AutoHotkey Community

@fasto
Unfortuntely, the slowness is due to UTF-8 <-> UTF-16 conversions. I tried to cache the conversion result of the haystack. Though it's much faster than the current approach, it's still much slower than ANSI version. The conversion is required because PCRE doesn't support UTF-16.

BTW, in the most cases there is no need to pass the huge data to RegEx*. In your example it can call RegExMatch one line at a time instead. It should give you acceptable performance (it should also run much faster in ANSI version, given larger data.).

Thx for sharing it.

In my case, Urldownloadtofile doesn't work.
Ansi version works well.
but, Unicode version doesn't work.
Always ErrorLevel is 1 and, not created a "downfile.htm" file.

Here is my test code.

I saved it with UTF16-LE BOM.

In fact, for UTF-16 and a zero-length file, this add the byte order mark (BOM) - (FF FE) - and then the specified string as a plain (ANSI) text.

Thanks for your report. It is fixed.

Just one thing:


Isn't that no longer valid as we moved to the TextStream classes?

I'm also thinking about hacking PCRE to handle TCHARs instead of UTF-8 through chars.

_________________
fincs
Highly recommended: AutoHotkey_L (see why) (all my code snippets require it)
Formal request to polyethene - I support the unity of the AutoHotkey community
Get SciTE4AutoHotkey v3.0.00 (Release Candidate)
[My project list]

I upload a new version including the AutoHotkeySC.bin.
This version also comes with XDebug supports.

I'm tinkering with the debugger and I've made some fixes.
As converting the variable contents from/to UTF-8 is hugely slow I've decided to add a build option to AutoHotkeyU that sends the variable contents directly in UTF-16. I've also updated XDebugClient to support this change.
Speaking of changes, the StrGet() function has been modified to accept a new optional parameter that specifies the encoding of the string (0=TCHAR, 1=UTF-8 for now).

PD: Yes, I'm thinking about porting my SciTE debugger to AutoHotkeyU

_________________
fincs
Highly recommended: AutoHotkey_L (see why) (all my code snippets require it)
Formal request to polyethene - I support the unity of the AutoHotkey community
Get SciTE4AutoHotkey v3.0.00 (Release Candidate)
[My project list]

Thank you

No need to replace the ANSI AutoHotkeySC.bin in its folder



With the parameter /NoDecompile there is a problem:
"Could not extract script from EXE."

Ahk2Exe and AutoHotkey use black magic to implement /NoDecompile. The public version of the source code doesn't include that black magic.

_________________
fincs
Highly recommended: AutoHotkey_L (see why) (all my code snippets require it)
Formal request to polyethene - I support the unity of the AutoHotkey community
Get SciTE4AutoHotkey v3.0.00 (Release Candidate)
[My project list]

it worked !
but when it complied it turn into ASCII

not for me:

everything is ok here
всичко е наред тук
Tout va bien ici
totul este ok aici
все ОК здесь
όλα είναι εντάξει εδώ
הכל בסדר כאן
كل شيء على ما يرام هنا
这里一切正常
sve je u redu ovdje
alles in Ordnung ist hier
모든게 괜찮다

The script must be saved in UTF-8 ...
This is the *only* encoding supported by the compiled version.

Good job!

One suggestion: a tutorial or example will be perferable for our common users.

Thanks for your work!

this is really great!

but the only problem i've is that the auto completion doesn't work if i change my keyboard;s layout using this.

for example
earlier
suppose there was something like this

<using the Send {U+@#$#$} kinda thing>

now i configure to change the keyboard layout using your script

now when i want

this doesn't replace!

by the way, I want to thank you from the bottom of my heart-
this is very very useful script! now i changed my entire keyboard to unicode Indian langauge with out trouble

God bless you and I pray for your long & Happy life!