Regular expression curvy brackets in Python - python

I have a string like this:
a = '{CGPoint={CGPoint=d{CGPoint=dd}}}{CGSize=dd}dd{CSize=aa}'
Currently I am using this re statement to get desired result:
filter(None, re.split("\\{(.*?)\\}", a))
But this gives me:
['CGPoint={CGPoint=d{CGPoint=dd', '}}', 'CGSize=dd', 'dd', 'CSize=aa']
which is incorrect for my current situation, I need a list like this:
['CGPoint={CGPoint=d{CGPoint=dd}}', 'CGSize=dd', 'dd', 'CSize=aa']


As #m.buettner points out in the comments, Python's implementation of regular expressions can't match pairs of symbols nested to an arbitrary degree. (Other languages can, notably current versions of Perl.) The Pythonic thing to do when you have text that regexs can't parse is to use a recursive-descent parser.
There's no need to reinvent the wheel by writing your own, however; there are a number of easy-to-use parsing libraries out there. I recommend pyparsing which lets you define a grammar directly in your code and easily attach actions to matched tokens. Your code would look something like this:
import pyparsing
lbrace = Literal('{')
rbrace = Literal('}')
contents = Word(printables)
expr = Forward()
expr << Combine(Suppress(lbrace) + contents + Suppress(rbrace) + expr)
for line in lines:
results = expr.parseString(line)


There's an alternative regex module for Python I really like that supports recursive patterns:
https://pypi.python.org/pypi/regex
pip install regex
Then you can use a recursive pattern in your regex as demonstrated in this script:
import regex
from pprint import pprint
thestr = '{CGPoint={CGPoint=d{CGPoint=dd}}}{CGSize=dd}dd{CSize=aa}'
theregex = r'''
(
{
(?<match>
[^{}]*
(?:
(?1)
[^{}]*
)+
|
[^{}]+
)
}
|
(?<match>
[^{}]+
)
)
'''
matches = regex.findall(theregex, thestr, regex.X)
print 'all matches:\n'
pprint(matches)
print '\ndesired matches:\n'
print [match[1] for match in matches]
This outputs:
all matches:
[('{CGPoint={CGPoint=d{CGPoint=dd}}}', 'CGPoint={CGPoint=d{CGPoint=dd}}'),
('{CGSize=dd}', 'CGSize=dd'),
('dd', 'dd'),
('{CSize=aa}', 'CSize=aa')]
desired matches:
['CGPoint={CGPoint=d{CGPoint=dd}}', 'CGSize=dd', 'dd', 'CSize=aa']


pyparsing has a nestedExpr function for matching nested expressions:
import pyparsing as pp
ident = pp.Word(pp.alphanums)
expr = pp.nestedExpr("{", "}") | ident
thestr = '{CGPoint={CGPoint=d{CGPoint=dd}}}{CGSize=dd}dd{CSize=aa}'
for result in expr.searchString(thestr):
print(result)
yields
[['CGPoint=', ['CGPoint=d', ['CGPoint=dd']]]]
[['CGSize=dd']]
['dd']
[['CSize=aa']]


Here is some pseudo code. It creates a stack of strings and pops them when a close brace is encountered. Some extra logic to handle the fact that the first braces encountered are not included in the array.
String source = "{CGPoint={CGPoint=d{CGPoint=dd}}}{CGSize=dd}dd{CSize=aa}";
Array results;
Stack stack;
foreach (match in source.match("[{}]|[^{}]+")) {
switch (match) {
case '{':
if (stack.size == 0) stack.push(new String()); // add new empty string
else stack.push('{'); // child, so include matched brace.
case '}':
if (stack.size == 1) results.add(stack.pop()) // clear stack add to array
else stack.last += stack.pop() + '}"; // pop from stack and concatenate to previous
default:
if (stack.size == 0) results.add(match); // loose text, add to results
else stack.last += match; // append to latest member.
}
}

Related

Complex string filtering with python

I have a long string that is a phylogenetic tree and I want to do a very specific filtering.
(Esy#ESY15_g64743_DN3_SP7_c0:0.0726396855636,Aar#AA_maker7399_1:0.137507902808,((Spa#Tp2g18720:0.0318934795022,Cpl#CP2_g48793_DN3_SP8_c:0.0273465005242):9.05326020871e-05,(((Bst#Bostr_13083s0053_1:0.0332592496158,((Aly#AL8G21130_t1:0.0328569260951,Ath#AT5G48370_1:0.0391706378372):0.0205924636564,(Chi#CARHR183840_1:0.0954469923893,Cru#Carubv10026342m:0.0570981548016):0.00998579652059):0.0150356382287):0.0340484449097,(((Hco#scaff1034_g23864_DN3_SP8_c_TE35_CDS100:0.00823215335663,Hlo#DN13684_c0_g1_i1_p1:0.0085462978729):0.0144626717872,Hla#DN22821_c0_g1_i1_p1:0.0225079453622):0.0206478928557,Hse#DN23412_c0_g1_i3_p1:0.048590776459):0.0372829371381):0.00859075940423,(Esa#Thhalv10004228m:0.0378509854703,Aal#Aa_G102140_t1:0.0712272454125):1.00000050003e-06):0.00328120860999):0.0129090235079):0.0129090235079;
Basically every x#y is a species#gene_id information. What I am trying to do is trimming this down so that I will only have x instead of x#y.
(Esy, Aar,(Spa,Cpl))...
I tried splitting the string first but the problem is string has different 'split points' for what I want to achieve i.e. some parts x#y is ending with a , and others with a ). I searched for a solution and saw regular expression operations, but I am new to Python and I couldn't be sure if that is what I should be focusing on. I also thought about strip() but it seems like I need to specify the characters to be stripped for this.
Main problem is there is no 'pattern' for me to tell Python to follow. Only thing is that all species ids are 3 letters and they are before an # character.
Is there a method that can do what I want? I will be really glad if you can help me out with my problem. Thanks in advance.

Give this a try:
import re:
pat = re.compile(r'(\w{3})#')
txt = "(Esy#ESY15_g64743_DN3_SP7_c0:0.0726396855636,Aar#AA_maker7399_1:0.137507902808,((Spa#Tp2g18720:0.0318934795022,Cpl#CP2_g48793_DN3_SP8_c:0.0273465005242):9.05326020871e-05,(((Bst#Bostr_13083s0053_1:0.0332592496158,((Aly#AL8G21130_t1:0.0328569260951,Ath#AT5G48370_1:0.0391706378372):0.0205924636564,(Chi#CARHR183840_1:0.0954469923893,Cru#Carubv10026342m:0.0570981548016):0.00998579652059):0.0150356382287):0.0340484449097,(((Hco#scaff1034_g23864_DN3_SP8_c_TE35_CDS100:0.00823215335663,Hlo#DN13684_c0_g1_i1_p1:0.0085462978729):0.0144626717872,Hla#DN22821_c0_g1_i1_p1:0.0225079453622):0.0206478928557,Hse#DN23412_c0_g1_i3_p1:0.048590776459):0.0372829371381):0.00859075940423,(Esa#Thhalv10004228m:0.0378509854703,Aal#Aa_G102140_t1:0.0712272454125):1.00000050003e-06):0.00328120860999):0.0129090235079):0.0129090235079;"
pat.findall(t)
Result:
['Esy', 'Aar', 'Spa', 'Cpl', 'Bst', 'Aly', 'Ath', 'Chi', 'Cru', 'Hco', 'Hlo', 'Hla', 'Hse', 'Esa', 'Aal']
If you need the structure intact, we can try to remove the unnecessary parts instead:
pat = re.compile(r'(#|:)[^/),]*')
pat.sub('',t).replace(',', ', ')
Result:
'(Esy, Aar, ((Spa, Cpl), (((Bst, ((Aly, Ath), (Chi, Cru))), (((Hco, Hlo), Hla), Hse)), (Esa, Aal))))'
Regex demo

How about this kind of function:
def parse_string(string):
new_string = ''
skip = False
for char in string:
if char == '#':
skip = True
if char == ',':
skip = False
if not skip or char in ['(', ')']:
new_string += char
return new_string
Calling it on your string:
string = '(Esy#ESY15_g64743_DN3_SP7_c0:0.0726396855636,Aar#AA_maker7399_1:0.137507902808,((Spa#Tp2g18720:0.0318934795022,Cpl#CP2_g48793_DN3_SP8_c:0.0273465005242):9.05326020871e-05,(((Bst#Bostr_13083s0053_1:0.0332592496158,((Aly#AL8G21130_t1:0.0328569260951,Ath#AT5G48370_1:0.0391706378372):0.0205924636564,(Chi#CARHR183840_1:0.0954469923893,Cru#Carubv10026342m:0.0570981548016):0.00998579652059):0.0150356382287):0.0340484449097,(((Hco#scaff1034_g23864_DN3_SP8_c_TE35_CDS100:0.00823215335663,Hlo#DN13684_c0_g1_i1_p1:0.0085462978729):0.0144626717872,Hla#DN22821_c0_g1_i1_p1:0.0225079453622):0.0206478928557,Hse#DN23412_c0_g1_i3_p1:0.048590776459):0.0372829371381):0.00859075940423,(Esa#Thhalv10004228m:0.0378509854703,Aal#Aa_G102140_t1:0.0712272454125):1.00000050003e-06):0.00328120860999):0.0129090235079):0.0129090235079;'
parse_string(string)
> '(Esy,Aar,((Spa,Cpl),(((Bst,((Aly,Ath),(Chi,Cru))),(((Hco,Hlo),Hla),Hse)),(Esa,Aal))))'

you can use regex:
import re
s = "(Esy#ESY15_g64743_DN3_SP7_c0:0.0726396855636,Aar#AA_maker7399_1:0.137507902808,((Spa#Tp2g18720:0.0318934795022,Cpl#CP2_g48793_DN3_SP8_c:0.0273465005242):9.05326020871e-05,(((Bst#Bostr_13083s0053_1:0.0332592496158,((Aly#AL8G21130_t1:0.0328569260951,Ath#AT5G48370_1:0.0391706378372):0.0205924636564,(Chi#CARHR183840_1:0.0954469923893,Cru#Carubv10026342m:0.0570981548016):0.00998579652059):0.0150356382287):0.0340484449097,(((Hco#scaff1034_g23864_DN3_SP8_c_TE35_CDS100:0.00823215335663,Hlo#DN13684_c0_g1_i1_p1:0.0085462978729):0.0144626717872,Hla#DN22821_c0_g1_i1_p1:0.0225079453622):0.0206478928557,Hse#DN23412_c0_g1_i3_p1:0.048590776459):0.0372829371381):0.00859075940423,(Esa#Thhalv10004228m:0.0378509854703,Aal#Aa_G102140_t1:0.0712272454125):1.00000050003e-06):0.00328120860999):0.0129090235079):0.0129090235079;"
p = "...?(?=#)|\(|\)"
result = re.findall(p, s)
and you have your result as a list, so you can make it string or do anything with it
for explaining what is happening :
p is regular expression pattern
so in this pattern:
. means matching any word
...?(?=#) means match any word until I get to a word ? wich ? is #, so this whole pattern means that you get any three words before #
| is or statement, I used it here to find another pattern
and the rest of them is to find ) and (

Try this regex if you need the brackets in the output:
import re
regex = r"#[A-Za-z0-9_\.:]+|[0-9:\.;e-]+"
phylogenetic_tree = "(Esy#ESY15_g64743_DN3_SP7_c0:0.0726396855636,Aar#AA_maker7399_1:0.137507902808,((Spa#Tp2g18720:0.0318934795022,Cpl#CP2_g48793_DN3_SP8_c:0.0273465005242):9.05326020871e-05,(((Bst#Bostr_13083s0053_1:0.0332592496158,((Aly#AL8G21130_t1:0.0328569260951,Ath#AT5G48370_1:0.0391706378372):0.0205924636564,(Chi#CARHR183840_1:0.0954469923893,Cru#Carubv10026342m:0.0570981548016):0.00998579652059):0.0150356382287):0.0340484449097,(((Hco#scaff1034_g23864_DN3_SP8_c_TE35_CDS100:0.00823215335663,Hlo#DN13684_c0_g1_i1_p1:0.0085462978729):0.0144626717872,Hla#DN22821_c0_g1_i1_p1:0.0225079453622):0.0206478928557,Hse#DN23412_c0_g1_i3_p1:0.048590776459):0.0372829371381):0.00859075940423,(Esa#Thhalv10004228m:0.0378509854703,Aal#Aa_G102140_t1:0.0712272454125):1.00000050003e-06):0.00328120860999):0.0129090235079):0.0129090235079;"
print(re.sub(regex,"",phylogenetic_tree))
Output:
(Esy,Aar,((Spa,Cpl),(((Bst,((Aly,Ath),(Chi,Cru))),(((Hco,Hlo),Hla),Hs)),(Esa,Aal))))

Because you are trying to parse a phylogenetic tree, I highly suggest to let BioPython do the heavy lifting for you.
You can easily parse and display a phylogenetic with Bio.Phylo. Then it is just iterating over all tree elements and splitting the names at the 'at'-sign.
Because Phylo expects the input to be in a file, we create an in-memory file-like object with io.StringIO. Getting the complete tree is then as easy as
Phylo.read(io.StringIO(s), 'newick')
In order to check if the parsed tree looks sane, I print it once with print(tree).
Now we want to change all node names that contain a '#'. With tree.find_elements we get access to all nodes. Some nodes don't have a name and some might not contain a '#'. So to be extra careful, we first check if n.name and '#' in n.name. Only then do we split each node's name at the '#' and take just the first part (index 0) of it:
n.name = n.name.split('#')[0]
In order to recreate the initial string representation, we use Phylo.write:
out = io.StringIO()
Phylo.write(tree, out, "newick")
print(out.getvalue())
Again, write wants to get a file argument - if we just want to get a string, we can use a StringIO object again.
Full code:
import io
from Bio import Phylo
if __name__ == '__main__':
s = '(Esy#ESY15_g64743_DN3_SP7_c0:0.0726396855636,Aar#AA_maker7399_1:0.137507902808,((Spa#Tp2g18720:0.0318934795022,Cpl#CP2_g48793_DN3_SP8_c:0.0273465005242):9.05326020871e-05,(((Bst#Bostr_13083s0053_1:0.0332592496158,((Aly#AL8G21130_t1:0.0328569260951,Ath#AT5G48370_1:0.0391706378372):0.0205924636564,(Chi#CARHR183840_1:0.0954469923893,Cru#Carubv10026342m:0.0570981548016):0.00998579652059):0.0150356382287):0.0340484449097,(((Hco#scaff1034_g23864_DN3_SP8_c_TE35_CDS100:0.00823215335663,Hlo#DN13684_c0_g1_i1_p1:0.0085462978729):0.0144626717872,Hla#DN22821_c0_g1_i1_p1:0.0225079453622):0.0206478928557,Hse#DN23412_c0_g1_i3_p1:0.048590776459):0.0372829371381):0.00859075940423,(Esa#Thhalv10004228m:0.0378509854703,Aal#Aa_G102140_t1:0.0712272454125):1.00000050003e-06):0.00328120860999):0.0129090235079):0.0129090235079;'
tree = Phylo.read(io.StringIO(s), 'newick')
print(' before '.center(20, '='))
print(tree)
for n in tree.find_elements():
if n.name and '#' in n.name:
n.name = n.name.split('#')[0]
print(' result '.center(20, '='))
out = io.StringIO()
Phylo.write(tree, out, "newick")
print(out.getvalue())
Output:
====== before ======
Tree(rooted=False, weight=1.0)
Clade(branch_length=0.0129090235079)
Clade(branch_length=0.0726396855636, name='Esy#ESY15_g64743_DN3_SP7_c0')
Clade(branch_length=0.137507902808, name='Aar#AA_maker7399_1')
Clade(branch_length=0.0129090235079)
Clade(branch_length=9.05326020871e-05)
Clade(branch_length=0.0318934795022, name='Spa#Tp2g18720')
Clade(branch_length=0.0273465005242, name='Cpl#CP2_g48793_DN3_SP8_c')
Clade(branch_length=0.00328120860999)
Clade(branch_length=0.00859075940423)
Clade(branch_length=0.0340484449097)
Clade(branch_length=0.0332592496158, name='Bst#Bostr_13083s0053_1')
Clade(branch_length=0.0150356382287)
Clade(branch_length=0.0205924636564)
Clade(branch_length=0.0328569260951, name='Aly#AL8G21130_t1')
Clade(branch_length=0.0391706378372, name='Ath#AT5G48370_1')
Clade(branch_length=0.00998579652059)
Clade(branch_length=0.0954469923893, name='Chi#CARHR183840_1')
Clade(branch_length=0.0570981548016, name='Cru#Carubv10026342m')
Clade(branch_length=0.0372829371381)
Clade(branch_length=0.0206478928557)
Clade(branch_length=0.0144626717872)
Clade(branch_length=0.00823215335663, name='Hco#scaff1034_g23864_DN3_SP8_c_TE35_CDS100')
Clade(branch_length=0.0085462978729, name='Hlo#DN13684_c0_g1_i1_p1')
Clade(branch_length=0.0225079453622, name='Hla#DN22821_c0_g1_i1_p1')
Clade(branch_length=0.048590776459, name='Hse#DN23412_c0_g1_i3_p1')
Clade(branch_length=1.00000050003e-06)
Clade(branch_length=0.0378509854703, name='Esa#Thhalv10004228m')
Clade(branch_length=0.0712272454125, name='Aal#Aa_G102140_t1')
==== result =====
(Esy:0.07264,Aar:0.13751,((Spa:0.03189,Cpl:0.02735):0.00009,(((Bst:0.03326,((Aly:0.03286,Ath:0.03917):0.02059,(Chi:0.09545,Cru:0.05710):0.00999):0.01504):0.03405,(((Hco:0.00823,Hlo:0.00855):0.01446,Hla:0.02251):0.02065,Hse:0.04859):0.03728):0.00859,(Esa:0.03785,Aal:0.07123):0.00000):0.00328):0.01291):0.01291;
The default format of Phylo uses less digits than in your original tree. In order to keep the numbers unchanged, just override the branch length format string with a '%s':
Phylo.write(tree, out, "newick", format_branch_length="%s")

Parsing code can be hard to follow. Tatsu lets you write readable parsing code by combining grammars and python:
text = "(Esy#ESY15_g64743_DN3_SP7_c0:0.0726396855636,Aar#AA_maker7399_1:0.137507902808,((Spa#Tp2g18720:0.0318934795022,Cpl#CP2_g48793_DN3_SP8_c:0.0273465005242):9.05326020871e-05,(((Bst#Bostr_13083s0053_1:0.0332592496158,((Aly#AL8G21130_t1:0.0328569260951,Ath#AT5G48370_1:0.0391706378372):0.0205924636564,(Chi#CARHR183840_1:0.0954469923893,Cru#Carubv10026342m:0.0570981548016):0.00998579652059):0.0150356382287):0.0340484449097,(((Hco#scaff1034_g23864_DN3_SP8_c_TE35_CDS100:0.00823215335663,Hlo#DN13684_c0_g1_i1_p1:0.0085462978729):0.0144626717872,Hla#DN22821_c0_g1_i1_p1:0.0225079453622):0.0206478928557,Hse#DN23412_c0_g1_i3_p1:0.048590776459):0.0372829371381):0.00859075940423,(Esa#Thhalv10004228m:0.0378509854703,Aal#Aa_G102140_t1:0.0712272454125):1.00000050003e-06):0.00328120860999):0.0129090235079):0.0129090235079;"
import sys
import tatsu
grammar = """
start = things ';'
;
things = thing [ ',' things ]
;
thing = x '#' y ':' number
| '(' things ')' ':' number
;
x = /\w+/
;
y = /\w+/
;
number = /[+-]?\d+\.?\d*(e?[+-]?\d*)/
;
"""
class Semantics:
def x(self, ast):
# the method name matches the rule name
print('X =', ast)
parser = tatsu.compile(grammar, semantics=Semantics())
parser.parse(text)

Python regex match anything enclosed in either quotations brackets braces or parenthesis

UPDATE
This is still not entirely the solution so far. It is only for preceding repeated closing characters (e.g )), ]], }}). I'm still looking for a way to capture enclosed contents and will update this.
Code:
>>> import re
>>> re.search(r'(\(.+?[?<!)]\))', '((x(y)z))', re.DOTALL).groups()
('((x(y)z))',)
Details:
r'(\(.+?[?<!)]\))'
() - Capturing group special characters.
\( and \) - The open and closing characters (e.g ', ", (), {}, [])
.+? - Match any character content (use with re.DOTALL flag)
[?<!)] - The negative lookbehind for character ) (replace this with the matching closing character). This will basically find any ) character where \) character does not precede (more info here).
I was trying to parse something like a variable assignment statement for this lexer thing I'm working with, just trying to get the basic logic behind interpreters/compilers.
Here's the basic assignment statements and literals I'm dealing with:
az = none
az_ = true
az09 = false
az09_ = +0.9
az_09 = 'az09_'
_az09 = "az09_"
_az = [
"az",
0.9
]
_09 = {
0: az
1: 0.9
}
_ = (
true
)
Somehow, I managed to parse those simple assignments like none, true, false, and numeric literals. Here's where I'm currently stuck at:
import sys
import re
# validate command-line arguments
if (len(sys.argv) != 2): raise ValueError('usage: parse <script>')
# parse the variable name and its value
def handle_assignment(index, source):
# TODO: handle quotations, brackets, braces, and parenthesis values
variable = re.search(r'[\S\D]([\w]+)\s+?=\s+?(none|true|false|[-+]?\d+\.?\d+|[\'\"].*[\'\"])', source[index:])
if variable is not None:
print('{}={}'.format(variable.group(1), variable.group(2)))
index += source[index:].index(variable.group(2))
return index
# parse through the source element by element
with open(sys.argv[1]) as file:
source = file.read()
index = 0
while index < len(source):
# checks if the line matches a variable assignment statement
if re.match(r'[\S\D][\w]+\s+?=', source[index:]):
index = handle_assignment(index, source)
index += 1
I was looking for a way to capture those values with enclosed quotations, brackets, braces, and parenthesis.
Probably, will update this post if I found an answer.

Use a regexp with multiple alternatives for each matching pair.
re.match(r'\'.*?\'|".*?"|\(.*?\)|\[.*?\]|\{.*?\}', s)
Note, however, that if there are nested brackets, this will match the first ending bracket, e.g. if the input is
(words (and some more words))
the result will be
(words (and some more words)
Regular expressions are not appropriate for matching nested structures, you should use a more powerful parsing technique.

Solution for #Barmar's recursive characters using the regex third-party module:
pip install regex
python3
>>> import regex
>>> recurParentheses = regex.compile(r'[(](?:[^()]|(?R))*[)]')
>>> recurParentheses.findall('(z(x(y)z)x) ((x)(y)(z))')
['(z(x(y)z)x)', '((x)(y)(z))']
>>> recurCurlyBraces = regex.compile(r'[{](?:[^{}]|(?R))*[}]')
>>> recurCurlyBraces.findall('{z{x{y}z}x} {{x}{y}{z}}')
['{z{x{y}z}x}', '{{x}{y}{z}}']
>>> recurSquareBrackets = regex.compile(r'[[](?:[^][]|(?R))*[]]')
>>> recurSquareBrackets.findall('[z[x[y]z]x] [[x][y][z]]')
['[z[x[y]z]x]', '[[x][y][z]]']
For string literal recursion, I suggest take a look at this.

Remove outer most curly bracket with Regex

I am trying to remove the outer most curly bracket while keeping only the inner string. My code almost works 100%, except when
expr = 'namespace P {\\na; b;}'
# I expect '\\na; b;'
# but I get 'namespace P {\na; b;}' instead
Any idea how to fix my regex string?
import doctest
import re
def remove_outer_curly_bracket(expr):
"""
>>> remove_outer_curly_bracket('P {')
'P {'
>>> remove_outer_curly_bracket('P')
'P'
>>> remove_outer_curly_bracket('P { a; b(); { c1(d,e); } }')
' a; b(); { c1(d,e); } '
>>> remove_outer_curly_bracket('a { }')
' '
>>> remove_outer_curly_bracket('')
''
>>> remove_outer_curly_bracket('namespace P {\\na; b;}')
'\\na; b;'
"""
r = re.findall(r'[.]*\{(.*)\}', expr)
return r[0] if r else expr
doctest.testmod()

This suffices:
def remove_outer_curly_bracket(expr):
r = re.search(r'{(.*)}', expr, re.DOTALL)
return r.group(1) if r else expr
The match will start as soon as possible, so the first { will indeed match the leftmost opening brace. Because * is greedy, .* will want to be as large as possible, which will ensure } will match the last closing brace.
Neither of the braces is a special character, and does not need escaping; also, [.]* matches any number of periods in a row, and will not help you at all in this task.
This will not work sensibly if the braces are not balanced; for example, for "{ { x }" will return " { x", but fortunately your examples do not include such.
EDIT: That said, this is just prettifying the original a bit. The functionality is unchanged. As blhsing says in comments, it seems your code is doing what it is supposed to. It even passes your tests.
EDIT2: There is nothing special in 'namespace P {\\na; b;}'. I believe you meant 'namespace P {\na; b;}'? With a line break inside? Indeed, that would not have worked. I changed my code so it does. The issue is that normally . matches every character except newline. We can modify that behaviour by supplying the flag re.DOTALL.

pythonic string syntax corrector

I wrote a script to catch and correct commands before they are read by a parser. The parser requires equal, not equal, greater, etc, entries to be separated by commas, such as:
'test(a>=b)' is wrong
'test(a,>=,b)' is correct
The script i wrote works fine, but i would love to know if there's a more efficient way to do this.
Here's my script:
# Correction routine
def corrector(exp):
def rep(exp,a,b):
foo = ''
while(True):
foo = exp.replace(a,b)
if foo == exp:
return exp
exp = foo
# Replace all instances with a unique identifier. Do it in a specific order
# so for example we catch an instance of '>=' before we get to '='
items = ['>=','<=','!=','==','>','<','=']
for i in range(len(items)):
exp = rep(exp,items[i],'###%s###'%i)
# Re-add items with commas
for i in range(len(items)):
exp = exp.replace('###%s###'%i,',%s,'%items[i])
# Remove accidental double commas we may have added
return exp.replace(',,',',')
print corrector('wrong_syntax(b>=c) correct_syntax(b,>=,c)')
// RESULT: wrong_syntax(b,>=,c) correct_syntax(b,>=,c)
thanks!

As mentioned in the comments, one approach would be to use a regular expression. The following regex matches any of your operators when they are not surrounded by commas, and replaces them with the same string with the commas inserted:
inputstring = 'wrong_syntax(b>=c) correct_syntax(b,>=,c)'
regex = r"([^,])(>=|<=|!=|==|>|<|=)([^,])"
replace = r"\1,\2,\3"
result = re.sub(regex, replace, inputstring)
print(result)
Simple regexes are relatively easy, but they can get complicated quickly. Check out the docs for more info:
http://docs.python.org/2/library/re.html

Here is a regex that will do what you asked:
import re
regex = re.compile(r'''
(?<!,) # Negative lookbehind
(!=|[><=]=?)
(?!,) # Negative lookahead
''', re.VERBOSE)
print regex.sub(r',\1,', 'wrong_expression(b>=c) or right_expression(b,>=,c)')
outputs
wrong_expression(b,>=,c) or right_expression(b,>=,c)

Python, how do I parse key=value list ignoring what is inside parentheses?

Suppose I have a string like this:
"key1=value1;key2=value2;key3=(key3.1=value3.1;key3.2=value3.2)"
I would like to get a dictionary corresponding to the above, where the value for key3 is the string
"(key3.1=value3.1;key3.2=value3.2)"
and eventually the corresponding sub-dictionary.
I know how to split the string at the semicolons, but how can I tell the parser to ignore the semicolon between parentheses?
This includes potentially nested parentheses.
Currently I am using an ad-hoc routine that looks for pairs of matching parentheses, "clears" its content, gets split positions and applies them to the original string, but this does not appear very elegant, there must be some prepackaged pythonic way to do this.
If anyone is interested, here is the code I am currently using:
def pparams(parameters, sep=';', defs='=', brc='()'):
'''
unpackages parameter string to struct
for example, pippo(a=21;b=35;c=pluto(h=zzz;y=mmm);d=2d3f) becomes:
a: '21'
b: '35'
c.fn: 'pluto'
c.h='zzz'
d: '2d3f'
fn_: 'pippo'
'''
ob=strfind(parameters,brc[0])
dp=strfind(parameters,defs)
out={}
if len(ob)>0:
if ob[0]<dp[0]:
#opening function
out['fn_']=parameters[:ob[0]]
parameters=parameters[(ob[0]+1):-1]
if len(dp)>0:
temp=smart_tokenize(parameters,sep,brc);
for v in temp:
defp=strfind(v,defs)
pname=v[:defp[0]]
pval=v[1+defp[0]:]
if len(strfind(pval,brc[0]))>0:
out[pname]=pparams(pval,sep,defs,brc);
else:
out[pname]=pval
else:
out['fn_']=parameters
return out
def smart_tokenize( instr, sep=';', brc='()' ):
'''
tokenize string ignoring separators contained within brc
'''
tstr=instr;
ob=strfind(instr,brc[0])
while len(ob)>0:
cb=findclsbrc(tstr,ob[0])
tstr=tstr[:ob[0]]+'?'*(cb-ob[0]+1)+tstr[cb+1:]
ob=strfind(tstr,brc[1])
sepp=[-1]+strfind(tstr,sep)+[len(instr)+1]
out=[]
for i in range(1,len(sepp)):
out.append(instr[(sepp[i-1]+1):(sepp[i])])
return out
def findclsbrc(instr, brc_pos, brc='()'):
'''
given a string containing an opening bracket, finds the
corresponding closing bracket
'''
tstr=instr[brc_pos:]
o=strfind(tstr,brc[0])
c=strfind(tstr,brc[1])
p=o+c
p.sort()
s1=[1 if v in o else 0 for v in p]
s2=[-1 if v in c else 0 for v in p]
s=[s1v+s2v for s1v,s2v in zip(s1,s2)]
s=[sum(s[:i+1]) for i in range(len(s))] #cumsum
return p[s.index(0)]+brc_pos
def strfind(instr, substr):
'''
returns starting position of each occurrence of substr within instr
'''
i=0
out=[]
while i<=len(instr):
try:
p=instr[i:].index(substr)
out.append(i+p)
i+=p+1
except:
i=len(instr)+1
return out

If you want to build a real parser, use one of the Python parsing libraries, like PLY or PyParsing. If you figure such a full-fledged library is overkill for the task at hand, go for some hack like the one you already have. I'm pretty sure there is no clean few-line solution without an external library.

Expanding on Sven Marnach's answer, here's an example of a pyparsing grammar that should work for you:
from pyparsing import (ZeroOrMore, Word, printables, Forward,
Group, Suppress, Dict)
collection = Forward()
simple_value = Word(printables, excludeChars='()=;')
key = simple_value
inner_collection = Suppress('(') + collection + Suppress(')')
value = simple_value ^ inner_collection
key_and_value = Group(key + Suppress('=') + value)
collection << Dict(key_and_value + ZeroOrMore(Suppress(';') + key_and_value))
coll = collection.parseString(
"key1=value1;key2=value2;key3=(key3.1=value3.1;key3.2=value3.2)")
print coll['key1'] # value1
print coll['key2'] # value2
print coll['key3']['key3.1'] # value3.1

You could use a regex to capture the groups:
>>> import re
>>> s = "key1=value1;key2=value2;key3=(key3.1=value3.1;key3.2=value3.2)"
>>> r = re.compile('(\w+)=(\w+|\([^)]+\));?')
>>> dict(r.findall(s))
This regex says:
(\w)+ # Find and capture a group with 1 or more word characters (letters, digits, underscores)
= # Followed by the literal character '='
(\w+ # Followed by a group with 1 or more word characters
|\([^)]+\) # or a group that starts with an open paren (parens escaped with '\(' or \')'), followed by anything up until a closed paren, which terminates the alternate grouping
);? # optionally this grouping might be followed by a semicolon.
Gotta say, kind of a strange grammar. You should consider using a more standard format. If you need guidance choosing one maybe ask another question. Good luck!

Categories