1
0
mirror of https://github.com/ilri/csv-metadata-quality.git synced 2024-11-26 23:58:18 +01:00
csv-metadata-quality/csv_metadata_quality/fix.py

204 lines
5.4 KiB
Python
Raw Normal View History

import re
import pandas as pd
2019-07-28 16:47:28 +02:00
def whitespace(field):
"""Fix whitespace issues.
Return string with leading, trailing, and consecutive whitespace trimmed.
"""
# Skip fields with missing values
if pd.isna(field):
return
# Initialize an empty list to hold the cleaned values
values = list()
# Try to split multi-value field on "||" separator
2019-08-29 00:10:39 +02:00
for value in field.split("||"):
# Strip leading and trailing whitespace
value = value.strip()
# Replace excessive whitespace (>2) with one space
2019-08-29 00:10:39 +02:00
pattern = re.compile(r"\s{2,}")
match = re.findall(pattern, value)
if match:
2019-08-29 00:10:39 +02:00
print(f"Excessive whitespace: {value}")
value = re.sub(pattern, " ", value)
# Save cleaned value
values.append(value)
# Create a new field consisting of all values joined with "||"
2019-08-29 00:10:39 +02:00
new_field = "||".join(values)
return new_field
def separators(field):
"""Fix for invalid multi-value separators (ie "|")."""
# Skip fields with missing values
if pd.isna(field):
return
# Initialize an empty list to hold the cleaned values
values = list()
# Try to split multi-value field on "||" separator
2019-08-29 00:10:39 +02:00
for value in field.split("||"):
# After splitting, see if there are any remaining "|" characters
2019-08-29 00:10:39 +02:00
pattern = re.compile(r"\|")
match = re.findall(pattern, value)
if match:
2019-08-29 00:10:39 +02:00
print(f"Fixing invalid multi-value separator: {value}")
2019-08-29 00:10:39 +02:00
value = re.sub(pattern, "||", value)
# Save cleaned value
values.append(value)
# Create a new field consisting of all values joined with "||"
2019-08-29 00:10:39 +02:00
new_field = "||".join(values)
return new_field
def unnecessary_unicode(field):
"""Remove and replace unnecessary Unicode characters.
Removes unnecessary Unicode characters like:
- Zero-width space (U+200B)
- Replacement character (U+FFFD)
- No-break space (U+00A0)
Replaces unnecessary Unicode characters like:
- Soft hyphen (U+00AD) hyphen
Return string with characters removed or replaced.
"""
# Skip fields with missing values
if pd.isna(field):
return
# Check for zero-width space characters (U+200B)
2019-08-29 00:10:39 +02:00
pattern = re.compile(r"\u200B")
match = re.findall(pattern, field)
if match:
2019-08-29 00:10:39 +02:00
print(f"Removing unnecessary Unicode (U+200B): {field}")
field = re.sub(pattern, "", field)
# Check for replacement characters (U+FFFD)
2019-08-29 00:10:39 +02:00
pattern = re.compile(r"\uFFFD")
match = re.findall(pattern, field)
if match:
2019-08-29 00:10:39 +02:00
print(f"Removing unnecessary Unicode (U+FFFD): {field}")
field = re.sub(pattern, "", field)
# Check for no-break spaces (U+00A0)
2019-08-29 00:10:39 +02:00
pattern = re.compile(r"\u00A0")
match = re.findall(pattern, field)
if match:
2019-08-29 00:10:39 +02:00
print(f"Removing unnecessary Unicode (U+00A0): {field}")
field = re.sub(pattern, "", field)
# Check for soft hyphens (U+00AD), sometimes preceeded with a normal hyphen
2019-08-29 00:10:39 +02:00
pattern = re.compile(r"\u002D*?\u00AD")
match = re.findall(pattern, field)
if match:
2019-08-29 00:10:39 +02:00
print(f"Replacing unnecessary Unicode (U+00AD): {field}")
field = re.sub(pattern, "-", field)
return field
2019-07-29 17:05:03 +02:00
def duplicates(field):
"""Remove duplicate metadata values."""
# Skip fields with missing values
if pd.isna(field):
return
# Try to split multi-value field on "||" separator
2019-08-29 00:10:39 +02:00
values = field.split("||")
2019-07-29 17:05:03 +02:00
# Initialize an empty list to hold the de-duplicated values
new_values = list()
# Iterate over all values
for value in values:
# Check if each value exists in our list of values already
if value not in new_values:
new_values.append(value)
else:
2019-08-29 00:10:39 +02:00
print(f"Dropping duplicate value: {value}")
2019-07-29 17:05:03 +02:00
# Create a new field consisting of all values joined with "||"
2019-08-29 00:10:39 +02:00
new_field = "||".join(new_values)
2019-07-29 17:05:03 +02:00
return new_field
def newlines(field):
"""Fix newlines.
Single metadata values should not span multiple lines because this is not
rendered properly in DSpace's XMLUI and even causes issues during import.
Implementation note: this currently only detects Unix line feeds (0x0a).
This is essentially when a user presses "Enter" to move to the next line.
Other newlines like the Windows carriage return are already handled with
the string stipping performed in the whitespace fixes.
Confusingly, in Vim '\n' matches a line feed when searching, but you must
use '\r' to *insert* a line feed, ie in a search and replace expression.
Return string with newlines removed.
"""
# Skip fields with missing values
if pd.isna(field):
return
# Check for Unix line feed (LF)
2019-08-29 00:10:39 +02:00
match = re.findall(r"\n", field)
if match:
2019-08-29 00:10:39 +02:00
print(f"Removing newline: {field}")
field = field.replace("\n", "")
return field
def comma_space(field, field_name):
"""Fix occurrences of commas missing a trailing space, for example:
Orth,Alan S.
This is a very common mistake in author and citation fields.
Return string with a space added.
"""
# Skip fields with missing values
if pd.isna(field):
return
# Check for comma followed by a word character
2019-08-29 00:10:39 +02:00
match = re.findall(r",\w", field)
if match:
2019-08-29 00:10:39 +02:00
print(f"Adding space after comma ({field_name}): {field}")
field = re.sub(r",(\w)", r", \1", field)
return field