Compiling a large scale application

I have recently aquired a large scale application with nearly 300 .java files. The problem I'm having is compiling such a large application where every class seems to be dependent on other classes and packages. Is there a simple solution to compiling such an app or do I have to use brute force and trial and error until I get all the packages and files compiled? The route I am taking now is going to each folder and making txt files that I pass to javac ("javac @currdir.txt") and if i get linking errors, track those files down and try to compile them first.
thanks in advance,
Bosco

There are numerous tools out there that will help you, and most IDE's have some notion of a "project"--a set of files that goes together and that may depend on each other.
Personally I like ant, from the fine people at Apache/Jakarta.
http://ant.apache.org/

Similar Messages

What design aids for large-scale application (using LV6i) exist?

I have a large-scale application that includes analog and digital I/O, motion control, multiple temperature readings, Ethernet communication, RS-232, DDE and ActiveX controls for communicating with other commercial software. We have to improve the system performance and ease the pain of maintaining and upgrading. What aids are there for large-scale application design and development?

This doesn't exactly count as development "tools" but I can send you copies of three papers that I found when I was just getting started with LV. They can show you how to think about your problem--and that is really the hardest (and most important) part.
Contact me directly and I'll email them to you. The archive is to large to post.
Mike...
Certified Professional Instructor
Certified LabVIEW Architect
LabVIEW Champion
"... after all, He's not a tame lion..."
Be thinking ahead and mark your dance card for NI Week 2015 now: TS 6139 - Object Oriented First Steps

Architecture of larger Flex applications

Hello,
I'm new to Flash and Flex development, having been a .NET developer for many years. I am now putting together a Flash application for the web and I suspect that I'll be using Flex and Flash Builder as my primary development tools.
I had a question about overall architecture of a larger scale application. Specifically, how should a Flex app be structured or architected to group UI elements and load and unload those elements as particular portions of the application are navigated to. The best analogy I can make would be pages in a web application, or Forms in a windows application. It seems pretty clear that the "state" structure from Flash Catalyst is not the right way to do this. And I don't know if multiple "stages" from Flash Pro can be created, saved, and loaded as needed by Flex.
Thanks in advance for your help, and guidance for a very new developer to these tools!
-Ben

Thanks for the fast reply Darrell!
That seems to be the right path, sometimes it's just a matter of vocabulary and knowing what words to Google search for
-Ben

Flex deployed on a large scale?

We plan on developing a new product and Flex popped in my
mind as a development platform. I know a good deal of Flex 1.5, but
only used it for personal sites.
My question is how well Flex behaves in a large scale
environment to those who have deployed it in such. Server load will
be at least thousand / day.
Thanks!

Hmm... I made one medium sized application in 1.5 (approx 10
screens, user access <1000 times per day) and it seems to be
working alright for the client.
Now I am working on a major application (over 20 main
screens, and definately access>1000 times per day) and it is not
going well. I am really worried about the bugs and memory issues of
Flex 2.0. I have also not found a sure-fire way to address these
potential issues. I can say this: for the size of application we
are making, Flex and Flash Player just aren't up to the job.
Compiled and executed as a single .swf application results in 755MB
ram usage and for some reason a constant CPU access of 60% (Pentium
4 proc.) after accessing every screen. And this is just FlashPlayer
doing what it is supposed to. Me, not being a computer engineer,
can't really address these problems. Flex and AS aren't C. I can't
control memory usage with my code. By breaking up the huge
application into smaller ones and then loading those via an
SWFLoader I may be able to avoid this rampant resource hogging but
it's sort of illogical from an application architecture standpoint
because this is ONE application.
As a developer, I can see plenty of places to streamline the
application but this simply isn't possible when dealing with the
client. They want this screen to look and act this way and that
screen to look and act the other way. I can talk about how if both
screens use the same layout and logic they can both use the same
template class, share static resources, blah blah until I am blue
in the face but it won't matter because they are the client and
they decide how the application is going to look--at the expense of
streamlining. That's just the real world. Then I have to somehow
make it work.
By the way... before you think "just use view states!", I do
use those--and bitwise logic flags for more complicated
configurations--it's still not enough, although it did cut approx
160 screens in documentation form down to just 20 in
implementation.
In worst case scenarios, I have to deny the client what they
want and if they ask me why, I have to reply "it can't be done with
Flex". Then their satisfaction in the product drops. Flex suddenly
isn't as incredible as it seemed at first. Doesn't matter how
pretty and animated the screens are when if you run them over an
hour your computer slows to a halt or .ttc fonts stop loading (HUGE
issue here in Japan).
I have yet to see a sample application that comes close to
the scale of our current project: A library book browser? neat but
that's just square one; A Commodore 64 emulator? cute. no place in
business; A real-estate browswer? in our project that would be the
equivalent of ONE SCREEN out of the entire application.
I like Flex. It's fun--on a small scale. But I never want to
develop a real world business application using it again. There are
way more (and way more skilled) Java, JSP, PHP, etc. etc.
developers out there than Flex developers who can make much more
robust applications. It's a shame the client got caught up in the
hype of Flex RIA before the technology was ready for the task.
Very long story short: Beware using Flex for an involved
application.
It's going to require exponentially more time than a smaller,
less ambitious project--especially if you don't purchase FDS. And
oh my god implementing a Flex application on a legacy Struts
framework... kill me now! As much as I hate "page-refresh"
applications, Flex (both 1.5 and 2.0) has not proven to be the
god-send that I had hoped and dreamed it would be as a developer.
What can you expect, though? It's only been out a few years.... And
as far as clients' perspectives go, the price for FDS also
certainly doesn't help make it appealing. That is why it is so
embarrassing to tell them their dream application is quickly
becoming an egregious memory hog.
Anyway, good luck if you take on your project with flex. Just
be careful!

Large Scale Digital Printing Guidelines

Hi,
I'm trying to get a getter handle on the principles and options for creating the best large and very large scale prints from digital files. I'm more than well versed in the basics of Photoshop and color management but there remain some issues I've never dealt with.
It would be very helpful if you could give me some advice about this issue that I've divided into four levels. In some cases I've stated principles as I understand them. Feel free to better inform me. In other cases I've posed direct questions and I'd really appreciate professional advice about these issues, or references to places where I can learn more.
Thanks alot,
Chris
Level one – Start with the maximum number of pixels possible.
Principle: Your goal is to produce a print without interpolation at no less than 240 dpi. This means that you need as many pixels as the capture device can produce at its maximum optical resolution.
Level two – Appropriate Interpolation within Photoshop
Use the Photoshop Image Size box with the appropriate interpolation setting (Bicubic Smoother) to increase the image size up to the maximum size of your ink jet printer.
What is the absolute minimum resolution that is acceptable when printing up to 44”?
What about the idea of increasing your print size in 10% increments? Does this make a real difference?
Level three - Resizing with vector-based applications like Genuine Fractals?
In your experience do these work as advertized and do you recommend them for preparing files to print larger than the Epson 9900?
Level four – Giant Digital Printing Methods
What are the options for creating extremely large digital prints?
Are there web sites or other resources you can point me to to learn more about this?
How do you prepare files for very large-scale digital output?

While what you say may be true, it is not always the case. I would view a 'painting' as more than a 'poster' in terms of output resolution, at least in the first stages of development. Definately get the info from your printer and then plan to use your hardware/software setup to give you the most creative flexibility. In other words - work as big as you can (within reason, see previous statement) to give yourself the most creative freedom. Things like subtle gradations and fine details will benefit from more pixels, and can with the right printer be transferred to hard copy at higher resolutions (a photo quality ink jet will take advantage of 600ppi) if that's what you're going for.
Additionally it's much easier to down scale than to wish you had a bigger image after a 100 hours of labor...

Large JNI application

Hi: I have a very large Fortran application, compiled as a shared object file, which I am trying to run from Java via the JNI. It contains some very large array declarations. If I reduce the size of the arrays, it works fine. But if I increase them to the size I need, it bombs out at the System.load() statement, with the message "Failed to map segment from shared object: Cannot allocate memory".
I can also run the shared object application from a small Fortran main program. When I do that, I get the same problem. I can fix this by increasing the stack size, with the command "ulimit -s 20480".
However, this solution does not seem to work with Java. I have tried issuing the "ulimit -s 20480" command before starting the JVM, but it doesn't help. I have also tried changing the -Xms -Xmx and -Xss options before starting the JVM, but no joy.
The Fortran application is compiled with g77 on a Redhat 9 system. I am using Java 1.4.2.
Any suggestions would be appreciated.
Cheers................Neil

Just run it as a seperate application.
Use sockets, files or streams to communicate with it.
Doing this will make it easier to debug both apps and in case there are bugs in the fortran side it won't take down your entire application.

Very-large-scale searching in J2EE

I'm looking to solve a very-large-scale searching problem. I am creating a site
where users can search a table with five million records, filtering and sorting
independantly on ten different columns. For example, the table might be five million
customers, and the user might choose "S*" for the last name, and sort ascending
on street name.
I have read up on a number of patterns to solve this problem, but anticipate some
performance issues. I'll explain below:
1) "Page-by-Page Iterator" or "Value List Handler"
In this pattern, it appears that all records that match the search criteria are
retrieved from the database and cached on the application server. The client (JSP)
can then access small pieces of the cached results at a time. Issues with this
include:
- If the customer record is 1KB, then wide search criteria (i.e. last name =
S*) will cause 1 GB transfer from the database server to app server, and then
1GB being stored on the app server, cached, waiting for the user (each user!)
to ask for the next 10 or 100 records. This is inefficient use of network and
memory resources.
- 99% of the data transfered from the database server will not by used ... most
users flip through a couple of pages and then choose a record or start a new search
2) Requery the database each time and ask for a subset
I haven't seen this formalized into a pattern yet, but the basic idea is this:
If a clients asks for records 1-100 first (i.e. page 1), only fetch that many
records from the db. If the user asks for the next page, requery the database
and use the JDBC API's ResultSet.absolute(int row) to start at record 101. Issue:
The query is re-performed, causing the Oracle server to do another costly "execute"
(bad on 5M records with sorting).
To solve this, I've beed trying to enhance the second strategy above by caching
the ResultSet object in a stateful session bean. Unfortunately, this causes a
"ResultSet already closed" SQLException, although I ensure that the Connection,
PreparedStatement, and ResultSet are all stored in the EJB and not closed. I've
seen this on newsgroups ... it appears that WebLogic is forcing the Connection
closed. If this is how J2EE and pooled connections work, then that's fine ...
there's nothing I can really do about it.
Another idea is to use "explicit cursors" in Oracle. I haven't fully explored
it yet, but it wouldn't be a great solution as it would be using Oracle-specific
functionality (we are trying to be db-agnostic).
More information:
- BEA WebLogic Server 8.1
- JDBC: Oracle's thin driver provided with WLS 8.1
- Platform: Sun Solaris 5.8
- Oracle 9i
Any other ideas on how I can solve this issue?

Michael McNeil wrote:
I'm looking to solve a very-large-scale searching problem. I am creating a site
where users can search a table with five million records, filtering and sorting
independantly on ten different columns. For example, the table might be five million
customers, and the user might choose "S*" for the last name, and sort ascending
on street name.
I have read up on a number of patterns to solve this problem, but anticipate some
performance issues. I'll explain below:
1) "Page-by-Page Iterator" or "Value List Handler"
In this pattern, it appears that all records that match the search criteria are
retrieved from the database and cached on the application server. The client (JSP)
can then access small pieces of the cached results at a time. Issues with this
include:
- If the customer record is 1KB, then wide search criteria (i.e. last name =
S*) will cause 1 GB transfer from the database server to app server, and then
1GB being stored on the app server, cached, waiting for the user (each user!)
to ask for the next 10 or 100 records. This is inefficient use of network and
memory resources.
- 99% of the data transfered from the database server will not by used ... most
users flip through a couple of pages and then choose a record or start a new search
2) Requery the database each time and ask for a subset
I haven't seen this formalized into a pattern yet, but the basic idea is this:
If a clients asks for records 1-100 first (i.e. page 1), only fetch that many
records from the db. If the user asks for the next page, requery the database
and use the JDBC API's ResultSet.absolute(int row) to start at record 101. Issue:
The query is re-performed, causing the Oracle server to do another costly "execute"
(bad on 5M records with sorting).
To solve this, I've beed trying to enhance the second strategy above by caching
the ResultSet object in a stateful session bean. Unfortunately, this causes a
"ResultSet already closed" SQLException, although I ensure that the Connection,
PreparedStatement, and ResultSet are all stored in the EJB and not closed. I've
seen this on newsgroups ... it appears that WebLogic is forcing the Connection
closed. If this is how J2EE and pooled connections work, then that's fine ...
there's nothing I can really do about it.
Another idea is to use "explicit cursors" in Oracle. I haven't fully explored
it yet, but it wouldn't be a great solution as it would be using Oracle-specific
functionality (we are trying to be db-agnostic).
More information:
- BEA WebLogic Server 8.1
- JDBC: Oracle's thin driver provided with WLS 8.1
- Platform: Sun Solaris 5.8
- Oracle 9i
Any other ideas on how I can solve this issue? Hi. Fancy SQL to the rescue! If the table has a unique key, you can simply send a
query per page, with iterative SQL that selects the next N rows beyond what was
selected last time. Eg:
Let variable X be the highest key value you've seen so far. Initially it would
be the lowest possible value.
select * from mytable M
where ... -- application-specific qualifications...
and M.key >= X
and (100 <= select count(*) from mytable MM where MM.key > X and MM.key < M.key and ...)
In English, this says, select all the qualifying rows higher than what I last saw, but
only those that have fewer than 100 qualifying rows between the last I saw and them (ie:
the next 100).
When processing this query, remember the highest key value you see, and use it for the
next query.
Joe

Has anyone ever used web start in large scale ?

Dear friends,
have you ever used some web start application in large scale ?
i.e., have you distributed some real application with web-start?
I�m doing such experience here and I�m a bit afraid about several problems in the final user machine - such as java version, browser incompatibility and mainly with the ugly WebStart cache problems.

can you give a bit more detail on what exactly you're afraid of? is this a www application? internal business app? etc.
for example you can force your java version in java web-start,etc...

Typical/Common large-scale ACE deployment or designs?

I am deploying several ACE devices and GSS devices to facilitate redundancy and site load balancing at a couple of data centers. Now that I have a bunch of experience with the ACE and GSS, are there typical or common ACE deployment methods? Are there reference designs? I have been looking, and haven't really found any.
Even if they are not Cisco 'official' methods, I'm wondering how most people, particularily those who deploy a lot of these or deploy them with large-scale systems, typically do it.
I'm using routed mode (not one-arm mode) and I'm wondering if most people use real server (in my case, web servers) with dual-NICs to support connectivity to back-end systems? Or do people commonly just route it all through the ACE?
Also, how many VIPs and real servers have been deployed in a single ACE 4710 device? I'm trying to deploy about 700 VIPs with about 1800 Real Servers providing content to those VIPs.
How do people configure VIPs, farms, and sticky? I'm looking for how someone who wants to put a large ammount of VIPs and real servers into the ACE would succeed at doing it. I have attempted to add a large number in the 'global' policy-map, but that uses too many PANs.
I have tried a few methods myself, and have run into the limit on Policy Action Nodes (PANs) in the ACE device. Has anyone else hit this issue? Any tips or tricks on how to use PANs more conservitively?
Any insight you can share would be appreciated.
- Erik

As far as i can see from your requirements i suggest you create 1 ear file for your portal and 1 ear file per module.
The ear file from your portal is the main application and the ear files of your modules are shared libraries that contain the taskflows. These taskflows can be consumed in the portal application.
This way, you can easily deploy 1 module without needing to deploy the main application or the other application.
It also let you devide your team of developer so everybody can work on a sepperate module without interfering.
On a sidenote: when you have deployed your main application, and later you create a new module, than you have to register that module to your application so then you will need to redeploy your portal but if you update an existing module, you won't need to redeploy your portal.
As for the security, all your modules will inherit the security model of your portal application.

Best and most secure tracking option for large scale iPad deployment?

Hi, we are deploying iPads to hotel rooms on a large scale basis and need to find an appropriate solution to securely track them in case they are stolen.
We are actually considering Find my iPhone for its simplicity, but the 100 devices limit and also the fact it is relatively easy to disable poses a problem.
On the other spectrum there is AirWatch which is very expensive.
I was wondering if any other options are available for this?
Best regards
Florian

MobileIron Sentry iPad app, enterprise IT departments can track iPhones and iPads, view device inventory, block offending or compromised devices, and remotely wipe stolen, lost or compromised units. The application also enables remotely suspending email access for active staff, just-terminated employees, and others.
Just read about it.

Large scale forte implementation

Dear Forte experts:
I am part of a team in an large insurance company in charge of developing an
enterprise-wide insurnace solution. We have been approached by few vendors, one
of which bases its architecture on Forte. We really like what we saw, however,
neither the vendor nor us are confident nor knowledgable enough about the
performance behaviour of Forte in a distributed computing environment which can
be characterized as, multiple islands of processing, with
millions-of-transactions/day, spread across wide area network.
I am afraid we won't be able to choose Forte route unless we gain confidence on
its performance capability in our typical environment. So any insight, examples,
case studies that I can get from this group collective knowledge is extremly
helpful and is greatly appreciated.
Sincerely,
Farhad Abar, Ph.D.

From: Inman, Kal
Sent: Thursday, June 12, 1997 7:07 AM
To: [email protected]
Subject: RE: large scale forte implementation
Farhad
At Andersen Windows, we have been running our Order Entry system over
a 56K frame relay network since the systems initial deployment in Nov
of 1994. We currently have a user base of approximately 120 PC & Mac
clients running over the frame, with an additional internal installed
base of approximately 50 PC & Mac workstations. This system runs on a
single Sequent server. We soon hope to add NT clients to this mix.
It has been our observation that Forte has not been a constraint to
performance. When we have performance problems, it has generally been
caused by poor design. One of are largest constraints to performance
is the amount of data we drag across the network.
Since the successful implementation of our Order Entry system,
Andersen has adopted Forte as our enterprise custom development tool.
It has allowed our development staff to concentrated on development of
business functionality while insulating us from the complexities of
operating systems, messaging, and maintaining platform specific code.
We currently have several additional systems deployed using Forte.
These systems include three Express applications, a standalone windows
application and a mobile client application.
I think Thomas Mercer Hursh asked a valid question "What are the
alternatives you are considering?". I don't think you will find one
to compare with Forte.
Kal Inman
Andersen Windows
From: [email protected][SMTP:[email protected]]
Sent: Wednesday, June 11, 1997 10:07 AM
To: [email protected]
Subject: large scale forte implementation
Dear Forte experts:
I am part of a team in an large insurance company in charge of
developing an
enterprise-wide insurnace solution. We have been approached by few
vendors, one
of which bases its architecture on Forte. We really like what we saw,
however,
neither the vendor nor us are confident nor knowledgable enough about
the
performance behaviour of Forte in a distributed computing environment
which can
be characterized as, multiple islands of processing, with
millions-of-transactions/day, spread across wide area network.
I am afraid we won't be able to choose Forte route unless we gain
confidence on
its performance capability in our typical environment. So any insight,
examples,
case studies that I can get from this group collective knowledge is
extremly
helpful and is greatly appreciated.
Sincerely,
Farhad Abar, Ph.D.

Hi, My printing has suddenly changed in adobe to a large scale, as in, what should be one page of print comes out as 24 pages? I havent changed anything, its happening on more than one document also, I have to stop my printer before all the pages spew o

Hi, My printing has suddenly changed in adobe to a large scale, as in, what should be one page of print comes out as 24 pages? I havent changed anything, its happening on more than one document also, I have to stop my printer before all the pages spew out. I have tried printing 'one single page' and it does exactly the same? Help?

Is the Poster Print feature turned ON?

When printing from an online PDF, the page prints in extra large scale. How do I fix this?

when printing from an online PDF, the page prints in extra large scale. How do I fix this?

This can happen when Firefox has misread the paper size from the information supplied by Windows. Clearing it can involve finding some obscure settings, but here goes:
(1) In a new tab, type or paste '''about:config''' in the address bar and press Enter. Click the button promising to be careful.
(2) In the search box above the list, type or paste '''print''' and pause while the list is filtered
(3) For each setting for the problem printer, right-click and Reset it. The fastest way is to right-click with the mouse and then press the r key on the keyboard with your other hand.
Note: In a couple other threads involving Brother printers, the preference '''printer_printer_name.print_paper_data''' was set to 256 and when the user edited it to 1 that fixed the paper size problem. If you see a 256 there, you can edit the value by doubling-clicking it or using right-click>Modify.

Line appears when applying drop shadow on large scale

Hello!
Some weeks ago I had to make a large scale graphic (800mmx2000mm) for a roll-up banner. I wanted to apply a drop shadow to a rounded shape, and ugly lines came up. Since it was a bit urgent, I decided not to use it.
But now I'm curious, so I quickly made an ellipse and added a shadow, so you know what I mean. This also happens when I save it as pdf or image.
Perhaps someday I will have to use a drop shadow on large scale. So, does anybody knows how to fix this or what could I do in case I need to use this effect in these conditions? I use Illustrator CS6 in Mac with Mavericks.
Thanks in advance.

Mike Gondek wrote:
I was able to create an ellipse to your dimension and got a good drop shadow. What happens if you manually make a drop shadow using appearance?
FYI I tried making the same using drop shadow filter in CS5 and got this error.
Incase your file was created in CS5 and opened in 6, I would recreate the drop shadow in CS6. I know they redid the gaussian blur in CS6, but not sure if that affected drop shadow.
CS6 is better on raster effects at large sizes.
My file was created and opened in CS6.
My ellipse is around 175cmx50cm. I tried it manually like you said and got the same results:
So, I guess I'm alone with such a problem. No idea what is wrong :/

Tweaking product prices on a large scale - how?

My Client has a software store on BC. His supplier is constantly changing their prices and my client wants to be able to quickly review prices, make changes to reflect supplier prices every few days
If I export the Product List the Excel export is unusable with it full of HTML markup from the product descriptions.
Apart from opening each product individually to check and tweak prices how is everyone ammending prices on a large scale.... My client only has 60 products at the moment but this is soon to quadruple and I have prospective clients looking at BC for their ecommerce solution and they have thousands of items.
Regards
Richard

If its just prices you want to input, see if you can just eliminate all the other columns that are not needed and only import the price column with its product identifier ofcourse, and see if it will just update the price and not have to deal with the descriptions...Just a thought...

Compiling a large scale application

Similar Messages

Maybe you are looking for