TYPO3 Core

[globalVar = TSFE:config|config|sys_language_uid = 1]
mycode
[end]

is not working at all either.

I noticed that the first call is not working but the following calls are working

is definitely wrong!

here another ts:

#Sprachvariabeln default setzen
config.sys_language_uid = 0
config.language = de
config.locale_all = german

1. Sprachparameter in Abhängigkeit der URL setzen
   [globalString = HTTP_HOST= *.com]
   config.sys_language_uid =1
   config.language = en
   config.locale_all = en {$mylangvar} = 1
   [global]

1. Sprachparameter in Abhängigkeit des Parameters L setzen
   [globalVar = GP:L=1]
   config.sys_language_uid =1
   config.language = en
   config.locale_all = en
   [global]

[globalVar = GP:L=0]
config.sys_language_uid = 0
config.language = de
config.locale_all = german
[global]

1. temp-Variable zum Drucken
   temp.PRINT = COA
   temp.PRINT.10 = TEXT
   temp.PRINT.10.value =Seite Drucken
   ...
   #Nur zum Test wird die Sprach-id ausgegeben
2. das funktioniert
   temp.PRINT.20 = TEXT
   temp.PRINT.20.data = TSFE:sys_language_uid
   temp.PRINT.20.wrap = --Sprach-id: | --

#Abfrage, wenn Sprach-id 1, dann schreibe Print funktioniert nicht, da anscheinend ein Bug
[globalVar= TSFE:sys_language_uid =1]
temp.PRINT.10.value = Print
[global]

My conclusion: its just impossible to check sys_language_uid with globalVars, because parsing of globalVars is before parsing of config array! So domainchecking is not very easy atm. Using stdWrap.data = TSFE:sys_language_uid is working but useless.

Final solution for domainswitching:

in the constants:

1. Sprachparameter in Abhängigkeit der URL setzen oder L-Parameter setzen (language var from http_host or L-parameter)

mylangvar = 0

[globalString = HTTP_HOST= *.com]
mylangvar = 1
[global]

[globalVar = GP:L=1]
mylangvar = 1
[global]

usw.

in the setup:

[globalVar = LIT:1 = {$mylangvar}]
das was eben sprachabhängig gemacht werden muss
(condition for english)
[global]

This bug can be closed now. Thanks to aga.

Hmm. I think the only way around this problem is to hack t3lib_matchcondition inside of the getGP_ENV_TSFE() function. However it seems that this condition was never needed before.

Since you already found a workaround for your case, I'll close this bug.

Summary: Bug exists bug is not going to be fixed.