Skip to content
GitLab
Explore
Sign in
Primary navigation
Search or go to…
Project
C
ckanext-odsh
Manage
Activity
Members
Labels
Plan
Issues
Issue boards
Milestones
Wiki
Code
Merge requests
Repository
Branches
Commits
Tags
Repository graph
Compare revisions
Snippets
Build
Pipelines
Jobs
Pipeline schedules
Artifacts
Deploy
Releases
Package registry
Container registry
Model registry
Operate
Environments
Terraform modules
Monitor
Incidents
Analyze
Value stream analytics
Contributor analytics
CI/CD analytics
Repository analytics
Model experiments
Help
Help
Support
GitLab documentation
Compare GitLab plans
Community forum
Contribute to GitLab
Provide feedback
Terms and privacy
Keyboard shortcuts
?
Snippets
Groups
Projects
Show more breadcrumbs
Transparenzportal
ckanext-odsh
Commits
4db5387c
Commit
4db5387c
authored
6 years ago
by
chbaeh
Browse files
Options
Downloads
Patches
Plain Diff
adjust harvesters
parent
a307ef53
No related branches found
No related tags found
No related merge requests found
Changes
2
Expand all
Show whitespace changes
Inline
Side-by-side
Showing
2 changed files
ckanext/odsh/harvesters/kielharvester.py
+32
-15
32 additions, 15 deletions
ckanext/odsh/harvesters/kielharvester.py
ckanext/odsh/tests/test_statistikNordHarvester.py
+628
-0
628 additions, 0 deletions
ckanext/odsh/tests/test_statistikNordHarvester.py
with
660 additions
and
15 deletions
ckanext/odsh/harvesters/kielharvester.py
+
32
−
15
View file @
4db5387c
...
@@ -39,23 +39,26 @@ class KielHarvester(ODSHBaseHarvester):
...
@@ -39,23 +39,26 @@ class KielHarvester(ODSHBaseHarvester):
try
:
try
:
used_identifiers
=
[]
used_identifiers
=
[]
ids
=
[]
ids
=
[]
package_ids_in_db
=
list
(
map
(
lambda
x
:
x
[
0
],
model
.
Session
.
query
(
HarvestObject
.
guid
)
\
package_ids_in_db
=
list
(
map
(
lambda
x
:
x
[
0
],
model
.
Session
.
query
(
HarvestObject
.
guid
)
.
filter
(
HarvestObject
.
current
==
True
)
\
.
filter
(
HarvestObject
.
current
==
True
)
.
filter
(
HarvestObject
.
harvest_source_id
==
harvest_job
.
source
.
id
).
all
()))
.
filter
(
HarvestObject
.
harvest_source_id
==
harvest_job
.
source
.
id
).
all
()))
log
.
info
(
"
Package IDs in DB: %s
"
%
str
(
package_ids_in_db
))
log
.
info
(
"
Package IDs in DB: %s
"
%
str
(
package_ids_in_db
))
for
dataset
in
datasets
:
for
dataset
in
datasets
:
guid
=
str
(
uuid
.
uuid3
(
uuid
.
NAMESPACE_URL
,
dataset
.
get
(
"
url
"
).
encode
(
'
ascii
'
,
'
ignore
'
)))
guid
=
str
(
uuid
.
uuid3
(
uuid
.
NAMESPACE_URL
,
dataset
.
get
(
"
url
"
).
encode
(
'
ascii
'
,
'
ignore
'
)))
if
guid
not
in
package_ids_in_db
:
if
guid
not
in
package_ids_in_db
:
obj
=
HarvestObject
(
job
=
harvest_job
,
guid
=
guid
)
obj
=
HarvestObject
(
job
=
harvest_job
,
guid
=
guid
)
obj
.
content
=
json
.
dumps
(
dataset
)
obj
.
content
=
json
.
dumps
(
dataset
)
obj
.
save
()
obj
.
save
()
log
.
info
(
"
harvest_object_id: %s, GUID: %s successfully gathered
"
%
(
str
(
obj
.
id
),
str
(
obj
.
guid
)))
log
.
info
(
"
harvest_object_id: %s, GUID: %s successfully gathered
"
%
(
str
(
obj
.
id
),
str
(
obj
.
guid
)))
used_identifiers
.
append
(
guid
)
used_identifiers
.
append
(
guid
)
ids
.
append
(
obj
.
id
)
ids
.
append
(
obj
.
id
)
except
Exception
as
e
:
except
Exception
as
e
:
self
.
_save_gather_error
(
self
.
_save_gather_error
(
'
Statistik-Nord-Harvester: Error gathering the identifiers from the source server [%s]
'
%
str
(
e
),
'
Kiel-Harvester: Error gathering the identifiers from the source server [%s]
'
%
str
(
e
),
harvest_job
)
harvest_job
)
log
.
error
(
e
)
log
.
error
(
e
)
return
None
return
None
...
@@ -68,7 +71,8 @@ class KielHarvester(ODSHBaseHarvester):
...
@@ -68,7 +71,8 @@ class KielHarvester(ODSHBaseHarvester):
return
ids
return
ids
else
:
else
:
log
.
error
(
"
No records received
"
)
log
.
error
(
"
No records received
"
)
self
.
_save_gather_error
(
"
Couldn
'
t find any metadata files
"
,
harvest_job
)
self
.
_save_gather_error
(
"
Couldn
'
t find any metadata files
"
,
harvest_job
)
return
None
return
None
@staticmethod
@staticmethod
...
@@ -79,6 +83,7 @@ class KielHarvester(ODSHBaseHarvester):
...
@@ -79,6 +83,7 @@ class KielHarvester(ODSHBaseHarvester):
return
False
return
False
def
import_stage
(
self
,
harvest_object
):
def
import_stage
(
self
,
harvest_object
):
log
.
debug
(
'
IMPORT
'
)
context
=
{
context
=
{
'
model
'
:
model
,
'
model
'
:
model
,
'
session
'
:
model
.
Session
,
'
session
'
:
model
.
Session
,
...
@@ -89,12 +94,14 @@ class KielHarvester(ODSHBaseHarvester):
...
@@ -89,12 +94,14 @@ class KielHarvester(ODSHBaseHarvester):
return
False
return
False
if
harvest_object
.
content
is
None
:
if
harvest_object
.
content
is
None
:
self
.
_save_object_error
(
'
Empty content for object %s
'
%
harvest_object
.
id
,
harvest_object
,
'
Import
'
)
self
.
_save_object_error
(
'
Empty content for object %s
'
%
harvest_object
.
id
,
harvest_object
,
'
Import
'
)
return
False
return
False
else
:
else
:
package_dict
=
json
.
loads
(
harvest_object
.
content
)
package_dict
=
json
.
loads
(
harvest_object
.
content
)
source_dataset
=
get_action
(
'
package_show
'
)(
context
.
copy
(),
{
'
id
'
:
harvest_object
.
source
.
id
})
source_dataset
=
get_action
(
'
package_show
'
)(
context
.
copy
(),
{
'
id
'
:
harvest_object
.
source
.
id
})
package_dict
[
'
owner_org
'
]
=
source_dataset
.
get
(
'
owner_org
'
)
package_dict
[
'
owner_org
'
]
=
source_dataset
.
get
(
'
owner_org
'
)
if
package_dict
[
'
type
'
]
==
'
datensatz
'
:
if
package_dict
[
'
type
'
]
==
'
datensatz
'
:
...
@@ -115,25 +122,35 @@ class KielHarvester(ODSHBaseHarvester):
...
@@ -115,25 +122,35 @@ class KielHarvester(ODSHBaseHarvester):
package_dict
[
'
groups
'
]
=
mapped_groups
package_dict
[
'
groups
'
]
=
mapped_groups
extras
=
package_dict
[
'
extras
'
]
extras
=
package_dict
[
'
extras
'
]
package_dict
[
'
extras
'
]
=
list
()
new_
extras
=
list
()
for
extra
in
extras
:
for
extra
in
extras
:
if
extra
[
'
key
'
]
in
[
'
temporal_start
'
,
'
temporal_end
'
,
'
issued
'
]:
if
extra
[
'
key
'
]
in
[
'
temporal_start
'
,
'
temporal_end
'
,
'
issued
'
]:
package_dict
[
extra
[
'
key
'
]]
=
extra
[
'
value
'
]
new_extras
.
append
(
extra
)
new_extras
.
append
(
{
'
spatial_uri
'
:
'
http://dcat-ap.de/def/politicalGeocoding/districtKey/01002
'
})
package_dict
[
'
extras
'
]
=
new_extras
package_dict
[
'
spatial_uri
'
]
=
'
http://dcat-ap.de/def/politicalGeocoding/districtKey/01002
'
log
.
debug
(
package_dict
[
'
extras
'
])
license_id
=
self
.
_get_license_id
(
package_dict
[
'
license_id
'
])
license_id
=
self
.
_get_license_id
(
package_dict
[
'
license_id
'
])
if
license_id
:
if
license_id
:
package_dict
[
'
license_id
'
]
=
license_id
package_dict
[
'
license_id
'
]
=
license_id
else
:
else
:
log
.
error
(
'
invalid license_id: %s
'
%
package_dict
[
'
license_id
'
])
log
.
error
(
'
invalid license_id: %s
'
%
self
.
_save_object_error
(
'
Invalid license_id: %s
'
%
package_dict
[
'
license_id
'
],
harvest_object
,
'
Import
'
)
package_dict
[
'
license_id
'
])
self
.
_save_object_error
(
'
Invalid license_id: %s
'
%
package_dict
[
'
license_id
'
],
harvest_object
,
'
Import
'
)
return
False
return
False
try
:
try
:
context
=
{
'
user
'
:
self
.
_get_user_name
(),
'
return_id_only
'
:
True
,
'
ignore_auth
'
:
True
}
context
=
{
'
user
'
:
self
.
_get_user_name
(
package_plugin
=
lib_plugins
.
lookup_package_plugin
(
package_dict
.
get
(
'
type
'
,
None
))
),
'
return_id_only
'
:
True
,
'
ignore_auth
'
:
True
}
package_plugin
=
lib_plugins
.
lookup_package_plugin
(
package_dict
.
get
(
'
type
'
,
None
))
package_schema
=
package_plugin
.
create_package_schema
()
package_schema
=
package_plugin
.
create_package_schema
()
context
[
'
schema
'
]
=
package_schema
context
[
'
schema
'
]
=
package_schema
log
.
debug
(
package_schema
)
self
.
_handle_current_harvest_object
(
harvest_object
,
harvest_object
.
guid
)
self
.
_handle_current_harvest_object
(
harvest_object
,
harvest_object
.
guid
)
result
=
toolkit
.
get_action
(
'
package_create
'
)(
context
,
package_dict
)
result
=
toolkit
.
get_action
(
'
package_create
'
)(
context
,
package_dict
)
return
result
return
result
...
...
This diff is collapsed.
Click to expand it.
ckanext/odsh/tests/test_statistikNordHarvester.py
0 → 100644
+
628
−
0
View file @
4db5387c
This diff is collapsed.
Click to expand it.
Preview
0%
Loading
Try again
or
attach a new file
.
Cancel
You are about to add
0
people
to the discussion. Proceed with caution.
Finish editing this message first!
Save comment
Cancel
Please
register
or
sign in
to comment