libcurl - how to decode special chars im mail

[flashbob]

I'm currently testing the libcurl library on macOS. The received messages contain characters instead of e.g. german umlauts as in the following example (Memory Viewer):

Content-Type: text/html; charset=UTF-8
Content-Transfer-Encoding: quoted-printable
...
Dies ist eine Testnachricht mit </span><b><span =[CR]
[LF]
style=3D'font-size:10.0pt;font-family:"Arial","sans-serif";color:red'>rot=[CR]
[LF]
em</span></b><span =[CR]
[LF]
style=3D'font-size:10.0pt;font-family:"Arial","sans-serif";color:red'> =[CR]
[LF]
</span><span =[CR]
[LF]
style=3D'font-size:10.0pt;font-family:"Arial","sans-serif";color:#002060'=[CR]
[LF]
>und <b>fetten</b> Text.<br><br>Mit freundlichen =[CR]
[LF]
Gr=FC=DFen<br><br><o:p></o:p></span></p></div></body></html>


As you can see there are special chars (=E4, 3D, =F6, =09, 0, ...)

Any ideas how to decode the received message ? I think its not the right way to replace strings such as:
msg_html = ReplaceString(msg_html, "=E4", "ä" )
msg_html = ReplaceString(msg_html, "=FC", "ü" )

thx in advance !

[idle]

Im on phone but there are two functions you can use Urldecode() and
Esacpehtml() or is that escapehtmlstring()

[infratec]

If you look in the header fields, you can see:

Content-Transfer-Encoding: quoted-printable

Look here:
https://en.wikipedia.org/wiki/Quoted-printable
https://de.wikipedia.org/wiki/Quoted-Pr ... -Kodierung

Such a decoder is not implemented in PB. You have to write it on your own.

[infratec]

A first try:

Code: Select all

Procedure.s Quotedprintable_Decoder(*Text.Character)
  
  Protected Result$, *Start
  
  *Start = *Text
  While *Text\c
    If *Text\c = '='
      Result$ + PeekS(*Start, (*Text - *Start) / SizeOf(Character))
      *Text + SizeOf(Character)
      Result$ + Chr(Val("$" + PeekS(*Text, 2)))
      *Text + 2 * SizeOf(Character)
      *Start = *Text
    Else
      *Text + SizeOf(Character)
    EndIf
  Wend
  
  If *Start < *Text
    Result$ + PeekS(*Start, (*Text - *Start) / SizeOf(Character))
  EndIf
  
  ProcedureReturn Result$
  
EndProcedure





QPText$ = ~"Dies ist eine Testnachricht mit </span><b><span =0Dstyle=3D'font-size:10.0pt;font-family:\"Arial\",\"sans-serif\";color:red'>rot=0Dem</span></b><span =0D"
QPText$ + ~"style=3D'font-size:10.0pt;font-family:\"Arial\",\"sans-serif\";color:red'> =0D</span><span =0Dstyle=3D'font-size:10.0pt;font-family:\"Arial\",\"sans-serif\";color:#002060'=0D"
QPText$ + ~">und <b>fetten</b> Text.<br><br>Mit freundlichen =0DGr=FC=DFen<br><br><o:p></o:p></span></p></div></body></html>"

Debug Quotedprintable_Decoder(@QPText$)

[flashbob]

... wow, that's a great solution. Thank you!

Using your code the converted HTML code is now formatted correctly on MAC and Windows (related to special characters like "3D", "=E4", ...)

Here some more test results using MAC OS:
content-Type: text/html; charset="iso-8859-1"
-> german umlauts are displayed correct in debugger and web gadget.
content="text/html; charset=utf-8"
-> Umlauts are displayed incorrectly in the debugger and web gadget (e.g. "ö" = "ö", "ü" instead of "ü")
-> I used a function to convert asccii to utf8 (see func below) -> ok for web gadget and debugger

Test results using Windows 7:
content-Type: text/html; charset="iso-8859-1"
-> german umlauts are displayed correct in debugger but NOT in web gadget (e.g. "Mit freundlichen Grüßen")
content="text/html; charset=utf-8"
-> german umlauts are NOT displayed correct in debugger and web gadget ("ü" instead of "ü", "ä" instead of "ä",..)
-> Using my function converts not all of german umlauts (e.g. "ß", big "Ü")

Her my function to convert ascii:

Code: Select all

Procedure.s AsciiToUTF8(String$)
  Protected Buffer$
  Buffer$ = Space(Len(String$))
  PokeS(@Buffer$, String$, -1, #PB_Ascii)
  String$ = PeekS(@Buffer$, -1, #PB_UTF8)
  ProcedureReturn String$
EndProcedure