When subtitling in Arabic punctuation marks positioned wrongly, and the whole line get messed if LTR (e.g. English) embedded within.
Hope to find solution soon.
>When subtitling in Arabic punctuation marks positioned wrongly
This problem may be solved by saving your subtitles in UTF-16 or UTF-8
witht BOM, and placing an invisible control code called
RLM (U+200F RIGHT-TO-LEFT MARK ) as the next (i.e. left) character
of that punctuation (e.g. "!").
It's true that general bidi is very hard to support and is
poorly-supported.
Some other problems, like ones related to "minor" Indic languages,
may be solved by just placing a newer version of USP10.dll in the same
directory as VSFilter.dll.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
The punctuations issue can be fixed with RLMs, but BiDi support in VSFilter.dll is almost absent. When adding LTR text in a RTL context, the line gets rearranged in a weird way. The same goes for SSA tags, which is quite irritating to work around. Also, having poor BiDi support results in making quick "bad" workarounds, which will not be displayed correctly when viewed on a platform other than Windows that has tools with better BiDi support. Is proper BiDi support planned for future releases?
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
If you (we) make sample files to demonstrate problems,
and report problems to the mpc-hc project ( https://sourceforge.net/projects/mpc-hc/ ),
they might eventually try to do something about them.
In mpc-hc they're still developing VSFilter actively there.
Here in Guliverkli2, development is rather passive, mostly based on patches
provided by others...
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
>When subtitling in Arabic punctuation marks positioned wrongly
This problem may be solved by saving your subtitles in UTF-16 or UTF-8
witht BOM, and placing an invisible control code called
RLM (U+200F RIGHT-TO-LEFT MARK ) as the next (i.e. left) character
of that punctuation (e.g. "!").
It's true that general bidi is very hard to support and is
poorly-supported.
Some other problems, like ones related to "minor" Indic languages,
may be solved by just placing a newer version of USP10.dll in the same
directory as VSFilter.dll.
The punctuations issue can be fixed with RLMs, but BiDi support in VSFilter.dll is almost absent. When adding LTR text in a RTL context, the line gets rearranged in a weird way. The same goes for SSA tags, which is quite irritating to work around. Also, having poor BiDi support results in making quick "bad" workarounds, which will not be displayed correctly when viewed on a platform other than Windows that has tools with better BiDi support. Is proper BiDi support planned for future releases?
If you (we) make sample files to demonstrate problems,
and report problems to the mpc-hc project ( https://sourceforge.net/projects/mpc-hc/ ),
they might eventually try to do something about them.
In mpc-hc they're still developing VSFilter actively there.
Here in Guliverkli2, development is rather passive, mostly based on patches
provided by others...
I made a tool as On-Screen Keyboard writes unicode control character RLE or LRE
http://www.mediafire.com/?m2h3kbhy0x0ymvo