Migrating from Indivo 1.0 to Indivo 2.0
Scott Gilroy
I’ve pushed changes to the CollaboRhythm repository to move from using our fork of Indivo 1.0 https://github.com/newmediamedicine/indivo_server_1_0 to Indivo 2.0 https://github.com/newmediamedicine/indivo_server
This means that if you pull upstream changes into your fork of CollaboRhythm you will need to be aware of some changes to schemas and other important aspects of working with Indivo 2.0.
For HW2013:
Most participants in the HW2013 event are advised to not pull upstream changes until further notice. I will help you with the migration, or perform the merge for you and let you know when your repository has been updated. If you want to get the latest version of CollaboRhythm compatible with Indivo 1.0, you will need to get a branch named “indivo_1_ha”.
To summarize, the following has changed:
1. VitalSign, CodedValue, and ValueAndUnit schemas renamed to CollaboRhythmVitalSign, CollaboRhythmCodedValue, and CollaboRhythmValueAndUnit. This was done to maintain compatibility with existing data while avoiding conflicts with the new SMART schemas with the same names.
2. Problem schema has changed significantly. The diagnosedBy field was removed, the name field changed from CollaboRhythmCodedValue to (SMART) CodedValue, and the other fields were renamed.
3. RecurrenceRule in HealthActionSchedule and MedicationScheduleItem was changed slightly but should not affect most people. The frequency and interval fields were changed from CodedValue to string and integer respectively.
4. Address of the default server has changed from indivo-demo to
The migration to 2.0 is not quite complete. There is one feature that is currently not working in CollaboRhythm: text messaging. This relies on custom API extensions that I have not finished porting over to Indivo 2.0. Don’t expect messaging to be working if you start using the Indivo 2.0 version of CollaboRhythm.
Scott Gilroy
Correction to the list of changes below: Address of the default server has changed from indivo-demo to http://indivo-dev.media.MIT.edu:9291 Scott Gilroy <sgi...@MIT.EDU> wrote:
You received this message because you are subscribed to the Google Groups "CollaboRhythm Developers" group.
To post to this group, send email to collaborhyth...@googlegroups.com.
To unsubscribe from this group, send email to collaborhythm-deve...@googlegroups.com.
Visit this group at http://groups.google.com/group/collaborhythm-developers?hl=en.
For more options, visit https://groups.google.com/groups/opt_out.
Scott Gilroy
The issues with text messaging have been resolved (thanks to Travers Franckle). No changes were required on the CollaboRhythm side. The changes to the server have been deployed to indivo-dev.
Also, I resolved an issue with clinician mode that was resulting in the “connecting…” message to display indefinitely. These changes have been pushed.
For the main CollaboRhythm repository, we are now completely migrated to Indivo 2.0 with all CollaboRhythm data types working, including the new SMART Problem model and our messaging extensions.
The next step for HW2013 participants will be to migrate the teams over to the updated code and have them switch servers. I will probably tackle this on a team-by-team basis, starting with the Hypertension team.
Scott Gilroy
I have pushed a changes to IndivoAccountCreator. The master (default) branch is now updated to Indivo 2.0 (I merged in the indivo2 branch and . The indivo_1_ha has been created for Indivo 1.0 support. Make sure you use the appropriate branch for the server you are using.
To be clear: the master (default) branch in the IndivoAccountCreator and CollaboRhythm repositories are updated to Indivo 2.0. Support for older servers is available in both repository under the indivo_1_ha branches.
Scott Gilroy
I have migrated the Hypertension and CHF repositories from Indivo 1.0 to Indivo 2.0. If you want help migrating your repository, please let me know. You should ensure that as much of your code as possible from all team members is committed and pushed (especially any data with VitalSign instances) before requesting a migration.
For reference (or in case you want to attempt the migration on you own) the steps are:
1. Pull upstream changes. See: Pulling Upstream Changes from the Main CollaboRhythm Repository
2. Resolve merge conflicts, if any.
3. Commit changes from the merge, if any (don’t push yet).
4. Migrate any data that you have. Search and replace all data files under CollaboRhythm.ExampleData as follows
|
VitalSign |
CollaboRhythmVitalSign |
case sensitive, whole word only |
|
dateOnset |
startDate |
case sensitive, whole word only, file mask Problem*.xml |
|
dateResolution |
endDate |
case sensitive, whole word only, file mask Problem*.xml |
|
comments |
notes |
case sensitive, whole word only, file mask Problem*.xml |
|
<name\ type="umls-snomed"\ value="(\d+)"\ abbrev=".*">([^<]+)</name> |
<name>\n <title>$2</title>\n <system>http://purl.bioontology.org/ontology/SNOMEDCT/</system>\n <identifier>$1</identifier>\n </name> |
case sensitive, regular expression, file mask Problem*.xml |
|
\n*\s*<diagnosedBy>([^<]+)</diagnosedBy> |
|
case sensitive, regular expression, file mask Problem*.xml (Replace by is blank to delete this element) |
5. Manually fix uses of the updated Problem class in your CollaboRhythm code (if any).
6. Manually fix uses of the updated Demographics class (not Contact is eliminated, and some fields have moved or been eliminated from the schemas).
7. Create a new account data file for use with IndivoAccountCreator. You can duplicate your existing file and change the server to the following: <indivoServerBaseURL>http://indivo-dev.media.mit.edu:9291</indivoServerBaseURL>
8. Ensure that email addresses in the file have a valid syntax (Indivo 2.0 added validation). For example, replace not_used@localhost with not_...@fake.org
9. Get the latest version of IndivoAccountCreator
10. Open your new data file in IndivoAccountCreator
11. Click File -> Database Reset to clear the record ids and other fields.
12. Create accounts and records as needed.
13. Iterate on the above as necessary until you can load all of your data and run CollaboRhythm
14. When you have successfully loaded your data and made any additional changes, commit and push your changes to your repository.
From: collaborhyth...@googlegroups.com [mailto:collaborhyth...@googlegroups.com] On Behalf Of Scott Gilroy
Sent: Monday, January 28, 2013 2:27 PM
To: collaborhyth...@googlegroups.com