact)-[:lives-at]->(:Address{country:'US'})
Closed. This question needs to be more focused. It is not currently accepting answers.
Want to improve this question? Update the question so it focuses on one problem only by editing this post.
Closed 8 years ago.
Improve this question
Note: This is a rough copy i didnt include constraints, weak entities, ..., etc yet. I still need to have a solid understanding of this question.
Questions:
To keep track of what theater company manages performer, what performer is in two theatre companies do i have to make a unique code for each entity set in other entity sets to keep track of them?
Can start_Location simply point to Place for the theatre company entity?
Can an Actor be Born in a place or does it have to have a attribute that points to place?
Do my relationships so far make sense?
Are there any redundant attributes such as Short_Descript in Plays?
Can i make an attribute in Place called "Town, State/Department/Province"? Or does it have to be a composed attribute?
Please note: I will be editing and updating my diagram if I have more questions and such...
I would appreciate any suggestions or hints.
ERD:
Question Information:
An actor is born in a place and he/she lives presently in a place (this information is mandatory).
We store in the database only the last known place where the actor lives.
We need the following information for an actor: actor number, actor name , date when actor was born, and date when actor died (check if died > born).
An actor is a performer, or/and a theater director.
We store for performer the date when he/she started to perform.
We store for theater director the date when starts his/her last employment as theater director
We consider in DBActors the following types of plays: drama, comedy and tragedy.
For each we like to store the following data: play’s number , play’s title , play’s short description , year when it was written ,date when it was first presented on stage(p_date_p, date).
For dramas we store also the drama type,name of the main positive character, and name of main negative character.
The drama type is one of the following:
“classical”, “medieval”, “renaissance”, “nineteenth-century”, “modern”, and
“contemporary”
For comedies we store the comedy type, the name of main
character , and the name of the second character
The comedy type is one of the following: “ancient mroman”, “ancient greek”, “farce”, “comedy of humors”, “comedy of manners”,
“commedia dell’arte”, and “theater of absurd”;
For tragedies we store the tragedy type(t_type, varchar(20)),and name of main
character
The tragedy type is one of the following: “Greek”, “Roman”, “Renaissance”, “Neo
-classical”, and “Modern”
A play is written by one or many dramatists
It is possible that we do not know the dramatist for certain plays.
We store in the database all known plays even if they were not performed (“closet plays”)
Some actors are also dramatists.
We store in the database all known mdramatists.
An actor is hired by a unique theater company at any timestamp
He/she will stay in the same company the whole year when he/she was hired.
We store in the database the year when he/she was hired by the theater company
(small integer)
It is possible that the actor changes the theater company where he/she is
working during his/her life many times. It is possible that an actor is hired by the same company many times in different years. He/she can perform in
one or many plays (at least one)
which are presented by theater companies.
It is possible that an actor is hired by a theater company and performs in a play presented by another theater company.
It is unusual but possible that the same performer plays in the same play
presented by different theater companies. A theater company performs/presents
one or many plays every year.
Same play can be performed by one or many distinct theater companies.
We like to store in the database the date when the play starts to be performed
by a theater company.
It is possible that the same play is performed by different theater companies starting at same date.
We need to store for a dramatist his/her dramatist number,his/her name.
A dramatist wrote one or many plays(at least one).
The information to be stored in the database for each theater company
is: theater company number,theater company name , date when the
theater company started.
For each theater company we store in the database
the first location (place) where the theater company started
There might be more than one theater company starting in the same place.
A theater company must hire at least one actor.
Each theater company has a unique theater director.
He/she starts his/her work at a specific date.
It is possible that the same theater company has different theater directors but at distinct times and the same theater director manages different
theater companies in distinct times(never at the same date).
It is possible that the same theater director manages the same
theater company at different dates.
The information to be stored for place is: place number, town and state/department/province, place country
Here are my responses to your questions:
Whenever you look at two tables and see a Many to Many relationship, you can solve the problem easily using a linker table. Also known as a junction table “is a database table that contains common fields from two or more other database tables within the same database. It is on the many side of a one-to-many relationship with each of the other tables. Junction tables are known under many names, among them cross-reference table, bridge table, join table, map table, intersection table, linking table, many-to-many resolver, link table, pairing table, transition table, crosswalk, associative entity or association table.” Wikipedia example You saw me use these tables in your previous question. In this case you are stating that an actor can be managed my many Theater Companies and A Theater Company and also manage many Actors. This is a many to many so if you created a link table in between those tables for every relastionship between the two you’d add a new row in the link table that only contains a theater Company id and an actor id. If an actor was managed by many theater companies then you’d add several rows to the link table each holding the same actor id but each row having a different theater company’s id.
Yes, you can have start_Location point directly to place. This means that that Start_Location attribute must be a Foreign Key (FK) pointing the theater company to the Primary Key (PK) of the related Place record.
By all means an actor can be born in a place, but just like above, you need a column in Actor, that is a FK to the Place Table’s PK. You could call this column Birth_Place and all it’d hold is the PK of the record in Place that relates to the actor’s birth place. This column would also need to be NOT NULL because all actor’s need a Birth_Place.
So far it seems like your diagram will work to solve this problem, yes. Just see question 1’s answer for that follow up addition.
You’re getting good at removing redundancies. Your diagram looks good. The only suggestion, I’d make is why do you have a play table and then 3 separate play type tables? Why not add them together in on Table called Play. It’d sit exactly where Play currently sits in your diagram and contain the same attributes it already does, but you also add the following:
a. Type – Would be a string that you could place “Drama”, “Comedy”, or “Tradegy” in so you’d know exactly what type of play it is. Also this would allow you to add future play types to the plays table and not have to add a whole new table to the DB.
b. Sub_Type – Would also be a string and hold the type that you currently have under the separate tables. They are all essentially the same attribute in each table and would just hold different type descriptors depending on the parent Type.
c. Main_Character – Would be a string that holds the main character, because in your three separate tables, you have main characters. You’re just calling them 3 separate things. (get the direction I’m going in here? )
d. Secondary_Character – Would be a string that holds the secondary character. You have a secondary character in your dramas and comedies, but non in your tradegies so in tradegy records this column would wind up being null. See what I did there? You now have one table where you used to have 4, and in that one table you can retrieve all the same information you had in those 4 separate tables. Hopefully that’ll make your life easier.
You can do whatever you like, but I’m assuming you mean by best practices and it would be generally considered best practice to separate this single attribute into it’s Simple attribute sub parts. I.E. make it a composed attribute.
(Sorry about the vagueness of the title; I can't think how to really say what I'm looking for without writing a book.)
So in our app, we allow users to change key pieces of data. I'm keeping records of who changed what when in a log schema, but now the problem presents itself: how do I best represent that data in a view for reporting?
An example will help: a customer's data (say, billing address) changed on 4/4/09. Let's say that today, 10/19/09, I want to see all of their 2009 orders, before and after the change. I also want each order to display the billing address that was current as of the date of the order.
So I have 4 tables:
Orders (with order data)
Customers (with current customer data)
CustomerOrders (linking the two)
CustomerChange (which holds the date of the change, who made the change (employee id), what the old billing address was, and what they changed it to)
How do I best structure a view to be used by reporting so that the proper address is returned? Or am I better served by creating a reporting database and denormalizing the data there, which is what the reports group is requesting?
There is no need for a separate DB if this is the only thing you are going to do. You could just create a de-normalized table/cube...and populate and retrieve from it. If your data is voluminous apply proper indexes on this table.
Personally I would design this so you don't need the change table for the report. It is a bad practice to store an order without all the data as of the date of the order stored in a table. You lookup the address from the address table and store it with the order (same for partnumbers and company names and anything that changes over time.) You never get information on an order by joining to customer, address, part numbers, price tables etc.
Audit tables are more for fixing bad changes or looking up who made them than for reporting.