# coding=utf8
# the above tag defines encoding for this document and is for Python 2.x compatibility
import re
regex = r">.*?<"
test_str = (" <title>WhatsApp</title>\n"
" <meta name=\"viewport\" content=\"edited text\">\n"
" <meta name=\"google\" content=\"edited text\">\n"
" <meta name=\"format-detection\" content=\"edited text\">\n"
" <meta name=\"description\" content=\"edited text\">\n"
" <meta name=\"og:description\" content=\"edited text\">\n"
" <meta name=\"og:url\" content=\"edited text\">\n"
" <meta name=\"og:title\" content=\"edited text\">\n"
" <meta name=\"og:image\" content=\"edited text\">\n"
" <link rel=\"apple-touch-icon\" sizes=\"194x194\" href=\"/apple-touch-icon.png\" type=\"image/png\">\n"
" <link rel=\"stylesheet\" href=\"/cssm_qr.1de71c5f94dde5fbf180fe974849b354.css\">\n"
" <style>\n"
" html,\n"
" body,\n"
" #app {\n"
" height: 100%;\n"
" width: 100%;\n"
" overflow: hidden;\n"
" padding: 0;\n"
" margin: 0;\n\n"
"----------------\n"
" <div tabindex=\"-1\" class=\"_2EoyP\">\n"
" <div class=\"_3qx7_\"><button class=\"_3e4VU\"><div class=\"_1MdKA gt-MT\"><span data-testid=\"back\" data-icon=\"back\" class=\"\"><svg xmlns=\"http://www.w3.org/2000/svg\" viewBox=\"0 0 24 24\" width=\"24\" height=\"24\"><path fill=\"currentColor\" d=\"M12 4l1.4 1.4L7.8 11H20v2H7.8l5.6 5.6L12 20l-8-8 8-8z\"></path></svg></span></div><div class=\"_1MdKA w-vsN\"><span data-testid=\"search\" data-icon=\"search\" class=\"\"><svg xmlns=\"http://www.w3.org/2000/svg\" viewBox=\"0 0 24 24\" width=\"24\" height=\"24\"><path fill=\"currentColor\" d=\"M15.009 13.805h-.636l-.22-.219a5.184 5.184 0 0 0 1.256-3.386 5.207 5.207 0 1 0-5.207 5.208 5.183 5.183 0 0 0 3.385-1.255l.221.22v.635l4.004 3.999 1.194-1.195-3.997-4.007zm-4.808 0a3.605 3.605 0 1 1 0-7.21 3.605 3.605 0 0 1 0 7.21z\"></path></svg></span></div></button><span></span>\n"
" <div class=\"J3VFH\">Pesquise ou crie uma conversa</div><label class=\"_3xpD_\"><div tabindex=\"-1\" class=\"_2FVVk cBxw-\"><div class=\"_2FbwG\" style=\"visibility: visible;\"></div><div class=\"_3FRCZ copyable-text selectable-text\" data-tab=\"3\" dir=\"ltr\" contenteditable=\"true\"></div></div></label></div>\n"
"---------------\n"
" <div class=\"_2kHpK\">\n"
" <div class=\"_3dtfX\">\n"
" <div class=\"_3CneP\"><span class=\"_357i8\"><span dir=\"auto\" title=\"edited text\" class=\"_3ko75 _5h6Y_ _3Whw5\">Some text</span>\n"
" <div class=\"_3XFan\"></div>\n"
" </span>\n"
" </div>\n"
" <div class=\"m61XR\">ontem</div>\n"
" </div>\n"
" <div class=\"_1582E\">\n"
" <div class=\"_3tBW6\"><span class=\"_2iq-U\" title=\"edited text\"><span dir=\"ltr\" class=\"_3ko75 _5h6Y_ _3Whw5\">Some text</span></span>\n"
" </div>\n"
" <div class=\"m61XR\"><span></span><span></span><span></span></div>\n"
" </div>\n"
" </div>\n"
"--------------\n"
" <div class=\"_2kHpK\">\n"
" <div class=\"_3dtfX\">\n"
" <div class=\"_3CneP\">\n"
" <div class=\"_357i8\"><span dir=\"auto\" title=\"edited text\" class=\"_3ko75 _5h6Y_ _3Whw5\">Some text<img crossorigin=\"anonymous\" src=\"data:image/gif;base64,R0lGODlhAQABAIAAAAAAAP///yH5BAEAAAAALAAAAAABAAEAAAIBRAA7\" alt=\"\" draggable=\"false\" style=\"background-position: 0px -20px;\" class=\"b3 emoji wa _3Whw5\"></span>\n"
" <div class=\"_3XFan\"></div>\n"
" </div>\n"
" </div>\n"
" <div class=\"m61XR\">08:41</div>\n"
" </div>\n"
" <div class=\"_1582E\">\n\n"
" <div class=\"_3tBW6\"><span class=\"_2iq-U\" title=\"Some text that\n"
"cross lines\n"
"more than twice.\n"
"\"><div class=\"zFnXi\"><span data-testid=\"status-check\" data-icon=\"status-check\" class=\"\"><svg xmlns=\"http://www.w3.org/2000/svg\" viewBox=\"0 0 14 18\" width=\"14\" height=\"18\"><path fill=\"currentColor\" d=\"M12.502 5.035l-.57-.444a.434.434 0 0 0-.609.076l-6.39 8.198a.38.38 0 0 1-.577.039l-2.614-2.556a.435.435 0 0 0-.614.007l-.505.516a.435.435 0 0 0 .007.614l3.887 3.8a.38.38 0 0 0 .577-.039l7.483-9.602a.435.435 0 0 0-.075-.609z\"></path></svg></span></div><span dir=\"ltr\" class=\"_3ko75 _5h6Y_ _3Whw5\">Some text that\n"
"cross lines\n"
"more than twice.\n"
"</span></span>\n"
" </div>\n")
matches = re.finditer(regex, test_str, re.MULTILINE)
for matchNum, match in enumerate(matches, start=1):
print ("Match {matchNum} was found at {start}-{end}: {match}".format(matchNum = matchNum, start = match.start(), end = match.end(), match = match.group()))
for groupNum in range(0, len(match.groups())):
groupNum = groupNum + 1
print ("Group {groupNum} found at {start}-{end}: {group}".format(groupNum = groupNum, start = match.start(groupNum), end = match.end(groupNum), group = match.group(groupNum)))
# Note: for Python 2.7 compatibility, use ur"" to prefix the regex and u"" to prefix the test string and substitution.
Please keep in mind that these code samples are automatically generated and are not guaranteed to work. If you find any syntax errors, feel free to submit a bug report. For a full regex reference for Python, please visit: https://docs.python.org/3/library/re.html