[checkmk-commits] Autorefactor: got auto_refactor.py von rh-refactoring-valuespec

Module: check_mk Branch: master Commit: e084d75d3d95590d3dbdad0e8077c69bb68bf3b3 URL: http://git.mathias-kettner.de/git/?p=check_mk.git;a=commit;h=e084d75d3d9559… Author: Roland Halbig &lt;rh(a)mathias-kettner.de&gt; Date: Thu Dec 15 12:59:10 2016 +0100 Autorefactor: got auto_refactor.py von rh-refactoring-valuespec Change-Id: I8263eb9c5594dc959afc81bde4b2f8174aec18a3 Autorefactor: Added try catch block to return statement. Change-Id: Ia710e58108614c8aa34737c4dfa140b6ec68b702 --- tests/web/auto_refactor.py | 155 +++++++++++++++++++++++++++++++-------------- 1 file changed, 106 insertions(+), 49 deletions(-) diff --git a/tests/web/auto_refactor.py b/tests/web/auto_refactor.py index f5b0063..7660e65 100644 --- a/tests/web/auto_refactor.py +++ b/tests/web/auto_refactor.py @@ -4,17 +4,59 @@ from bs4 import BeautifulSoup as bs from bs4 import NavigableString +def append_to_html(html, indent, addendum): + return "%s\n%s%s" %(html, ' ' * indent, "%s" % addendum) + + +def split_html(text): + index = 1 + while index < len(text) and text[index - 1] != "(": + index += 1 + open_braces = 1 + while index < len(text) and open_braces > 0: + if text[index] == "(": + open_braces += 1 + elif text[index] == ")": + open_braces -= 1 + index += 1 + while index < len(text) and text[index] != "\n": + index += 1 + return text[:index], text[index:] + + +def stripper(x): + x = re.sub(r' %\s+\(.*', '', x) + index = len(x) - 1 + while index >= 0 and x[index] in ['n', ' ', ')', '"', '\'']: + if x[index] == 'n' and index > 0 and x[index - 1] == '\\': + index -= 1 + index -= 1 + return x[:index+1] + + +def preprocess_tag(tag_str): + index = 1 + while index + 1< len(tag_str): + if tag_str[index - 1] == '=' and tag_str[index] not in ['\"', '\'']: + next_index = index + 1 + while next_index < len(tag_str) and re.match(r'\w', tag_str[next_index]) is not None: + next_index += 1 + tag_str = tag_str[:index] + '\"' + tag_str[index:(next_index)] + '\"' + tag_str[next_index:] + index = next_index + 1 + else: + index += 1 + return tag_str + + def eval_tag(tag_str, next_one, next_inbetween): + skip_next = False children = list(bs(tag_str, 'html5lib').body.children) if tag_str[1] == '/': addendum = "html.close_%s()" % re.sub("<|[/]|>", '' , tag_str) -# elif '%' + 's' in tag_str: -# addendum = '' - elif children and not isinstance(children[0], NavigableString): tag_name = '' attrs = '' @@ -33,7 +75,7 @@ def eval_tag(tag_str, next_one, next_inbetween): attrs += "%s=\"%s\"" % (key, val) # See if we can close the tag right away - if next_one and next_one[1] == '/' and next_one == ("</%s>" % tag_name): + if next_one and next_one == ("</%s>" % tag_name): addendum = "html.%s(%s%s)" % (tag_name, next_inbetween, ", " + attrs if attrs else '') skip_next = True elif tag_name in ['br', 'hr', 'img']: @@ -42,45 +84,48 @@ def eval_tag(tag_str, next_one, next_inbetween): addendum = "html.open_%s(%s)" % (tag_name, attrs) else: tag_name = tag_str.lstrip(' ').lstrip('<').split(' ')[0].rstrip("/").rstrip(">") - attrs = tag_str.lstrip(' ').lstrip('<').rstrip(' ').rstrip('>').lstrip(tag_name) + attrs = tag_str.lstrip(' ').lstrip('<').rstrip(' ').rstrip('>').lstrip(tag_name).lstrip(' ') + attrs = re.sub(r'class=', "class_=", attrs) + attrs = re.sub(r'id=', "id_=", attrs) + attrs = re.sub(r'type=', "type_=", attrs) + attrs = re.sub(r'for=', "for_=", attrs) addendum = "html.open_%s(%s)" % (tag_name, attrs) return addendum, skip_next -def append_to_html(html, indent, addendum): - return "%s\n%s%s" %(html, ' ' * indent, "%s" % addendum) - - -def split_html(text): - index = 1 - while index < len(text) and text[index - 1] != "(": - index += 1 - open_braces = 1 - while index < len(text) and open_braces > 0: - if text[index] == "(": - open_braces += 1 - elif text[index] == ")": - open_braces -= 1 - index += 1 - while index < len(text) and text[index] != "\n": - index += 1 - return text[:index], text[index:] - +def replace_inputs(text, inputs): + for index, input in enumerate(inputs): + text = re.sub("%" + "s", '[[[%s]]]' % index, text, 1) + return text + + +def insert_inputs(text, inputs): + if not inputs: + return text + replacements = [] + text_index = 0 + for counter, input in enumerate(inputs): + if "_(" in input and input.count("(") > input.count(")"): + input += ")" + if "[[[%s]]]" % counter in text: + index = text.index("[[[%s]]]" % counter) + if text[:(index+1)].count('\"') + text[:(index+1)].count('\'') % 2 == 0: + text = re.sub(r"[[]{3}%s[]]{3}" % counter, input, text, 1) + else: + text = re.sub(r"[[]{3}%s[]]{3}" % counter, '%' + 's' + '\" % ' + input + "\"", text, 1) + return text -def stripper(x): - x = re.sub(r' %\s+\(.*', '', x) - index = len(x) - 1 - while index >= 0 and x[index] in ['n', ' ', ')', '"', '\'']: - if x[index] == 'n' and index > 0 and x[index - 1] == '\\': - index -= 1 - index -= 1 - return x[:index+1] +def test_replace_inputs(): + text = "<tag class=\"ein %s\" id=%s>%s</tag>" + inputs = ["test", "id", "Hallo Welt!"] + print replace_inputs(text, inputs) # this function does a big chunk of the refactoring for me def replace_tags(html, indent = 0): + # I want to refactor only lines with html.write if not html.lstrip(' ').startswith('html.write('): return html @@ -93,24 +138,24 @@ def replace_tags(html, indent = 0): orig_html = html # strip all comments - html = re.sub(r'#.*', '', html) - - if len(html.split(" % ")) == 2: - html, string_input = html.split(" % ") - string_input = string_input.lstrip("(").rstrip("$").rstrip("\n").rstrip(' ').rstrip(")") - for input in string_input.split(","): - if ("%" + "s") in html: - html = re.sub("%" + "s", input, html, 1) + inputs = [] + if " % " in html or ' %\n' in html: + if " % " in html: + html, string_input = html.split(" % ") + else: + html, string_input = html.split(" %\n") + string_input = string_input.lstrip(' ').lstrip("(").rstrip("$").rstrip("\n").rstrip(' ').rstrip(")") + inputs = string_input.split(",") + html = replace_inputs(html, inputs) + ")" tags = re.findall(r'<[^<]*>', html) - if len(tags) == 0: - return orig_html + rest - inbetween = re.split(r'<[^<]*>', re.sub(r'\s*html\.write\([\'|"]?', '', html, 1)) inbetween = map(stripper, inbetween) - html = orig_html + "\n(new)" + if len(tags) == 0: + return orig_html + rest + html = '' if inbetween[0].strip(' ') not in ['', '\n']: html = append_to_html(html, indent, "html.write(%s)" % inbetween[0]) @@ -118,24 +163,34 @@ def replace_tags(html, indent = 0): counter = 0 skip_next = False while counter < len(tags): - tag_str = tags[counter] + tag_str = preprocess_tag(tags[counter]) next_one = tags[counter + 1] if counter + 1 < len(tags) else None addendum, skip_next = eval_tag(tag_str, next_one, inbetween[counter + 1]) - if not skip_next and addendum.strip(' '): - html = append_to_html(html, indent, addendum) + html = append_to_html(html, indent, addendum) counter += 1 if not skip_next and inbetween[counter].strip(' ') not in ['', '\n']: html = append_to_html(html, indent, "html.write(%s)" % inbetween[counter]) + elif skip_next and addendum.strip(' '): + counter += 1 + + html = insert_inputs(html, inputs) + for index, input in enumerate(inputs): + if "[[[%s]]]" % index in html: + html = re.sub("[[[%s]]]" % index, input, html) + try: - return html + "\n(/new)" + rest + return orig_html + "\n(new)" + html + "\n" + rest except: + print "orig_html:\n", orig_html + print "__________________________________________" print "html:\n", html print "__________________________________________" print "rest:\n", rest - return html + "\n(/new)" + rest + return orig_html + "\n(new)" + html + "\n" + rest + import re html = sys.argv[1] @@ -156,6 +211,8 @@ if html.endswith('.py'): indent = 0 while indent < len(part) and part[len(part) - 1 - indent] == ' ': indent += 1 +elif html == "test": + test_replace_inputs() else: print replace_tags(html)