Hi all, as part of a mail with Vacek, I made a list of apps from https://www.ibiblio.org/pub/micro/pc-stuff/freedos/files/distributions/test/report.html which MIGHT have some sort of Unicode and codepage awareness: By that, I mean that those apps can process input and/or output which are encoded using Unicode or some codepage and then show or otherwise process it either in Unicode or using the current codepage (which could get autodetected), or graphically, maybe including a custom non-codepage font. It surprises me that there are more than 30 suspects, but only a fraction of those will ACTUALLY have the features I hope them to have. Maybe you can help me to make the list more exact :-) HTMLHELP shows input (HTML with Unicode and entity support) using awareness of which chars exist in the current codepage. DN2 (DOS Navigator file manager, Ritlabs and Necromancer forks) may be able to handle file names or view files beyond simple "treat as 8-bit, assume it fits codepage". Same for the DOSZIP file manager and PGME (which even comes with fonts, I think). The SQLITE database engine may still contain Unicode support even though it may be of limited use in DOS. Like file managers, some archivers may be aware of filenames supporting encodings beyond the current codepage: 7ZIP just distinguishes DOS, WIN and UTF, whatever that means. 7ZDEC may just assume that Unicode chars 0 to 255 are your codepage? CABEXTRACT seems to rely on ICONV for Unicode? ZIP and UNZIP may or may not support encodings in their Infozip DOS ports? I do not expect any of the other archivers to ponder encodings. Some of the larger programming languages, often ports using 32-bit compilers for DOS, could support Unicode in some way: DOJS (JavaScript), Euphoria, FreeBASIC (FBC), FreePascal (FPC), Lua, Regina Rexx, Perl, OpenWatcom C, OpenWatcom Fortran maybe? I suspect filesystem drivers to have Unicode or codepage awareness, suspects are: DOSLFN, LFNDOS, NTFS, USBDOX :-) Among text editors, MinEd seems to be as Unicode- and codepage- aware as HTMLHELP: http://towo.net/mined/term-dos.png Blocek even comes with a graphical Unicode font. SETEDIT, ELVIS and VIM are powerful enough to possibly support various encodings? The FOXTYPE viewer explicitly supports Unicode. GNUCHCP is a bit of an alternative to the DISPLAY/MODE/CPI font ecosystem. UNRTF converts RTF to other text formats. Likewise, internet apps such as Arachne, Dillo, Lynx, Links, SSHDOS and SSH2DOS could support Unicode and other encodings? Media player MPLAYER probably does, too. Maybe also OPENCP? Last but not least, the OPENGEM GUI distro could contain encoding-aware apps or infrastructure? What are your thoughts? There might be more Unicode in FreeDOS than I had intuitively expected. Even when support is minimal, it would be cool to know that multiple apps grasp the concept of, say, UTF-8 and codepages being able to show a tiny subset of Unicode space and that a few apps even come with fonts with far more than 256 different chars already :-) Thanks for your insights! Regards, Eric PS: For all things NOT mentioned above, I expect no support for Unicode or conversions at all. I expect those to just assume an 8-bit encoding in text (and file names) matching your codepage. _______________________________________________ Freedos-user mailing list Freedos-user@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/freedos-user
