Skip to content
Snippets Groups Projects
Commit c18d94ee authored by Daniel Hornung's avatar Daniel Hornung
Browse files

Merge branch 'f-fix-table-converter-list' into 'dev'

FIX: `to_table` failed on lists

See merge request !99
parents a36c3319 25c6e3c0
No related branches found
No related tags found
2 merge requests!100WIP: Filling XLSX: Seems to be working.,!99FIX: `to_table` failed on lists
Pipeline #50266 passed with warnings
...@@ -16,6 +16,9 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0 ...@@ -16,6 +16,9 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
### Changed ### ### Changed ###
* `table_converter.to_table` now returns an empty DataFrame instead of raising a
ValueError when called with an empty container.
### Deprecated ### ### Deprecated ###
### Removed ### ### Removed ###
...@@ -27,6 +30,7 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0 ...@@ -27,6 +30,7 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
### Fixed ### ### Fixed ###
- Json schema exporter handles reference properties better. - Json schema exporter handles reference properties better.
- [#59](https://gitlab.com/linkahead/linkahead-advanced-user-tools/-/issues/59) `to_table` failed on lists as values.
### Security ### ### Security ###
......
...@@ -158,7 +158,7 @@ def setup_package(): ...@@ -158,7 +158,7 @@ def setup_package():
install_requires=["linkahead>=0.13.1", install_requires=["linkahead>=0.13.1",
"jsonref", "jsonref",
"jsonschema[format]>=4.4.0", "jsonschema[format]>=4.4.0",
"numpy>=1.17.3", "numpy>=1.24.0",
"openpyxl>=3.0.7", "openpyxl>=3.0.7",
"pandas>=1.2.0", "pandas>=1.2.0",
"xlrd>=2.0", "xlrd>=2.0",
......
...@@ -25,6 +25,7 @@ import re ...@@ -25,6 +25,7 @@ import re
import sys import sys
import caosdb as db import caosdb as db
import numpy as np
import pandas as pd import pandas as pd
...@@ -48,27 +49,25 @@ def generate_property_name(prop): ...@@ -48,27 +49,25 @@ def generate_property_name(prop):
def to_table(container): def to_table(container):
""" creates a table from the records in a container """ """Create a table from the records in a container."""
if len(container) == 0: if len(container) == 0:
raise ValueError("Container is empty") return pd.DataFrame()
properties = set() rts = {p.name for p in container[0].parents}
data = []
for rec in container: for rec in container:
properties.update([generate_property_name(p) if {p.name for p in rec.parents} != rts:
for p in container[0].get_properties()])
df = pd.DataFrame(columns=list(properties))
rts = set([p.name for p in container[0].parents])
for ii, rec in enumerate(container):
if set([p.name for p in rec.parents]) != rts:
raise ValueError("Parents differ") raise ValueError("Parents differ")
for p in rec.get_properties(): row_dict = {}
for prop in rec.get_properties():
df.loc[ii, generate_property_name(p)] = p.value propname = generate_property_name(prop)
row_dict[propname] = prop.value
data.append(row_dict)
result = pd.DataFrame(data=data)
return df return result
def from_table(spreadsheet, recordtype): def from_table(spreadsheet, recordtype):
......
...@@ -27,6 +27,7 @@ from tempfile import NamedTemporaryFile ...@@ -27,6 +27,7 @@ from tempfile import NamedTemporaryFile
import caosdb as db import caosdb as db
import pandas as pd import pandas as pd
from caosdb.apiutils import compare_entities from caosdb.apiutils import compare_entities
from numpy import nan
from caosadvancedtools.table_converter import (from_table, from_tsv, to_table, from caosadvancedtools.table_converter import (from_table, from_tsv, to_table,
to_tsv) to_tsv)
...@@ -42,7 +43,8 @@ class TableTest(unittest.TestCase): ...@@ -42,7 +43,8 @@ class TableTest(unittest.TestCase):
def test_empty(self): def test_empty(self):
c = db.Container() c = db.Container()
self.assertRaises(ValueError, to_table, c) df = to_table(c)
assert df.shape == (0, 0)
def test_different_props(self): def test_different_props(self):
r1 = db.Record() r1 = db.Record()
...@@ -65,6 +67,36 @@ class TableTest(unittest.TestCase): ...@@ -65,6 +67,36 @@ class TableTest(unittest.TestCase):
c.extend([r1, r2]) c.extend([r1, r2])
self.assertRaises(ValueError, to_table, c) self.assertRaises(ValueError, to_table, c)
def test_list(self):
r1 = db.Record()
r1.add_parent("no1")
r1.add_property("p1", value=1)
r1.add_property("p3", value=23)
r1.add_property("p4", value=[1])
r2 = db.Record()
r2.add_parent("no1")
r2.add_property("p1")
r2.add_property("p2", value=[20, 21])
r2.add_property("p3", value=[30, 31])
r2.add_property("p4", value=[40.0, 41.0])
r3 = db.Record()
r3.add_parent("no1")
r3.add_property("p5", value=[50, 51])
c = db.Container()
c.extend([r1, r2, r3])
result = to_table(c)
# NaN is hard to compare, so we replace it by -999
# autopep8: off
assert result.replace(to_replace=nan, value=-999).to_dict() == {
'p1': {0: 1, 1: -999, 2: -999}, # noqa: E202
'p3': {0: 23, 1: [30, 31], 2: -999}, # noqa: E202
'p4': {0: [1], 1: [40.0, 41.0], 2: -999}, # noqa: E202
'p2': {0: -999, 1: [20, 21], 2: -999}, # noqa: E202
'p5': {0: -999, 1: -999, 2: [50, 51]}
}
# autopep8: on
assert list(result.dtypes) == [float, object, object, object, object]
class FromTsvTest(unittest.TestCase): class FromTsvTest(unittest.TestCase):
def test_basic(self): def test_basic(self):
......
0% Loading or .
You are about to add 0 people to the discussion. Proceed with caution.
Please register or to comment