)]}'
{"elements/svc-map/extra-data.d/10-merge-svc-map-files":[{"author":{"_account_id":10035,"name":"greghaynes","email":"greg@greghaynes.net","username":"greghaynes"},"change_message_id":"e9a659b690b8b677b0719a72232b86a2a6fa5c9d","unresolved":false,"context_lines":[{"line_number":44,"context_line":"    family \u003d os_family(distro)"},{"line_number":45,"context_line":"    for servicename, mapping in six.iteritems(source):"},{"line_number":46,"context_line":"        if servicename in result:"},{"line_number":47,"context_line":"            if allow_dupes:"},{"line_number":48,"context_line":"                print(\"WARNING: service %s is redefined in a duplicate \""},{"line_number":49,"context_line":"                      \"element %s, assuming you intend to override the \""},{"line_number":50,"context_line":"                      \"existing element.\" % (servicename, element))"}],"source_content_type":"application/octet-stream","patch_set":2,"id":"3aaa91ec_50354144","line":47,"updated":"2016-06-28 15:31:25.000000000","message":"I think we don\u0027t want to lose this sanity check as a whole, but only to allow a duplicate definition if it is identical. In other words - can we get the result of the conditional block below and compare that with what is in result[servicename] and still raise if they are not equal?","commit_id":"de8d2e4e546c757107652898c58fd8a3a91c6ed1"},{"author":{"_account_id":12459,"name":"Stephanie Miller","email":"stephane@alum.mit.edu","username":"cinerama"},"change_message_id":"05089f60c69e6e30f68c46e3589929e96a950107","unresolved":false,"context_lines":[{"line_number":44,"context_line":"    family \u003d os_family(distro)"},{"line_number":45,"context_line":"    for servicename, mapping in six.iteritems(source):"},{"line_number":46,"context_line":"        if servicename in result:"},{"line_number":47,"context_line":"            if allow_dupes:"},{"line_number":48,"context_line":"                print(\"WARNING: service %s is redefined in a duplicate \""},{"line_number":49,"context_line":"                      \"element %s, assuming you intend to override the \""},{"line_number":50,"context_line":"                      \"existing element.\" % (servicename, element))"}],"source_content_type":"application/octet-stream","patch_set":2,"id":"3aaa91ec_d1b929ee","line":47,"in_reply_to":"3aaa91ec_50354144","updated":"2016-06-28 19:15:17.000000000","message":"I\u0027m just thinking about the case where we\u0027re overriding a deprecated in-tree element and the service name changes in the newer version - I think it\u0027s reasonable to allow that.","commit_id":"de8d2e4e546c757107652898c58fd8a3a91c6ed1"},{"author":{"_account_id":12459,"name":"Stephanie Miller","email":"stephane@alum.mit.edu","username":"cinerama"},"change_message_id":"f74b7602a2945b6083d831e64be5cabedead9750","unresolved":false,"context_lines":[{"line_number":44,"context_line":"    family \u003d os_family(distro)"},{"line_number":45,"context_line":"    for servicename, mapping in six.iteritems(source):"},{"line_number":46,"context_line":"        if servicename in result:"},{"line_number":47,"context_line":"            if allow_dupes:"},{"line_number":48,"context_line":"                print(\"WARNING: service %s is redefined in a duplicate \""},{"line_number":49,"context_line":"                      \"element %s, assuming you intend to override the \""},{"line_number":50,"context_line":"                      \"existing element.\" % (servicename, element))"}],"source_content_type":"application/octet-stream","patch_set":2,"id":"1aa78d24_f9c0792c","line":47,"in_reply_to":"3aaa91ec_6c7d896c","updated":"2016-07-05 16:34:35.000000000","message":"I\u0027m inclined to agree that a fix at a higher level is better. I think we should focus our efforts at the moment on trying to get some of these broader changes to land.","commit_id":"de8d2e4e546c757107652898c58fd8a3a91c6ed1"},{"author":{"_account_id":10035,"name":"greghaynes","email":"greg@greghaynes.net","username":"greghaynes"},"change_message_id":"6cb921d3c6e40f53ac451b224925b3fc4c9594c8","unresolved":false,"context_lines":[{"line_number":44,"context_line":"    family \u003d os_family(distro)"},{"line_number":45,"context_line":"    for servicename, mapping in six.iteritems(source):"},{"line_number":46,"context_line":"        if servicename in result:"},{"line_number":47,"context_line":"            if allow_dupes:"},{"line_number":48,"context_line":"                print(\"WARNING: service %s is redefined in a duplicate \""},{"line_number":49,"context_line":"                      \"element %s, assuming you intend to override the \""},{"line_number":50,"context_line":"                      \"existing element.\" % (servicename, element))"}],"source_content_type":"application/octet-stream","patch_set":2,"id":"3aaa91ec_6c7d896c","line":47,"in_reply_to":"3aaa91ec_a71a5f7a","updated":"2016-06-30 17:52:41.000000000","message":"It seems better for us to fail closed rather than open which is my main concern here -this is a nice sanity check for us in other places which would be bad to lose. My hope would be that the deprecation period is temporary and that although we lose some flexibility during that period (we also cant rename files in the element without issues) it is a temporary cost we are ok with - once we pass the deprecation period and the deprecated element is removed were back to normal.\n\n\nIf we can find a way for either the depending or dependee element to specify that it wants to override the svc map that would be awesome. I don\u0027t have a great idea for how to implement that though...\n\n\nAlternatively, ianw has an interesting patch up which fixes this issue at a higher level: https://review.openstack.org/#/c/334785/. If we were to use that this issue goes away entirely but I suspect its going to be a bit before it merges.","commit_id":"de8d2e4e546c757107652898c58fd8a3a91c6ed1"},{"author":{"_account_id":11655,"name":"Julia Kreger","email":"juliaashleykreger@gmail.com","username":"jkreger","status":"Flying to the moon with a Jetpack!"},"change_message_id":"38b75141a7cf5274cd665ec1687f93b1cb99211a","unresolved":false,"context_lines":[{"line_number":44,"context_line":"    family \u003d os_family(distro)"},{"line_number":45,"context_line":"    for servicename, mapping in six.iteritems(source):"},{"line_number":46,"context_line":"        if servicename in result:"},{"line_number":47,"context_line":"            if allow_dupes:"},{"line_number":48,"context_line":"                print(\"WARNING: service %s is redefined in a duplicate \""},{"line_number":49,"context_line":"                      \"element %s, assuming you intend to override the \""},{"line_number":50,"context_line":"                      \"existing element.\" % (servicename, element))"}],"source_content_type":"application/octet-stream","patch_set":2,"id":"3aaa91ec_a71a5f7a","line":47,"in_reply_to":"3aaa91ec_d1b929ee","updated":"2016-06-28 20:07:34.000000000","message":"Off the wall idea: What about a way of flagging an element as deprecated or being able to have an optional processing priority so the element maintainers can explicitly state intention for duplicate handling if necessary.","commit_id":"de8d2e4e546c757107652898c58fd8a3a91c6ed1"},{"author":{"_account_id":21741,"name":"Andreas Florath","email":"andreas@florath.net","username":"florath"},"change_message_id":"fe43aaeddc4e83b5feff08e13823c308c8e771cb","unresolved":false,"context_lines":[{"line_number":45,"context_line":"    for servicename, mapping in six.iteritems(source):"},{"line_number":46,"context_line":"        if servicename in result:"},{"line_number":47,"context_line":"            if allow_dupes:"},{"line_number":48,"context_line":"                print(\"WARNING: service %s is redefined in a duplicate \""},{"line_number":49,"context_line":"                      \"element %s, assuming you intend to override the \""},{"line_number":50,"context_line":"                      \"existing element.\" % (servicename, element))"},{"line_number":51,"context_line":"            else:"}],"source_content_type":"application/octet-stream","patch_set":2,"id":"3aaa91ec_3ea35c51","line":48,"updated":"2016-06-27 19:16:39.000000000","message":"Maybe this can be done silently as long as the definition in both files are the same?","commit_id":"de8d2e4e546c757107652898c58fd8a3a91c6ed1"},{"author":{"_account_id":7118,"name":"Ian Wienand","email":"iwienand@redhat.com","username":"iwienand"},"change_message_id":"cd754025af568958090110cab1c1835cec00aee6","unresolved":false,"context_lines":[{"line_number":94,"context_line":""},{"line_number":95,"context_line":""},{"line_number":96,"context_line":"def main():"},{"line_number":97,"context_line":"    elements \u003d os.environ.get(\"IMAGE_ELEMENT\").split(\u0027 \u0027)"},{"line_number":98,"context_line":"    base_element_dir, element_paths \u003d get_element_paths()"},{"line_number":99,"context_line":"    service_names \u003d dict()"},{"line_number":100,"context_line":"    seen_elements \u003d set()"}],"source_content_type":"application/octet-stream","patch_set":2,"id":"3aaa91ec_67187776","line":97,"updated":"2016-06-28 19:50:23.000000000","message":"So I\u0027m actually fairly happy with how I7092e1845942f249175933d67ab121188f3511fd worked out ...\n\nWith that it seems here you could just decode IMAGE_ELEMENT_YAML and walk the dictionary?  it has element name and full path, and has already had overrides processed","commit_id":"de8d2e4e546c757107652898c58fd8a3a91c6ed1"},{"author":{"_account_id":12459,"name":"Stephanie Miller","email":"stephane@alum.mit.edu","username":"cinerama"},"change_message_id":"f91cfa22ac34f377b6c114a744893d3e6ef2f0ee","unresolved":false,"context_lines":[{"line_number":94,"context_line":""},{"line_number":95,"context_line":""},{"line_number":96,"context_line":"def main():"},{"line_number":97,"context_line":"    elements \u003d os.environ.get(\"IMAGE_ELEMENT\").split(\u0027 \u0027)"},{"line_number":98,"context_line":"    base_element_dir, element_paths \u003d get_element_paths()"},{"line_number":99,"context_line":"    service_names \u003d dict()"},{"line_number":100,"context_line":"    seen_elements \u003d set()"}],"source_content_type":"application/octet-stream","patch_set":2,"id":"3aaa91ec_147fb115","line":97,"in_reply_to":"3aaa91ec_67187776","updated":"2016-06-30 15:52:02.000000000","message":"Testing with your patch series, it seems like this pretty much takes care of our needs and is a more general solution. I\u0027d specifically made this change limited in scope to avoid too many objections and land it in a timely fashion so I could proceed with a dependent change. \n\nI don\u0027t really see the value in continuing with this unless we also want to allow service definition duplication between non-overridden elements (which seems like a poor choice).","commit_id":"de8d2e4e546c757107652898c58fd8a3a91c6ed1"}],"releasenotes/notes/svc-map-extra-paths-ab2c585dc69e3405.yaml":[{"author":{"_account_id":11655,"name":"Julia Kreger","email":"juliaashleykreger@gmail.com","username":"jkreger","status":"Flying to the moon with a Jetpack!"},"change_message_id":"38b75141a7cf5274cd665ec1687f93b1cb99211a","unresolved":false,"context_lines":[{"line_number":1,"context_line":"---"},{"line_number":2,"context_line":"upgrade:"},{"line_number":3,"context_line":"  - svc-map now supports duplicate service definitions,"},{"line_number":4,"context_line":"    under limited conditions. If an element with the same"},{"line_number":5,"context_line":"    name exists outside of BASE_ELEMENT_DIR, its svc-map"}],"source_content_type":"text/x-yaml","patch_set":2,"id":"3aaa91ec_6469e025","line":2,"updated":"2016-06-28 20:07:34.000000000","message":"nit: I kind of wonder if this is more a bug fix or feature than an upgrade note item.","commit_id":"de8d2e4e546c757107652898c58fd8a3a91c6ed1"},{"author":{"_account_id":11655,"name":"Julia Kreger","email":"juliaashleykreger@gmail.com","username":"jkreger","status":"Flying to the moon with a Jetpack!"},"change_message_id":"38b75141a7cf5274cd665ec1687f93b1cb99211a","unresolved":false,"context_lines":[{"line_number":4,"context_line":"    under limited conditions. If an element with the same"},{"line_number":5,"context_line":"    name exists outside of BASE_ELEMENT_DIR, its svc-map"},{"line_number":6,"context_line":"    can redefine services."},{"line_number":7,"context_line":""},{"line_number":8,"context_line":""}],"source_content_type":"text/x-yaml","patch_set":2,"id":"3aaa91ec_440b047e","line":8,"range":{"start_line":7,"start_character":0,"end_line":8,"end_character":0},"updated":"2016-06-28 20:07:34.000000000","message":"nit: Excess lines","commit_id":"de8d2e4e546c757107652898c58fd8a3a91c6ed1"}]}
