public:lta_tricks

Differences

This shows you the differences between two versions of the page.

Link to this comparison view

Both sides previous revision Previous revision
Next revision		 Previous revision
public:lta_tricks [2016-03-31 11:37] – [AstroWise Python Interface] Luciano Cerrigonepublic:lta_tricks [2023-07-17 08:51] (current) – Robbie Luijben
Line 6: Line 6:
  
   * You can use colons in numeric queries, to select ranges. This will for example give all observations and pipelines that have a SAS/Observation ID in the range from 432000 to 432190:   * You can use colons in numeric queries, to select ranges. This will for example give all observations and pipelines that have a SAS/Observation ID in the range from 432000 to 432190:
-{{ :public:lta_range_selection.png |}} 
  
-In textual entries, wildcards can be used. +{{  :public:lta_range_selection.png  }} 
-{{ :public:lta_wildcard_selection.png |}}+ 
 +In textual entries, wildcards can be used. {{  :public:lta_wildcard_selection.png  }}
  
   * You can put a list of SAS/Observation IDs in the query:   * You can put a list of SAS/Observation IDs in the query:
  
-{{ :public:lta_list_query.png |}}+{{  :public:lta_list_query.png  }}
  
 ===== Viewing data ===== ===== Viewing data =====
Line 19: Line 19:
 When you are looking at the results of a query you might see something like this: When you are looking at the results of a query you might see something like this:
  
-{{ :public:lta_observation_with_no_archived_data.png |}}+{{  :public:lta_observation_with_no_archived_data.png  }}
  
 This means that the observation is known in the LTA, it knows what data was produced, the produced data was not archived, but further processing happened on the raw data and the results of some of those pipelines were archived. If you click on the zero, you will see something like this: This means that the observation is known in the LTA, it knows what data was produced, the produced data was not archived, but further processing happened on the raw data and the results of some of those pipelines were archived. If you click on the zero, you will see something like this:
  
-{{ :public:lta_dataproduct_is_link.png |}}+{{  :public:lta_dataproduct_is_link.png  }}
  
 This allows you to navigate from a pipeline back to the original observation, or from the observation to any pipelines that have run on the raw data. This allows you to navigate from a pipeline back to the original observation, or from the observation to any pipelines that have run on the raw data.
Line 31: Line 31:
   * You can retrieve data on the Observation and Pipeline level, you don't have to select all files individually.   * You can retrieve data on the Observation and Pipeline level, you don't have to select all files individually.
  
-{{ :public:lta_observation_selection.png |}} +{{  :public:lta_observation_selection.png  }}
  
   * If you have a query with more than 1000 results, you can open the multiple pages each in a separate tab/window.   * If you have a query with more than 1000 results, you can open the multiple pages each in a separate tab/window.
  
-{{ :public:lta_page_selection.png |}} +{{  :public:lta_page_selection.png  }}
  
   * With the small triangle next to a list, you can fold or unfold the list to get a better overview.   * With the small triangle next to a list, you can fold or unfold the list to get a better overview.
 +
 == Folded entries == == Folded entries ==
-{{ :public:lta_folded.png |}} 
  
 +{{  :public:lta_folded.png  }}
  
 == Unfolded entries == == Unfolded entries ==
-{{ :public:lta_unfolded.png |}}+ 
 +{{  :public:lta_unfolded.png  }}
  
 ===== DBView ===== ===== DBView =====
  
-There is a server that gives the option to run your own queries on the database http://lofar-dbview.target.rug.nl/+There is a server that gives the option to run your own queries on the database [[https://lta-dbview.lofar.eu/|https://lta-dbview.lofar.eu/]]
  
 A useful query might be this one, that gives you all files for a certain Obs Id (SAS VIC tree ID). A useful query might be this one, that gives you all files for a certain Obs Id (SAS VIC tree ID).
-  SELECT fo.URI, dp."dataProductType", dp."dataProductIdentifier", +<code>
-   dp."processIdentifier" +
-  FROM AWOPER."DataProduct+" dp, +
-       AWOPER.FileObject fo, +
-       AWOPER."Process+" pr +
-  WHERE dp."processIdentifier" = pr."processIdentifier" +
-    AND pr."observationId" = '123456' +
-    AND fo.data_object = dp."object_id" +
-    AND dp."isValid" > 0+
  
-In this '123456' should be replaced with the Obs Id of an Observation/Pipeline you're looking for.+SELECT fo.URI, dp."dataProductType", dp."dataProductIdentifier", 
 + dp."processIdentifier" 
 +FROM AWOPER."DataProduct+" dp, 
 +     AWOPER.FileObject fo, 
 +     AWOPER."Process+" pr 
 +WHERE dp."processIdentifier" = pr."processIdentifier" 
 +  AND pr."observationId" = '123456' 
 +  AND fo.data_object = dp."object_id" 
 +  AND dp."isValid"> 0
  
-===== AstroWise Python Interface =====+</code>
  
-There is also a python interface to the LTA. With this, you can script some advanced queries. +In this '123456' should be replaced with the Obs Id of an Observation/Pipeline you're looking for. Pipelines also have an "observationId" == the SAS Id, even though that's a but confusing. To be able to run this query, you have to go to the link above, login as the right user, select the right project, and then put this query into the "Manual SQL".
-To have this working, you first need to install the [[lta:client_installation|LTA client]] in your machine. +
-Once you have installed the client, set up your user name and password. These are the same as for MoM. Remember that this is just a different interface to the LTA catalogue: you will need the same credentials as for the web interface.+
  
-After installing the LTA client, add the following to the file .awe/Environment.cfg in your home directory+**Example**  You can also modify these queries. for example if you want to also know the MD5 checksum, you can run:
  
-<file> +<code> 
-database_user : <your username> +SELECT fo.URI, fo.hash_md5, dp."dataProductType", dp."dataProductIdentifier", 
-database_password : <your password> + dp."processIdentifier" 
-</file>+FROM AWOPER."DataProduct+" dp, 
 +     AWOPER.FileObject fo, 
 +     AWOPER."Process+" pr 
 +WHERE dp."processIdentifier" = pr."processIdentifier" 
 +  AND pr."observationId" = '123456' 
 +  AND fo.data_object = dp."object_id" 
 +  AND dp."isValid"> 0
  
-then create the variable awetarget and set it to awlofar. In a bash shell, you can do so by adding the following to your .profile file:+</code>
  
-<file> +===== AstroWise Python Interface =====
-export AWETARGET=awlofar +
-</file>+
  
-Finally, your hostname may cause an error, if it does not contain a full domain. In this case, check your /etc/hosts file. You should find a line that looks like this+There is a Python client library for accessing the LTA. With this library, you can script your own queries. The installation description can be found here: [[:lta:client_installation|LTA Client installation]]. Be sure to have the latest version installed. Note that since January 2018 this library uses python3, python2 is no longer supported. 
 + 
 +Once you have installed the client, set up your user name and password. These are the same as for MoM. Remember that this is just a different interface to the LTA catalogue: you will need the same credentials as for the web interface.
  
 +After installing the LTA client, the file .awe/Environment.cfg will appear in your home directory (if not, then create one). Make sure the file at least contains the following lines:
 <file> <file>
-127.0.0.1 localhost  
-</file> 
  
-Change that line into+[global] 
 +database_user       : <your username> 
 +database_password   : <your password>
  
-<file> 
-127.0.0.1 localhost <your_host_name> 
 </file> </file>
  
-If you do not know your hostname, just type hostname in a shell and you will get it as an output.+The following script can be used to test your installation:
  
-Now, you can use the following script as a test. It may still give you some warnings, but if it prints out a list of pointings, then you are ready to go. You may need to kill the script, because it will print out all the observations in a certain patch of the sky archived in the LTA. +<code python> 
- +# Python3 code
-<file> +
-# python code+
 from pprint import pprint from pprint import pprint
 +from awlofar.main.aweimports import Observation, Pointing, SubArrayPointing
 from common.database.Context import context from common.database.Context import context
-from awlofar.main.aweimports import Observation, Pointing, SubArrayPointing 
 result = {} result = {}
 for project in sorted(context.get_projects()) : for project in sorted(context.get_projects()) :
-    print "Project %(project)s" % vars()+    print("Project %(project)s" % vars())
     ok = context.set_project(project)     ok = context.set_project(project)
     # do your query     # do your query
Line 114: Line 115:
             (Pointing.declination    > 20)  & \             (Pointing.declination    > 20)  & \
             (Pointing.declination    < 30)             (Pointing.declination    < 30)
-    print "Total Pointings %d" % len(query)+    print("Total Pointings %d" % len(query))
     for pointing in query :     for pointing in query :
-        print "Pointing found RA %f DEC %f" % (pointing.rightAscension, pointing.declination)+        print("Pointing found RA %f DEC %f" % (pointing.rightAscension, pointing.declination))
         query_subarr = SubArrayPointing.pointing == pointing         query_subarr = SubArrayPointing.pointing == pointing
         for subarr in query_subarr:         for subarr in query_subarr:
Line 123: Line 124:
                 obs_ids.add(obs.observationId)                 obs_ids.add(obs.observationId)
     result[project] = sorted(list(obs_ids))     result[project] = sorted(list(obs_ids))
-    print result[project]+    print(result[project])
  
 pprint(result) pprint(result)
-</file> 
  
-If you get errors and do not manage to view the list of pointings, there may be the need to open some port on the firewall at your institution. Specifically, port 1521 should be open. In case of trouble, get in contact with Science Support.+</code>
  
-Once you have tested that your connection to the catalogue is working, you are ready to browse the archive and stage the data you need. +It should print out a list of pointings (note that in this example the library was installed in $HOME/tmp):
-Here we will list a few examples of python scripts that can be used to access the LTA. All of them will need to import some modules:+
  
-<file>+<code> 
 +$ env PYTHONPATH=$HOME/tmp/lib/python3.5/site-packages python3 lta_test.py 
 +Project ALL 
 +Total Pointings 202 
 +Pointing found RA 95.003499 DEC 24.838742 
 +Pointing found RA 95.174754 DEC 28.660087 
 +Pointing found RA 95.220000 DEC 29.140000 
 +Pointing found RA 95.546250 DEC 23.331750 
 +Pointing found RA 95.561458 DEC 24.584056 
 +..etc.. 
 + 
 +</code> 
 + 
 +You may need to kill the script, because it will print out all the observations in a certain patch of the sky archived in the LTA. 
 + 
 +In case of errors, there may be the need to open some port on the firewall at your institution. Specifically, port 1521 should be open. Also make sure that the LTA client library can be found in your PYTHONPATH (see [[:lta:client_installation|LTA Client installation]] for more details). In case of trouble, get in contact with Science Operations and Support. 
 + 
 +==== Examples ==== 
 + 
 +Once you have tested that your connection to the catalogue is working, you are ready to browse the archive and stage the data you need. Here we will list a few examples of python scripts that can be used to access the LTA. All of them will need to import some modules: 
 + 
 +<code python>
 from datetime import datetime from datetime import datetime
 from awlofar.database.Context import context from awlofar.database.Context import context
Line 140: Line 160:
     Observation     Observation
 from awlofar.toolbox.LtaStager import LtaStager, LtaStagerError from awlofar.toolbox.LtaStager import LtaStager, LtaStagerError
-</file>+ 
 + 
 +</code>
  
 The lines above must be added to each of the scripts below for these to work. The lines above must be added to each of the scripts below for these to work.
  
-This simple script will allow you to find and stage all data within a single project LCX_YYY.+=== Ex: get staging URI's ===
  
-<file> +This script will allow you to find all data within a single project, for example LC2_035. Please change the project name to the code of a project of yours. If you also want to stage the data you found, just set the do_stage variable to True. Be careful with how many files you stage and what size they have: the same limits as for the web interface apply here. 
-do_stage = True + 
-project = 'LCX_YYY'+<code python> 
 +# Should the found files be staged ? 
 +do_stage = False 
 +# The project to query, LC2_035 has public data 
 +project = 'LC2_035' 
 +# The class of data to query
 cls = CorrelatedDataProduct cls = CorrelatedDataProduct
-if not context.set_project(project) : +# Query for private data of the project, you must be member of the project 
-    raise Exception("You are not member of project %s" % project)+private_data = False
  
-query_observations = Observation.select_all().project_only()+# To see private data of this project, you must be member of this project 
 +if private_data : 
 +    context.set_project(project) 
 +    if project != context.get_current_project().name: 
 +        raise Exception("You are not member of project %s" % project) 
 + 
 +query_observations = Observation.select_all().project_only(project)
 uris = set() # All URIS to stage uris = set() # All URIS to stage
 for observation in query_observations : for observation in query_observations :
Line 175: Line 208:
     stager = LtaStager()     stager = LtaStager()
     stager.stage_uris(uris)     stager.stage_uris(uris)
-</file> 
  
-The following will find (but not stage) subbands 301 and 302 for all targets within two different projects. Pay attention to the difference between the keys subband and stationSubband; the former is a sequential number assigned to each subband in an observation, while the latter is linked to the frequency at which the observation was performed. Example: an observation was set up covering the range 30-77.3 MHz with two simultaneous beams using 244 subbands each. In this case, subband will range from 0 to 487, while stationSubband from 153 to 396. 
  
-<file>+</code> 
 + 
 +=== Ex: filter on subbands === 
 + 
 +The following script will find subbands 301 and 302 for all targets within two different projects. 
 + 
 +Pay attention to the difference between the keys subband and stationSubband; the former is a sequential number assigned to each subband in an observation, while the latter is linked to the frequency at which the observation was performed. Example: an observation was set up covering the range 30-77.3 MHz with two simultaneous beams using 244 subbands each. In this case, subband will range from 0 to 487, while stationSubband from 153 to 396. The stationSubband information is stored in the observation, but not in the pipeline products (which instead contain the frequency). If you want to search on stationSubband, you must perform your search on observations first, then fetch the pipelines linked to those observations. If you use frequency, you can search directly on pipelines. 
 + 
 +As a general advise, before performing a search, you need to **understand thoroughly the meaning of the keywords that you are using and where their values are stored**, otherwise you may not find the data you are looking for. 
 + 
 +<code python>
 do_stage = False do_stage = False
-project1 = 'LCX_YYY' +project1 = 'LC2_016' 
-project2 = 'LCZ_VVV'+project2 = 'LC2_012'
 subband1 = 301 subband1 = 301
 subband2 = 302 subband2 = 302
 cls = CorrelatedDataProduct cls = CorrelatedDataProduct
 +# Query for private data of the project, you must be member of the project
 +private_data = False
  
 # All URIS to stage # All URIS to stage
Line 195: Line 238:
 for project in (project1, project2) : for project in (project1, project2) :
     print("Using project %s" % project)     print("Using project %s" % project)
-    if not context.set_project(project) : +    if private_data : 
-        raise Exception("You are not member of project %s" % project) +        context.set_project(project) 
-    query_observations = Observation.select_all().project_only()+        if project != context.get_current_project().name: 
 +            raise Exception("You are not member of project %s" % project) 
 +    query_observations = Observation.select_all().project_only(project)
     for observation in query_observations :     for observation in query_observations :
         print("Querying ObservationID %s" % observation.observationId)         print("Querying ObservationID %s" % observation.observationId)
Line 219: Line 264:
  
 stager = LtaStager() stager = LtaStager()
-for project in (project1, project2) : +if do_stage : 
-    if do_stage :+    for project in (project1, project2) :
         stager.stage_uris(uris[project])         stager.stage_uris(uris[project])
-</file> 
  
-Here, we find and stage data between freq1 and freq2 taken within one project between day1 and day2 
  
-<file> +</code> 
-do_stage = True + 
-project = 'LCX_YYY'+=== Ex: filter on frequency and observation date === 
 + 
 +Here, we find data between freq1 and freq2 taken within one project between day1 and day2 
 + 
 +<code python> 
 +do_stage = False 
 +project = 'LC2_033'
 freq1 = 172.0 freq1 = 172.0
 freq2 = 178.0 freq2 = 178.0
-day1 = datetime(2016,1,20) # this could include time; ie hours, minutes, secondes +day1 = datetime(2014,8,26) # this could include time; ie hours, minutes, secondes 
-day2 = datetime(2016,1,31) # idem+day2 = datetime(2014,8,29) # idem
 # DataProduct class to query; CorrelatedDataProduct, SkyImageDataProduct, etc ... # DataProduct class to query; CorrelatedDataProduct, SkyImageDataProduct, etc ...
 cls = CorrelatedDataProduct cls = CorrelatedDataProduct
 +# Query for private data of the project, you must be member of the project
 +private_data = False
  
-if not context.set_project(project) : +# To see private data of this project, you must be member of this project 
-    raise Exception("You are not member of project %s" % project)+if private_data : 
 +    context.set_project(project) 
 +    if project != context.get_current_project().name: 
 +        raise Exception("You are not member of project %s" % project) 
 + 
 +query_observations = ( 
 +    (Observation.startTime >= day1) & 
 +    (Observation.endTime   <  day2) ).project_only(project)
  
-query_observations = ((Observation.startTime >= day1) & 
-                      (Observation.endTime   <  day2)).project_only() 
 uris = set() uris = set()
 for observation in query_observations : for observation in query_observations :
Line 263: Line 319:
     stager = LtaStager()     stager = LtaStager()
     stager.stage_uris(uris)     stager.stage_uris(uris)
-</file>+ 
 + 
 +</code> 
 + 
 +=== Ex: query public data === 
 + 
 +Querying public data in projects you are not member of. First set project ALL, then construct a query and optionally limit the query to a certain project : 
 + 
 +<code python> 
 +context.set_project('ALL') 
 +query = CorrelatedDataProduct.select_all() 
 +query &= query.project_only('LC0_017') 
 +print(len(query)) 
 +# 1800 
 + 
 + 
 +</code> 
 + 
 +=== Ex: get release dates === 
 + 
 +<code python> 
 +from awlofar.main.aweimports import Observation, PipelineRun, DataProduct 
 +from common.database.Context import context 
 + 
 +project = 'LC2_035' 
 + 
 +# Query for private data of the project, you must be member of the project 
 +private_data = True 
 + 
 +# To see private data of this project, you must be member of this project 
 +if private_data : 
 +    context.set_project(project) 
 +    if project != context.get_current_project().name: 
 +        raise Exception("You are not member of project %s" % project) 
 + 
 +# Observations 
 +query_observations = Observation.select_all().project_only(project) 
 +for observation in query_observations : 
 +    print("Querying ObservationID %s, %s" % (observation.observationId, observation.releaseDate)) 
 + 
 +# Pipelines 
 +query_pipelines = PipelineRun.select_all().project_only(project) 
 +for pipeline in query_pipelines : 
 +    print("Pipeline: %s, %s, %s" % (type(pipeline).__name__, pipeline.pipelineName, pipeline.releaseDate)) 
 + 
 +# Data products 
 +query_products = DataProduct.select_all().project_only(project) 
 +query_products &= DataProduct.isValid == 1 
 +for product in query_products : 
 +    print("Product: %s, %s, %s, %s" % (product.dataProductIdentifier, product.dataProductIdentifierSource, product.dataProductType, product.releaseDate)) 
 + 
 + 
 +</code> 
 + 
 +===== Python Module for Staging ===== 
 + 
 +The python interaction with the LTA catalog can be complemented with the use of a specific module developed to give users more control over their staging requests. 
 + 
 +Current released version 2.0 ([[http://git.astron.nl/astron-sdc/lofar_stager_api/-/releases/2.0|tagged on master branch]]) is to be used with the new LTA stager (stageit), it is not backwards compatible with the old LTA stager. Older versions of this script (i.e., 1.7 and older) have become obsolete. Please see **"Version 2.0 usage notes"**  listed below for documentation (or check the README file in the repository linked above). 
 + 
 +  * User documentation for __//stageit //__  can be found at: [[https://support.astron.nl/confluence/display/SDCP/User+documentation|https://support.astron.nl/confluence/display/SDCP/User+documentation]] 
 +  * Version 2.0 release can be found at: [[https://git.astron.nl/astron-sdc/lofar_stager_api/-/releases/2.0|https://git.astron.nl/astron-sdc/lofar_stager_api/-/releases/2.0]] 
 + 
 +==== Version 2.0 usage notes ==== 
 + 
 +The module is made available [[http://git.astron.nl/astron-sdc/lofar_stager_api/-/releases/2.0|here]] (tagged on master branch). Simply checkout the tagged commit and use the script. Please see the last note in the list below with regards to required dependencies. 
 + 
 +__Notes:__ 
 + 
 +  * You need an access token to the stageit api. Please refer to the user guide linked above to sign up and login to stageit. After logging in, a token can be obtained in one of two ways: 
 +      * Visit [[https://sdc.astron.nl/stageit/api/staging/get-token|https://sdc.astron.nl/stageit/api/staging/get-token]] 
 +      * From anywhere in the application, click on your account name in the top right to access your profile. From your profile page, click the "Request token" button to receive a token. 
 +  * The token is valid indefinitely. Requesting a token multiple times will yield the same token. 
 +  * Make sure the token is available in your ~/.stagingrc file: 
 +      * api_token=YOUR_TOKEN_HERE 
 +      * remove the old username and password from the ''.stagingrc''  file 
 +  * The script is Python2 compatible, there is a Dockerfile available for Python2 testing in ''./tests/docker'' 
 +  * The requests library is a required dependency. If you care about Python2 compatability, you can use at most version 2.22.0 of requests. Otherwise, you can install any version (note: you can also ''pip install -r 'requirements.txt''', which will install version 2.22.0) 
 + 
 +Also note that some functions are not supported in the new LTA stager. The states that a request can be in have been simplified. As such, there is no need for these functions anymore. Upon use, they will display an error stating that the function is deprecated. Please look at the ''stager_access.py''  file for more information. 
 + 
 +==== Functionality ==== 
 + 
 +For a description of what the user can do, we list here the functions that are available. 
 + 
 +**stage(surls)** \\ It takes in a list of surls, queues a staging request for those urls, and outputs the ID of the request. 
 + 
 +**get_status(stageid)** \\ It tells the user if a request is queued, in progress or finished (success). Possible statuses: "new", "scheduled", "in progress", "aborted", "failed", "partial success", "success", "on hold" 
 + 
 +**abort(stageid)** \\ It allows users to end a staging request. 
 + 
 +**get_surls_online(stageid)** \\ It gives a list of the surls that have been staged for the relative request. The list is updated whenever a new surl comes on line. 
 + 
 +**get_srm_token(stageid)** \\ The srm token is useful to interact directly with the SRM site through GRID/SRM tools. 
 + 
 +**reschedule(stageid)** \\ If a request failed, it can be rescheduled. 
 + 
 +**get_progress()** \\ No input needed. It returns the statuses of all the requests owned by the user. 
 + 
 +Below is an example of how to use this: 
 +<code> 
 +> python 
 +Python 2.7.10 (default, Oct 23 2015, 19:19:21) 
 +[GCC 4.2.1 Compatible Apple LLVM 7.0.0 (clang-700.0.59.5)] on darwin 
 +Type "help", "copyright", "credits" or "license" for more information.>>> import stager_access as sa 
 + 
 +2016-11-24 16:39:55.865000 stager_access: Parsing user credentials from /Users/renting/.stagingrc 
 +2016-11-24 16:39:55.865111 stager_access: Creating proxy>>> sa.prettyprint(sa.get_progress()) 
 + 
 ++ 12227 
 +  - File count   ->     100 
 +  - Files done      ->     40 
 +  - Flagged abort      ->     false 
 +  - Location      ->     fz-juelich 
 +  - Percent done      ->     40 
 +  - Status      ->     on hold 
 +  - User id      ->     1919 
 + 
 +</code> 
  
  • Last modified: 2016-03-31 11:37
  • by Luciano Cerrigone