File: percent-encoding.py

package info (click to toggle)
firefox-esr 140.5.0esr-1
  • links: PTS, VCS
  • area: main
  • in suites: forky, sid
  • size: 4,538,920 kB
  • sloc: cpp: 7,381,527; javascript: 6,388,905; ansic: 3,710,087; python: 1,393,776; xml: 628,165; asm: 426,916; java: 184,004; sh: 65,744; makefile: 19,302; objc: 13,059; perl: 12,912; yacc: 4,583; cs: 3,846; pascal: 3,352; lex: 1,720; ruby: 1,226; exp: 762; php: 436; lisp: 258; awk: 247; sql: 66; sed: 54; csh: 10
file content (23 lines) | stat: -rw-r--r-- 990 bytes parent folder | download | duplicates (25) 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
 
import base64
from wptserve.utils import isomorphic_decode

# Use numeric references to let the HTML parser take care of inserting the correct code points
# rather than trying to figure out the necessary bytes for each encoding. (The latter can be
# especially tricky given that Python does not implement the Encoding Standard.)
def numeric_references(input):
    output = b""
    for cp in input:
        output += b"&#x" + format(ord(cp), u"X").encode(u"utf-8") + b";"
    return output

def main(request, response):
    # Undo the "magic" space with + replacement as otherwise base64 decoding will fail.
    value = request.GET.first(b"value").replace(b" ", b"+")
    encoding = request.GET.first(b"encoding")

    output_value = numeric_references(base64.b64decode(value).decode(u"utf-8"))
    return (
        [(b"Content-Type", b"text/html;charset=" + encoding)],
        b"""<!doctype html>
<a href="https://doesnotmatter.invalid/?%s#%s">test</a>
""" % (output_value, output_value))