From ae6c7d0aee1ab371d8c1890375ce7de375d7e667 Mon Sep 17 00:00:00 2001 From: Geoffrey Sneddon Date: Tue, 12 Jul 2016 17:51:03 +0100 Subject: [PATCH 1/3] Fix parse.py after #257 --- parse.py | 13 ++++--------- 1 file changed, 4 insertions(+), 9 deletions(-) diff --git a/parse.py b/parse.py index 2ed8f1c2..705995b0 100755 --- a/parse.py +++ b/parse.py @@ -55,7 +55,7 @@ def parse(): tokenizer = HTMLTokenizer - p = html5parser.HTMLParser(tree=treebuilder, tokenizer=tokenizer, debug=opts.log) + p = html5parser.HTMLParser(tree=treebuilder, debug=opts.log) if opts.fragment: parseMethod = p.parseFragment @@ -96,7 +96,7 @@ def parse(): def run(parseMethod, f, encoding, scripting): try: - document = parseMethod(f, encoding=encoding, scripting=scripting) + document = parseMethod(f, override_encoding=encoding, scripting=scripting) except: document = None traceback.print_exc() @@ -117,16 +117,14 @@ def printOutput(parser, document, opts): document.writexml(sys.stdout, encoding="utf-8") elif tb == "lxml": import lxml.etree - sys.stdout.write(lxml.etree.tostring(document)) + sys.stdout.write(lxml.etree.tostring(document, encoding="unicode")) elif tb == "etree": - sys.stdout.write(utils.default_etree.tostring(document)) + sys.stdout.write(utils.default_etree.tostring(document, encoding="unicode")) elif opts.tree: if not hasattr(document, '__getitem__'): document = [document] for fragment in document: print(parser.tree.testSerializer(fragment)) - elif opts.hilite: - sys.stdout.write(document.hilite("utf-8")) elif opts.html: kwargs = {} for opt in serializer.HTMLSerializer.options: @@ -188,9 +186,6 @@ def getOptParser(): parser.add_option("", "--no-html", action="store_false", default=True, dest="html", help="Don't output html") - parser.add_option("", "--hilite", action="store_true", default=False, - dest="hilite", help="Output as formatted highlighted code.") - parser.add_option("-c", "--encoding", action="store_true", default=False, dest="encoding", help="Print character encoding used") From f232990e752f9b0c2a73800c26383c0b4634043d Mon Sep 17 00:00:00 2001 From: Geoffrey Sneddon Date: Wed, 13 Jul 2016 22:54:32 +0100 Subject: [PATCH 2/3] fixup! Fix parse.py after #257 --- parse.py | 2 -- 1 file changed, 2 deletions(-) diff --git a/parse.py b/parse.py index 705995b0..0edcca74 100755 --- a/parse.py +++ b/parse.py @@ -53,8 +53,6 @@ def parse(): treebuilder = treebuilders.getTreeBuilder(opts.treebuilder) - tokenizer = HTMLTokenizer - p = html5parser.HTMLParser(tree=treebuilder, debug=opts.log) if opts.fragment: From b94b100c0a6a0ba70717362b359f2ae6fdf8692e Mon Sep 17 00:00:00 2001 From: Geoffrey Sneddon Date: Wed, 13 Jul 2016 23:04:25 +0100 Subject: [PATCH 3/3] fixup! Fix parse.py after #257 --- parse.py | 1 - 1 file changed, 1 deletion(-) diff --git a/parse.py b/parse.py index 0edcca74..d5087fb8 100755 --- a/parse.py +++ b/parse.py @@ -9,7 +9,6 @@ from optparse import OptionParser from html5lib import html5parser -from html5lib.tokenizer import HTMLTokenizer from html5lib import treebuilders, serializer, treewalkers from html5lib import constants from html5lib import utils pFad - Phonifier reborn

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.

Note: This service is not intended for secure transactions such as banking, social media, email, or purchasing. Use at your own risk. We assume no liability whatsoever for broken pages.

Alternative Proxies:

Alternative Proxy

pFad Proxy

pFad v3 Proxy

pFad v4 Proxy