How to tell Googletrans to ignore certain parts?

Question

I would like to use googletrans to make use of the Google translate API. However, there are strings where are variable names in it:

User "%(first_name)s %(last_name)s (%(email)s)" has been deleted.

If I use this via googletrans I get

from googletrans import Translator
translator = Translator()
translator.translate(u'User "%(first_name)s %(last_name)s (%(email)s)" has been assigned.', src='en', dest='fr').text

I get the following:

L'utilisateur "% (first_name) s% (last_name) s (% (email) s)" a été affecté.

However, the "%(first_name) s% (last_name)s (%(email)s)" has some strings introduced. Is there a way around this? I've already tried:

u'User "<span class="notranslate">%(first_name)s %(last_name)s (%(email)s)</span>" has been assigned.'

Answer 1

It seems Googletrans leaves, e.g., __1__ untouched. So you can replace %(first_name)s with __0__, %(last_name)s with __1__, etc. before you translate, and then restore the variables afterwards. Here code to do this:

from googletrans import Translator
import re

translator = Translator()
txtorig = u'User "%(first_name)s %(last_name)s (%(email)s)" has been assigned.'

# temporarily replace variables of format "%(example_name)s" with "__n__" to
#  protect them during translate()
VAR, REPL = re.compile(r'%\(\w+\)s'), re.compile(r'__(\d+)__')
varlist = []
def replace(matchobj):
  varlist.append(matchobj.group())
  return "__%d__" %(len(varlist)-1)
def restore(matchobj):
  return varlist[int(matchobj.group(1))]

txtorig = VAR.sub(replace, txtorig)
txttrans = translator.translate(txtorig, src='en', dest='fr').text
txttrans = REPL.sub(restore, txttrans)

print(txttrans)

Here the result:

L'utilisateur "%(first_name)s %(last_name)s (%(email)s)" a été attribué.