Best way to decode multilanguage Character

[rootx]

On my script I work with a multiple languages, but I'm not sure why somethimes the result of binarytostring fail. My source is an XML file UTF-8

Ōgaki-shi = Ogaki ???

中央区 = Chūō  = Chuo ???

Zürich = Zurich --------> BinaryToString($dBinary,4) works

THX

 

[jchd]

You post doesn't make much sense. Post code to reproduce a problem.

AutoIt uses Unicode, just like Windows and there is no such beast as "decoding multilanguage Character". If your input uses a different encoding, there are functions able to convert to/from anything you need.

[rootx]

expandcollapse popup

<?xml version="1.0" encoding="UTF-8"?>
<GeocodeResponse>
 <status>OK</status>
 <result>
  <type>political</type>
  <type>sublocality</type>
  <type>sublocality_level_4</type>
  <formatted_address>1 Chome-37-1 Asakusa, Taitō-ku, Tōkyō-to 111-0032, Giappone</formatted_address>
  <address_component>
   <long_name>1</long_name>
   <short_name>1</short_name>
   <type>political</type>
   <type>sublocality</type>
   <type>sublocality_level_4</type>
  </address_component>
  <address_component>
   <long_name>37</long_name>
   <short_name>37</short_name>
   <type>political</type>
   <type>sublocality</type>
   <type>sublocality_level_3</type>
  </address_component>
  <address_component>
   <long_name>1 Chome</long_name>
   <short_name>1 Chome</short_name>
   <type>political</type>
   <type>sublocality</type>
   <type>sublocality_level_2</type>
  </address_component>
  <address_component>
   <long_name>Asakusa</long_name>
   <short_name>Asakusa</short_name>
   <type>political</type>
   <type>sublocality</type>
   <type>sublocality_level_1</type>
  </address_component>
  <address_component>
   <long_name>Taitō-ku</long_name>
   <short_name>Taitō-ku</short_name>
   <type>locality</type>
   <type>political</type>
  </address_component>
  <address_component>
   <long_name>Tōkyō-to</long_name>
   <short_name>Tōkyō-to</short_name>
   <type>administrative_area_level_1</type>
   <type>political</type>
  </address_component>
  <address_component>
   <long_name>Giappone</long_name>
   <short_name>JP</short_name>
   <type>country</type>
   <type>political</type>
  </address_component>
  <address_component>
   <long_name>111-0032</long_name>
   <short_name>111-0032</short_name>
   <type>postal_code</type>
  </address_component>
  <geometry>
   <location>
    <lat>35.7127796</lat>
    <lng>139.7966054</lng>
   </location>
   <location_type>ROOFTOP</location_type>
   <viewport>
    <southwest>
     <lat>35.7114306</lat>
     <lng>139.7952564</lng>
    </southwest>
    <northeast>
     <lat>35.7141286</lat>
     <lng>139.7979544</lng>
    </northeast>
   </viewport>
  </geometry>
  <place_id>ChIJ46VqO8GOGGAR-UzmJAzGWN0</place_id>
 </result>
 <result>
  <type>political</type>
  <type>sublocality</type>
  <type>sublocality_level_3</type>
  <formatted_address>1 Chome-37 Asakusa, Taitō-ku, Tōkyō-to 111-0032, Giappone</formatted_address>
  <address_component>
   <long_name>37</long_name>
   <short_name>37</short_name>
   <type>political</type>
   <type>sublocality</type>
   <type>sublocality_level_3</type>
  </address_component>
  <address_component>
   <long_name>1 Chome</long_name>
   <short_name>1 Chome</short_name>
   <type>political</type>
   <type>sublocality</type>
   <type>sublocality_level_2</type>
  </address_component>
  <address_component>
   <long_name>Asakusa</long_name>
   <short_name>Asakusa</short_name>
   <type>political</type>
   <type>sublocality</type>
   <type>sublocality_level_1</type>
  </address_component>
  <address_component>
   <long_name>Taitō-ku</long_name>
   <short_name>Taitō-ku</short_name>
   <type>locality</type>
   <type>political</type>
  </address_component>
  <address_component>
   <long_name>Tōkyō-to</long_name>
   <short_name>Tōkyō-to</short_name>
   <type>administrative_area_level_1</type>
   <type>political</type>
  </address_component>
  <address_component>
   <long_name>Giappone</long_name>
   <short_name>JP</short_name>
   <type>country</type>
   <type>political</type>
  </address_component>
  <address_component>
   <long_name>111-0032</long_name>
   <short_name>111-0032</short_name>
   <type>postal_code</type>
  </address_component>
  <geometry>
   <location>
    <lat>35.7126408</lat>
    <lng>139.7960277</lng>
   </location>
   <location_type>GEOMETRIC_CENTER</location_type>
   <viewport>
    <southwest>
     <lat>35.7112751</lat>
     <lng>139.7948124</lng>
    </southwest>
    <northeast>
     <lat>35.7139731</lat>
     <lng>139.7975104</lng>
    </northeast>
   </viewport>
   <bounds>
    <southwest>
     <lat>35.7124158</lat>
     <lng>139.7956557</lng>
    </southwest>
    <northeast>
     <lat>35.7128324</lat>
     <lng>139.7966671</lng>
    </northeast>
   </bounds>
  </geometry>
  <place_id>ChIJ1yS7JsGOGGARSmB0sZggAXk</place_id>
 </result>
</GeocodeResponse>

Thx, I read this xml <formatted_address>1 Chome-37-1 Asakusa, Taitō-ku, Tōkyō-to 111-0032, Giappone</formatted_address>

Taito-ku and not Taitō-ku   <------------------- UTF-8 ō to ASCII o

Tokyo-to 111-0032 and not Tōkyō-to 111-0032

or in spanish Placa de la Sagrada Família and not Plaça de la Sagrada Família  <------------ UTF-8 ç to ASCII c    Etc...

and this for the other languages, my goal is write a legible words. thx

 

 

Edited September 7, 2017 by rootx

[iamtheky]

Are you converting it to the symbol that most closely visually matches the shape of the character, or converting to the symbol that matches phonetically?

[jchd]

You mean you want to ignore diacritics in roman languages and convert accented and otherwise decorated characters to their "closest" unaccented base char?

If you want to perform this without embarquing the whole ICU library (a monster) yet obtain a relatively correct result you can do the following: use a Windows primitive to convert the UTF16 AutoIt string having normalization form C (NFC) into an UTF16 string having form D (NFD, or better form NFKD if this looks better for you). Then process this string to drop any codepoint > 0x127. This last step can be easily and elegantly done by a regexp.

Consult this primer: https://en.wikipedia.org/wiki/Unicode_equivalence

For how to persuade Windows to convert from one form into another, see: https://msdn.microsoft.com/en-us/library/windows/desktop/dd374126(v=vs.85).aspx

 

 

[rootx]

Yes I was looking for this!!! THX. Why I ask this? because i created a script that uses exiftool to fetch exif and GPS data, the problem is that when Exiftool encounters folders with umlaut or other accented characters or in other languages it fails to read the directory. I follow the help but doesn't work. Now I evaluating a valid approach to solving the problem. https://sno.phy.queensu.ca/~phil/exiftool/faq.html#Q10

Edited September 8, 2017 by rootx

[jchd]

You say this Windows program is incompatible with Windows? Gosh! Find anoher one.

[rootx]

I found this  _FileGetProperty, I have tried to read GPS lat and long, but on Win10 doesn't works... Do you have another way to do it? Thx

 

[jchd]

Googling exif extract should answer the question