I'm at college for the summer not for summer session but for a programming internship. The goal is for me to design an application that connects with a SQL server that is to read in comma delimited data and generate the appropriate tables and columns. The data is also to be represented via a standard x/y coordinate graph. All of this is to work together, with the application grabbing data from the server and exporting it to be graphed (in my plan anyway..)
Tools I'm using are Visual Studio 2008 Express as well as sql server 2005 express. I plan on using crystal reports to handle the graphs. I've taken python and an intro class to c++ back in high school so I think I can handle this. I'm planning on learning C# and some T-SQL to complete said project by september. Is this realistic? any suggestions?
The express site came with some good tutorials that basically sets up the code foundation for the application -> server integration, but I can only see long nights from this point onward
Shouldn't be too hard. See if your college doesn't have the full VC++ already. Crystal reports sounds a little bit ENTERPRISE, however... if you're going to use C# I'm sure you'll find an easier way to do a simple x/y plot, without having to use an external program.
Writing program to read shit from text files and put it into sql tables should take two hours at most. Graphs should take a day.
With proper tools, which you won't have because you are using Visual Studio 2008 Express as well as sql server 2005 express.
Name:
Anonymous2008-07-02 6:35
>>5
How would you draw a graph using just the .NET libraries?
Name:
Anonymous2008-07-02 8:53
My software platform doesn't have any graph drawing APIs.
How do you draw graphs using that platform? Terrible!
Name:
Anonymous2008-07-02 13:07
>>5
Be aware the most stuff I've done was in each program's little IDE, like a command prompt. Simple read in data is fine and easy as well as manipulating the data with variables, etc. Simple intro shit. Now I'm working with a creating apps with a MS GUI and the .NET framework which seems a bit daunting. I'm sure most of you can do this stuff in a day, but the time I'm there is mainly for learning and I need it.
Basically, I can do a bulk insert to generate the tables automatically with a stored procedure, but it's supposed to be automatic every time you start. Fire up the program, looks for new data, generates it, etc. I'm guessing I need to set up the tables accordingly (3NF, foreign keys?) for it to work correctly. Crystal Reports is a bitch.. would graphs be in reporting services/tools?
What the fuck, this is at maximum 12 lines of Python with an external call to gnuplot. I question the need for a database (it's just x,y coordinates you dolt) but I'm sure it was some other idiot who wrote such a retarded specification.
Enjoy working on a 4-hour project for 3 months. I hope you milk the retards that are paying you for all they're worth.
>>17 >>18
The first steps include building the basic graph, yes. Afterwards it's supposed to look like a scatterplot of different colors to show the density of the certain variables being read in. Certain ranges would match certain colors, so the entire graph would have different shades of green, yellow, red to represent the data. I don't have the data yet so I dunno what is being accounted for. I'm pretty sure the graph is distance in relation to time, so afterwards the program would have to take any point in the distance axis and take an end point and draw a line of best fit, so to speak, and represent the change based on the scatterplot color (neutral would be normal, higher density would be a bit curvy or zigzag). The user would choose the range. This would also include a report of point A to B, with the difference in time versus actual difference in time (distances, etc are provided in another data file), and so on.
Since the files will probably have thousands of lines of code, wouldn't doing a bulk insert into a SQL server save processing time by pulling out the specific info after being read in once instead of reading in a big file every single time from scratch? Also, the data will be kept on a server nearby, and the exe would have to work on any machine in the comp lab
Also, regarding the scatterplot, think of the heat vision in splinter cell.
thousands of lines of code
That's it? If you're working with that small of a dataset, just write the code in a method which is easy and quick to write. Doing anything else is a premature optimization, especially for a one-time insert. If you were dealing with hundreds of thousands or millions of datapoints, then it might be a different issue.
I still don't get why the fuck you need a RDBMS backend for something as simple as a list of (x,y,z) tuples where each tuple is a numeric. A plaintext format would be a lot simpler and have considerably less overhead. If you've got a lot more metadata associated with each entry, then there might be reason, otherwise you're taking an ENTERPRISE route.
Name:
Anonymous2008-07-03 10:58
Also, regarding the scatterplot, think of the heat vision in splinter cell.
Python + matplotlib + ~50 lines of code. For the whole project.
If you insist on a db you should look into Berkeley DB and SQLite
Name:
Anonymous2008-07-03 16:28
>>23
Because those work wonderfully over a network, right?
Name:
Anonymous2008-07-03 16:31
Maybe he has to use an RDBMS 'cause the data will be used in other projects in different ways, did you think of that? And if queried intelligently, let the server do some of the calculation work.
OP here. The application is to load any given day's data file, ranging from 00:00:00 to 23:59:59 with various information on each line that's comma delimited. Again I don't have the data yet, but it seems like it's traffic data based on the overview. First a basic graph is to be generated in relation to time and distance (x and y axis). The thousands of x,y plots are to be a certain color based on a third variable (lol if/else statements) on each line. Hence a graph of multiple colored pixels is generated.
Afterwards, the user is to pick a point A and B anywhere along the y-axis. I have a separate data file with locations in miles/feet. A resultant line is to be drawn from each time interval from t=0 until a gradual line is formed. Not quite a line of regression, but it'll just do a bunch of slopes and connect them at the end based on the average between the time intervals. Later, reports will be generated to show how long it took using real time data in this simulation versus theoretical data / "what-ifs", etc. Multiple lines are to be shown for different locations anywhere on the y-axis, and multiple graphs of multiple days must be aggregated into one if needed and the user has to choose (could be M/W/F, every wednesday, an entire week, etc).
I'm thinking of throwing out the database. Looks like you can call gnuplot from Visual Studio, so I'll just make a network share with the data files and let the program search for them like a calendar or something. It just seemed to make more sense importing the data files into tables and let a server handle the queries, but right now gnuplot can read in this sample data I found online extremely fast, so whatever, the end user will have to wait for it to generate a graph. I've done a little python in the past, and matplotlib looks good. thanks for the advice so far
Name:
Anonymous2008-07-03 19:26
You need to take a course in basic software design right now. You do know that we have software modelling tools that are used for the purpose of abstracting a system model.
>>28 >>30
Thanks, been playing around with some, very useful!
Name:
Anonymous2008-07-04 1:27
>>28
I'm glad there's OMG. It unites all kinds of bullshit, so it gets easier to avoid. Anything coming from OMG is pure, utter bullshit. I don't know what kind of enterprise fucktards work there, or if it's all just a big prank or an inside joke. At times, OMG's shit really looks like a prank, the kind of joke paragraph about professional multi-tier mission-critical scalable enterprise-ready solutions for your business we write over here. I'd be somewhat surprised if they really took all this for serious; I mean, how could human beings working with something related to computers be this useless! My goodness, they are a waste of air and food.
Name:
Anonymous2008-07-04 2:37
>>32
I took a class in ENTERPRISE, and the impression I got from the instructor is that he was blissfully unaware that he wasn't taking this stuff seriously. Everything about his attitude said he even enjoyed it, but when he had us make some of the graphs involved in whatever methodology we were on, it became clear that he didn't have the haziest idea what it was supposed to look like or whether we were close: indeed, he didn't seem to be aware that it was supposed to conform to any standard in particular.
My working theory is that the people who keep this sort of thing afloat operate in much the same manner — some sort of Enterprise Free-association Autopilot engages during enterprise communication and then they go do whatever it is they meant to do, secure in the knowledge that their hindbrain is following best practices.
OMG would be some meta-organism using their best-practice glands as its substrate. That leaves the question of what kind of person actually cooks these things up in the air, but I think it's a start.
Name:
Anonymous2008-07-04 2:46
>>34
Also, how about a Star Trek parody called Star Trek: ENTERPRISE? Who needs a Prime Directive when you have United Federation of Planets data collection best practice processes for ensuring natural development of scalable pre-warp civilizations during Starfleet scientific investigation?
Name:
Anonymous2008-07-04 3:32
>>34 My software development teacher didn't fully comprehend what he was talking about. Therefore, all structured software development methods are non-comprehendable.
Great logic there.!
First of all, I didn't say it was a software class (it wasn't). I also didn't say he didn't comprehend what he was talking about. Least of all did I say that enterprise methodology cannot be comprehended. I didn't even use that word or any of its synonyms. You seem to be posting based on what you wish I had written. In the real world, what I said is that while he was all about design methods, the details were irrelevant. And doesn't that explain a lot, if the stuff we rage over pretty much flies under their radar?
Name:
Anonymous2008-07-04 7:51
Anyone here uses UML, DFDs, use case diagrams, etc? I've found CRC useful but not the 'strict' diagrams.
Are you saying that /prog/ is moderated ? As in, with moderators, thread deletions, and bans ?
That sounds preposterous to me.
Name:
Post deleted by user.2008-07-04 14:38
Name:
ThE_DaRkNeSs[MOD]2008-07-04 14:45
I’'ve removed content from this thread because promotion of illegal activities (for example, drug use) is a violation of the Code of Conduct. You can review the Code of Conduct here: http://yhbt.net
Name:
Anonymous2008-07-04 16:03
i think the website must be down or something what does hand mean
>>39-51
Same person, and we have beet tolling Constantine.
Name:
Anonymous2009-02-25 8:05
Windows file The question is What kind of database provider will allow me behind towards the basement On the drum so that the user can Add tab object in code then access its method so if some flag is thrown when there are about a dozen gigabit connections through.
Name:
Anonymous2009-03-06 8:07
Put Windows are organized hierarchically i e Vans Very comfortable like walking with pillows attached to the web browser beating IE Chrome and Opera in speed tests 2 Internet Explorer is able to render every webpage correctly unlike IE Chrome and Opera in speed tests 2 Microsoft Internet Explorer is not financed by evil tyrannical corporations intent on abusing their monopolies.
All work and no play makes Jack a dull boy
All work and no play makes Jack a dull boy
All work and no play makes Jack a dull boy
All work and no play makes Jack a dull boy
All work and no play makes Jack a dull boy
All work and no play makes Jack a dull boy
All work and no play makes Jack a dull boy
All work and no play makes Jack a dull boy
All work and no play makes Jack a dull boy
All work and no play makes Jack a dull boy
All work and no play makes Jack a dull boy
All work and no play makes Jack a dull boy
All work and no play makes Jack a dull boy
All work and no play makes Jack a dull boy
All work and no play makes Jack a dull boy
All work and no play makes Jack a dull boy
All work and no play makes Jack a dull boy
All work and no play makes Jack a dull boy