1

I want to replace a very long string according to dictionary. My code is like that:

def rep(self, mystr, dict): new_pstr = '' for char in mystr: try: new_pstr += dict[char] except: continue return new_pstr

This code is not as fast as I expected. Maybe without loop, it will be more faster. But I don't know how to do it. Last but not least, I can't replace all same character together, each time I should replace only one character. Therefore, replace function may not be my option. To be more clear, I give an example:

if d = {'A':'C', 'C':'B'} and mystr = 'AC', then the new_pstr = 'CB'

(If your way return me 'BB', which is not what I expected)

In actual case, my dictionary looks like that:

d = {u'q': [u'k'], u'v': [u'v'], u'e': [u'e'], u'\xe7': [u'\xe7'], u'\xe9': [u'e'], u'h': [u'y'], u'j': [u'z'], u'o': [u'u'], u'\xf1': [u'g'], u'i': [u'i'], u'\u015f': [u's'], u'\xf6': [u'u'], u'x': [u'x'], u'\xfc': [u'v'], u'\u011f': [u'g']}

and my string like that:

str = "tériniñ yiriklişip kétişi havadiki nemlikniñ tövenlep ketkenlikidin bolup ، bu vaqitta tére téximu qurğaqlişip kétidu ، tériniñ ilastikiliqi acizlap ، xünük bolup qalidu. şuña xanim – qizlar bundaq vaqitta tére qurğaqlişişniñ aldini alidiğan çare– tedbirlerni qolliniş kérek. nemlikni saqlaşta yuquri dericilik su toluqlaş yüzlüki، hesel ve örük méğizi méyiğa muvapiq miqdarda un arilaşturup melhem qilip yüzge çaplap bers e، yaki nemxuşluqi yuquri bolğan tére nemleştürüş vazilin méyi sürüp berse، qurğaq térige su toluqlaşqa paydiliq."

The reason I use try.. except ... because my code sometimes return error like that UnicodeEncodeError: 'ascii' codec can't encode character u'\xe7' in position 2: ordinal not in range(128)

Improve this question
5
  • 2
    Any reason not to use string.translate()? Commented Jan 30, 2016 at 9:22
  • use array append and join, instead of string concatenation in python. Commented Jan 30, 2016 at 9:38
  • not related to question, but do NOT use dict or other builtin types as variable names, mydict would be more consistent with mystr Commented Jan 30, 2016 at 9:38
  • any reason why do you need to delete the characters not present in the dictionary? (that is what except: continue is doing) Commented Jan 30, 2016 at 9:54
  • @neoyghur, do you actually have multiple characters as values? Commented Jan 30, 2016 at 11:08

3 Answers 3

Reset to default
1

Consulte this Tutorial. It may help you.

The method translate() returns a copy of the string in which all characters have been translated using table (constructed with the maketrans() function in the string module), optionally deleting all characters found in the string deletechars.

Example code:

Following is the example to delete 'x' and 'm' characters from the string:

#!/usr/bin/python from string import maketrans # Required to call maketrans function. intab = "aeiou" outtab = "12345" trantab = maketrans(intab, outtab) str = "this is string example....wow!!!"; print str.translate(trantab, 'xm')

This will produce following result −

th3s 3s str3ng 21pl2....w4w!!!

The source of this code.

Improve this answer
Sign up to request clarification or add additional context in comments.

2 Comments

my dict include unicode character, so I can't use maketrans function.
Oh ok, sorry, I didn't know that part.
1

What you seem to want to do is change the chars to their ascii equivalent, the Unidecode lib will do that for you, all you need to do with the string is decode and pass it to unidecode.unidecode:

In [8]: s = "tériniñ yiriklişip kétişi havadiki nemlikniñ tövenlep ketkenlikidin bolup ، bu vaqitta tére téximu qurğaqlişip kétidu ، tériniñ ilastikiliqi acizlap ، xünük bolup qalidu. şuña xanim – qizlar bundaq vaqitta tére qurğaqlişişniñ aldini alidiğan çare– tedbirlerni qolliniş kérek. nemlikni saqlaşta yuquri dericilik su toluqlaş yüzlüki، hesel ve örük méğizi méyiğa muvapiq miqdarda un arilaşturup melhem qilip yüzge çaplap bers e، yaki nemxuşluqi yuquri bolğan tére nemleştürüş vazilin méyi sürüp berse، qurğaq térige su toluqlaşqa paydiliq." In [9]: unidecode.unidecode(s.decode("utf-8")) Out[9]: 'terinin yiriklisip ketisi havadiki nemliknin tovenlep ketkenlikidin bolup , bu vaqitta tere teximu qurgaqlisip ketidu , terinin ilastikiliqi acizlap , xunuk bolup qalidu. suna xanim - qizlar bundaq vaqitta tere qurgaqlisisnin aldini alidigan care- tedbirlerni qollinis kerek. nemlikni saqlasta yuquri dericilik su toluqlas yuzluki, hesel ve oruk megizi meyiga muvapiq miqdarda un arilasturup melhem qilip yuzge caplap bers e, yaki nemxusluqi yuquri bolgan tere nemlesturus vazilin meyi surup berse, qurgaq terige su toluqlasqa paydiliq.'

Or if you have multiple characters as values, this is a faster working version of your own logic:

In [27]: from itertools import chain In [28]: d = {k:v[0] for k,v in d.items()} In [29]: "".join([d[ch] if ch in d else ch for ch in chain.from_iterable(s)]) Out[29]: 'terinig yiriklisip ketisi yavadiki nemliknig tuvenlep ketkenlikidin bulup ، bu vakitta tere teximu kurgaklisip ketidu ، terinig ilastikiliki acizlap ، xvnvk bulup kalidu. suga xanim – kizlar bundak vakitta tere kurgaklisisnig aldini alidigan çare– tedbirlerni kullinis kerek. nemlikni saklasta yukuri dericilik su tuluklas yvzlvki، yesel ve urvk megizi meyiga muvapik mikdarda un arilasturup melyem kilip yvzge çaplap bers e، yaki nemxusluki yukuri bulgan tere nemlestvrvs vazilin meyi svrvp berse، kurgak terige su tuluklaska paydilik.'

Also the correct way to use str.translate with unicode is to use the ord of the characters

table =({ord(k):ord(ch) for k ,v in d.items() for ch in v}) s.translate(table)
Improve this answer

Comments

1

I get the final answer by combining the answers from @Padraic Cunningham @PRVS, this version is 100 times faster than my original code.

 new_d = {ord(k): ord(v[0]) for k, v in d.items()} # ord for Unicode characters mystr.translate(d)

If your code don't have any Unicode characters, please check @PRVS answer.

Improve this answer

Comments

Start asking to get answers

Find the answer to your question by asking.

Ask question

Explore related questions

See similar questions with these tags.