Total Commander

1. How can I batch-Edit file comments?
2. Would TC read/write 4NT 8's DESCRIPT.ION? (Unicode Format)

cheming wrote:1. How can I batch-Edit file comments?

Using TC-internal commands? - Currently not at all.
But such a feature has been requested before.
Oops. Seems as if my knowledge of the TC7 internal custom columns is somewhat limited. See Lefteous explanation below.

2. Would TC read/write 4NT 8's DESCRIPT.ION? (Unicode Format)

The TC descripit.ion file implements the old 4NT standard, i.e. it will read/write 8-bit Ascii/Ansi characters at the time being.
Full Unicode will be implemented in TC in a later version, TC7.5, TC8??? Who knows? If anyone does, this will be Christian Ghisler.

Karl

2cheming

How can I batch-Edit file comments?

If you are looking for assigning a comment to multiple files at once press Files/Change Attributes and add tc.Comment (TC 7 required).

Would TC read/write 4NT 8's DESCRIPT.ION? (Unicode Format)

A plug-in author could create a content plug-in which supports other comment formats but Unicode is currently not supported (see Karlchens comment).

I tested this in 7.5pb1 and it is not supported too. TC can't read descript.ion file with UNICODE format created by 4NT 8.xx

I have never heard about 4NT Unicode comments. Where can I find the specifications?

Found this PDF by googling this "description 4NT 8"

Am I wrong or this PDF describes how to store details only and there is nothing about filenames there?

In that case we are back again to Unicode filenames problem. Also: since Christian has his own ID, XML tags are just waste of space IMHO.

I know that if we want comments following "standard" it could be nice to have but it is still not enough

Perhaps I was to fast. I checked the file only a little.
Read Unicode and it looked well...

The only official document I know of is this one:
http://www.jpsoft.com/ascii/descfile.txt

Unfortunately there is no "unicode" folder (like the ascii folder) on that site, so if anyone knows about such comments for Unicode file names, let me know!

How about sending them an email?

ghisler(Author) wrote:The only official document I know of is this one:
http://www.jpsoft.com/ascii/descfile.txt

Unfortunately there is no "unicode" folder (like the ascii folder) on that site, so if anyone knows about such comments for Unicode file names, let me know!

I think the format of 4nt descript.ion file is simple UNICODE text file which is begin with 0xFF 0xFE.

I just make two screenshots for you:

http://tinyurl.com/c3tt7b

Hi to all!

AFAIK Total Commander had never granted 100% support for commenting files with Unicode characters from a different codepage in the name (say, accented letters with umlauts in German words in the OS with non-Unicode language = Cyrillic).

However till the version 7.50 it was possible to create and view comments for such files and comments usually (not always) were preserved when transferring file to another location.

Since the version TC 7.50 and till the current 8.0 the situation has got worse, because now TC gives a warning message: "Unfortunately this function does not support Unicode characters" every time on pressing Ctrl+Z in try to edit or even simply view existing comment. Yes, you can press [OK] and than get the comment editor dialog, but this warning every time is really VERY annoying when you need to edit or view several tens of comments one by one

Is there any option to disable this limitation for Unicode filenames (from other codepage). IMHO the fact, that it will not follow the common standard, is not a problem at all, because we can regard such comments as a TC-specific feature and it has not be compatible wit any other software or OS internal functions. Currently TC has many such unique options, thus one more would not make great difference. I'm quite sure, that the most of users will agree with me.

However if this concept is not correct from the Christian's point of view. is there any workaround to omit this annoying warning? Say there could be some key in wincmd.ini for advanced users or an additional flag "Do not warn me about it again" in the warning message window or any other solution.

Many thanks in advance!

P.S.
Dear moderator, please regard changing the title of the topic by adding some reference to Unicode-problem. It was not a trivial task to find this thread.

Sorry, descript.ion files are ANSI and do not support Unicode UTF-8 or UTF-16.

2ghisler(Author)

OK, I see. However there are two different problems here which we should not mix-up.

1. More obvious and technically simpler question is, why TC does not allow a user to disable this annoying warning. Yes, the warning itself is helpful to remind a user, that descript.ion is not Unicode-compatible. But please trust me, after I've seen this message 10 times for the last hour, I AM really acknowledged about the problem. Even Microsoft allows to disable such type of warnings in their software.
Well, it might be some "hidden" key in wincmd.ini to assure, that disabling this warning is accessible onöy for advanced users, who understand the risk.

2. The second question is whether descript.ion format supports non-ASCII characters from another codepage. Now I tell you a news: it does... more or less.
I'm not coder neither I know much about how it works, but I trust my own eyes. You could make the following simple test.
Assign the default code page of Widnows (Regional Settings - > Language for non-Unicode programs), say, to Winows-1251 (Russian Cyrillic) and reload OS. Than create a filename with symbols from another codepage, say, with German accented letters Ä, Ö, Ü etc.
Or make vice versa - use Cyrillic letters (say Ж, Ш, И, Г) under Cyrillic codepage.

Now try to create a comment in descript.ion format and save it. Now press Ctrl+Z or hover mouse over the file to see the tooltip. You'll see, that the comment is associated with the file. Furthermore, when you rename or copy or move the file to another location, the comment stay with the file.

Regarding, that descript.ion format does actually not support Unicode, the fact seems to be a miracle, but the reason become obvious, when you look inside descript.ion file. It replaces every accented non-ASCII character by the vowels, that is base English (ASCII) character, say German [München] is converted to [Munchen]. However TC is intellectual enough to connect the word "Munchen" in the descript.ion with the file München.txt. in the current directory. The only situation when it does not work properly is the case, when a file with a similar ASCII name, that is Munchen.txt in my example, happens to reside in the same directory. But such coincidence is rather abstract speculation, so IMHO most users would ready to accept this risk.

Yes, it supports different codepages - but not mixed up within the same file! There would be no way to know which codepage is used for a file name.