Req: text search both normal and unicode

[Sheepdog]

I would find it very handy if there were another checkbox (maybe ANSI) to let TC search normal Text as well as unicode.

I was looking for a particular text into an *.inf file. The search found several files but not the one I was looking for. After a few times I had the idea to check 'Unicode' and bingo: the file was found - while the other files were not.

I think it should be possible to search both unicode and ususal text in one step. How should I know if the file is saved as unicode or not as the display doesn't show me any difference.

sheepdog[/i]

[nevidimka]

2Sheepdog

How should I know if the file is saved as unicode or not as the display doesn't show me any difference.

If you're using TC internal lister you have to look menu/Option.

[Sheepdog]

2Sheepdog

How should I know if the file is saved as unicode or not as the display doesn't show me any difference.

If you're using TC internal lister you have to look menu/Option.

Thanks, I know how to distinguish between unicode and normal text, but
how should you know which format the wanted file uses before you have found it ?

sheepdog

[SanskritFritz]

I support the idea. But is it possible to program? That is a crucial question.

[Clo]

2Sheepdog
Hello Stefan !

but how should you know which format the wanted file uses before you have found it ?

Maybe using a crystal bowl ?
- I support your idea too, whether it's possible, like SanskritFritz says…

 V G
Claude
Clo

[Sheepdog]

- I support your idea too, whether it's possible, like SanskritFritz says…

I think TC should search internally 2 times but remember the result of the first search and present both.

sheepdog

[Clo]

2Sheepdog
Hi Stefan !

I think TC should search internally 2 times but remember the result of the first search and present both

.
¤ I agree. I noticed that TC founds the text in UTF8, it doesn't distinguish from the usual *.txt (ANSI - ASCII)
We have not an UTF8 tick box (I don't know if it should be useful ???)

VG
Claude
clo

[Flint]

I was going to create a new thread, but found this one...

At present TC supports searching for files with some text in ANSI, ASCII, Unicode (UTF-16) and UTF-8 encodings, but all they are exclusive. There is a confusing thing that the encoding options are designed as checkboxes, so the user get an illusion that it's possible to select several of them, but when he tries, he fails.

My suggestion is to implement one of the following two ideas:
1. (more preferable, but more complex to implement) To make it possible selecting several different encodings, so that the user could quickly find the file even if he doesn't remember its encoding (See the fake screenshot.)
2. (more easy, but less preferable) To make radio-buttons instead of checkboxes, so that users knew at once that it's impossible to search text in several encodings. (See the fake screenshot.)

Of course, in the first variant there is a problem that searching the text will take longer time with several encodings selected than with only one, but if I need to search in different encodings, it will take long time in any case, but in addition I will need to make several different searches with different parameters... Why not automate it?

[Sheepdog]

1. (more preferable, but more complex to implement) To make it possible selecting several different encodings, so that the user could quickly find the file even if he doesn't remember its encoding (See the fake screenshot.)

Very good idea.

100% support+++

sheepdog

[ghisler(Author)]

Well, I could add this, but checking 3 options would mean 3 separate searches then, which would mean a considerable slowdown...

[Flint]

ghisler(Author)

but checking 3 options would mean 3 separate searches then, which would mean a considerable slowdown...

Of course, we understand it. But it's much better than making 3 separate searches by hand. Thank you for considering this feature!

[gigaman]

Besides, if those 3 searches won't be completely sequential (searching all files for ANSI text first, then searching all files for UNICODE text, etc.), but rather somehow parallel (searching one file for ANSI, then again for UNICODE, etc., then moving to another file - or possibly even by blocks?), the slowdown shouldn't be that bad for "bigger searches" - the file caching should help (compared to 3 separate searches), IMHO.

I certainly vote for this feature!

[StatusQuo]

2ghisler(Author)

but checking 3 options

As this should be optional to the user, this sounds like a good way of implementation.

Although it would be 4 possible options from the current state:
- Unicode
- UTF8
- ANSI (which is standard now, not having a checkbox yet)
- DOS/ASCII

Optional, because when you know which encoding the searched file has, you don't mind about "matching" files in another encoding. E.g. *.LNK seem to be some kind of Unicode/UTF16, while *.URL are not (in my experience). Also, when searching for "CompanyName" you probably don't want every single office file to be listed...

I agree that option this would make searching more comfortable in other cases, so:

Support+

[d]

unicode Tracing: is T r a c i n g : in ansi/utf-8. they could be searched in parallel.

[d]

Sheepdog>I think it should be possible to search both unicode and ususal text in one step.
ghisler(Author)>Well, I could add this, but checking 3 options would mean 3 separate searches then, which would mean a considerable slowdown...
gigaman>won't be completely sequential (searching all files for ANSI text first, then searching all files for UNICODE text, etc.), but rather somehow parallel
d(I)>unicode Tracing: is T r a c i n g : in ansi/utf-8. they could be searched in parallel.

i mean searching like "ignore case" searching.
it is already available! - with file search tool - how? - set "find text", set "RegEx", and write RegEx with that meaning:
(but don't set "utf8" nor "unicode")
..RegEx with that meaning:
if i search for russian word "здраствуй" (hello)
you should write that:
"здраствуй" OR "здраствуй" OR "74@0AB2C9",
first of them is ansi(windows-1251), second - utf-8, and third -unicode.
what is RexEx for that?:
здраствуй|здраствуй|74@0AB2C9

and, "case sensitive" is faster.

how can you know that здраствуй ? save with notepad as utf-8 and look with lister as ansi.