def test_import(self):
if dataent.db.exists("Blog Category", "test-category"):
dataent.delete_doc("Blog Category", "test-category")
exporter.export_data("Blog Category", all_doctypes=True, template=True)
content = read_csv_content(dataent.response.result)
content.append(["", "test-category", "Test Cateogry"])
importer.upload(content)
self.assertTrue(
dataent.db.get_value("Blog Category", "test-category", "title"),
"Test Category")
# export with data
exporter.export_data("Blog Category",
all_doctypes=True,
template=True,
with_data=True)
content = read_csv_content(dataent.response.result)
# overwrite
content[-1][3] = "New Title"
importer.upload(content, overwrite=True)
self.assertTrue(
dataent.db.get_value("Blog Category", "test-category", "title"),
"New Title")
def import_csv(context,
path,
only_insert=False,
submit_after_import=False,
ignore_encoding_errors=False,
no_email=True):
"Import CSV using data import"
from dataent.core.doctype.data_import import importer
from dataent.utils.csvutils import read_csv_content
site = get_site(context)
if not os.path.exists(path):
path = os.path.join('..', path)
if not os.path.exists(path):
print('Invalid path {0}'.format(path))
sys.exit(1)
with open(path, 'r') as csvfile:
content = read_csv_content(csvfile.read())
dataent.init(site=site)
dataent.connect()
try:
importer.upload(content,
submit_after_import=submit_after_import,
no_email=no_email,
ignore_encoding_errors=ignore_encoding_errors,
overwrite=not only_insert,
via_console=True)
dataent.db.commit()
except Exception:
print(dataent.get_traceback())
dataent.destroy()
def get_transaction_entries(filename, headers):
header_index = {}
rows, transactions = [], []
if (filename.lower().endswith("xlsx")):
from dataent.utils.xlsxutils import read_xlsx_file_from_attached_file
rows = read_xlsx_file_from_attached_file(file_id=filename)
elif (filename.lower().endswith("csv")):
from dataent.utils.file_manager import get_file_path
from dataent.utils.csvutils import read_csv_content
filepath = get_file_path(filename)
with open(filepath,'rb') as csvfile:
rows = read_csv_content(csvfile.read())
elif (filename.lower().endswith("xls")):
rows = get_rows_from_xls_file(filename)
else:
dataent.throw(_("Only .csv and .xlsx files are supported currently"))
stmt_headers = headers.values()
for row in rows:
if len(row) == 0 or row[0] == None or not row[0]: continue
#print("Processing row {0}".format(row))
if header_index:
transaction = get_transaction_info(stmt_headers, header_index, row)
transactions.append(transaction)
elif is_headers_present(stmt_headers, row):
header_index = get_header_index(stmt_headers, row)
return transactions
def test_export_with_data(self):
exporter.export_data("User",
all_doctypes=True,
template=True,
with_data=True)
content = read_csv_content(dataent.response.result)
self.assertTrue(content[1][1], "User")
self.assertTrue(
'"Administrator"' in [c[1] for c in content if len(c) > 1])
def test_import_only_children(self):
user_email = "*****@*****.**"
if dataent.db.exists("User", user_email):
dataent.delete_doc("User", user_email, force=True)
dataent.get_doc({
"doctype": "User",
"email": user_email,
"first_name": "Test Import UserRole"
}).insert()
exporter.export_data("Has Role",
"User",
all_doctypes=True,
template=True)
content = read_csv_content(dataent.response.result)
content.append(["", "*****@*****.**", "Blogger"])
importer.upload(content)
user = dataent.get_doc("User", user_email)
self.assertTrue(
dataent.db.get_value("Has Role",
filters={
"role": "Blogger",
"parent": user_email,
"parenttype": "User"
}))
self.assertTrue(user.get("roles")[0].role, "Blogger")
# overwrite
exporter.export_data("Has Role",
"User",
all_doctypes=True,
template=True)
content = read_csv_content(dataent.response.result)
content.append(
["", "*****@*****.**", "Website Manager"])
importer.upload(content, overwrite=True)
user = dataent.get_doc("User", user_email)
self.assertEqual(len(user.get("roles")), 1)
self.assertTrue(user.get("roles")[0].role, "Website Manager")
def test_export_with_all_doctypes(self):
exporter.export_data("User",
all_doctypes="Yes",
template=True,
with_data=True)
content = read_csv_content(dataent.response.result)
self.assertTrue(content[1][1], "User")
self.assertTrue(
'"Administrator"' in [c[1] for c in content if len(c) > 1])
self.assertEqual(content[13][0], "DocType:")
self.assertEqual(content[13][1], "User")
self.assertTrue("Has Role" in content[13])
def test_import_with_children(self): #pylint: disable=R0201
if dataent.db.exists("Event", "EV00001"):
dataent.delete_doc("Event", "EV00001")
exporter.export_data("Event", all_doctypes="Yes", template=True)
content = read_csv_content(dataent.response.result)
content.append([None] * len(content[-2]))
content[-1][1] = "__Test Event with children"
content[-1][2] = "Private"
content[-1][3] = "2014-01-01 10:00:00.000000"
importer.upload(content)
dataent.get_doc("Event", {"subject": "__Test Event with children"})
def import_file_by_path(path,
ignore_links=False,
overwrite=False,
submit=False,
pre_process=None,
no_email=True):
from dataent.utils.csvutils import read_csv_content
print("Importing " + path)
with open(path, "r") as infile:
upload(rows=read_csv_content(infile.read()),
ignore_links=ignore_links,
no_email=no_email,
overwrite=overwrite,
submit_after_import=submit,
pre_process=pre_process)
def get_csv_contents(files_path):
csv_content = {}
for filepath in files_path:
fname = os.path.basename(filepath)
for file_type in [
"account.account.template", "account.account.type",
"account.chart.template"
]:
if fname.startswith(file_type) and fname.endswith(".csv"):
with open(filepath, "r") as csvfile:
try:
csv_content.setdefault(file_type, [])\
.append(read_csv_content(csvfile.read()))
except Exception as e:
continue
return csv_content
def _bulk_rename(context, doctype, path):
"Rename multiple records via CSV file"
from dataent.model.rename_doc import bulk_rename
from dataent.utils.csvutils import read_csv_content
site = get_site(context)
with open(path, 'r') as csvfile:
rows = read_csv_content(csvfile.read())
dataent.init(site=site)
dataent.connect()
bulk_rename(doctype, rows, via_console=True)
dataent.destroy()
def upload_bank_statement():
if getattr(dataent, "uploaded_file", None):
with open(dataent.uploaded_file, "rb") as upfile:
fcontent = upfile.read()
else:
from dataent.utils.file_manager import get_uploaded_content
fname, fcontent = get_uploaded_content()
if dataent.safe_encode(fname).lower().endswith("csv".encode('utf-8')):
from dataent.utils.csvutils import read_csv_content
rows = read_csv_content(fcontent, False)
elif dataent.safe_encode(fname).lower().endswith("xlsx".encode('utf-8')):
from dataent.utils.xlsxutils import read_xlsx_file_from_attached_file
rows = read_xlsx_file_from_attached_file(fcontent=fcontent)
columns = rows[0]
rows.pop(0)
data = rows
return {"columns": columns, "data": data}
def test_export(self):
exporter.export_data("User", all_doctypes=True, template=True)
content = read_csv_content(dataent.response.result)
self.assertTrue(content[1][1], "User")
def upload(rows=None,
submit_after_import=None,
ignore_encoding_errors=False,
no_email=True,
overwrite=None,
update_only=None,
ignore_links=False,
pre_process=None,
via_console=False,
from_data_import="No",
skip_errors=True,
data_import_doc=None,
validate_template=False,
user=None):
"""upload data"""
# for translations
if user:
dataent.cache().hdel("lang", user)
dataent.set_user_lang(user)
if data_import_doc and isinstance(data_import_doc, string_types):
data_import_doc = dataent.get_doc("Data Import", data_import_doc)
if data_import_doc and from_data_import == "Yes":
no_email = data_import_doc.no_email
ignore_encoding_errors = data_import_doc.ignore_encoding_errors
update_only = data_import_doc.only_update
submit_after_import = data_import_doc.submit_after_import
overwrite = data_import_doc.overwrite
skip_errors = data_import_doc.skip_errors
else:
# extra input params
params = json.loads(dataent.form_dict.get("params") or '{}')
if params.get("submit_after_import"):
submit_after_import = True
if params.get("ignore_encoding_errors"):
ignore_encoding_errors = True
if not params.get("no_email"):
no_email = False
if params.get('update_only'):
update_only = True
if params.get('from_data_import'):
from_data_import = params.get('from_data_import')
if not params.get('skip_errors'):
skip_errors = params.get('skip_errors')
dataent.flags.in_import = True
dataent.flags.mute_emails = no_email
def get_data_keys_definition():
return get_data_keys()
def bad_template():
dataent.throw(
_("Please do not change the rows above {0}").format(
get_data_keys_definition().data_separator))
def check_data_length():
if not data:
dataent.throw(
_("No data found in the file. Please reattach the new file with data."
))
def get_start_row():
for i, row in enumerate(rows):
if row and row[0] == get_data_keys_definition().data_separator:
return i + 1
bad_template()
def get_header_row(key):
return get_header_row_and_idx(key)[0]
def get_header_row_and_idx(key):
for i, row in enumerate(header):
if row and row[0] == key:
return row, i
return [], -1
def filter_empty_columns(columns):
empty_cols = list(filter(lambda x: x in ("", None), columns))
if empty_cols:
if columns[-1 * len(empty_cols):] == empty_cols:
# filter empty columns if they exist at the end
columns = columns[:-1 * len(empty_cols)]
else:
dataent.msgprint(_(
"Please make sure that there are no empty columns in the file."
),
raise_exception=1)
return columns
def make_column_map():
doctype_row, row_idx = get_header_row_and_idx(
get_data_keys_definition().doctype)
if row_idx == -1: # old style
return
dt = None
for i, d in enumerate(doctype_row[1:]):
if d not in ("~", "-"):
if d and doctype_row[i] in (None, '', '~', '-',
_("DocType") + ":"):
dt, parentfield = d, None
# xls format truncates the row, so it may not have more columns
if len(doctype_row) > i + 2:
parentfield = doctype_row[i + 2]
doctypes.append((dt, parentfield))
column_idx_to_fieldname[(dt, parentfield)] = {}
column_idx_to_fieldtype[(dt, parentfield)] = {}
if dt:
column_idx_to_fieldname[(dt,
parentfield)][i +
1] = rows[row_idx +
2][i + 1]
column_idx_to_fieldtype[(dt,
parentfield)][i +
1] = rows[row_idx +
4][i + 1]
def get_doc(start_idx):
if doctypes:
doc = {}
attachments = []
last_error_row_idx = None
for idx in range(start_idx, len(rows)):
last_error_row_idx = idx # pylint: disable=W0612
if (not doc) or main_doc_empty(rows[idx]):
for dt, parentfield in doctypes:
d = {}
for column_idx in column_idx_to_fieldname[(
dt, parentfield)]:
try:
fieldname = column_idx_to_fieldname[(
dt, parentfield)][column_idx]
fieldtype = column_idx_to_fieldtype[(
dt, parentfield)][column_idx]
if not fieldname or not rows[idx][column_idx]:
continue
d[fieldname] = rows[idx][column_idx]
if fieldtype in ("Int", "Check"):
d[fieldname] = cint(d[fieldname])
elif fieldtype in ("Float", "Currency",
"Percent"):
d[fieldname] = flt(d[fieldname])
elif fieldtype == "Date":
if d[fieldname] and isinstance(
d[fieldname], string_types):
d[fieldname] = getdate(
parse_date(d[fieldname]))
elif fieldtype == "Datetime":
if d[fieldname]:
if " " in d[fieldname]:
_date, _time = d[fieldname].split()
else:
_date, _time = d[
fieldname], '00:00:00'
_date = parse_date(d[fieldname])
d[fieldname] = get_datetime(_date +
" " +
_time)
else:
d[fieldname] = None
elif fieldtype in ("Image", "Attach Image",
"Attach"):
# added file to attachments list
attachments.append(d[fieldname])
elif fieldtype in ("Link", "Dynamic Link",
"Data") and d[fieldname]:
# as fields can be saved in the number format(long type) in data import template
d[fieldname] = cstr(d[fieldname])
except IndexError:
pass
# scrub quotes from name and modified
if d.get("name") and d["name"].startswith('"'):
d["name"] = d["name"][1:-1]
if sum([0 if not val else 1 for val in d.values()]):
d['doctype'] = dt
if dt == doctype:
doc.update(d)
else:
if not overwrite and doc.get("name"):
d['parent'] = doc["name"]
d['parenttype'] = doctype
d['parentfield'] = parentfield
doc.setdefault(d['parentfield'], []).append(d)
else:
break
return doc, attachments, last_error_row_idx
else:
doc = dataent._dict(zip(columns, rows[start_idx][1:]))
doc['doctype'] = doctype
return doc, [], None
# used in testing whether a row is empty or parent row or child row
# checked only 3 first columns since first two columns can be blank for example the case of
# importing the item variant where item code and item name will be blank.
def main_doc_empty(row):
if row:
for i in range(3, 0, -1):
if len(row) > i and row[i]:
return False
return True
def validate_naming(doc):
autoname = dataent.get_meta(doctype).autoname
if autoname:
if autoname[0:5] == 'field':
autoname = autoname[6:]
elif autoname == 'naming_series:':
autoname = 'naming_series'
else:
return True
if (autoname not in doc) or (not doc[autoname]):
from dataent.model.base_document import get_controller
if not hasattr(get_controller(doctype), "autoname"):
dataent.throw(
_("{0} is a mandatory field".format(autoname)))
return True
users = dataent.db.sql_list("select name from tabUser")
def prepare_for_insert(doc):
# don't block data import if user is not set
# migrating from another system
if not doc.owner in users:
doc.owner = dataent.session.user
if not doc.modified_by in users:
doc.modified_by = dataent.session.user
def is_valid_url(url):
is_valid = False
if url.startswith("/files") or url.startswith("/private/files"):
url = get_url(url)
try:
r = requests.get(url)
is_valid = True if r.status_code == 200 else False
except Exception:
pass
return is_valid
def attach_file_to_doc(doctype, docname, file_url):
# check if attachment is already available
# check if the attachement link is relative or not
if not file_url:
return
if not is_valid_url(file_url):
return
files = dataent.db.sql(
"""Select name from `tabFile` where attached_to_doctype='{doctype}' and
attached_to_name='{docname}' and (file_url='{file_url}' or thumbnail_url='{file_url}')"""
.format(doctype=doctype, docname=docname, file_url=file_url))
if files:
# file is already attached
return
save_url(file_url, None, doctype, docname, "Home/Attachments", 0)
# header
filename, file_extension = ['', '']
if not rows:
from dataent.utils.file_manager import get_file # get_file_doc
fname, fcontent = get_file(data_import_doc.import_file)
filename, file_extension = os.path.splitext(fname)
if file_extension == '.xlsx' and from_data_import == 'Yes':
from dataent.utils.xlsxutils import read_xlsx_file_from_attached_file
rows = read_xlsx_file_from_attached_file(
file_id=data_import_doc.import_file)
elif file_extension == '.csv':
from dataent.utils.csvutils import read_csv_content
rows = read_csv_content(fcontent, ignore_encoding_errors)
else:
dataent.throw(_("Unsupported File Format"))
start_row = get_start_row()
header = rows[:start_row]
data = rows[start_row:]
try:
doctype = get_header_row(get_data_keys_definition().main_table)[1]
columns = filter_empty_columns(
get_header_row(get_data_keys_definition().columns)[1:])
except:
dataent.throw(_("Cannot change header content"))
doctypes = []
column_idx_to_fieldname = {}
column_idx_to_fieldtype = {}
if skip_errors:
data_rows_with_error = header
if submit_after_import and not cint(
dataent.db.get_value("DocType", doctype, "is_submittable")):
submit_after_import = False
parenttype = get_header_row(get_data_keys_definition().parent_table)
if len(parenttype) > 1:
parenttype = parenttype[1]
# check permissions
if not dataent.permissions.can_import(parenttype or doctype):
dataent.flags.mute_emails = False
return {
"messages": [_("Not allowed to Import") + ": " + _(doctype)],
"error": True
}
# Throw expception in case of the empty data file
check_data_length()
make_column_map()
total = len(data)
if validate_template:
if total:
data_import_doc.total_rows = total
return True
if overwrite == None:
overwrite = params.get('overwrite')
# delete child rows (if parenttype)
parentfield = None
if parenttype:
parentfield = get_parent_field(doctype, parenttype)
if overwrite:
delete_child_rows(data, doctype)
import_log = []
def log(**kwargs):
if via_console:
print(
(kwargs.get("title") + kwargs.get("message")).encode('utf-8'))
else:
import_log.append(kwargs)
def as_link(doctype, name):
if via_console:
return "{0}: {1}".format(doctype, name)
else:
return getlink(doctype, name)
# publish realtime task update
def publish_progress(achieved, reload=False):
if data_import_doc:
dataent.publish_realtime(
"data_import_progress", {
"progress": str(int(100.0 * achieved / total)),
"data_import": data_import_doc.name,
"reload": reload
},
user=dataent.session.user)
error_flag = rollback_flag = False
batch_size = dataent.conf.data_import_batch_size or 1000
for batch_start in range(0, total, batch_size):
batch = data[batch_start:batch_start + batch_size]
for i, row in enumerate(batch):
# bypass empty rows
if main_doc_empty(row):
continue
row_idx = i + start_row
doc = None
publish_progress(i)
try:
doc, attachments, last_error_row_idx = get_doc(row_idx)
validate_naming(doc)
if pre_process:
pre_process(doc)
original = None
if parentfield:
parent = dataent.get_doc(parenttype, doc["parent"])
doc = parent.append(parentfield, doc)
parent.save()
else:
if overwrite and doc.get("name") and dataent.db.exists(
doctype, doc["name"]):
original = dataent.get_doc(doctype, doc["name"])
original_name = original.name
original.update(doc)
# preserve original name for case sensitivity
original.name = original_name
original.flags.ignore_links = ignore_links
original.save()
doc = original
else:
if not update_only:
doc = dataent.get_doc(doc)
prepare_for_insert(doc)
doc.flags.ignore_links = ignore_links
doc.insert()
if attachments:
# check file url and create a File document
for file_url in attachments:
attach_file_to_doc(doc.doctype, doc.name, file_url)
if submit_after_import:
doc.submit()
# log errors
if parentfield:
log(
**{
"row":
doc.idx,
"title":
'Inserted row for "%s"' %
(as_link(parenttype, doc.parent)),
"link":
get_absolute_url(parenttype, doc.parent),
"message":
'Document successfully saved',
"indicator":
"green"
})
elif submit_after_import:
log(
**{
"row":
row_idx + 1,
"title":
'Submitted row for "%s"' %
(as_link(doc.doctype, doc.name)),
"message":
"Document successfully submitted",
"link":
get_absolute_url(doc.doctype, doc.name),
"indicator":
"blue"
})
elif original:
log(
**{
"row":
row_idx + 1,
"title":
'Updated row for "%s"' %
(as_link(doc.doctype, doc.name)),
"message":
"Document successfully updated",
"link":
get_absolute_url(doc.doctype, doc.name),
"indicator":
"green"
})
elif not update_only:
log(
**{
"row":
row_idx + 1,
"title":
'Inserted row for "%s"' %
(as_link(doc.doctype, doc.name)),
"message":
"Document successfully saved",
"link":
get_absolute_url(doc.doctype, doc.name),
"indicator":
"green"
})
else:
log(
**{
"row": row_idx + 1,
"title": 'Ignored row for %s' % (row[1]),
"link": None,
"message": "Document updation ignored",
"indicator": "orange"
})
except Exception as e:
error_flag = True
# build error message
if dataent.local.message_log:
err_msg = "\n".join([
'<p class="border-bottom small">{}</p>'.format(
json.loads(msg).get('message'))
for msg in dataent.local.message_log
])
else:
err_msg = '<p class="border-bottom small">{}</p>'.format(
cstr(e))
error_trace = dataent.get_traceback()
if error_trace:
error_log_doc = dataent.log_error(error_trace)
error_link = get_absolute_url("Error Log",
error_log_doc.name)
else:
error_link = None
log(
**{
"row":
row_idx + 1,
"title":
'Error for row %s' %
(len(row) > 1 and dataent.safe_decode(row[1]) or ""),
"message":
err_msg,
"indicator":
"red",
"link":
error_link
})
# data with error to create a new file
# include the errored data in the last row as last_error_row_idx will not be updated for the last row
if skip_errors:
if last_error_row_idx == len(rows) - 1:
last_error_row_idx = len(rows)
data_rows_with_error += rows[row_idx:last_error_row_idx]
else:
rollback_flag = True
finally:
dataent.local.message_log = []
start_row += batch_size
if rollback_flag:
dataent.db.rollback()
else:
dataent.db.commit()
dataent.flags.mute_emails = False
dataent.flags.in_import = False
log_message = {"messages": import_log, "error": error_flag}
if data_import_doc:
data_import_doc.log_details = json.dumps(log_message)
import_status = None
if error_flag and data_import_doc.skip_errors and len(data) != len(
data_rows_with_error):
import_status = "Partially Successful"
# write the file with the faulty row
from dataent.utils.file_manager import save_file
file_name = 'error_' + filename + file_extension
if file_extension == '.xlsx':
from dataent.utils.xlsxutils import make_xlsx
xlsx_file = make_xlsx(data_rows_with_error,
"Data Import Template")
file_data = xlsx_file.getvalue()
else:
from dataent.utils.csvutils import to_csv
file_data = to_csv(data_rows_with_error)
error_data_file = save_file(file_name, file_data, "Data Import",
data_import_doc.name,
"Home/Attachments")
data_import_doc.error_file = error_data_file.file_url
elif error_flag:
import_status = "Failed"
else:
import_status = "Successful"
data_import_doc.import_status = import_status
data_import_doc.save()
if data_import_doc.import_status in [
"Successful", "Partially Successful"
]:
data_import_doc.submit()
publish_progress(100, True)
else:
publish_progress(0, True)
dataent.db.commit()
else:
return log_message
