Page 1 of 4 1234 LastLast
Results 1 to 10 of 31

Thread: Feature request: NOT all your source is UTF-8

Hybrid View

Previous Post Previous Post   Next Post Next Post
  1. #1

    Default Feature request: NOT all your source is UTF-8

    Hi,

    I know this is probably not up for debate, as my guess is that currently codemax dictates how it works.
    Yet I have to ask as this is how it is handled in most other programming languages on windows... and I like it better that way.

    When I edit code in Visual Studio it stays in OEM until the Studio decides it cannot save it unless it is saved in unicode.

    IOW. I do NOT get a BOM in each source file that I touch unless I add a character that requires it to be changed.

    See this is what happens when I try to type in Thai in my source in VS.

    Click image for larger version. 

Name:	MustSaveInUnicodeNow.png 
Views:	172 
Size:	12.9 KB 
ID:	13226

    If the Studio doesn't change EACH and EVERY file over to unicode as soon as I type a character then it would mean a lot more peace in our source control world.

    FWIW, delphi does something similar too.

    edit: and now you say "so don't type in your source"...
    --
    Wil

  2. #2

    Default Re: Feature request: NOT all your source is UTF-8

    Let me give you another reason...

    This will make syncing code between different versions of DataFlex a bit of a nightmare. (well.. not a bit)

    Just tried to open a DF20 source file in Studio 19 and it shows the BOM marker as characters. So the DF19.1 Studio can't cope with it.

    Compiling that chokes on the bom marker too.

    --
    Wil

  3. #3
    Join Date
    Feb 2009
    Location
    The Netherlands
    Posts
    4,674

    Default Re: Feature request: NOT all your source is UTF-8

    If you want your source to compile in older versions then the easiest method is to maintain in older versions. Also note that the compiler & CDS internally work with UTF-8 data so they have to perform conversions on old source code.

    FYI: Of course we'll take your request into consideration..
    Last edited by Harm Wibier; 15-Nov-2019 at 07:41 AM.

  4. #4

    Default Re: Feature request: NOT all your source is UTF-8

    Harm,

    Thanks for your answers.

    Quote Originally Posted by Harm Wibier View Post
    If you want your source to compile in older versions then the easiest method is to maintain in older versions. Also note that the compiler & CDS internally work with UTF-8 data so they have to perform conversions on old source code.

    FYI: Of course we'll take your request into consideration..
    Thanks for taking my request into consideration.

    If you have to keep editing sources in an older version of DataFlex then that would be a HUGE reason for people not to upgrade.
    It would hamper progress so to speak, so I do not like that suggestion much as I prefer it to be easy to go forward.

    For most DataFlex developers unicode support is not high on their list of needs at this moment.
    For the developers who do want unicode support, it is mostly about being able to have the data in unicode.
    IOW it probably still is fine to have the source in OEM.
    If a multilingual UI is offered than that translated UI is likely also coming from a database...

    How about the following workaround otherwise?

    A hopefully easy workaround - I think - would be to have an option to have the Studio edit in OEM.
    So you can keep your source in OEM unless you toggle a switch somewhere after which it only edits in UTF-8.

    It doesn't have to be as fancy as it is in VS.
    --
    Wil

  5. #5
    Join Date
    Feb 2009
    Location
    Queens, NY, NY
    Posts
    7,429

    Default Re: Feature request: NOT all your source is UTF-8

    And if Wil's suggestion of edit in OEM doesn't work, then we need a utility to convert code back as a pre-compile option we can run in the earlier studios. It can either just strip out the BOM marker, or be more complex in that it parses the who;e file looking for actual UTF content, and complains if it's not there.

    There is not much worse than trying to maintain the same code in two entirely separate versions. Stuff ALWAYS gets updated in only one branch, and then everybody is unhappy.

    MM
    Michael Mullan.
    Danes Bridge Enterprises.

    ++++++++++++++++++++++++++++
    There is just today. Tomorrow is a concept
    that is mostly theoretical. -- GM Wylie
    ++++++++++++++++++++++++++++

  6. #6
    Join Date
    Feb 2009
    Posts
    5,470

    Default Re: Feature request: NOT all your source is UTF-8

    Quote Originally Posted by Michael Mullan View Post
    There is not much worse than trying to maintain the same code in two entirely separate versions. Stuff ALWAYS gets updated in only one branch, and then everybody is unhappy.
    Just curious in your current code do you have lots of If (!@....)
    Success consists of going from failure to failure without loss of enthusiasm - Winston Churchill

  7. #7
    Join Date
    Feb 2009
    Location
    Queens, NY, NY
    Posts
    7,429

    Default Re: Feature request: NOT all your source is UTF-8

    Some,

    Before source control this was worse, but there are still one or two clients with an 18.0 runtime.

    MM
    Michael Mullan.
    Danes Bridge Enterprises.

    ++++++++++++++++++++++++++++
    There is just today. Tomorrow is a concept
    that is mostly theoretical. -- GM Wylie
    ++++++++++++++++++++++++++++

  8. #8
    Join Date
    Feb 2009
    Posts
    1,270

    Default Re: Feature request: NOT all your source is UTF-8

    Harm,

    My normal way of working with an alpha or beta version of Dataflex is to copy all my sources to new directories with a new version number. Convert all workspaces of my libraries and work as much as possible in the new, well mainly only in the new, test in the new and bring back sources with a compare tool back to the current development release if I do not think the new version of DF will be released before I am done with my new release.

    So being able to compare sources is major important to be able to test in new. If that is not possible I will have to make a choice to not work in the new release as often, in fact it possible will be much more minor. At this moment of time after a new test release I would have compiles a few project under the new release. I did not do that with this release.

    I do not need UTF8 characters in my sources.

    So please consider a solution.


    Peter
    Peter H. van Wijk
    X-Sources International B.V.
    X-Organize Consultancy N.V.

  9. #9
    Join Date
    Feb 2009
    Location
    The Netherlands
    Posts
    4,674

    Default Re: Feature request: NOT all your source is UTF-8

    Quote Originally Posted by phvwijk View Post
    Harm,

    My normal way of working with an alpha or beta version of Dataflex is to copy all my sources to new directories with a new version number. Convert all workspaces of my libraries and work as much as possible in the new, well mainly only in the new, test in the new and bring back sources with a compare tool back to the current development release if I do not think the new version of DF will be released before I am done with my new release.

    So being able to compare sources is major important to be able to test in new. If that is not possible I will have to make a choice to not work in the new release as often, in fact it possible will be much more minor. At this moment of time after a new test release I would have compiles a few project under the new release. I did not do that with this release.

    I do not need UTF8 characters in my sources.

    So please consider a solution.


    Peter
    Don't know which compare tool you use? But Beyond Compare is very capable of comparing UTF-8 with BOM files and OEM files. If you put it on rules based comparison it will actually tell you both files are the same if the BOM is the only difference..

  10. #10

    Default Re: Feature request: NOT all your source is UTF-8

    Yeah found that Beyond Compare does alleviate the merge problem by quite a bit.

    The feature request still stands though.

    (Hey you guys got half a year or more of a head start on us, so we will bump into things and be surprised about it)
    --
    Wil

Page 1 of 4 1234 LastLast

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •