Develop Reference

Python 3.6 script surprisingly slow on Windows 10 but not on Ubuntu 17.10

I recently had to write a challenge for a company that was to merge 3 CSV files into one based on the first attribute of each (the attributes were repeating in all files).
I wrote the code and sent it to them, but they said it took 2 minutes to run. That was funny because it ran for 10 seconds on my machine. My machine had the same processor, 16GB of RAM, and had an SSD as well. Very similar environments.
I tried optimising it and resubmitted it. This time they said they ran it on an Ubuntu machine and got 11 seconds, while the code ran for 100 seconds on the Windows 10 still.
Another peculiar thing was that when I tried profiling it with the Profile module, it went on forever, had to terminate after 450 seconds. I moved to cProfiler and it recorded it for 7 seconds.
EDIT: The exact formulation of the problem is
Write a console program to merge the files provided in a timely and
efficient manner. File paths should be supplied as arguments so that
the program can be evaluated on different data sets. The merged file
should be saved as CSV; use the id column as the unique key for
merging; the program should do any necessary data cleaning and error
checking.
Feel free to use any language you’re comfortable with – only
restriction is no external libraries as this defeats the purpose of
the test. If the language provides CSV parsing libraries (like
Python), please avoid using them as well as this is a part of the
test.
Without further ado here's the code:
#!/usr/bin/python3
import sys
from multiprocessing import Pool
HEADERS = ['id']
def csv_tuple_quotes_valid(a_tuple):
"""
checks if a quotes in each attribute of a entry (i.e. a tuple) agree with the csv format
returns True or False
"""
for attribute in a_tuple:
in_quotes = False
attr_len = len(attribute)
skip_next = False
for i in range(0, attr_len):
if not skip_next and attribute[i] == '\"':
if i < attr_len - 1 and attribute[i + 1] == '\"':
skip_next = True
continue
elif i == 0 or i == attr_len - 1:
in_quotes = not in_quotes
else:
return False
else:
skip_next = False
if in_quotes:
return False
return True
def check_and_parse_potential_tuple(to_parse):
"""
receives a string and returns an array of the attributes of the csv line
if the string was not a valid csv line, then returns False
"""
a_tuple = []
attribute_start_index = 0
to_parse_len = len(to_parse)
in_quotes = False
i = 0
#iterate through the string (line from the csv)
while i < to_parse_len:
current_char = to_parse[i]
#this works the following way: if we meet a quote ("), it must be in one
#of five cases: "" | ", | ," | "\0 | (start_of_string)"
#in case we are inside a quoted attribute (i.e. "123"), then commas are ignored
#the following code also extracts the tuples' attributes
if current_char == '\"':
if i == 0 or (to_parse[i - 1] == ',' and not in_quotes): # (start_of_string)" and ," case
#not including the quote in the next attr
attribute_start_index = i + 1
#starting a quoted attr
in_quotes = True
elif i + 1 < to_parse_len:
if to_parse[i + 1] == '\"': # "" case
i += 1 #skip the next " because it is part of a ""
elif to_parse[i + 1] == ',' and in_quotes: # ", case
a_tuple.append(to_parse[attribute_start_index:i].strip())
#not including the quote and comma in the next attr
attribute_start_index = i + 2
in_quotes = False #the quoted attr has ended
#skip the next comma - we know what it is for
i += 1
else:
#since we cannot have a random " in the middle of an attr
return False
elif i == to_parse_len - 1: # "\0 case
a_tuple.append(to_parse[attribute_start_index:i].strip())
#reached end of line, so no more attr's to extract
attribute_start_index = to_parse_len
in_quotes = False
else:
return False
elif current_char == ',':
if not in_quotes:
a_tuple.append(to_parse[attribute_start_index:i].strip())
attribute_start_index = i + 1
i += 1
#in case the last attr was left empty or unquoted
if attribute_start_index < to_parse_len or (not in_quotes and to_parse[-1] == ','):
a_tuple.append(to_parse[attribute_start_index:])
#line ended while parsing; i.e. a quote was openned but not closed
if in_quotes:
return False
return a_tuple
def parse_tuple(to_parse, no_of_headers):
"""
parses a string and returns an array with no_of_headers number of headers
raises an error if the string was not a valid CSV line
"""
#get rid of the newline at the end of every line
to_parse = to_parse.strip()
# return to_parse.split(',') #if we assume the data is in a valid format
#the following checking of the format of the data increases the execution
#time by a factor of 2; if the data is know to be valid, uncomment 3 lines above here
#if there are more commas than fields, then we must take into consideration
#how the quotes parse and then extract the attributes
if to_parse.count(',') + 1 > no_of_headers:
result = check_and_parse_potential_tuple(to_parse)
if result:
a_tuple = result
else:
raise TypeError('Error while parsing CSV line %s. The quotes do not parse' % to_parse)
else:
a_tuple = to_parse.split(',')
if not csv_tuple_quotes_valid(a_tuple):
raise TypeError('Error while parsing CSV line %s. The quotes do not parse' % to_parse)
#if the format is correct but more data fields were provided
#the following works faster than an if statement that checks the length of a_tuple
try:
a_tuple[no_of_headers - 1]
except IndexError:
raise TypeError('Error while parsing CSV line %s. Unknown reason' % to_parse)
#this replaces the use my own hashtables to store the duplicated values for the attributes
for i in range(1, no_of_headers):
a_tuple[i] = sys.intern(a_tuple[i])
return a_tuple
def read_file(path, file_number):
"""
reads the csv file and returns (dict, int)
the dict is the mapping of id's to attributes
the integer is the number of attributes (headers) for the csv file
"""
global HEADERS
try:
file = open(path, 'r');
except FileNotFoundError as e:
print("error in %s:\n%s\nexiting...")
exit(1)
main_table = {}
headers = file.readline().strip().split(',')
no_of_headers = len(headers)
HEADERS.extend(headers[1:]) #keep the headers from the file
lines = file.readlines()
file.close()
args = []
for line in lines:
args.append((line, no_of_headers))
#pool is a pool of worker processes parsing the lines in parallel
with Pool() as workers:
try:
all_tuples = workers.starmap(parse_tuple, args, 1000)
except TypeError as e:
print('Error in file %s:\n%s\nexiting thread...' % (path, e.args))
exit(1)
for a_tuple in all_tuples:
#add quotes to key if needed
key = a_tuple[0] if a_tuple[0][0] == '\"' else ('\"%s\"' % a_tuple[0])
main_table[key] = a_tuple[1:]
return (main_table, no_of_headers)
def merge_files():
"""
produces a file called merged.csv
"""
global HEADERS
no_of_files = len(sys.argv) - 1
processed_files = [None] * no_of_files
for i in range(0, no_of_files):
processed_files[i] = read_file(sys.argv[i + 1], i)
out_file = open('merged.csv', 'w+')
merged_str = ','.join(HEADERS)
all_keys = {}
#this is to ensure that we include all keys in the final file.
#even those that are missing from some files and present in others
for processed_file in processed_files:
all_keys.update(processed_file[0])
for key in all_keys:
merged_str += '\n%s' % key
for i in range(0, no_of_files):
(main_table, no_of_headers) = processed_files[i]
try:
for attr in main_table[key]:
merged_str += ',%s' % attr
except KeyError:
print('NOTE: no values found for id %s in file \"%s\"' % (key, sys.argv[i + 1]))
merged_str += ',' * (no_of_headers - 1)
out_file.write(merged_str)
out_file.close()
if __name__ == '__main__':
# merge_files()
import cProfile
cProfile.run('merge_files()')
# import time
# start = time.time()
# print(time.time() - start);
Here is the profiler report I got on my Windows.
EDIT: The rest of the csv data provided is here. Pastebin was taking too long to process the files, so...
It might not be the best code and I know that, but my question is what slows down Windows so much that doesn't slow down an Ubuntu? The merge_files() function takes the longest, with 94 seconds just for itself, not including the calls to other functions. And there doesn't seem to be anything too obvious to me for why it is so slow.
Thanks
EDIT: Note: We both used the same dataset to run the code with.

It turns out that Windows and Linux handle very long strings differently. When I moved the out_file.write(merged_str) inside the outer for loop (for key in all_keys:) and stopped appending to merged_str, it ran for 11 seconds as expected. I don't have enough knowledge on either of the OS's memory management systems to be able to give a prediction on why it is so different.
But I would say that the way that the second one (the Windows one) is the more fail-safe method because it is unreasonable to keep a 30 MB string in memory. It just turns out that Linux sees that and doesn't always try to keep the string in cache, or to rebuild it every time.
Funny enough, initially I did run it a few times on my Linux machine with these same writing strategies, and the one with the large string seemed to go faster, so I stuck with it. I guess you never know.
Here's the modified code
for key in all_keys:
merged_str = '%s' % key
for i in range(0, no_of_files):
(main_table, no_of_headers) = processed_files[i]
try:
for attr in main_table[key]:
merged_str += ',%s' % attr
except KeyError:
print('NOTE: no values found for id %s in file \"%s\"' % (key, sys.argv[i + 1]))
merged_str += ',' * (no_of_headers - 1)
out_file.write(merged_str + '\n')
out_file.close()

When I run your solution on Ubuntu 16.04 with the three given files, it seems to take ~8 seconds to complete. The only modification I made was to uncomment the timing code at the bottom and use it.
$ python3 dimitar_merge.py file1.csv file2.csv file3.csv
NOTE: no values found for id "aaa5d09b-684b-47d6-8829-3dbefd608b5e" in file "file2.csv"
NOTE: no values found for id "38f79a49-4357-4d5a-90a5-18052ef03882" in file "file2.csv"
NOTE: no values found for id "766590d9-4f5b-4745-885b-83894553394b" in file "file2.csv"
8.039648056030273
$ python3 dimitar_merge.py file1.csv file2.csv file3.csv
NOTE: no values found for id "38f79a49-4357-4d5a-90a5-18052ef03882" in file "file2.csv"
NOTE: no values found for id "766590d9-4f5b-4745-885b-83894553394b" in file "file2.csv"
NOTE: no values found for id "aaa5d09b-684b-47d6-8829-3dbefd608b5e" in file "file2.csv"
7.78482985496521
I rewrote my first attempt without using csv from the standard library and am now getting times of ~4.3 seconds.
$ python3 lettuce_merge.py file1.csv file2.csv file3.csv
4.332579612731934
$ python3 lettuce_merge.py file1.csv file2.csv file3.csv
4.305467367172241
$ python3 lettuce_merge.py file1.csv file2.csv file3.csv
4.27345871925354
This is my solution code (lettuce_merge.py):
from collections import defaultdict
def split_row(csv_row):
return [col.strip('"') for col in csv_row.rstrip().split(',')]
def merge_csv_files(files):
file_headers = []
merged_headers = []
for i, file in enumerate(files):
current_header = split_row(next(file))
unique_key, *current_header = current_header
if i == 0:
merged_headers.append(unique_key)
merged_headers.extend(current_header)
file_headers.append(current_header)
result = defaultdict(lambda: [''] * (len(merged_headers) - 1))
for file_header, file in zip(file_headers, files):
for line in file:
key, *values = split_row(line)
for col_name, col_value in zip(file_header, values):
result[key][merged_headers.index(col_name) - 1] = col_value
file.close()
quotes = '"{}"'.format
with open('lettuce_merged.csv', 'w') as f:
f.write(','.join(quotes(a) for a in merged_headers) + '\n')
for key, values in result.items():
f.write(','.join(quotes(b) for b in [key] + values) + '\n')
if __name__ == '__main__':
from argparse import ArgumentParser, FileType
from time import time
parser = ArgumentParser()
parser.add_argument('files', nargs='*', type=FileType('r'))
args = parser.parse_args()
start_time = time()
merge_csv_files(args.files)
print(time() - start_time)
I'm sure this code could be optimized even further but sometimes just seeing another way to solve a problem can help spark new ideas.

Compare XML documents for equality [duplicate]

I am new to programming in python,´and i have some troubles understanding the concept. I wish to compare two xml files. These xml files are quite large.
I will give an example for the type of files i wish to compare.
xmlfile1:
<xml>
<property1>
<property2>
<property3>
</property3>
</property2>
</property1>
</xml>
xml file2:
<xml>
<property1>
<property2>
<property3>
<property4>
</property4>
</property3>
</property2>
</property1>
</xml>
the property1,property2 that i have named are different from the ones that are actually in the file. There are a lot of properties within the xml file.
ANd i wish to compare the two xml files.
I am using an lxml parser to try to compare the two files and to print out the difference between them.
I do not know how to parse it and compare it automatically.
I tried reading through the lxml parser, but i couldnt understand how to use it to my problem.
Can someone please tell me how should i proceed with this problem.
Code snippets can be very useful
One more question, Am i following the right concept or i am missing something else? Please correct me of any new concepts that you knwo about

This is actually a reasonably challenging problem (due to what "difference" means often being in the eye of the beholder here, as there will be semantically "equivalent" information that you probably don't want marked as differences).
You could try using xmldiff, which is based on work in the paper Change Detection in Hierarchically Structured Information.

My approach to the problem was transforming each XML into a xml.etree.ElementTree and iterating through each of the layers.
I also included the functionality to ignore a list of attributes while doing the comparison.
The first block of code holds the class used:
import xml.etree.ElementTree as ET
import logging
class XmlTree():
def __init__(self):
self.hdlr = logging.FileHandler('xml-comparison.log')
self.formatter = logging.Formatter('%(asctime)s %(levelname)s %(message)s')
#staticmethod
def convert_string_to_tree( xmlString):
return ET.fromstring(xmlString)
def xml_compare(self, x1, x2, excludes=[]):
"""
Compares two xml etrees
:param x1: the first tree
:param x2: the second tree
:param excludes: list of string of attributes to exclude from comparison
:return:
True if both files match
"""
if x1.tag != x2.tag:
self.logger.debug('Tags do not match: %s and %s' % (x1.tag, x2.tag))
return False
for name, value in x1.attrib.items():
if not name in excludes:
if x2.attrib.get(name) != value:
self.logger.debug('Attributes do not match: %s=%r, %s=%r'
% (name, value, name, x2.attrib.get(name)))
return False
for name in x2.attrib.keys():
if not name in excludes:
if name not in x1.attrib:
self.logger.debug('x2 has an attribute x1 is missing: %s'
% name)
return False
if not self.text_compare(x1.text, x2.text):
self.logger.debug('text: %r != %r' % (x1.text, x2.text))
return False
if not self.text_compare(x1.tail, x2.tail):
self.logger.debug('tail: %r != %r' % (x1.tail, x2.tail))
return False
cl1 = x1.getchildren()
cl2 = x2.getchildren()
if len(cl1) != len(cl2):
self.logger.debug('children length differs, %i != %i'
% (len(cl1), len(cl2)))
return False
i = 0
for c1, c2 in zip(cl1, cl2):
i += 1
if not c1.tag in excludes:
if not self.xml_compare(c1, c2, excludes):
self.logger.debug('children %i do not match: %s'
% (i, c1.tag))
return False
return True
def text_compare(self, t1, t2):
"""
Compare two text strings
:param t1: text one
:param t2: text two
:return:
True if a match
"""
if not t1 and not t2:
return True
if t1 == '*' or t2 == '*':
return True
return (t1 or '').strip() == (t2 or '').strip()
The second block of code holds a couple of XML examples and their comparison:
xml1 = "<note><to>Tove</to><from>Jani</from><heading>Reminder</heading><body>Don't forget me this weekend!</body></note>"
xml2 = "<note><to>Tove</to><from>Daniel</from><heading>Reminder</heading><body>Don't forget me this weekend!</body></note>"
tree1 = XmlTree.convert_string_to_tree(xml1)
tree2 = XmlTree.convert_string_to_tree(xml2)
comparator = XmlTree()
if comparator.xml_compare(tree1, tree2, ["from"]):
print "XMLs match"
else:
print "XMLs don't match"
Most of the credit for this code must be given to syawar

If your intent is to compare the XML content and attributes, and not just compare the files byte-by-byte, there are subtleties to the question, so there is no solution that fits all cases.
You have to know something about what is important in the XML files.
The order of attributes listed in an element tag is generally not supposed to matter. That is, two XML files that differ only in the order of element attributes generally ought to be judged the same.
But that's the generic part.
The tricky part is application-dependent. For instance, it may be that white-space formatting of some elements of the file doesn't matter, and white-space might be added to the XML for legibility. And so on.
Recent versions of the ElementTree module have a function canonicalize(), which can take care of simpler cases, by putting the XML string into a canonical format.
I used this function in the unit tests of a recent project, to compare a known XML output with output from a package that sometimes changes the order of attributes. In this case, white space in the text elements was unimportant, but it was sometimes used for formatting.
import xml.etree.ElementTree as ET
def _canonicalize_XML( xml_str ):
""" Canonicalizes XML strings, so they are safe to
compare directly.
Strips white space from text content."""
if not hasattr( ET, "canonicalize" ):
raise Exception( "ElementTree missing canonicalize()" )
root = ET.fromstring( xml_str )
rootstr = ET.tostring( root )
return ET.canonicalize( rootstr, strip_text=True )
To use it, something like this:
file1 = ET.parse('file1.xml')
file2 = ET.parse('file2.xml')
canon1 = _canonicalize_XML( ET.tostring( file1.getroot() ) )
canon2 = _canonicalize_XML( ET.tostring( file2.getroot() ) )
print( canon1 == canon2 )
In my distribution, the Python 2 doesn't have canonicalize(), but Python 3 does.

Another script using xml.etree. Its awful but it works :)
#!/usr/bin/env python
import sys
import xml.etree.ElementTree as ET
from termcolor import colored
tree1 = ET.parse(sys.argv[1])
root1 = tree1.getroot()
tree2 = ET.parse(sys.argv[2])
root2 = tree2.getroot()
class Element:
def __init__(self,e):
self.name = e.tag
self.subs = {}
self.atts = {}
for child in e:
self.subs[child.tag] = Element(child)
for att in e.attrib.keys():
self.atts[att] = e.attrib[att]
print "name: %s, len(subs) = %d, len(atts) = %d" % ( self.name, len(self.subs), len(self.atts) )
def compare(self,el):
if self.name!=el.name:
raise RuntimeError("Two names are not the same")
print "----------------------------------------------------------------"
print self.name
print "----------------------------------------------------------------"
for att in self.atts.keys():
v1 = self.atts[att]
if att not in el.atts.keys():
v2 = '[NA]'
color = 'yellow'
else:
v2 = el.atts[att]
if v2==v1:
color = 'green'
else:
color = 'red'
print colored("first:\t%s = %s" % ( att, v1 ), color)
print colored("second:\t%s = %s" % ( att, v2 ), color)
for subName in self.subs.keys():
if subName not in el.subs.keys():
print colored("first:\thas got %s" % ( subName), 'purple')
print colored("second:\thasn't got %s" % ( subName), 'purple')
else:
self.subs[subName].compare( el.subs[subName] )
e1 = Element(root1)
e2 = Element(root2)
e1.compare(e2)

Method returns Unicode object, gets assigned into a NoneType. In Python

Whis scripts read from a source with lines consisting of artist names followed by a parenthesis with information about whether the artists cancelled and which country they come from.
A normal sentence may look like:
Odd Nordstoga (NO) (Cancelled), 20-08-2012, Blå
As I import the data I decode them into UTF-8 and this works fine. Uncommenting the second comment in the else block of the remove_extra() method shows that all variables are of type Unicode.
However, when a value is returned and put into another variable and the value of this is tested, the majority of the variables seems to be of NoneType.
Why does this happen? And how can it be corrected? Seems to be an error happening between the method return and assignment of the new variable.
# -*- charset: utf-8 -*-
import re
f1 = open("oya_artister_2011.csv")
artister = []
navnliste = []
PATTERN = re.compile(r"(.*)(\(.*\))")
TEST_PAT = re.compile(r"\(.*\)")
def remove_extra(tekst):
if re.search(PATTERN, tekst) > 1:
after = re.findall(PATTERN, tekst)[0][0]
#print "tekst is: %s " % tekst
#print "and of type: %s" % type(tekst)
remove_extra(after)
else:
#print "will return: ", tekst
#print "of type: %s" % type(tekst)
return tekst
for line in f1:
navn, _rest = line.split(",",1)
navn = navn.decode("utf-8")
artister.append(navn)
for artist in artister:
ny_artist = remove_extra(artist)
#print "%s" % ny_artist
print "of type: %s" % type(ny_artist)

Try
return remove_extra(after)
instead of just
remove_extra(after)

split a pdf based on outline

i would like to use pyPdf to split a pdf file based on the outline where each destination in the outline refers to a different page within the pdf.
example outline:
main --> points to page 1
sect1 --> points to page 1
sect2 --> points to page 15
sect3 --> points to page 22
it is easy within pyPdf to iterate over each page of the document or each destination in the document's outline; however, i cannot figure out how to get the page number where the destination points.
does anybody know how to find the referencing page number for each destination in the outline?

I figured it out:
class Darrell(pyPdf.PdfFileReader):
def getDestinationPageNumbers(self):
def _setup_outline_page_ids(outline, _result=None):
if _result is None:
_result = {}
for obj in outline:
if isinstance(obj, pyPdf.pdf.Destination):
_result[(id(obj), obj.title)] = obj.page.idnum
elif isinstance(obj, list):
_setup_outline_page_ids(obj, _result)
return _result
def _setup_page_id_to_num(pages=None, _result=None, _num_pages=None):
if _result is None:
_result = {}
if pages is None:
_num_pages = []
pages = self.trailer["/Root"].getObject()["/Pages"].getObject()
t = pages["/Type"]
if t == "/Pages":
for page in pages["/Kids"]:
_result[page.idnum] = len(_num_pages)
_setup_page_id_to_num(page.getObject(), _result, _num_pages)
elif t == "/Page":
_num_pages.append(1)
return _result
outline_page_ids = _setup_outline_page_ids(self.getOutlines())
page_id_to_page_numbers = _setup_page_id_to_num()
result = {}
for (_, title), page_idnum in outline_page_ids.iteritems():
result[title] = page_id_to_page_numbers.get(page_idnum, '???')
return result
pdf = Darrell(open(PATH-TO-PDF, 'rb'))
template = '%-5s %s'
print template % ('page', 'title')
for p,t in sorted([(v,k) for k,v in pdf.getDestinationPageNumbers().iteritems()]):
print template % (p+1,t)

This is just what I was looking for. Darrell's additions to PdfFileReader should be part of PyPDF2.
I wrote a little recipe that uses PyPDF2 and sejda-console to split a PDF by bookmarks. In my case there are several Level 1 sections that I want to keep together. This script allows me to do that and give the resulting files meaningful names.
import operator
import os
import subprocess
import sys
import time
import PyPDF2 as pyPdf
# need to have sejda-console installed
# change this to point to your installation
sejda = 'C:\\sejda-console-1.0.0.M2\\bin\\sejda-console.bat'
class Darrell(pyPdf.PdfFileReader):
...
if __name__ == '__main__':
t0= time.time()
# get the name of the file to split as a command line arg
pdfname = sys.argv[1]
# open up the pdf
pdf = Darrell(open(pdfname, 'rb'))
# build list of (pagenumbers, newFileNames)
splitlist = [(1,'FrontMatter')] # Customize name of first section
template = '%-5s %s'
print template % ('Page', 'Title')
print '-'*72
for t,p in sorted(pdf.getDestinationPageNumbers().iteritems(),
key=operator.itemgetter(1)):
# Customize this to get it to split where you want
if t.startswith('Chapter') or \
t.startswith('Preface') or \
t.startswith('References'):
print template % (p+1, t)
# this customizes how files are renamed
new = t.replace('Chapter ', 'Chapter')\
.replace(': ', '-')\
.replace(': ', '-')\
.replace(' ', '_')
splitlist.append((p+1, new))
# call sejda tools and split document
call = sejda
call += ' splitbypages'
call += ' -f "%s"'%pdfname
call += ' -o ./'
call += ' -n '
call += ' '.join([str(p) for p,t in splitlist[1:]])
print '\n', call
subprocess.call(call)
print '\nsejda-console has completed.\n\n'
# rename the split files
for p,t in splitlist:
old ='./%i_'%p + pdfname
new = './' + t + '.pdf'
print 'renaming "%s"\n to "%s"...'%(old, new),
try:
os.remove(new)
except OSError:
pass
try:
os.rename(old, new)
print' succeeded.\n'
except:
print' failed.\n'
print '\ndone. Spliting took %.2f seconds'%(time.time() - t0)

Small update to #darrell class to be able to parse UTF-8 outlines, which I post as answer because comment would be hard to read.
Problem is in pyPdf.pdf.Destination.title which may be returned in two flavors:
pyPdf.generic.TextStringObject
pyPdf.generic.ByteStringObject
so that output from _setup_outline_page_ids() function returns also two different types for title object, which fails with UnicodeDecodeError if outline title contains anything then ASCII.
I added this code to solve the problem:
if isinstance(title, pyPdf.generic.TextStringObject):
title = title.encode('utf-8')
of whole class:
class PdfOutline(pyPdf.PdfFileReader):
def getDestinationPageNumbers(self):
def _setup_outline_page_ids(outline, _result=None):
if _result is None:
_result = {}
for obj in outline:
if isinstance(obj, pyPdf.pdf.Destination):
_result[(id(obj), obj.title)] = obj.page.idnum
elif isinstance(obj, list):
_setup_outline_page_ids(obj, _result)
return _result
def _setup_page_id_to_num(pages=None, _result=None, _num_pages=None):
if _result is None:
_result = {}
if pages is None:
_num_pages = []
pages = self.trailer["/Root"].getObject()["/Pages"].getObject()
t = pages["/Type"]
if t == "/Pages":
for page in pages["/Kids"]:
_result[page.idnum] = len(_num_pages)
_setup_page_id_to_num(page.getObject(), _result, _num_pages)
elif t == "/Page":
_num_pages.append(1)
return _result
outline_page_ids = _setup_outline_page_ids(self.getOutlines())
page_id_to_page_numbers = _setup_page_id_to_num()
result = {}
for (_, title), page_idnum in outline_page_ids.iteritems():
if isinstance(title, pyPdf.generic.TextStringObject):
title = title.encode('utf-8')
result[title] = page_id_to_page_numbers.get(page_idnum, '???')
return result

Darrell's class can be modified slightly to produce a multi-level table of contents for a pdf (in the manner of pdftoc in the pdftk toolkit.)
My modification adds one more parameter to _setup_page_id_to_num, an integer "level" which defaults to 1. Each invocation increments the level. Instead of storing just the page number in the result, we store the pair of page number and level. Appropriate modifications should be applied when using the returned result.
I am using this to implement the "PDF Hacks" browser-based page-at-a-time document viewer with a sidebar table of contents which reflects LaTeX section, subsection etc bookmarks. I am working on a shared system where pdftk can not be installed but where python is available.

A solution 10 years later for newer python and PyPDF:
from PyPDF2 import PdfReader, PdfWriter
filename = "main.pdf"
with open(filename, "rb") as f:
r = PdfReader(f)
bookmarks = list(map(lambda x: (x.title, r.get_destination_page_number(x)), r.outline))
print(bookmarks)
for i, b in enumerate(bookmarks):
begin = b[1]
end = bookmarks[i+1][1] if i < len(bookmarks) - 1 else len(r.pages)
# print(len(r.pages[begin:end]))
name = b[0] + ".pdf"
print(f"{name=}: {begin=}, {end=}")
with open(name, "wb") as f:
w = PdfWriter(f)
for p in r.pages[begin:end]:
w.add_page(p)
w.write(f)

Django, custom template filters - regex problems

I'm trying to implement a WikiLink template filter in Django that queries the database model to give different responses depending on Page existence, identical to Wikipedia's red links. The filter does not raise an Error but instead doesn't do anything to the input.
WikiLink is defined as: [[ThisIsAWikiLink | This is the alt text]]
Here's a working example that does not query the database:
from django import template
from django.template.defaultfilters import stringfilter
from sites.wiki.models import Page
import re
register = template.Library()
#register.filter
#stringfilter
def wikilink(value):
return re.sub(r'\[\[ ?(.*?) ?\| ?(.*?) ?\]\]', r'\2', value)
wikilink.is_safe = True
The input (value) is a multi-line string, containing HTML and many WikiLinks.
The expected output is substituting [[ThisIsAWikiLink | This is the alt text]] with
This is the alt text
or if "ThisIsAWikiLink" doesn't exist in the database:
This is the alt text
and returning value.
Here's the non-working code (edited in response to comments/answers):
from django import template
from django.template.defaultfilters import stringfilter
from sites.wiki.models import Page
import re
register = template.Library()
#register.filter
#stringfilter
def wikilink(value):
m = re.match(r'\[\[ ?(.*?) ?\| ?(.*?) ?\]\]', value)
if(m):
page_alias = m.group(2)
page_title = m.group(3)
try:
page = Page.objects.get(alias=page_alias)
return re.sub(r'(\[\[)(.*)\|(.*)(\]\])', r'\3', value)
except Page.DoesNotExist:
return re.sub(r'(\[\[)(.*)\|(.*)(\]\])', r'\3', value)
else:
return value
wikilink.is_safe = True
What the code needs to do is:
extract all the WikiLinks in value
query the Page model to see if the page exists
substitute all the WikiLinks with normal links, styled dependent on each wikipage existence.
return the altered value
The updated question is:
What regular expression (method) can return a python List of WikiLinks, which can be altered and used to substitute the original matches (after being altered).
Edit:
I'd like to do something like this:
def wikilink(value):
regex = re.magic_method(r'\[\[ ?(.*?) ?\| ?(.*?) ?\]\]', value)
foreach wikilink in regex:
alias = wikilink.group(0)
text = wikilink.group(1)
if(alias exists in Page):
regex.sub(""+ text +"")
else:
regex.sub("<a href="+alias+" class='redlink'>"+ text +"</a>")
return value

If your string contains other text in addition to the wiki-link, your filter won't work because you are using re.match instead of re.search. re.match matches at the beginning of the string. re.search matches anywhere in the string. See matching vs. searching.
Also, your regex uses the greedy *, so it won't work if one line contains multiple wiki-links. Use *? instead to make it non-greedy:
re.search(r'\[\[(.*?)\|(.*?)\]\]', value)
Edit:
As for tips on how to fix your code, I suggest that you use re.sub with a callback. The advantages are:
It works correctly if you have multiple wiki-links in the same line.
One pass over the string is enough. You don't need a pass to find wiki-links, and another one to do the replacement.
Here is a sketch of the implmentation:
import re
WIKILINK_RE = re.compile(r'\[\[(.*?)\|(.*?)\]\]')
def wikilink(value):
def wikilink_sub_callback(match_obj):
alias = match_obj.group(1).strip()
text = match_obj.group(2).strip()
if(alias exists in Page):
class_attr = ''
else:
class_attr = ' class="redlink"'
return '<a href="%s"%s>%s</a>' % (alias, class_attr, text)
return WIKILINK_RE.sub(wikilink_sub_callback, value)

This is the type of problem that falls quickly to a small set of unit tests.
Pieces of the filter that can be tested in isolation (with a bit of code restructuring):
Determining whether or not value contains the pattern you're looking for
What string gets generated if there is a matching Page
What string gets generated is there isn't a matching Page
That would help you isolate where things are going wrong. You'll probably find that you'll need to rewire the regexps to account for optional spaces around the |.
Also, on first glance it looks like your filter is exploitable. You're claiming the result is safe, but you haven't filtered the alt text for nasties like script tags.

Code:
import re
def page_exists(alias):
if alias == 'ThisIsAWikiLink':
return True
return False
def wikilink(value):
if value == None:
return None
for alias, text in re.findall('\[\[\s*(.*?)\s*\|\s*(.*?)\s*\]\]',value):
if page_exists(alias):
value = re.sub('\[\[\s*%s\s*\|\s*%s\s*\]\]' % (alias,text), '%s' % (alias, text),value)
else:
value = re.sub('\[\[\s*%s\s*\|\s*%s\s*\]\]' % (alias,text), '%s' % (alias, text), value)
return value
Sample results:
>>> import wikilink
>>> wikilink.wikilink(None)
>>> wikilink.wikilink('')
''
>>> wikilink.wikilink('Test')
'Test'
>>> wikilink.wikilink('[[ThisIsAWikiLink | This is the alt text]]')
'This is the alt text'
>>> wikilink.wikilink('[[ThisIsABadWikiLink | This is the alt text]]')
'This is the alt text'
>>> wikilink.wikilink('[[ThisIsAWikiLink | This is the alt text]]\n[[ThisIsAWikiLink | This is another instance]]')
'This is the alt text\nThis is another instance'
>>> wikilink.wikilink('[[ThisIsAWikiLink | This is the alt text]]\n[[ThisIsAWikiLink | This is another instance]]')
General comments:
findall is the magic re function you're looking for
Change page_exists to run whatever query you want
Vulnerable to HTML injection (as mentioned by Dave W. Smith above)
Having to recompile the regex on each iteration is inefficient
Querying the database each time is inefficient
I think you'd run into performance issues pretty quickly with this approach.

This is the working code in case someone needs it:
from django import template
from django.template.defaultfilters import stringfilter
from sites.wiki.models import Page
import re
register = template.Library()
#register.filter
#stringfilter
def wikilink(value):
WIKILINK_RE = re.compile(r'\[\[ ?(.*?) ?\| ?(.*?) ?\]\]')
def wikilink_sub_callback(match_obj):
alias = match_obj.group(1).strip()
text = match_obj.group(2).strip()
class_attr = ''
try:
Page.objects.get(alias=alias)
except Page.DoesNotExist:
class_attr = ' class="redlink"'
return '<a href="%s"%s>%s</a>' % (alias, class_attr, text)
return WIKILINK_RE.sub(wikilink_sub_callback, value)
wikilink.is_safe = True
Many thanks for all the answers!