Best practice dealing with mixed footage types

I will soon embark on my first proper FCP project. The film will be shot at the end of the week on a Sony PMW-EX1 in 1080p25. We already shot some footage a few weeks ago, alas this was mistakenly shot interlaced 1080i50. What would be the best way to go about combining these two sets of footage? Should I de-interlace the 1080i50 stuff first or just drop it all into the timeline and allow FCP to deal with it?

OK, I found a better solution.Future<String> errString = executorService.submit(
    new Callable<String>() {
        public String call() throws Exception {
            StringWriter errWriter = new StringWriter();
            IOUtil.copy(process.getErrorStream(), errWriter, "UTF-8");
            return errWriter.toString();
int exitValue = process.waitFor();
getLog().info("exitValue = " + exitValue);
try {
    getLog().info("errString =\n" + errString.get());
} catch (ExecutionException e) {
    throw new MojoExecutionException("proxygen: ExecutionException");
} The problem I was having before seemed to be that the call to Apache's IOUtil.copy(errorStream, errWriter, "UTF-8"); was not working right, it did not seem to be terminating on EOS. But now it seems to be working fine, so I must have been chasing some other problem (or non-problem).
So, it does seem the best thing to do is read the error and output streams from the process on their own daemon threads, and then call process.waitFor(). The ExecutorService API makes this easy, and using a Callable to return a future value does the right thing. Also, Callable is a little nicer as the call method can throw an Exception, so my code does not need to worry about that (and the readability is better).
Thanks for helping to clarify my thoughts and finding a good solution :-)
Now, it would be really nice if the Process API had a method like process.getFutureErrorString() which does what my code does.
Cheers, Eric

Similar Messages

  • JSF Best Practices: Dealing with phone numbers

    I am sure this would be benefitial to everyone so... I was wondering what would be the best practice to show/edit/search phone numbers in JSF-based UI? I guess it all depends on the table/attribute definition? Suppose in your table you have
    COUNTRY_CODE
    AREA_CODE
    PHONE (xxx-xxxx)
    EXT
    how would you design the 'most' user friendly UI? Is there any component with 'pre-built' phone mask?

    Hello,
    It's funny since I got the same question today in a meeting.
    I would write a PhoneNumber and PhoneNumberConverter class. The getAsString would simply returns the (xxx) xxx-xxxx format, the getAsObject method would check to string value against a regex, probably something like: \s*((\(\d{3}\))|(\d{3}\s*-))?\s*(\d{3})\s*-?\s*(\d{4})
    That way the user will always see the format xxx-xxxx but will able to enter all those formats:
    xxx-xxx-xxxx
    xxx - xxx - xxxx
    (xxx)xxx-xxxx
    (xxx) xxx-xxxx
    xxx-xxxx
    xxx - xxxx
    Regards,
    Simon Lessard
    Message was edited by:
    Simon Lessard

  • What is the best practice dealing with process.getErrorStream()

    I've been playing around creating Process objects with ProcessBuilder. I can use getErrorStream() and getOutputStream() to read the output from the process, but it seems I have to do this on another thread. If I simply call process.waitFor() and then try to read the streams that doesn't work. So I do something like final InputStream errorStream = process.getErrorStream();
    final StringWriter errWriter = new StringWriter();
    ExecutorService executorService = Executors.newCachedThreadPool();
    executorService.execute(
        new Runnable() {
            public void run() {
                try {
                    IOUtils.copy(errorStream, errWriter, "UTF-8");
             } catch (IOException e) {
                    getLog().error(e.getMessage(), e);
    int exitValue = process.waitFor();
    getLog().info("exitValue = " + exitValue);
    getLog().info("errString =\n" + errWriter); This works, but it seems rather inelegant somehow.
    The basic problem is that the Runnable never completes on its own. Through experimentation, I believe that when the process is actually done, errorStream is never closed, or never gets an end-of-file. My current code works because when it goes to read errWriter it just reads what is currently in the buffer. However, if I wanted to clean things up and use executorService.submit() to submit a Callable and get back a Future, then a lot more code is needed because "IOUtils.copy(errorStream, errWriter, "UTF-8");" never terminates.
    Am I misunderstanding something, or is process.getErrorStream() just a crappy API?
    What do other people do when they want to get the error and output results from running a process?
    Edited by: Eric Kolotyluk on Aug 16, 2012 5:26 PM

    OK, I found a better solution.Future<String> errString = executorService.submit(
        new Callable<String>() {
            public String call() throws Exception {
                StringWriter errWriter = new StringWriter();
                IOUtil.copy(process.getErrorStream(), errWriter, "UTF-8");
                return errWriter.toString();
    int exitValue = process.waitFor();
    getLog().info("exitValue = " + exitValue);
    try {
        getLog().info("errString =\n" + errString.get());
    } catch (ExecutionException e) {
        throw new MojoExecutionException("proxygen: ExecutionException");
    } The problem I was having before seemed to be that the call to Apache's IOUtil.copy(errorStream, errWriter, "UTF-8"); was not working right, it did not seem to be terminating on EOS. But now it seems to be working fine, so I must have been chasing some other problem (or non-problem).
    So, it does seem the best thing to do is read the error and output streams from the process on their own daemon threads, and then call process.waitFor(). The ExecutorService API makes this easy, and using a Callable to return a future value does the right thing. Also, Callable is a little nicer as the call method can throw an Exception, so my code does not need to worry about that (and the readability is better).
    Thanks for helping to clarify my thoughts and finding a good solution :-)
    Now, it would be really nice if the Process API had a method like process.getFutureErrorString() which does what my code does.
    Cheers, Eric

  • What is best way dealing with large tiff file in OSX Lion?

    I'm working with large tiff  file (engineering drawing), but preview couldnt handle (unresponsive) this situation.
    What is best way dealing with large tiff file in OSX Lion? (Viewing only or simple editing)
    Thx,
    54n9471

    Use an iPad and this app http://itunes.apple.com/WebObjects/MZStore.woa/wa/viewSoftware?id=400600005&mt=8

  • Best Practices Web Services - Complex Data Types

    Can someone provide some best practices documentation or info that for customers using CR against web services? Speciffically any information on complex data types such as  String[] or Address.
    Thanks
    Ian S

    And That's what I did.
    name.cfc is the component that has complex data types
    created.

  • SAP Best Practice Integrated with Solution manager

    We have a server in which we installed SAP Best practice baseline package and we have the solution manager 7.01 SP 25
    We maintained the logical port but when we try to check connectivity to solution manager we got the following error:
    Connectivity check to sap solution manager system not successful
    Message no. /SMB/BB_INSTALLER375
    Can anyone guide us how to solve the problem and also if there is another way to upload the solution defined on the best practice solution builder into sap solution manager as a template project
    Thanks,
    Heba Hesham

    Hi,
    Patches for SAPGUI 7.10 can be found at the following location:
    http://service.sap.com/patches
    -> Entry by Application Group -> SAP Frontend Components
    -> SAP GUI FOR WINDOWS -> SAP GUI FOR WINDOWS 7.10 CORE
    -> SAP GUI FOR WINDOWS 7.10 CORE -> Win 32
    -> gui710_2-10002995.exe

  • SAP Best Practice: Problems with Loading of Transaction Data

    Hi!
    I am about to implement SAP Best Practices scenario "B34: Accounts Receivable Analysis".
    Therefore I load the data from SAP ERP IDES system into SAP NetWeaver 2004s system.
    My problems are:
    when I try to load the Transaction data for Infosources 0FI_AR_4 and 0FI_AR_6 I get the following errors/warnings:
    when I start the "schedule process" the status getting still "yellow 14:27: 31(194 from 0 records)"
    On the right side I see some actions that are also "yellow", e.g. "DataStore Activation (Change Lo ): not yet activated".
    As a result I cannot see any data in tcode "RSRT" executing the queries "0FIAR_C03/...".
    The problems there
    1) Input help of the web template of query don't contain any values
    2) no data will be shown
    Can some one help me to solve this problem?
    Thank you very much!
    Jürgen

    Be in the monitor window where u got the below issue
    when I start the "schedule process" the status getting still "yellow 14:27: 31(194 from 0 records)"
    and go to environment in the menu options TransactRFC--->in the sourcesystem...
    give the logon details and enter and from there give the correct target destination as ur BI server and execute...
    if u find some idoc's pending there push that manually using F6..
    and come back to ur load and refresh....
    if still it doen't turn green u can manully change status to red in STATUS tab and come to processing tab and expand ur processing details and right click on ur data packet which was not yet updated and select manual update...
    it shows busy status and when it comes out of that once again refresh...
    rgds,

  • Requesting guidance on how best to deal with removal of CreateElementSteps for tables when option to 'script validation for new constraints' is enabled during schema compare

    I have a DeploymentPlanModifer subclass that is responsible for removing certain tables from a deployment plan under specific conditions. It is relatively trivial to find the
    CreateElementSteps I need and subsequently remove them via
    DeploymentPlanModifier.Remove(), but...
    ... if in my comparison I have enabled the 'Script validation for new constraints' option, the deployment plan will contain a
    DeploymentScriptDomStep with a Batch containing AlterTableConstraintModificationStatements for tables with foreign key constraints. My problem starts here - there will be an orphaned
    AlterTableConstraintModificationStatement for each of the tables that I removed. Obviously, execution of the generated script comes to a grinding halt when asking SQL Server to to alter a table that is never created.
    I'm able to get around this by digging around in the aforementioned batches and removing the orphaned alter statements, but this seems really hacky, which makes me think I'm missing the proper way of dealing with this.
    So... if anyone is aware of a more correct way of avoiding this problem, I would really appreciate finding out more about it.
    Thanks in advance for any help. :-)

    Hi Greg. Unfortunately there is not an easy solution here. Walking the deployment model, spotting potential issues and excluding them really is what you have as an option here. The alternative is to pre-process the dacpac to have it in the form you want,
    but I'm not sure if this is an option in your case and it also has limitations.
    Regards,
    Kevin

  • How to deal with unwanted footage in encoded video

    It seems to happen more and more that I will receive pre-encoded MPEG-2 video that was originally designed for VHS. Segments that contain static images were by necessity part of the video for the VHS version but now are to be replaced by actual static images on the DVD.
    I know how to "avoid" those sections of the video by using chapter markers and stories, but viewers can still get to them inadvertently unless I disable the chapter forward/back buttons. Sometimes this is an acceptable compromise but other times less so.
    What other options do I have? I know I can trim these sections out, but I am limited by GOP boundaries so sometimes the edits are not clean, not to mention that if I have subtitles I have to shift them by the amount of video that I have removed which is an added hassle.
    I am curious how others approach this issue. Thanks in advance.
    Todd Nocera

    Hi Todd - almost always the way around this is to extract the DV footage and re-encode. The problems (as I'm sure you know) are that the result is a loss in quality.
    However, if the source for the original encode was VHS then the quality of the MPEG file given to you isn't going to be too hot anyway.
    Trimming the clip in the timeline is another very good option, but as you say is a little limited - especially if timings within the clip are precise.
    What you could do is trim the timeline to just before the section, then save that as an MPEG (keeping the quality), then trim again to the other side of the section (use another track) then export that clip as a second MPEG. You can then extract the footage for the section needing change and perform the fine edits as DV then replace the image with the footage you want and re-encode that section only. There will be *very short* pieces either side of the replaced footage that are slightly lower quality, but ultimately these will be hardly noticeable.
    The bigger issue for you doing it this way is to match the luminance and colour depth of the original with the new section.
    It won't be easy - so you may save time and energy by extracting the whole lot and re-encoding, accepting the slight loss in quality.

  • Best practice for two differing encryption types

    I'm using a Aironet 1232g and I want to use WPA2-PSK. I have the XP clients connected appropriately (via WPA2-AES CCMP), however I also have a Ricoh Digital Sender (copier) that only supports WEP. What is the best security practice in this scenario? Can I enable two SSID's one with WPA2-PSK and the other with WEP? If so, do I need to establish a VLAN on the AP? Any help is appreciated.

    Hi Scott,
    your thougts are right.
    You have to establish 2 seperate VLANs on the AP. In the "encryption Manager" you setup the VLAN to encryption type association.
    Then you need to define the VLANs to SSIDs in the "SSID Manager".
    If you have a WDS running setup the "Server groups" in the WDS Section of "Wireless Services"
    Do not forget to change the switchport (where the AP is connected to) to trunking mode.
    I hope that helps.
    Best regards,
    Frank
    P.S. Please rate helpful posts.

  • Best Practice Solution with 2 Internet Connections

    Good day everyone,
    We used to have single ADSL connections at our clients that provide the internet connection for the network. We have recently partnered up with a Fiber provider and are slowly busy rolling out Fiber connections at our clients. We are also offering redundancy
    with the ADSL connection as a back up for the Fiber connection.
    I would like to know if anyone has suggestions as to what would be the best practise configuration on various Windows Server platforms to make this possible? The idea is that if the Fiber connection fails (for whatever reason) that the ADSL connection takes
    over. This must be an automated process.
    I am open for any form of suggestions.
    Thanks for your time.
    Rudi

    Hi,
    Our current solution is to have to NICs and have both connect to the server. They then have 2 different IPs and we have the DHCP give out the IP as the gateway. The only problem with that is that we cannot control the automated change of gateway IP if the
    main connection fails. 
    We are also willing to look into other hardware solutions that could control this.
    Regards,
    Rudi

  • How to deal with LONG data type in RTF template

    Hi,
    I'm developing a PO request report in which I should print some attachments. The attachments are stored in the database in table fnd_attachments_long_text in column long_text which data type is LONG. In my case the content of the column long_text is always plain text.
    The query I use to get the text from the table is something like:
    select long_text
    from fnd_documents_long_text
    where <some_conditions....>
    The query runs fine in SQL developer but when I run the XML report I get following error: "Stream has already been closed". The XML file generated is complete in its structure but the XML node which should contain the text of the attachment is empty, therefore I cannot print the attachment in the report.
    I made some researches on metalink and I've found note 832903.1 which explains the same error I'm getting. Looks like I cannot execute more than 1 query using the same JDBC connection if one of these queries retrieves a LONG data type.
    Any advice on how can I get the attachment text to show correctly in my report?
    Thank you

    Hi,
    Thank you for your reply,
    Yes, I tried to get the text as described in that article but maybe I'm missing something because my data template doesn't show the attachment text.
    The article talks about support for BLOB, CLOB and RAW and at the end it states "but we do not have support for RAW and LONG column types at the moment - we're looking into that for a future release". So this makes me think that I won't be albe to retrieve my text since it's stored in a LONG column.
    I think that the error I'm getting is not strictly related to XML publisher but it is in some way related to the JDBC connection used to get the data. Infact I have an old version of the same report I'm trying do develop which is made in Report Builder and it is retrieving the attachment text correctly.
    Right now I'm trying to use the Report Builder engine to get the XML data and then use XML Publisher to create the PDF output. Anyway, if it is possible, I would like to avoid Report Builder and use only XMLP.
    Thanks

  • Problems dealing with Boolean data types(SQL).

    Greetings.
    Below is an sql code i am trying to execute.
    select Band, Memb_contact =
    CASE 
    when contact_person >= 0 then 'Yes'
    else 'No' 
    end, Club_Name, Club_contact
    from Members, Club
    where contact_person = 'Yes'
    My column contact_person has a data type of boolean. It stores 1 and -1 only.
    I'm am trying to execute the select statement to return The Band Name and contact number only if the person
    is identified as a contact person in the members table.
    It return the following error:
    Msg 245, Level 16, State 1, Line 1
    Conversion failed when converting the varchar value 'Yes' to data type bit.
    Please help.

    Try the below: You can not have 'Yes' instead use 1.
    select Band, Memb_contact =
    CASE 
    when contact_person >= 0 then 'Yes'
    else 'No' 
    end, Club_Name, Club_contact
    from Members, Club
    where contact_person = 1

  • Best practices - Advertising with Flash CC ?

    What is it with these recommendations?
    http://help.adobe.com/en_US/flash/cs/using/WSd60f23110762d6b883b18f10cb1fe1af6-7b29a.html
    How is it possible that the new tool Adobe Flash CC Adobe does not provide standards for creating ads?
    I'm using the latest version of Flash in this is not possible. The client shows me the way Adobe and we have a misunderstanding and losses of time. Service using Flash ads or uses the old information.
    How can I create a new Flash creations CC using these official recommendations?
    What's going on?
    Regards!

    You're right.  I never noticed that.  I was on an older version of Flash when I was doing ads.  What a major screwup.  There are many people complaining about this in other threads.  I found the solution from kglad in this thread: http://forums.adobe.com/message/5750996
    You can get Flash CC to publish to version 9 player by putting these two files in these locations:
    This file: http://forums.adobe.com/servlet/JiveServlet/download/5510986-147069/FlashPlayer9_0.xml
    Goes here: C:\Program Files\Adobe\Adobe Flash CC\Common\Configuration\Players\FlashPlayer9_0.xml
    This file:  http://forums.adobe.com/servlet/JiveServlet/download/5510986-147068/playerglobal.swc.zip
    Goes here: C:\Program Files\Adobe\Adobe Flash CC\Common\Configuration\ActionScript 3.0\FP9\playerglobal.swc
    You may need to create the FP9 folder.
    Restart flash and you should see Player 9 in your publish options.  You'll have to find the AS3 clickTag code.

  • Malware on my Imac. How best to deal with it?

    I seem to have got malware on my Imac OS10.7.5 which appears with selected words linking to third party sites. How do I go about cleaning the Imac from this - is there a recommended and trustworthy software available?
    thanks
    Nicholas 

    Download and run the Adware Removal tool from the site below.
    http://www.thesafemac.com/arg/
    The author is a respected member of the Apple Support Community.

Maybe you are looking for

  • Disable Ad-hoc networking

    Due to recent security concerns, my company's networking team has begun looking into how one might disable ad-hoc networking on both Windows and Mac computers. Windows computers appears to be fairly easy - it's a check box in the Windows wireless man

  • Substitution in value date

    Hello all, I can not make substitution in the field of value date (bseg-valut). I use the typical ggb1. Do you have any clue if there is any special condition with this field? Thank you, Dimitris Edited by: dimit7 on Jul 12, 2011 12:58 PM

  • Importing duplicate file names

    To Whom it might concern: I have upgraded from LR4 to LR5. LR4 allowed duplictes to be imported - one a jepg the other same file name but raw (CR2). LR5 does not allow me to do this. Am I mistaken or am I over looking something? Albert F Schwartz Jr

  • CF10: zeroDateTimeBehavior Problem

    I'm setting up a MySQL data source. For legacy compatibility I need to handle dates with the value 0000-00-00. I can use the connection string zeroDateTimeBehavior=convertToNull to active that. However, this results in a rather odd error: java.sql.SQ

  • KT3 power/mem probs

    I have a KT3 Ultra2 board and when I switch on after shutting down I get 3 beeps and the J-LED hangs on memory init, However If after shutdown I switch off at the mains and switch back on it boots no probs. I also have this problem when using S3 stan