def test_rh_put():
rh.post(
"/metadata/1324_testcorpus_5768/" + os.environ["TESTUSER"] +
"/xml/en/html/ajokielto.xml", {
"uid": os.environ["TESTUSER"],
"testkey": "testvalue"
})
xml_metadata = rh.get(
"/metadata/1324_testcorpus_5768/" + os.environ["TESTUSER"] +
"/xml/en/html/ajokielto.xml", {"uid": os.environ["TESTUSER"]})
parser = xml_parser.XmlParser(xml_metadata.split("\n"))
metadata = parser.getMetadata()
assert metadata["testkey"] == "testvalue"
rh.put(
"/metadata/1324_testcorpus_5768/" + os.environ["TESTUSER"] +
"/xml/en/html/ajokielto.xml", {
"uid": os.environ["TESTUSER"],
"testkey": "testvalue2"
})
xml_metadata = rh.get(
"/metadata/1324_testcorpus_5768/" + os.environ["TESTUSER"] +
"/xml/en/html/ajokielto.xml", {"uid": os.environ["TESTUSER"]})
parser = xml_parser.XmlParser(xml_metadata.split("\n"))
metadata = parser.getMetadata()
assert "testvalue" in metadata["testkey"] and "testvalue2" in metadata[
"testkey"]
def test_corpus_settings(client):
login(client, os.environ["TESTUSER"], os.environ["TESTPW"])
rh.delete("/storage/1233test_corpus5678", {"uid": os.environ["TESTUSER"]})
post_data = {
"name": "1233test_corpus5678",
"group": "group",
"domain": "domain",
"origin": "origin",
"description": "description",
"pdf_reader": "pdf_reader",
"document_alignment": "document_alignment",
"sentence_alignment": "sentence_alignment",
"sentence_splitter": "sentence_splitter",
"autoalignment": "on"
}
rv = client.post('/create_corpus', data=post_data, follow_redirects=True)
rv = client.get('/corpus_settings/1233test_corpus5678')
assert b'Corpus Settings' in rv.data
xml_metadata = rh.get(
"/metadata/1233test_corpus5678/" + os.environ["TESTUSER"],
{"uid": os.environ["TESTUSER"]})
parser = xml_parser.XmlParser(xml_metadata.split("\n"))
metadata = parser.getMetadata()
field_dict = opusrepository.initialize_field_dict()
for key in post_data.keys():
if key != "name":
assert post_data[key] == metadata[field_dict[key][0]]
post_data = {
"name": "1233test_corpus5678",
"group": "group2",
"domain": "domain2",
"origin": "origin2",
"description": "description2",
"pdf_reader": "pdf_reader2",
"document_alignment": "document_alignment2",
"sentence_alignment": "sentence_alignment2",
"sentence_splitter": "sentence_splitter2",
"autoalignment": "off"
}
rv = client.post('/corpus_settings/1233test_corpus5678',
data=post_data,
follow_redirects=True)
xml_metadata = rh.get(
"/metadata/1233test_corpus5678/" + os.environ["TESTUSER"],
{"uid": os.environ["TESTUSER"]})
parser = xml_parser.XmlParser(xml_metadata.split("\n"))
metadata = parser.getMetadata()
field_dict = opusrepository.initialize_field_dict()
for key in post_data.keys():
if key != "name":
assert post_data[key] == metadata[field_dict[key][0]]
rh.delete("/storage/1233test_corpus5678", {"uid": os.environ["TESTUSER"]})
def test_getFileContent():
xml_data = [
'<letsmt-ws version="56">',
'<list path="/testcorpus/testuser/xml/en/html/ajokielto.xml">',
'<entry><?xml version="1.0" encoding="utf-8"?>',
'<letsmt version="1.0">',
'<p id="1">',
'<s id="1">If a person is sentenced to a punishment for causing a serious traffic hazard, '+\
'driving while intoxicated or driving while seriously intoxicated, the court also imposes a '+\
'driving ban for at most five years.</s>',
'</p>',
'</entry>',
'</list>',
'<status code="0" location="/storage/testcorpus/testuser/xml/en/html/ajokielto.xml"'+\
'operation="GET" type="ok"></status>',
'</letsmt-ws>'
]
parser = xml_parser.XmlParser(xml_data)
content = parser.getFileContent()
correct = '<?xml version="1.0" encoding="utf-8"?>\n'+\
'<letsmt version="1.0">\n'+\
'<p id="1">\n'+\
'<s id="1">If a person is sentenced to a punishment for causing a serious traffic '+\
'hazard, driving while intoxicated or driving while seriously intoxicated, the court '+\
'also imposes a driving ban for at most five years.</s>\n'+\
'</p>\n'
assert content == correct
def test_recursiveCorpora():
parser = xml_parser.XmlParser([])
parser.recursiveCorpora(ET.fromstring(
'<list path="">'\
'<entry directory="testcorpus2/testuser" />'\
'</list>'
))
assert len(parser.elementList) == 0
parser.recursiveCorpora(ET.fromstring(
'<list path="">'\
'<entry path="testcorpus/testuser" />'\
'<entry path="testcorpus2/testuser" />'\
'</list>'
))
assert parser.elementList[0] == "testcorpus"
assert parser.elementList[1] == "testcorpus2"
parser.recursiveCorpora(ET.fromstring(
'<list path="">'\
'<entry path="testcorpus/testuser" />'\
'<entry path="testcorpus2/testuser" />'\
'</list>'
))
assert len(parser.elementList) == 2
assert parser.elementList[0] == "testcorpus"
assert parser.elementList[1] == "testcorpus2"
def test_getGroupOwner():
xml_data = [
'<letsmt-ws version="56">', '<list path="/group/">',
'<entry id="testuser" kind="group" owner="testuser" testattr="testvalue">',
'<user>testuser</user>', '</entry>', '</list>',
'<status code="0" location="/group/testuser" operation="GET" type="ok"></status>',
'</letsmt-ws>'
]
parser = xml_parser.XmlParser(xml_data)
assert parser.getGroupOwner() == "testuser"
def test_recursiveAttrTag():
element = ET.fromstring(
'<list path="/group/">'\
'<entry id="testuser" kind="group" owner="testuser" testattr="testvalue">'\
'<user>testuser</user>'\
'</entry>'\
'</list>'
)
parser = xml_parser.XmlParser([])
parser.recursiveAttrTag(element, "entry", "testattr")
assert parser.elementString == "testvalue"
def test_groupsForUser():
xml_data = [
'<letsmt-ws version="56">', '<list path="group/testuser">',
'<entry id="testuser" kind="user info">', '<member>testuser</member>',
'<member_of>testuser,public</member_of>',
'<my_group>testuser</my_group>', '</entry>', '</list>',
'<status code="0" location="/group/testuser/" operation="GET" type="ok"></status>',
'</letsmt-ws>'
]
parser = xml_parser.XmlParser(xml_data)
groups = parser.groupsForUser()
assert groups[0] == "testuser"
assert groups[1] == "public"
def test_itemExists():
xml_data = [
'<letsmt-ws version="56">', '<list path="/testcorpus">',
'<entry kind="branch" path="/testcorpus/testuser">',
'<name>testuser</name>', '<group>public</group>',
'<owner>testuser</owner>', '</entry>', '</list>',
'<status code="0" location="/storage/testcorpus" operation="GET" type="ok"></status>',
'</letsmt-ws>'
]
parser = xml_parser.XmlParser(xml_data)
assert parser.itemExists() == True
xml_data = [
'<letsmt-ws version="56">',
'''<status code="6" location="/storage/testcorpusdoesnotexist" operation="GET"\
type="error">Cannot find/read slot 'testcorpusdoesnotexist'</status>''',
'</letsmt-ws>'
]
parser = xml_parser.XmlParser(xml_data)
assert parser.itemExists() == False
def test_getJobs():
xml_data = [
'<letsmt-ws version="56">', '<list path="jobs">',
'<entry name="job_1543831841_642742498" file="testcorpus/testuser/uploads/html.tar.gz"\
id="890" job="testcorpus/testuser/jobs/import/uploads/html.tar.gz.xml" status="RUNNING" />',
'</list>',
'<status code="0" location="job" operation="GET" type="ok" />',
'</letsmt-ws>'
]
parser = xml_parser.XmlParser(xml_data)
jobs = parser.getJobs()
assert jobs[0][0] == "testcorpus/testuser/uploads/html.tar.gz"
assert jobs[0][1] == "RUNNING"
def test_collectToList():
xml_data = [
'<letsmt-ws version="56">', '<list path="/testcorpus">',
'<entry kind="branch" path="/testcorpus/testuser">',
'<name>testuser</name>', '<group>public</group>',
'<owner>testuser</owner>', '</entry>', '</list>',
'<status code="0" location="/storage/testcorpus" operation="GET" type="ok"></status>',
'</letsmt-ws>'
]
parser = xml_parser.XmlParser(xml_data)
elementlist = parser.collectToList("group")
assert len(elementlist) == 1
assert elementlist[0] == "public"
def test_branchesForCorpus():
xml_data = [
'<letsmt-ws version="56">', '<list path="/testcorpus">',
'<entry kind="branch" path="/testcorpus/testuser">',
'<name>testuser</name>', '<group>public</group>',
'<owner>testuser</owner>', '</entry>', '</list>',
'<status code="0" location="/storage/testcorpus" operation="GET" type="ok"></status>',
'</letsmt-ws>'
]
parser = xml_parser.XmlParser(xml_data)
branches = parser.branchesForCorpus()
assert len(branches) == 1
assert branches[0] == "testuser"
def get_webview_roi(self,xml_str):
x1 = x2 = y1 = y2 = None
xml_parser = xp.XmlParser()
# xml_str = xml_parser.xml2str(xml_file)
node_list1 = xml_parser.read_from_string(xml_str)
if node_list1 is None:
log.info('Read xml file FAILURE!!!')
for n in node_list1:
if n.get('class') == 'android.webkit.WebView':
bound_value = n.get('bounds')
x1, y1, x2, y2 = xml_parser.coord_transform(bound_value)
break
return (x1, y1, x2, y2)
def test_recursiveCollect():
element = ET.fromstring(
'<list path="/testcorpus">'\
'<entry kind="branch" path="/testcorpus/testuser">'\
'<name>testuser</name>'\
'<group>public</group>'\
'<owner>testuser</owner>'\
'</entry>'\
'</list>'
)
parser = xml_parser.XmlParser([])
parser.recursiveCollect(element, "name")
assert parser.elementList[-1] == "testuser"
parser.recursiveCollect(element, "group")
assert parser.elementList[-1] == "public"
def test_getUsers():
xml_data = [
'<letsmt-ws version="56">', '<list path="/group/">',
'<entry id="public" kind="group" owner="admin">', '<user>user1</user>',
'<user>user2</user>', '<user>user3</user>', '<user>user4</user>',
'</entry>', '</list>',
'<status code="0" location="/group/public" operation="GET" type="ok"></status>',
'</letsmt-ws>'
]
parser = xml_parser.XmlParser(xml_data)
users = parser.getUsers()
assert len(users) == 4
assert users[0] == "user1"
assert users[3] == "user4"
def test_corporaForUser():
xml_data = [
'<letsmt-ws version="56">'\
'<list path="">'\
'<entry path="testcorpus2/testuser" />'\
'<entry path="testcorpus/testuser" />'\
'</list>'\
'<status code="0" location="/metadata" operation="GET" type="ok">Found 2 matching entries</status>'\
'</letsmt-ws>'
]
parser = xml_parser.XmlParser(xml_data)
corpora = parser.corporaForUser()
assert corpora[0] == "testcorpus2"
assert corpora[1] == "testcorpus"
def test_recursiveGroups():
parser = xml_parser.XmlParser([])
parser.recursiveGroups(
ET.fromstring('<entry id="testuser" kind="user info">\
<member>testuser</member>\
<my_group>testuser</my_group>\
</entry>'))
assert len(parser.elementList) == 0
parser.recursiveGroups(
ET.fromstring('<entry id="testuser" kind="user info">\
<member>testuser</member>\
<member_of>testuser,public</member_of>\
<my_group>testuser</my_group>\
</entry>'))
assert parser.elementList[0] == "testuser"
assert parser.elementList[1] == "public"
def get_filecontent():
if session:
username = session["username"]
path = request.args.get("path", "", type=str)
#content = get_from_api_and_parse("/storage"+path, {"uid": username, "action": "cat", "to": "1000"}, "getFileContent")
content = get_from_api_and_parse("/storage" + path, {
"uid": username,
"action": "cat"
}, "getFileContent")
parser = xml_parser.XmlParser(content.split("\n"))
if "/tmx/" in path:
content = parser.parseTMX()
else:
content = parser.parseDocXML()
return jsonify(content=content)
def get_from_api_and_parse(path, parameters, function):
rawXml = rh.get(path, parameters)
parser = xml_parser.XmlParser(rawXml.split("\n"))
parser_functions = {
"corporaForUser": parser.corporaForUser,
"groupsForUser": parser.groupsForUser,
"getMetadata": parser.getMetadata,
"getUsers": parser.getUsers,
"branchesForCorpus": parser.branchesForCorpus,
"navigateDirectory": parser.navigateDirectory,
"getMonolingualAndParallel": parser.getMonolingualAndParallel,
"getAlignCandidates": parser.getAlignCandidates,
"getJobs": parser.getJobs,
"getJobPath": parser.getJobPath,
"getFileContent": parser.getFileContent,
"itemExists": parser.itemExists
}
data = parser_functions[function]()
return data
def dom_check_login_ui(j_dict):
xml_dom_url = j_dict['dom_tree_url']
xml_parser = xp.XmlParser()
xml_parser.get_xml_doc(xml_dom_url)
node_list = xml_parser.get_node_list()
# assert len(node_list) != 0, "Dom node list is EMPTY!!!"
if len(node_list) == 0:
return None
detect_logon_txt_list = []
detect_user_account_txt_list = []
detect_pwd_txt_list = []
for n in node_list:
attr_class_val = n.get('class')
attr_text_val = n.get('text')
attr_rc_id_val = n.get('resource-id')
if check_in_cfg_itm_lis(attr_class_val, LOGON_WIDGET_CLASS) \
and (check_in_cfg_itm_lis(attr_text_val, LOGON_TEXT_ATTR) or check_in_cfg_itm_lis(attr_rc_id_val,
LOGON_RC_ID_ATTR)):
# print("logon button: ", n.get('bounds'))
detect_logon_txt_list.append(n)
if check_in_cfg_itm_lis(attr_class_val, USER_WIDGET_CLASS) \
and (check_in_cfg_itm_lis(attr_text_val, USER_TEXT_ATTR) or check_in_cfg_itm_lis(attr_rc_id_val,
USER_RC_ID_ATTR)):
# print("user edit box: ", n.get('bounds'))
# detect_user_account_txt_list.append(n.get('bounds'))
detect_user_account_txt_list.append(n)
if check_in_cfg_itm_lis(attr_class_val, PWD_WIDGET_CLASS) \
and (check_in_cfg_itm_lis(attr_text_val, PWD_TEXT_ATTR) or check_in_cfg_itm_lis(attr_rc_id_val,
PWD_RC_ID_ATTR)):
# print("password edit box: ", n.get('bounds'))
# detect_pwd_txt_list.append(n.get('bounds'))
detect_pwd_txt_list.append(n)
# assert len(detect_user_account_txt_list) == 1 \
# and len(detect_pwd_txt_list) == 1, "Detect ERROR !!!"
return detect_user_account_txt_list, detect_pwd_txt_list, detect_logon_txt_list
def test_getMonolingualAndParallel():
xml_data = [
'<letsmt-ws version="56">', '<list path="">',
'<entry path="testcorpus/testuser">', '<name>testuser</name>',
'<gid>public</gid>', '<import_queue></import_queue>',
'<langs>fi,en,sv</langs>', '<modif>2018-11-27 14:05:14</modif>',
'<origin></origin>', '<owner>testuser</owner>',
'<parallel-langs>fi-sv,en-fi,en-sv</parallel-langs>', '</entry>',
'</list>',
'<status code="0" location="/metadata/testcorpus/testuser" operation="GET"\
type="ok">Found matching path ID. Listing all of its properties</status>',
'</letsmt-ws>'
]
parser = xml_parser.XmlParser(xml_data)
mopa = parser.getMonolingualAndParallel()
assert len(mopa[0]) == 3
assert len(mopa[1]) == 3
for lan in [["fi", "dir"], ["sv", "dir"], ["en", "dir"]]:
assert lan in mopa[0]
for lan in [["fi-sv", "dir"], ["en-sv", "dir"], ["en-fi", "dir"]]:
assert lan in mopa[1]
def test_recursiveMetadata():
element = ET.fromstring(
'<entry path="testcorpus/testuser/uploads/html.tar.gz">'\
'<description>test</description>'\
'<direction>unknown</direction>'\
'<gid>public</gid>'\
'<owner>testuser</owner>'\
'<status>job canceled</status>'\
'</entry>'
)
parser = xml_parser.XmlParser([])
parser.recursiveMetadata(element)
correct = {
"description": "test",
"direction": "unknown",
"gid": "public",
"owner": "testuser",
"status": "job canceled"
}
for key in correct.keys():
assert parser.elementDict[key] == correct[key]
def test_getMetadata():
xml_data = [
'<letsmt-ws version="56">', '<list path="">',
'<entry path="testcorpus/testuser/uploads/html.tar.gz">',
'<description>test</description>', '<direction>unknown</direction>',
'<gid>public</gid>', '<owner>testuser</owner>',
'<status>job canceled</status>', '</entry>', '</list>',
'<status code="0" location="/metadata/testcorpus/testuser/uploads/html.tar.gz"\
operation="GET" type="ok">Found matching path ID. Listing all of its properties</status>',
'</letsmt-ws>'
]
parser = xml_parser.XmlParser(xml_data)
metadata = parser.getMetadata()
correct = {
"description": "test",
"direction": "unknown",
"gid": "public",
"owner": "testuser",
"status": "job canceled"
}
for key in correct.keys():
assert metadata[key] == correct[key]
def test_getAlignmentCandidates():
xml_data = [
'<letsmt-ws version="56">', '<list path="testcorpus/testuser/">',
'<entry path="testcorpus/testuser/xml/en/html/ajokielto.xml">',
'<align-candidates>xml/fi/html/ajokielto.xml</align-candidates>',
'<aligned_with>xml/fi/html/ajokielto.xml,xml/sv/html/ajokielto.xml</aligned_with>',
'<gid>public</gid>',
'<imported_from>uploads/html.tar.gz:html/en/ajokielto.html</imported_from>',
'<language>en</language>', '<owner>testuser</owner>',
'<parsed>ud/en/html/ajokielto.xml</parsed>',
'<resource-type>corpusfile</resource-type>', '<size>38</size>',
'<status>successfully aligned with testcorpus/testuser/xml/sv/html/ajokielto.xml</status>',
'</entry>', '</list>',
'<status code="0" location="/metadata/testcorpus/testuser" operation="GET" type="ok">Found 1 matching\
entries</status>', '</letsmt-ws>'
]
parser = xml_parser.XmlParser(xml_data)
candidates = parser.getAlignCandidates()
correct = {'en/html/ajokielto.xml': ['fi/html/ajokielto.xml']}
for key in correct.keys():
for item in correct[key]:
assert item in candidates[key]
def test_create_corpus(client):
login(client, os.environ["TESTUSER"], os.environ["TESTPW"])
rh.delete("/storage/1233test_corpus5678", {"uid": os.environ["TESTUSER"]})
rv = client.get('/create_corpus')
assert b'Create Corpus' in rv.data
post_data = {
"name": "",
"group": "group",
"domain": "domain",
"origin": "origin",
"description": "description",
"pdf_reader": "pdf_reader",
"document_alignment": "document_alignment",
"sentence_alignment": "sentence_alignment",
"sentence_splitter": "sentence_splitter",
"autoalignment": "on"
}
rv = client.post('/create_corpus', data=post_data, follow_redirects=True)
assert b'Name must be ASCII only and must not contain spaces' in rv.data
post_data["name"] = "with space"
rv = client.post('/create_corpus', data=post_data, follow_redirects=True)
assert b'Name must be ASCII only and must not contain spaces' in rv.data
post_data["name"] = "1233test_corpus5678"
rv = client.post('/create_corpus', data=post_data, follow_redirects=True)
assert b'Corpus "1233test_corpus5678" created!' in rv.data
xml_metadata = rh.get(
"/metadata/1233test_corpus5678/" + os.environ["TESTUSER"],
{"uid": os.environ["TESTUSER"]})
parser = xml_parser.XmlParser(xml_metadata.split("\n"))
metadata = parser.getMetadata()
field_dict = opusrepository.initialize_field_dict()
for key in post_data.keys():
if key != "name":
assert post_data[key] == metadata[field_dict[key][0]]
rv = client.post('/create_corpus', data=post_data, follow_redirects=True)
assert b'Corpus "1233test_corpus5678" already exists!' in rv.data
rh.delete("/storage/1233test_corpus5678", {"uid": os.environ["TESTUSER"]})
def test_navigateDirectory():
xml_data = [
'<letsmt-ws version="56">',
'<list path="/testcorpus/testuser/uploads">', '<entry kind="file">',
'<name>html.tar.gz</name>', '<commit revision="HEAD">',
'<author>testuser</author>', '<date>unknown</date>', '</commit>',
'<group>public</group>', '<owner>testuser</owner>',
'<size>14210</size>', '</entry>', '<entry kind="dir">',
'<name>html</name>', '<commit revision="HEAD">',
'<author>testuser</author>', '<date>unknown</date>', '</commit>',
'<group>public</group>', '<owner>testuser</owner>', '</entry>',
'</list>',
'<status code="0" location="/storage/testcorpus/testuser/uploads" operation="GET" type="ok"></status>',
'</letsmt-ws>'
]
parser = xml_parser.XmlParser(xml_data)
dirs = parser.navigateDirectory()
assert len(dirs) == 2
assert dirs[0][0] == "html.tar.gz"
assert dirs[0][1] == "file"
assert dirs[1][0] == "html"
assert dirs[1][1] == "dir"
def __init__(self):
# self.urls = url_manager.UrlManager()
self.downloader = xml_downloader.XmlDownloader()
self.parser = xml_parser.XmlParser()
self.outputer = xml_outputer.XmlOutputer()
import hashlib
import cv2
if __name__ == '__main__':
# # xml_doc = minidom.parse(xml_full_path)
# # # xmlDoc = minidom.loadXMLDoc("books.xml")
# #
# # x = xml_doc.getElementsByTagName("title")[0]
# # parent = x.parentNode
# #
# # document.write("Parent node: " + parent.nodeName)
# xml_dom_url = json_dict['dom_tree_url']
# shot_img_url = json_dict['screen_shot_url']
xml_parser = xp.XmlParser()
xml_dom_url1 = '/root/workspace/devicepass-ai/devicepassai/devicepassai/ImageDiffer/1_1.xml'
xml_dom_url2 = '/root/workspace/devicepass-ai/devicepassai/devicepassai/ImageDiffer/1_2.xml'
shot_img_url = '/root/workspace/devicepass-ai/devicepassai/devicepassai/ImageDiffer/1_2.jpg'
img_shot = cv2.imread(shot_img_url, cv2.IMREAD_COLOR)
node_list1 = xml_parser.get_xml_data(xml_dom_url1)
if node_list1 is None:
print " Read xml file FAILURE!!!"
leaf_node_list1 = []
for n in node_list1:
children_node = n.getchildren()
if len(children_node) == 0:
leaf_node_list1.append(n)
def main_entry(j_dict):
trace_logger()
xml_dom_url = j_dict['dom_tree_url']
shot_img_url = j_dict['screen_shot_url']
logging.info(
"--------Auto Login Verification Start----------------------------")
logging.info("dom tree file:" + xml_dom_url)
logging.info("screen shot file:" + shot_img_url)
# ocr识别默认采用英文，分别用英文和中文字符集做2次识别
is_chinese_sim = False
img_proc = image_processor.ImageProcessor(is_chinese_sim)
# 读入全图和XML数据,XML数据返回全部node节点数据
screen_img_color, height, width = img_proc.read_screen_img(shot_img_url)
if screen_img_color is None:
return [" Read image file FAILURE!!!"]
xml_parser = xp.XmlParser()
node_list = xml_parser.get_xml_data(xml_dom_url)
if node_list is None:
return [" Read xml file FAILURE!!!"]
##########################################################################################
edit_text_widgets = xml_parser.get_edit_text_widgets(
node_list, USER_WIDGET_CLASS)
'''读取xml控件类型:编辑框可能有其他类型'''
btn_widgets = xml_parser.get_btn_widget(node_list, LOGON_WIDGET_CLASS)
'''读取xml控件类型:登录框类型可能有其他类型'''
for n in btn_widgets:
logging.info("candidate logon widget1: " + str(n.get('bounds')))
estm_login_button = None
if len(btn_widgets) == 0:
'''如果没有指定类型的按钮.则以android.view.View进行ocr识别'''
btn_widgets = xml_parser.get_btn_widget(node_list,
LOGON_WIDGET_CLASS_SUPPLEMENT)
estm_login_button = img_proc.ocr_detect(True, shot_img_url, LOGON_KEY,
btn_widgets)
flag = xml_parser.is_login_ui(edit_text_widgets, btn_widgets)
if flag:
logging.info("current ui is logon dialog !")
else:
logging.info("current ui is NOT logon dialog !!!")
return None
estm_user_edits = xml_parser.xml_estm_edit(edit_text_widgets,
USER_TEXT_ATTR, USER_RC_ID_ATTR)
'''根据xml读取属性信息不可靠'''
estm_user_edit = None
is_logon_btn = False
if len(estm_user_edits) == 0:
estm_user_edit = img_proc.ocr_detect(is_logon_btn, shot_img_url,
USER_ACCOUNT_KEY,
edit_text_widgets)
elif len(estm_user_edits) > 1:
estm_user_edit = img_proc.ocr_detect(is_logon_btn, shot_img_url,
USER_ACCOUNT_KEY, estm_user_edits)
elif len(estm_user_edits) == 1:
estm_user_edit = estm_user_edits[0]
# if estm_user_edit is None:
# return None
if estm_user_edit is not None:
logging.info("FINAL detect user edit: " +
str(estm_user_edit.get('bounds')))
num_edit_text_widgets = len(edit_text_widgets)
estm_pwd_edit = None
if num_edit_text_widgets >= 1:
# if num_edit_text_widgets > 1:
estm_pwd_edits = xml_parser.xml_estm_edit(edit_text_widgets,
PWD_TEXT_ATTR,
PWD_RC_ID_ATTR)
if len(estm_pwd_edits) == 0:
estm_pwd_edit = img_proc.ocr_detect(is_logon_btn, shot_img_url,
PWD_KEY, edit_text_widgets)
elif len(estm_pwd_edits) > 1:
estm_pwd_edit = img_proc.ocr_detect(is_logon_btn, shot_img_url,
PWD_KEY, estm_pwd_edits)
elif len(estm_pwd_edits) == 1:
estm_pwd_edit = estm_pwd_edits[0]
# assert estm_pwd_edit is not None
# if estm_pwd_edit is None:
# return None
if estm_pwd_edit is not None:
logging.info("FINAL detect password edit: " +
str(estm_pwd_edit.get('bounds')))
candid_login_buttons = xml_parser.xml_estm_button(estm_user_edit,
estm_pwd_edit,
btn_widgets, LOGON_KEY)
'''xml读取属性信息不可靠'''
# assert len(candid_login_buttons) >= 1
# if len(candid_login_buttons) == 0:
# return None
for n in candid_login_buttons:
logging.info("candidate logon widget2: " + str(n.get('bounds')))
estm_login_buttons = None
edit_widget = None
if len(candid_login_buttons) == 1:
estm_login_button = candid_login_buttons[0]
elif len(candid_login_buttons) > 1 or len(candid_login_buttons) == 0:
if estm_user_edit is not None and estm_pwd_edit is not None:
edit_widget = estm_pwd_edit
if estm_user_edit is not None and estm_pwd_edit is None:
edit_widget = estm_user_edit
if estm_user_edit is None and estm_pwd_edit is not None:
edit_widget = estm_pwd_edit
if estm_user_edit is None and estm_pwd_edit is None:
return None
estm_login_button = img_proc.ocr_detect_logon_button(
True, edit_widget, shot_img_url, LOGON_KEY, candid_login_buttons)
# assert estm_login_button is not None
if estm_login_button is None:
return None
logging.info("FINAL detect logon button: " +
str(estm_login_button.get('bounds')))
return verify_result(estm_user_edit, estm_pwd_edit, estm_login_button)
def main_entry2(json_dict):
trace_logger()
xml_dom_url = json_dict['dom_tree_url']
shot_img_url = json_dict['screen_shot_url']
logging.info(
"--------Auto Login Verification Start----------------------------")
logging.info("dom tree file:" + xml_dom_url)
logging.info("screen shot file:" + shot_img_url)
# ocr识别默认采用英文，分别用英文和中文字符集做2次识别
is_chinese_sim = False
img_proc = image_processor.ImageProcessor(is_chinese_sim)
# 读入全图和XML数据,XML数据返回全部node节点数据
screen_img_color, height, width = img_proc.read_screen_img(shot_img_url)
if screen_img_color is None:
return [" Read image file FAILURE!!!"]
xml_parser = xp.XmlParser()
node_list = xml_parser.get_xml_data(xml_dom_url)
if node_list is None:
return [" Read xml file FAILURE!!!"]
'''
判断是否是登录框的标准依据2条， 一个是关键字信息，另一个是通过ocr获取控件位置关联信息
比如帐号编辑框， 如返回结果有多个，需要结合关键字和控件位置信息进行筛选。
ocr的识别准确度是关键,登录按钮可能需要裁边后再做ocr
一次遍历获得所有的候选信息，再依据位置关系进行筛选
矩形识别判断登录按钮.
针对ui是webview的特例的情况，直接调用图形检测和ocr, 当前逻辑如果根节点ocr最后没有返回正确信息，则调用图形检测和ocr模块
'''
# 全图矩形检测，找出登录按钮, 再进行ocr检测
'''
ocr检测帐号和密码编辑框， 不用xml的控件类型等不可靠的属性信息
结果将控件属性值和ocr结果进行关联
关键字信息采用NLP语义分析
'''
leaf_node_list = []
for n in node_list:
children_node = n.getchildren()
if len(children_node) == 0:
leaf_node_list.append(n)
account_candidate = []
pwd_candidate = []
login_candidate = []
for n in leaf_node_list:
coord = n.get('bounds')
if coord is None:
continue
eng_ocr_detect_failure = False
for i in range(0, 2):
if eng_ocr_detect_failure:
is_chinese_sim = True
img_proc.set_language(is_chinese_sim)
flag_account = img_proc.ocr_detect_match(False, screen_img_color,
coord, USER_ACCOUNT_KEY)
flag_pwd = img_proc.ocr_detect_match(False, screen_img_color,
coord, PWD_KEY)
flag_login = img_proc.ocr_detect_match(True, screen_img_color,
coord, LOGON_KEY)
if not flag_account and not flag_pwd and not flag_login:
eng_ocr_detect_failure = True
continue
else:
if flag_account:
account_candidate.append(n)
elif flag_pwd:
pwd_candidate.append(n)
elif flag_login:
login_candidate.append(n)
break
print 'Account Candidates:'
for n1 in account_candidate:
print n1.get('bounds')
print 'Password Candidates:'
for n2 in pwd_candidate:
print n2.get('bounds')
print 'Login Candidates:'
for n3 in login_candidate:
print n3.get('bounds')
def test_parseLine():
parser = xml_parser.XmlParser([])
result = parser.parseLine('<div test="attribute">hello</div>')
assert result == ('"div"', '"hello"', '"div"', {'test': 'attribute'})
